Report to the Governor and General Assembly of Virginia, Spending on Inmate Health Care, 2018

The Habeas Citebook: Prosecutorial Misconduct - Header

Report to the Governor and General Assembly of Virginia, Spending on Inmate Health Care, 2018

• Nov. 13, 2018 • Locations: United States of America -> Virginia • Topics: Cost of Prison Systems

Download original document:
Brief thumbnail

Document text

This text is machine-read, and may contain errors. Check the original document to verify accuracy.

Commonwealth of Virginia
November 13, 2018

Report to the Governor and the General Assembly of Virginia

Spending on Inmate Health Care
2018

JLARC

JOINT LEGISLATIVE AUDIT
AND REVIEW COMMISSION

Joint Legislative Audit and Review Commission
Senator Thomas K. Norment, Jr., Chair
Delegate R. Steven Landes, Vice-Chair
Delegate Terry Austin
Delegate Betsy Carr
Delegate M. Kirkland Cox
Senator Emmett W. Hanger, Jr.
Delegate Charniele L. Herring
Senator Janet D. Howell
Delegate S. Chris Jones
Senator Ryan T. McDougle
Delegate Robert D. Orrock, Sr.
Delegate Kenneth R. Plum
Senator Frank M. Ruff, Jr.
Delegate Christopher P. Stolle
Martha S. Mavredes, Auditor of Public Accounts

JLARC staff
Hal E. Greer, Director
Justin Brown, Senior Associate Director
Jeff Lunardi, Project Leader
Kate Agnelli
Danielle Childress
Nick Galvin
Information graphics by Nathan Skreslet

JLARC Report 511

Contents
Summary

Recommendations and Options

Chapters
1. Inmate Health Care in Virginia

2. Spending on Inmate Health Care

3. Strategies to Reduce Spending

4. Staffing and Risk Management

Appendixes
A: Study resolution

B: Research activities and methods

C: Inmate health care spending analysis

D: Map of VADOC facilities

E: Offsite and prescription drug expenditure analysis

F: Compassionate release policies in other states

G: Agency responses

Summary: Spending on Inmate Health Care
VADOC spends about the same on inmate health care as other states
and Medicaid, though spending is growing more quickly
Virginia spends about the same on inmate health care as other states and on its own
Medicaid program; however, spending growth has been higher. The Virginia Department of Corrections (VADOC) spending on health care per inmate has grown 7.6
percent annually over the past decade, nearly twice the
rate of Virginia’s Medicaid program and of health care
WHY WE DID THIS STUDY
spending nationwide.
JLARC directed staff to study the rising cost of inmate
VADOC is increasingly relying on contracts to help
manage spending growth. Though using contracts has
made spending more predictable in the short term, it
may not be saving the state money in the long term.
JLARC analysis found no evidence that contract facilities spend less than non-contract facilities on prescription drugs and offsite services.
VADOC pays more than other public payers
for certain services and medications

health care. From FY07 to FY17, the Virginia Department
of Corrections (VADOC) medical budget increased by
$56.8 million (40 percent).

ABOUT INMATE HEALTH CARE
Correctional systems are required to provide “reasonably adequate” health care to inmates. Inmates are
typically less healthy than the general population.
VADOC spent $201 million on medical and clinical
services in FY17, or about $6,500 per inmate. This accounted for 17 percent of VADOC’s $1.2 billion operating budget, nearly all of which comes from the general
fund. VADOC uses a mix of state employees and
contracts to care for approximately 30,000 stateresponsible inmates at its 41 facilities.

Despite being a public purchaser of health care,
VADOC pays rates higher than Medicare and Medicaid. Using a lower rate structure based on Medicare
could save $9 million annually. There would be, though, the need to balance the pursuit
of these cost savings with ensuring access to care. This will be especially challenging
as Virginia expands Medicaid, which will pose a substantial test of access in some areas
of the state.
VADOC also pays more for certain prescription drugs. Furthering its partnership with
VCU Health, the UVA health system, or others could reduce spending on biologic
medications, inhalers, and insulin by $1.5 million to $6.6 million annually.
Virginia’s restrictive compassionate release policies contribute to
higher health care spending
Virginia has one of the nation’s most restrictive compassionate release policies for
inmates with terminal diagnoses. Only one state (Kansas) has a more stringent time
requirement than Virginia’s. VADOC spends much more on inmates near the end of
their lives than on inmates generally. VADOC spent more than $61,000, on average,
to provide end-of-life care for 65 inmates who died in FY17. These 65 inmates comprised only 0.2 percent of all inmates, yet their end-of-life care accounted for 4.7 percent of VADOC health care spending on services and medications.

Summary: Spending on Inmate Health Care

Virginia also has one of the nation’s most restrictive release policies for inmates with
complex long-term medical conditions. Virginia is the only state that does not have a
policy under which inmates who have complex health conditions or are permanently
incapacitated may be considered for release. VADOC spends more on inmates with
complex long-term medical conditions, which often require high-cost services or medications, than on inmates generally. VADOC spent more than $20,000, on average, for
810 inmates with complex long-term medical conditions in FY17. These 810 inmates
comprised only two percent of all inmates, yet their care accounted for nearly 20 percent of VADOC health care spending on services and medications.
Spending for inmates with terminal illness or complex conditions is far higher
than spending for all other inmates (FY17)
VADOC spending,
average for inmates w ith

Terminal illness

Complex long-term
conditions

Neither terminal illness
nor complex conditions
(all other inmates)

$1,612

~-------- - - - - - - - - - - - - - - - - - - - - - 0

10,000

20,000

30,000

40,000

50,000

60,000

$70,000

SOURCE: JLARC analysis of VADOC claims data.

Aligning Virginia’s compassionate release policies with other states’ could reduce
VADOC inmate health care spending—which is almost entirely state general funds—
by anywhere from $1.5 million to $16.9 million. Some of these costs, though, would
eventually be incurred through other parts of the state budget. Many inmates who
would be released would be covered through other public programs, especially Medicaid, which lower general fund requirements.
Problems with staffing, record keeping, and monitoring pose legal
and financial risk
Staffing problems, especially at facilities currently managed through contracts, pose
the risk of a court finding that the state is not providing adequate health care to inmates. Stable, effective health administrative leadership is essential to managing a facility’s health care operations. However, facilities operated by contractors struggle to
retain health administrators; nine of the 12 contract facilities lost their health administrators during FY17 alone. Similarly, key front-line staff, especially registered nurses
and licensed practical nurses, left contract facilities twice as often as they left VADOC
facilities.

Summary: Spending on Inmate Health Care

Challenges with VADOC’s medical records system and monitoring also potentially leave
the state legally and financially vulnerable. VADOC’s use of paper-based medical records
hinders its ability to efficiently and reliably demonstrate that it is providing adequate care,
which is essential when inmates allege they have been denied adequate care. VADOC
does have a monitoring program that helps manage risk. The monitoring program,
though, does not comprehensively address the full range of risks to adequate care at
facilities (especially at the facilities the department manages itself). Issues identified
through monitoring are also not fully tracked to resolution; this lack of follow-through
also makes the state vulnerable.
WHAT WE RECOMMEND
Legislative action


Direct VADOC to design a pilot project to pay Medicare rates to providers
who treat inmates.



Direct VADOC to work with the VCU Health Authority and the UVA
health system to (i) treat inmates with chronic conditions that require ongoing or high-cost prescription drugs and (ii) implement a clinical pharmacy services pilot project.



Direct VADOC to work with the UVA and VCU health systems to develop and propose a pilot project to provide inmate health care for at least
one VADOC facility.



Option: Amend the Code of Virginia to allow inmates with serious illness
to petition the Virginia Parole Board to be considered for release.

Executive action
 Extend the time requirement for a terminal diagnosis from 3 months to 12
months for an inmate to be considered for release.
 Modify contracts to incentivize retaining health administrators and key
front-line staff at contract facilities.
 Develop and implement a risk-based monitoring program.
The complete list of recommendations and options is available on page v.

iii

Summary: Spending on Inmate Health Care

Recommendations and Options: Spending on Inmate
Health Care
RECOMMENDATION 1

The General Assembly may wish to consider including language in the Appropriation
Act directing the Virginia Department of Corrections, VCU Health Authority, and the
University of Virginia Health System to develop and implement a plan to treat inmates
with chronic conditions that require long-term or high-cost prescription drugs through
a 340B-eligible provider. (Chapter 3)
RECOMMENDATION 3

The General Assembly may wish to consider including language in the Appropriation
Act directing the Virginia Department of Corrections and the VCU Health Authority
to undertake a pilot project to provide clinical pharmacy services to a specific population of inmates. (Chapter 3)
RECOMMENDATION 4

The governor should extend the life expectancy requirement for terminally ill inmates
to be considered for medical clemency to 12 months. (Chapter 3)
RECOMMENDATION 5

The Virginia Department of Corrections should seek to ensure stable health administrator and front-line staffing at contract facilities by modifying contracts to incentivize
and ensure stability. (Chapter 4)
RECOMMENDATION 7

Recommendations and Options: Spending on Inmate Health Care

RECOMMENDATION 8

The Virginia Department of Corrections should modify its comprehensive health services contracts to increase the fines, and reduce the 90-day grace period, for not meeting critical standard of care requirements. (Chapter 4)
RECOMMENDATION 9

The Virginia Department of Corrections should develop and implement a formal riskbased monitoring program as part of its existing continuous quality improvement program. The program should (i) identify risk factors related to access and follow-up; (ii)
monitor risk on a regular basis across all facilities; (iii) use the results of monitoring to
address the problems identified; and (iv) track the resolution of the problems identified through monitoring activities. (Chapter 4)
RECOMMENDATION 11

The General Assembly may wish to consider including language in the Appropriation
Act directing the initiation of a pilot partnership program for a university health system to provide comprehensive medical care for at least one Virginia Department of
Corrections (VADOC) facility. The program should be jointly developed by (i) the
director of VADOC; (ii) the chief executive officer of the VCU Health System; and
(iii) the executive vice president for health affairs at the University of Virginia. The
plan should be submitted to the House Appropriations and Senate Finance Committees no later than November 1, 2020. (Chapter 4)
OPTION 1

The General Assembly could amend Title 53.1 of the Code of Virginia to allow inmates to petition the Virginia Parole Board for conditional release based on serious
illness. (Chapter 3)
OPTION 2

The Virginia Department of Corrections could make health administrator positions
state employee positions, if the stability of staffing at contract facilities does not sufficiently improve. (Chapter 4)

Inmate Health Care in Virginia

SUMMARY The Virginia Department of Corrections (VADOC) is constitutionally obligated to
provide inmates “reasonably adequate” health care. In FY17, VADOC spent at least $201 million to provide medical, dental, and behavioral health care to approximately 30,000 inmates
at 41 facilities. VADOC contracts with several private companies to provide care and manage
health care claims, both in facilities and at community providers’ offices and hospitals. Virginia and other states have struggled to maintain successful, long-term contractual relationships with private health care companies. Some states are moving toward partnering with
university health systems to provide cost-effective care to inmates.

Correctional systems are required to provide “reasonably adequate” health care to
inmates under a 1976 U.S. Supreme Court decision, Estelle v. Gamble. The case established that constitutional protections under the Eighth Amendment, which guarantees freedom from cruel and unusual punishment, include that all inmates have a
right to a standard of health care that is available in the community. In Virginia, the
cost of providing this health care has been increasing faster than the total corrections
budget. From FY07 to FY17, the Virginia Department of Corrections (VADOC)
medical budget increased by $57.5 million (40 percent), adjusted for inflation. Over
the same time period, the total VADOC budget increased by $66.7 million (six percent).
This trend of increased spending prompted JLARC in 2017 to direct staff to study
the rising cost of inmate health care. The study resolution specifically called for staff
to identify factors contributing to cost increases and assess whether VADOC provides
health care efficiently and cost-effectively. The resolution also directed JLARC staff to
assess VADOC’s partnerships with contractors and other government entities to provide care. (See Appendix A for study resolution.)
To address the study resolution, JLARC staff interviewed VADOC health services
staff, contractor staff, clinicians working in VADOC facilities, and correctional department staff in other states. JLARC staff also surveyed clinicians and health services
administrators in VADOC facilities, and analyzed data provided by VADOC and its
various health services contractors. (See Appendix B for more detail on the research
methods used in this study.)

Chapter 1: Inmate Health Care in Virginia

VADOC spent over $200 million on inmate health
care in FY17
VADOC spent $201 million on medical and clinical services in FY17, or about $6,500
per inmate. This accounted for 17 percent of VADOC’s $1.2 billion operating budget,
nearly all of which comes from the general fund. (See Appendix C for more information on inmate health care spending in FY17.) Health care is the second-largest
portion of VADOC’s budget, behind only security and management of inmates. It is
more than twice as large as the budget for probation and parole and four times larger
than the budget for food.
Three categories of spending—onsite care, offsite care, and prescription drugs—account for most of inmate health care costs. Some of these costs are paid for through
comprehensive health services contractors, to whom VADOC pays a per-inmate capitated rate to provide care at several facilities. The contractor then is responsible for
paying the direct costs of health care, including onsite care, offsite care, and prescription drugs for the inmates at those facilities. Payments to comprehensive service contractors accounted for 44 percent of VADOC’s health care costs. Spending on onsite
care accounts for 17 percent of VADOC health care spending. The charges for offsite
services account for 20 percent of spending. Prescription drugs account for another
10 percent of spending (Figure 1-1).
VADOC is responsible for
providing health care to
all state-responsible inmates, or those with a
sentence longer than one
year. Some state responsible inmates are housed
in local jails rather than
VADOC prisons due to
space limitations. The
Compensation Board
pays local jails a per-day
fee for housing state-responsible inmates. When
an inmate’s health care
costs are high, local jails
may request VADOC to
pay for certain services.

FIGURE 1-1
Four largest spending categories account for 91 percent of inmate health care
costs (FY17)
Administration and other

Payments to jails ($6M)
Payments to VCU Health ($4M) - - - -

$SM

$10M
Prescription drugs

$20M -

Onsite care

•

, ..

Total inmate
health spendi

$33M

Comprehensive health
services contracts

$89M

$201M

20%
Offsite care

$41 M _ _ ___,

SOURCE: JLARC analysis of VADOC expenditure data, FY17.
NOTE: Payments to jails are for offsite care services and prescription drugs not covered in the daily rate paid by the
Compensation Board (sidebar). Payment to VCU Health was an annual payment to make up for lower reimbursement
rates, but this payment ended in FY17. Prescription drugs, onsite care, and offsite care categories include spending on
services at non-contract facilities as well as payment for inpatient care and some prescription drugs at contract facilities.

Chapter 1: Inmate Health Care in Virginia

Onsite and offsite services include medical, behavioral health, and dental care. VADOC provides several services in facilities, including health screenings, primary medical care, preventative dental care, and psychological and psychiatric services. Most
spending on onsite care is for the staff who deliver these services, as well as supplies
and equipment. If onsite clinicians cannot provide the services an inmate needs,
VADOC will schedule, transport, and pay for the inmate to visit a local hospital or
provider’s office for offsite care. VADOC also provides medication to inmates as prescribed by a physician (Figure 1-2).
FIGURE 1-2
VADOC provides a range of medical, behavioral, and dental services to inmates

Medical

Behavioral health

00
Dental

• Primary ca re

• Individual counseling

• Preventive care (cleanings)

• Urgent care

• Group counseling

• Chronic care

• Medi cation management

• Treatment of low-level
cond itions (fillings, caps)

• Medication management

• Crisis management

• Specialist visits

• Psychiatric emergency
that cannot be addressed
onsite

Onsite

• Emergency room visits
• Outpatient surgery
Offsite

• Inpat ient stays

• Oral surgery

• Psychiatric emergency
that results in medical
injury requiring
emergency care

SOURCE: VADOC health care providers and department operating procedures.
NOTE: Chronic care refers to regular treatment for chronic diseases such as hypertension and diabetes. Chronic care
services include blood work, testing, and medication management. Some facilities are able to provide specialized
services like dialysis onsite as well.

Many other costs of providing inmate health care are not as easily tracked and accounted for as part of VADOC’s health care spending. Taken together, these costs
totaled more than $10 million in FY17. Virginia’s Medicaid program may pay for inpatient care when an eligible inmate is admitted to a hospital for more than 24 hours.
Virginia’s Medicaid program paid $5.5 million for eligible inpatient claims in FY17,
half of which came from the general fund. VADOC also transported inmates for
offsite care and back more than 25,000 times at an estimated cost of between $1.3 and
$1.6 million for fuel and vehicle maintenance. Transportation sometimes requires
overtime costs for security personnel, but these are difficult to estimate because some
transportation is done by security staff during regular working hours. VADOC also
provides some substance abuse and behavioral health treatment in intensive therapeutic communities, at an additional cost of $3.2 million in FY17.

Chapter 1: Inmate Health Care in Virginia

VADOC inmates have higher health care needs than
people in the community
VADOC calculates average daily population
(ADP) by totaling the inmate population counted
in periodic head-counts
and dividing by the number of head-counts taken.

VADOC provides health care to approximately 30,000 inmates. The number of inmates in VADOC custody has decreased slightly over the past decade, from 31,756 in
FY07 to 30,455 in FY17 (four percent decrease). There were approximately 11,000
inmates who entered VADOC custody and another 11,000 who were released in FY17,
which is typical for a given year. Because most inmates eventually return to the community, effective health care—particularly for mental health disorders and infectious
diseases—has positive impacts on public health and recidivism.
Inmates generally have more health problems than individuals in the
community
Inmates typically have worse health than the community population. Nationally, inmates are more likely to have chronic health conditions, which drive up health care
costs. Inmates are also much more likely to have a history of mental illness or drug
or alcohol dependency. In 2010, approximately 65 percent of adults in prison nationwide met the medical criteria for an alcohol or drug use disorder, and 33 percent suffered from mental illness. In comparison, 18.3 percent of adults in the U.S. had a mental health condition (2016).
The higher health needs of Virginia’s inmates mirror the nationwide trends. In FY17,
27 percent of VADOC inmates were identified as having a mental health condition.
Approximately 65 percent of inmates reported a history of alcohol use or abuse and
43 percent reported a history of drug use or abuse (2015).
VADOC’s high-cost populations are growing
Like most health systems, VADOC spends a substantial amount of money on a small
number of inmates with high health care needs. Two populations with high health care
needs and costs—female inmates and those over age 55—are growing in VADOC.
Overall health care costs are much higher for older and female inmates than for
younger and male inmates.
VADOC’s inmate population is predominately male, but the proportion of female inmates is increasing. In FY17, 8.0 percent of VADOC’s inmates were female, compared
to 6.0 percent in FY1997 (a 33 percent increase). Many other states are experiencing
growth in their female inmate populations as well. Research suggests that this increase
in the female inmate population is a result of changes in drug enforcement policies
and the rise of the opioid epidemic.
While female inmates are still a relatively small proportion of the inmate population,
per-inmate spending is much higher for female inmates. In FY17, VADOC spent
$6,204 per male inmate and $10,543 per female inmate. According to VADOC health
services staff, care for female inmates is more expensive for several reasons: (1) female

Chapter 1: Inmate Health Care in Virginia

inmates generally have greater health care needs when they enter VADOC custody; (2)
female inmates are more likely to request care; and (3) more female inmates have a
diagnosed mental health disorder (Table 1-1).
Table 1-1
VADOC spends more on health care for female inmates than for male inmates
(FY17)
Percent of inmates with

Male

Female

Serious mental health diagnosis
Hepatitis diagnosis (A, B, or C)

17.7%
12.1%

53.4%
21.9%

Offsite utilization
Pharmacy utilization
Average spending

21.2%
64.9%
$6,204

30.8%
75.8%
$10,543

SOURCE: VADOC health and spending data, FY17.
NOTE: VADOC assigns each inmate a mental health code from 0-4 that indicates the presence and severity of mental
illness. A code of two or higher indicates a serious mental health diagnosis.

Virginia’s inmate population is also getting older. The percentage of inmates over age
55 has more than doubled in the past 10 years, from 5.6 percent in FY07 to 12.4 percent in FY17. Two factors have led to this increase. First, Virginia’s 1995 truth-insentencing reform has led to inmates serving longer portions of their sentences, so
that more inmates are aging within the prison system (sidebar). Second, the average
age of an inmate entering VADOC custody for the first time is increasing. This mirrors
the population of Virginia overall.
Older inmates, like older individuals in society, have higher health care costs than
younger inmates. The annual cost of providing prescription drugs and offsite care to
older inmates with chronic health conditions is more than four times that of younger,
healthier inmates. In FY17, inmates over age 55 accounted for about 12 percent of the
VADOC population but 33 percent of the offsite and prescription drug spending (Table
1-2).
Table 1-2
VADOC spends more on health care for inmates over age 55 than for younger
inmates (FY17)
Percentage of inmates with
Cardiovascular diagnosis

Age 55 and under
29.9%

Over age 55
60.7%

Hepatitis diagnosis (A, B, or C)

11.5%

19.6%

Diabetes diagnosis

3.0%

10.2%

Offsite utilization

20.8%

40.4%

Pharmacy utilization

66.3%

80.9%

$1,878

$9,140

Average spending on offsite services & prescription drugs

SOURCE: VADOC health and spending data, FY17.
NOTE: Onsite spending could not be separated by age group, so only offsite and pharmacy spending is reported.

In 1995, Virginia abolished parole. Under the
reforms, an inmate must
serve at least 85 percent
of his sentence. This
“Truth-in-Sentencing”
policy resulted in inmates
serving longer portions of
the full sentence.

Chapter 1: Inmate Health Care in Virginia

Virginia and other states use a combination of state
employees and contracts to provide care
Nearly every state’s department of corrections uses contracts to some extent to provide inmate health care. Virginia employs medical and other staff to provide care to
about half of its inmate population while contracting with comprehensive health services vendors to provide care to the other half. Other states use contractors to varying
degrees to provide care. In some states, like Maryland, all inmate health care is provided by contractors, who are overseen by state employees. Maryland contracts with
separate private companies to provide distinct types of care: medical, dental, behavioral health, and pharmaceutical. In other states, like North Carolina, the state provides
care directly but supplements state-employed clinicians with individually contracted
staff when needed.
VADOC uses mix of state employees and contracts to provide care
VADOC provides medical, dental, and behavioral health services at 41 facilities across
the state. The vast majority of inmates (90 percent) are housed in 27 major facilities,
and the remaining 10 percent are housed in smaller work centers and field units. Multiple types of clinicians—physicians, nurses, dentists, psychiatrists, and mental health
professionals—provide care in the facilities. Physicians, dentists, and psychiatrists prescribe medication and refer inmates offsite for specialist care as needed. Onsite medical services are often provided in a dedicated medical unit that is overseen by a health
administrator, who provides clinical supervision to a team of nurses, and administrative supervision to an onsite physician (Figure 1-3).
When inmates require offsite care, most are treated by the Virginia Commonwealth
University Health System (VCU Health), at either the secure care unit at its primary
hospital or by its affiliated physicians. This strategy of fully contracting out some
facilities while directly operating others is unique among state correctional systems.
Most states contract out for certain providers or services, rather than some full facilities.

Chapter 1: Inmate Health Care in Virginia

FIGURE 1-3
A health administrator typically oversees staff of nurses and onsite physician to provide care

Health administrator

---------------------------------~

Director of nursing

Physician

RNs

LPNs

SOURCE: VADOC staffing reports and HR data.
NOTE: The number of clinicians varies between facilities, due to differences in facility size and onsite health care capabilities. This is
an example of the medical unit staffing at one facility, where staffing levels are reasonably representative of facilities across the state.

Chapter 1: Inmate Health Care in Virginia

VADOC provides care at majority of facilities but contracts out care at highneed, hard-to-staff facilities
340B drug pricing is a
federal program that requires drug manufacturers to provide outpatient
drugs to eligible health
care organizations at significantly reduced prices.
Eligible organizations include those that serve
vulnerable, low-income,
and uninsured patient
populations. VCU Health is
eligible for 340B pricing.

At 25 facilities with half of all inmates, VADOC directly administers health care.
VADOC hires state-employed clinicians and pays for all offsite care and prescription
drugs. At 15 facilities, VADOC pays two correctional health care vendors a capitated,
per-inmate rate to provide comprehensive health services. The contractors hire clinicians, provide care in facilities, pay for outpatient offsite care, and purchase most prescription drugs. Contractors are not financially responsible for inpatient hospital care
and certain high-cost medications. This arrangement allows VADOC to seek reimbursement through the Medicaid program for eligible inpatient stays and obtain substantially reduced prices on high-cost drugs through the federal 340B prescription drug
program (sidebar). Virginia also has one fully privatized prison; one private company
provides all services, including medical care, at the correctional facility in Lawrenceville
(Figure 1-4). (See Appendix D for a detailed map of facilities.)

FIGURE 1-4
VADOC provides services directly at most major facilities
CORRECTIONAL FACILITIES
■

Non-contract

■ Contract

Secure medical facilities

SOURCE: JLARC analysis of VADOC website and contract documents, FY17.
NOTE: Figure only shows major correctional facilities.

According to VADOC, contracts were introduced to facilities that were particularly
hard for VADOC to staff, either because of the facility’s location or because the facility
provided intensive, specialized services that required a larger number of highly qualified clinicians. Seven correctional facilities have specialized health care capabilities in
addition to the primary care provided in all facilities; five of these acquire services
under contract. Five facilities have infirmaries for acute and chronic conditions, three
facilities provide dialysis services, four facilities provide specialized behavioral health
services, and one facility has an assisted living unit (Table 1-3).

Chapter 1: Inmate Health Care in Virginia

TABLE 1-3
At most facilities that provide specialized health care, services are contracted
out (FY17)
Facility
Greensville
Sussex II
Fluvanna
Deerfield
Deep Meadow
Powhatan
Marion

ADP
2,961
1,247
1,198
1,064
718
331
290

Health care specialty
Infirmary, behavioral health services, dialysis
Dialysis
Infirmary, behavioral health services, dialysis
Infirmary, assisted living unit
Infirmary
Infirmary, behavioral health services
Behavioral health services

Contract status
Yes
Yes
Yes
Yes
Partial
Yes
No

VADOC has recently insourced health care provision at Fluvanna, but
has a staffing contract
through which the former
comprehensive services
contractor provides clinicians to deliver onsite
care.

SOURCE: Reports by VADOC management and the Joint Commission on Health Care.
NOTE: ADP and contract status is provided for FY17. Dental care is not outsourced at any facility. Deep Meadow is a
non-contract facility but has an infirmary that is run by a contractor.

Additional contracts facilitate purchase of prescription drugs and payment of
offsite claims

VADOC and its two comprehensive contract vendors use third-party contractors to
manage offsite care claims and prescription drug purchasing. All offsite care claims,
regardless of facility, are managed by a claims administrator, who negotiates rates with
providers and bills either VADOC or the comprehensive contractor for the services.
Most prescription drugs are purchased through a “prescription fill” vendor that buys
drugs from manufacturers and wholesalers and delivers them to VADOC facilities.
VADOC and both comprehensive health services contractors use the same prescription fill vendor, but through separate contracts. VADOC and its contractors also use
smaller subcontracts as needed, including individual staffing contracts with providers
and a contract with a company that provides dialysis services at two facilities.
Many states, including Virginia, have struggled to maintain stable,
long-term contracts with private health care companies
Many states, including Virginia, have struggled to maintain long-term contractual relationships with private comprehensive health care contractors. This is largely due to
two types of conflict: (1) the inherent tension between the state’s need to reduce
spending and the contractor’s need to make a profit, and (2) a disconnect between the
party that provides care (contractor) and the party that is legally accountable if care is
not adequate (the state).
Both of these conflicts have played out in Virginia in recent years. In 2013, a private
company secured a comprehensive health care services contract for 17 facilities by
significantly underbidding other vendors. The contract price ultimately was an estimated $15 million lower than the actual cost to provide care, and the company terminated the contract less than a year later due to financial losses. In 2012, a class-action
lawsuit was brought against VADOC claiming inadequate health care at a contract facility, Fluvanna Correctional Center for Women. The parties settled two years later, but
litigation over the settlement is ongoing, and health care at the Fluvanna facility has

JLARC did not assess
whether VADOC is
providing adequate care.
The Joint Commission on
Health Care has produced
research on the quality of
care provided in state
prisons and local jails.
This reports aims to evaluate spending growth
and identify ways to reduce spending without
reducing quality of care.

Chapter 1: Inmate Health Care in Virginia

been under continual, intense scrutiny. VADOC recently moved to take back administrative control of health care at Fluvanna.
Challenges with contracting are not unique to Virginia; several other states have struggled to maintain successful relationships with private contractors. Florida contracted
with two separate companies in 2012 to provide health services across the state, but
both contracts were terminated early. One contract was terminated by the contractor,
citing cost concerns, and the other was terminated by the state, citing concerns about
adequacy of care.
Idaho contracted with a private company from 2014 to 2017 to provide health care services, but the state did not renew the contract because of lawsuits alleging inadequate care.
The state signed a contract with a new vendor in 2017, but a court monitor cited “grossly
insufficient and extremely poor quality of psychiatric services” less than a year later.
Many states, including Virginia, benefit from contracts with state
academic hospitals
VADOC partners with VCU Health to provide some health care to inmates. VCU
Health has a secure care unit within its medical center that allows inmates to be seen
by physicians for outpatient services as well as be admitted for inpatient stays in a
secure setting. VCU staffs the secure care unit with medical personnel, and VADOC
provides the security. As a result, two-thirds of inpatient and outpatient hospital care
is provided by VCU Health. VCU Health physicians in over a dozen specialties diagnose and treat inmates via telemedicine as well, so inmates do not need to leave the
facility to receive care. Additionally, VADOC and VCU Health have a memorandum
of understanding that allows VADOC to purchase certain high-cost drugs—those for
HIV, hepatitis C, and hemophilia—from VCU Health at 340B prices. University of
Virginia physicians in several specialties also use telemedicine to treat inmates.
Other states have partnered with state academic hospitals to an even greater extent. In
Texas and New Jersey, the state correctional departments have contracts with state university hospital systems to provide staffing, offsite care, and prescription drugs. Both
states transitioned to partnering with university health systems after settling significant
lawsuits pertaining to inmate health care. The medical expertise and associated credibility
of a university can improve confidence in a state’s ability to provide adequate care.
Other states indicate that partnering with an academic health system can also improve
staffing and reduce costs. Many states have difficulties staffing health care positions
because prisons typically have more challenging patients, both in behavior and health
status, and because prisons are often located outside of major metropolitan areas.
Working with an academic system can give the state access to a wider pool of clinicians
for inmate health care, while providing additional clinical settings for students to observe and learn. Formal partnerships with university hospital systems have allowed
states to access 340B pricing for all outpatient prescription drugs and to pay actual
costs of care, rather than commercial rates, for offsite services.

Spending on Inmate Health Care

SUMMARY Virginia spends a similar amount per person on inmate health care as others do
on comparable populations; however, the growth in spending over time has been higher.
There are several likely reasons for this persistently higher spending growth, including the
changing demographics of the inmate population, the reliance on more expensive prescription drugs, and general advancements in correctional health care standards. Within the last
five years, the major drivers of VADOC’s spending growth have been higher spending on
prescription drugs and offsite care services. VADOC has also increased its reliance on comprehensive health services contracts, and although this has made spending more predictable
in the short term, it is likely not saving the state money in the long term.

VADOC spent a total of $201 million to provide health care to nearly 30,000 inmates
in FY17. While the amount VADOC spends per inmate on health care is similar to
what others spend on comparable populations, the rate at which this spending has
grown over time is significantly higher than those other populations. VADOC’s per
person spending on inmate health care has grown by 76 percent over the past 10 years.
Over that same time period, the percentage of VADOC’s total budget dedicated to
medical services increased from 13 percent to 17 percent.

Health care spending on VADOC inmates is similar
to spending on comparable populations, per person
The amount VADOC spends per inmate on health care is similar to what the state
spends on a comparable subset of Medicaid recipients and what other states spend on
their inmate population. Per-person spending was used for this comparison because
the sizes of these populations differ significantly.
VADOC health care spending is about the same per person as state
Medicaid spending
Virginia spent nearly the same per inmate as it did per comparable Medicaid recipient
in FY17. VADOC spent $6,507, on average, to provide health care to inmates in
VADOC custody. The state’s Medicaid program spent $6,582 to provide care for a
subset of recipients, which is $75 (1.2 percent) more than VADOC spends per inmate
(Figure 2-1). In order to calculate a useful comparison, JLARC staff used per-person
spending for a subset of Medicaid recipients with similar characteristics to the state’s
inmate population, adjusted for differences in age, gender, and health status. (See Appendix B for more detail on this analysis.)

Virginia DOC spent
$201 million in primarily
general funds to provide
care to about 30,000 inmates (not including
state-responsible inmates
in local/regional jails). Virginia’s Medicaid program
spent $9.2B in federal and
general funds to provide
health care to about
1.13 million individuals in
FY17.

Chapter 2: Spending on Inmate Health Care

FIGURE 2-1
Virginia spent 1.2 percent less per inmate than per Medicaid recipient (FY17)
Medical spending
per individual

: 1.2% ($75} :
~-------11-----I
,,

VADOC

I
I
I
I
I

I
I
I

VA Medicaid

(adjusted)

VA Medicaid

(not adjusted)
The National Health Expenditure (NHE) measure is an annual estimate
of total health care
spending in the US, published annually by the
Center for Medicare and
Medicaid Services (CMS).

1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 $9,000

SOURCE: JLARC analysis of expenditure data from VADOC and DMAS, FY17.
NOTE: Figures are weighted to account for differences in gender, age, and health status between the two populations.

More broadly, the amount VADOC spends per person is considerably less than the
nationwide average per person. In FY17, the U.S. collectively spent $7,565, on average,
per person to provide health care, according to the National Health Expenditure measure, which is $1,058 (16.3 percent) more than VADOC spends per person (sidebar).
Virginia spends about the same per inmate on health care as other
states
Virginia’s spending per inmate was very close to the nationwide median in FY15 (latest
available data). Virginia’s spending was $217 (3.8 percent) more per inmate than the
nationwide median (Figure 2-2). Virginia ranked 24th out of 49 states that reported
data on inmate health care spending, according to the Pew Charitable Trusts. Several
states spent considerably more, especially California, which spends by far the most of
any state (nearly $20,000), in part because its entire prison system is under a court
mandate to provide a certain level of care. Other states that spend considerably more
than Virginia (Vermont, New Mexico, and Wyoming) tend to have relatively few inmates; consequently, these states have difficulty gaining economies of scale in providing care that can reduce the cost per inmate.

Chapter 2: Spending on Inmate Health Care

FIGURE 2-2
Virginia’s health care spending per inmate was very close to median of other states (FY15)
Spend ing per inmate

$20,000
18,000
16,000
14,000
12,000
10,000
8,000
6,000

Virginia spending per inmate +3.8% ($217)
• I•

Nationwide median

4,000
2,000
0

SOURCE: Prison Health Care: Costs and Quality, Pew Charitable Trusts, 2017.
NOTE: Does not include data for New Hampshire, because the state did not provide Pew with spending data.

Virginia’s neighboring states spent between 23 percent more, and 33 percent less, than
Virginia. Because of the differences in how states operate their correctional systems,
it is difficult to know precisely what accounts for these differences in spending across
states. Virginia spends 33 percent more than West Virginia, which mandates the use
of Medicaid rates for inmate health care; this policy likely allows West Virginia to contain spending. (See Chapter 3 on the possibility of using this approach in Virginia.)
Virginia spends about the same per inmate as Tennessee, 17 percent less than North
Carolina, and 23 percent less than Maryland. The latter difference is part of a larger
pattern: Virginia spends less than Maryland, per person, on many governmental functions including Medicaid.
Virginia’s spending has consistently been in the middle range across all states. The Pew
Charitable Trusts have collected inmate health care spending from 49 other states between FY10 and FY15. Virginia’s per-inmate spending has ranked between 22nd and
27th during that time period.

Chapter 2: Spending on Inmate Health Care

Health care spending on VADOC inmates has grown
faster than spending on similar populations
Comparing the level of spending provides insight into a single year, but a comprehensive assessment necessitates also comparing the rate of growth in spending over time.
If historical rates of growth are trending higher, policymakers can take action to moderate growth. (See Chapter 3 for potential cost-saving strategies.) The need to moderate future growth is especially important for services, such as inmate health care, that
rely heavily on state general funds.
VADOC health care spending is rising faster than state Medicaid
spending and nationwide health care spending, per person
VADOC spending on health care per inmate has grown 76 percent from FY07 to
FY17. This represents an annualized growth rate of 7.6 percent. This increase in health
care spending is outpacing VADOC’s operating budget. As a result, health care spending accounts for 17 percent of the total VADOC budget, up from 13 percent in 2007.
VADOC’s spending per inmate has risen faster than several reasonable comparator
groups (Figure 2-3). VADOC’s spending increased at nearly twice the rate of growth
of Virginia’s Medicaid program. VADOC spending also increased at nearly twice the
rate of growth in health care spending nationwide. Per-inmate spending in Virginia
has also grown slightly faster than per-inmate spending in other states, ranking 17th in
growth in per-inmate spending from FY10 to FY15 (latest available data).
VADOC spending has also been growing faster than medical services nationwide as
calculated in the Consumer Price Index. Medical services costs grew 35 percent between 2007 and 2017. This was less than half the rate of growth in VADOC inmate
health care spending during the same time period.

Chapter 2: Spending on Inmate Health Care

FIGURE 2-3
Virginia’s per-inmate spending has been increasing more than its spending on Medicaid and
health care spending across the U.S. (2007-2017)
Medical spending, per individual

$8,000

National health expenditures +41%
VA Medicaid +42%
VADOC +76%

7,000
6,000
5,000

~~~~=---- --

1--:-~~~::____-----::;,__

4,000
3,000
2,000
1,000
0 '-----------------------

2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
SOURCE: JLARC analysis of expenditure data from VADOC and DMAS, and CMS National Health Expenditures.
NOTE: Figures not adjusted for inflation.

There is no quantifiably reliable way to isolate exactly why VADOC spending is growing more quickly than comparison groups. There are, however, several factors that are
at least partially contributing to this faster growth. VADOC’s population is, on average,
less healthy than the general population, and specifically has a higher rate of infectious
diseases. Many of these infectious diseases, including hepatitis C and HIV, require expensive medications and treatment.
The cost to treat these infectious diseases may continue to grow even faster, as treatment protocols change in response to emerging legal challenges. Two pending lawsuits
against VADOC allege that VADOC has failed to provide adequate treatment to inmates with hepatitis C. Judges in similar lawsuits in several other states have ruled in
favor of inmates and required states to provide newer, high-cost treatments for hepatitis C to all inmates with a diagnosis on a schedule dictated by the court.
Changing demographics have also contributed to increased spending, as the proportion of female inmates and inmates over the age of 55, both of which are associated
with higher health care costs, continues to grow. From FY12 to FY17, the percentage
of female inmates increased from 7.1 percent to 8.0 percent and the percentage of
inmates over 55 increased from 8.5 percent to 12.4 percent. If VADOC had the same
proportion of elderly inmates in FY17 as it did five years ago, total spending would
have been about $11 million less.

Chapter 2: Spending on Inmate Health Care

According to VADOC’s projections of the future inmate population, the trends that
contribute to faster growth are likely to continue in the near future. For example, offender forecasting reports predict the number of females to increase by 11.3 percent
between FY17 and FY23, compared to a 2.9 percent increase among males. Although
forecasting is not conducted for older inmates, the number of inmates age 50 or older
increased by almost 40 percent from 2010 to 2016, and there is no indication that this
growth rate is slowing.
This trend of significantly higher spending growth continued in FY18 but may be moderating, based on projected spending. Spending increased by more than 10 percent in
FY18, but the amount appropriated to VADOC for FY19 increases by only 1.4 percent
from FY18 and actually decreases in FY20 by 4.2 percent. This moderation in the growth
rate is largely the result of Virginia’s decision to expand its Medicaid program, which
will shift most of the cost of inpatient hospital stays to non-general fund sources in
FY19 and FY20. It is unclear, though, how much a potential judicial requirement to test
and treat all inmates for hepatitis C would offset this moderation in growth.
Advancements in the
“professionalization
and
sophistication of care
provided … brought care
for prisoners into closer
alignment … with health
care provided in the
community.

”

– Prison Health Care:
Costs and Quality
Pew 2017

The continuing growth in spending on inmate health care should be viewed in the
context of an ongoing effort to bring correctional health care in line with community
standards. This effort is the result of Estelle v. Gamble, the 1976 Supreme Court decision
that granted inmates a constitutional right to health care. In the decades since this
decision, as the expected community standard for care has changed, so too have correctional systems had to adapt, while still trying to control costs. The dilemma facing
correctional systems was highlighted in a 2008 article published in the Journal of Correctional Health Care:
Estelle results in one essential difference from the community: A constitutional
guarantee of fulfillment of all required health care needs will outweigh management of the bottom line.
Failing to meet such standards has, in many states, resulted in litigation, which in turn
has led to the imposition of court-mandated standards, further increasing correctional
health care budgets. Consider California, where per-inmate spending increased by 25
percent (from $15,827 in FY10 to $19,796 in FY15) after the state’s correctional system was placed under a court order to improve general conditions in their prisons,
which has required them to invest substantially in health care facilities and staff.

In 2014, new high-cost
drugs were approved to
treat hepatitis C. The use
of these drugs was a significant driver of VADOC’s
increase in prescription
drug spending over the
past five years.

Prescription drugs and offsite care, and more inmates at contract
facilities, are drivers of recent spending increases
Growth in health care spending at VADOC has primarily been driven by the increasing
cost of prescription drugs and offsite care, which make up nearly half of the increase
in total health care spending (Figure 2-4). These categories of spending have gone up
primarily due to more expensive medications and high-cost inpatient stays; utilization
rates have been fairly stable during this time period. Spending on comprehensive con-

Chapter 2: Spending on Inmate Health Care

tracts makes up over half of the remaining increase in total health care spending. However, this is primarily because VADOC added more facilities and inmates to the scope
of the contracts, not because of an increase in the capitated rate.
FIGURE 2-4
More inmates at contract facilities, and increase in prescription drugs and offsite care,
account for nearly all spending growth (FY12-FY17)
,-----

+$27.lM net growth.in spending

--:

$173.SM

■·--1,
Onsite

+$9.0M
33%

:
+$15.3M
56%

:'
I

I
I
I

Contracts

-$!~M

Other

Onsite

,:r- ■,
I

_____ !

,____ _
Offsite

Offsite

+$l ZM
4%
'

Prescription
drugs

Other ($2.8M)

·-

+$4.lM
15%

$200.GM

•

■

!fiiif1i

Prescriptio
drugs

Contracts

FY12

FY17

SOURCE: JLARC analysis of VADOC expenditure data, FY12 and FY17.
NOTE: FY12 figures have been adjusted for inflation. The increase in spending on contracts is largely the result of additional facilities being
added to the contracts, not an increase in per-person spending under the contracts. Excludes other categories of spending that combined
accounted for four percent of growth.

These spending increases may have been even larger without VADOC’s recent efforts
to manage spending. VADOC has reduced the price it pays for some high-cost prescription drugs by accessing federal 340B drug pricing through the VCU Health system. VADOC also changed its reimbursement method for inpatient hospital stays at
the primary hospital it uses, VCU Health, reducing the average cost per stay. Further,
as VADOC’s health care costs at its non-contract facilities have continued to rise, the
per-inmate capitated rate included in the comprehensive contracts has remained fixed.
This has helped control spending growth over the short term.
One of the typical drivers of increased health care spending is an increase in the number
of people who need and receive care. However, providing care to more inmates is not
one of the drivers of recent VADOC spending increases. The number of state-responsible inmates in VADOC custody has remained fairly stable over the past 10 years, ranging between 33,691 and 29,767, with an annual average population of 30,970.

From 2006 to 2012, VADOC used a risk-sharing
contract model. VADOC
and the contractor split
small-to-moderate profits
and losses equally. Extensive profits and losses
(greater than 17 percent
for offsite care, 10 percent for staffing) were incurred by VADOC.

Chapter 2: Spending on Inmate Health Care

Use of contracts stabilizes spending in short term
but likely does not reduce spending in long term
As noted above, the largest driver of recent VADOC spending increases is the expanded scope of contracting. VADOC pays comprehensive health services contractors a capitated rate to incentivize contractors to find efficiencies and reduce costs.
Using a capitated rate shifts the financial risk or potential for financial reward to the
contractor. If the contractor can provide services for less than the capitated rate, the
contractor profits. If the contractor spends more than the capitated rate, the contractor incurs the financial loss.
VADOC’s current contract structure is designed to incentivize contractors to find efficiencies specifically in offsite and pharmacy services. Offsite care, prescription drugs,
and onsite staffing are the largest spending categories in VADOC health care, but onsite staffing costs are largely fixed due to staffing level requirements built into the contract.
The capitated rate paid
by VADOC under its comprehensive contracts includes the cost for all onsite care, outpatient visits,
and most prescription
drugs.

VADOC uses capitated rate structure for contracts to stabilize spending
VADOC’s use of capitated rates has helped manage annual costs increases and made
spending more predictable in the short term. VADOC negotiated a fixed capitation
rate with its contractors for the first three years of the current contract. Contractors
have therefore had to absorb some of the increasing costs of prescription drugs and
offsite care, limiting the contractors’ ability to cover administrative costs and corporate
overhead. The largest contractor reported financial losses in 2016 and 2017, citing
offsite care and prescription drugs as the largest drivers of spending growth.
Over the long term, though, it is unlikely that rates in the current contracts will remain
the same. The capitated rates for VADOC’s comprehensive health services contracts
were renegotiated in November 2018, and can be renegotiated in each of the following
two, one-year renewal periods. Because of the cost increases that the contractors have
had to absorb, it is highly likely they will seek higher capitated rates during the renegotiation.
This short-term versus long-term dynamic is inherent in the use of capitated rates. In
the short term, entities like VADOC can stabilize spending increases through the use
of capitated rates. However, VADOC will only save money in the long term on its
health services contracts if contractors are able to find efficiencies and deliver care at
a lower cost than VADOC could. The contractor must cover its direct costs, administrative expenses, and make a profit to sustain a long-term contractual relationship. The
payment of administrative costs and profits is only financially beneficial for the state
if the contractor saves money on direct costs.

Chapter 2: Spending on Inmate Health Care

No evidence that contract facilities spend less than non-contract
facilities on pharmacy and offsite services
One of the goals of contracting with private companies is to incentivize cost savings;
however, per-inmate spending is higher at contract facilities. Contract facilities spend
slightly more, on average, than non-contract facilities on prescription drugs ($92 vs.
$83 per person per month) and substantially more on offsite services ($202 vs. $130
per person per month). This spending data, though, does not account for the fact that
facilities with specialized health care capabilities and services are more likely to be contracted-out. Consequently, contract facilities house inmates with higher health care
needs, compared to non-contract facilities. To account for this key difference between
facilities, JLARC conducted statistical analysis to control for patient demographic and
health factors (sidebar).
JLARC analysis found no evidence that contract facilities spend less than non-contract
facilities. Contract and non-contract facilities’ pharmacy and offsite spending was very
similar after controlling for patient differences (Figure 2-5). About one-third of both
non-contract and contract facilities spent less than expected after controlling for demographics and health. Based on the very similar results for non-contract and contract
facilities, the use of contracts may not be reducing spending.

JLARC developed four
regression models to
predict pharmacy utilization and spending, and
offsite utilization and
spending for each facility.
The models controlled for
inmate demographic information including age,
race, and gender, as well
as health characteristics
such as mental health
status and chronic disease diagnoses. (See Appendix E for more detail
on this analysis.)

Spending reductions are not the only potential benefits of using contracts, though.
Correctional systems also use contracts for other purposes, including to more effectively recruit and retain medical staff. (See Chapter 4 for additional information on
how contractors are used to staff facilities.)
FIGURE 2-5
Spending is similar at contract and non-contract facilities, adjusted for inmate
characteristics (FY15-FY17)
Contract facility spending

Non-contract facility spending

Less tha n
predicted

Lesstha n
predicted

More tha n
pred icted

More than
predicted

SOURCE: JLARC analysis of prescription drug and offsite care claims data, FY15-FY17. Spending is adjusted for each
facility based on inmates’ demographic characteristics and health status.
NOTE: This analysis excludes one contract facility, Powhatan, which is an infirmary.

Chapter 2: Spending on Inmate Health Care

Strategies to Reduce Spending

SUMMARY Providing health care to inmates, who typically are less healthy than the general
population, will always necessitate substantial public investment. Virginia can, however, make
several changes that will reduce its spending over time. Despite being a public purchaser of
health care, VADOC pays higher rates than other larger public health programs: Medicare
and Medicaid. Using lower, Medicare rates as the basis for reimbursement could reduce what
VADOC pays for these services by $9 million annually. VADOC also pays more for certain
prescription drugs. Furthering its partnership with the VCU or UVA health systems could reduce total spending on biologic medications, various types of inhalers, and insulin by
$1.5 million to $6.6 million annually. The most substantial potential savings could come from
implementing a less restrictive compassionate release policy. Virginia has one of the nation’s
most restrictive compassionate release policies. Revising its policy to mirror other states could
reduce VADOC spending by $1.5 million to $16.9 million.

All health care providers, including correctional health care systems, have three primary ways to manage or reduce spending:
 paying providers less for services and prescription medications,
 reducing utilization of services or medications, and
 providing services or medications to fewer individuals.
JLARC staff were directed to assess how efficiently VADOC is providing care, in particular the use of practices known to manage or reduce spending.

VADOC pays more than other public payers for
certain services and prescription drugs
VADOC could reduce the prices it pays for many drugs and services, but this must be
balanced against increased administrative costs and potential impacts on access to services and provider revenue. Obtaining the best possible price for necessary services
and prescriptions is the most direct way for VADOC to reduce spending on inmate
health care. The cost of health care services, particularly prescription drugs and hospital services, can vary widely depending on the provider of the service and who is
paying for it.

Chapter 3: Strategies to Reduce Spending

VADOC pays more than other public purchasers pay for physician and
hospital services
VADOC spent $63 million in FY17 on offsite care for inmates, including inpatient
hospital stays, outpatient procedures, and consultations with specialists such as oral
surgeons and cardiologists. VADOC currently pays providers the rate that is negotiated by the vendor that administers its claims for offsite care. These rates are often
higher than the rates Medicare and Medicaid pay. For example, VADOC pays about
$130 on average for a radiologist to conduct a type of CT scan, while Medicare pays
$92 (29 percent lower) and Medicaid pays about $80 (38 percent lower).
Consistent with federal
guidelines, inmates must
agree to be covered by
Medicaid. VADOC staff
indicate that it is likely
some inmates will not
agree to be enrolled in
Medicaid, or will be ineligible due to their immigration status or financial
circumstances. For these
inmates, the current
funding stream of primarily state general funds will
be used for inpatient hospital stays.

Medicaid expansion will shift the vast majority of costs for inpatient hospital services
out of VADOC’s budget. Medicaid pays for the cost of inpatient hospital services for
any inmate who would otherwise be eligible for Medicaid but is incarcerated. In prior
years this was a small number of elderly and disabled inmates, but after Medicaid expansion, any low income adult will be potentially eligible, and VADOC will be able to
enroll nearly all inmates in Medicaid (sidebar).
This shift will result in substantial savings to the state general fund. In FY17, VADOC
spent $30 million on inpatient hospital stays, and the total cost for these services will
be lower under Medicaid rates. Savings to the general fund will occur primarily because
VADOC’s budget is appropriated almost entirely from the general fund, and the costs
for Medicaid expansion will be paid for by federal funds and revenue generated
through a new tax on private hospitals.
While Medicaid expansion will allow VADOC to pay lower rates and shift much of
the spending to non-general funds for inpatient hospital stays, expansion does not
address physician services or outpatient hospital stays. The state could, though, take
the further step of reducing VADOC rates to the level of either Medicaid or Medicare
for these other aspects of care not addressed through Medicaid expansion.
The state could save the most by using Medicaid rates for physician services and outpatient hospital stays. Medicaid rates would be the lowest in most instances, and therefore result in the largest potential savings. JLARC estimates that VADOC would have
saved approximately $13 million on outpatient and physician services in FY17 (37 percent) had it reimbursed providers using Medicaid rates (Figure 3-1). The state could
also use Medicare rates, which are typically higher than Medicaid but lower than the
rates VADOC currently pays. Using Medicare rates in FY17 would have resulted in an
estimated $10 million in savings on outpatient and physician services (28 percent).

Chapter 3: Strategies to Reduce Spending

FIGURE 3-1
VADOC spending on outpatient and physician services would have been lower
using Medicare or Medicaid rates (FY17)
I

I
I
I

l
I
I
I

28%

I
I
I
I
I
I
I
I
I
I
I
I
I

37%

_________,
I

VADOC

Medicare

Medicaid

SOURCE: JLARC analysis of VADOC claims data, DMAS data on Medicaid provider rates, DMAS comparison of Medicaid and Medicare rates, and Medicaid hospital cost reports, FY17.
NOTE: Actual savings would vary depending on the mix of services used because the difference in rates vary for each
service.

Some other states use Medicare or Medicaid rates for inmate services

At least 10 states use either Medicare or Medicaid rates as the basis for reimbursement
for inmate health care. Five of these states (Arizona, Georgia, Maine, New Jersey, and
West Virginia) use Medicaid, while two (Oregon and North Carolina) set a higher rate
that is tied to Medicaid. Three states use either Medicare rates or a rate that is slightly
higher than Medicare (Florida, Indiana, and Texas).
It is unclear whether these states’ use of lower rates has reduced the number of providers willing to see inmates as patients. Reimbursing at lower rates always raises the
risk that at least some providers will stop accepting patients. States that used Medicare
or Medicaid rates for inmate health care have either not attempted to measure the
impact on access when making this change, or anecdotally indicated they did not notice
any major reductions in access to care.
Role of VCU Health and UVA health system, as well as broader Medicaid
expansion, complicate decision to lower VADOC rates

Reducing reimbursement rates could reduce the number of providers willing to treat
inmates, particularly in more remote areas of the state that have fewer providers. There
were eight VADOC facilities located in regions of the state with low Medicaid participation by specialist providers (based on analysis from a 2013 JLARC study), but seven
of these facilities already send a majority of inmates to VCU Health for offsite care.

Oregon pays a rate
higher than Medicaid for
inmate health care services, but indicated this
creates a perverse incentive for hospitals. Hospitals receive Medicaid
rates for inpatient services under Medicaid expansion, but higher rates
for outpatient services.

JLARC’s 2013 study, Review of the Impact of
Medicaid Rates on
Health Care Access in
Virginia, analyzed Medicaid provider participation rates across the state.

Chapter 3: Strategies to Reduce Spending

The remaining facility, Coffeewood, is the most likely to experience challenges scheduling offsite appointments if fewer providers are willing to treat inmates.
Although using lower rates may not substantially reduce access, it is clear that lowering
rates would reduce VADOC payments to VCU Health and the UVA health system.
Reducing spending on inmate health care through lower rates is to some extent a “zero
sum” game. More than 75 percent of hospital fees are paid to VCU and UVA medical
centers, with the vast majority being handled through the secure care unit at VCU.
Physicians from both of these health systems also provide a large amount of specialist
services to inmates. Reducing reimbursement rates would reduce the revenues of almost all providers, with a disproportionate reduction for these two state teaching hospitals. Based on FY17 spending, these hospitals may see reductions of up to $4 million
in total for outpatient services.
The impact that shifting to lower rates would have on providers is complicated by
Virginia’s impending Medicaid expansion. One question that the General Assembly
and the Department of Medical Assistance Services are currently analyzing is whether
Virginia’s health care market will be able to accommodate and provide care for the
estimated 400,000 individuals newly covered under Medicaid expansion. This major
market shift will generate a substantial amount of potential new patients for providers,
increasing revenues but also straining the ability of providers to treat all patients, particularly in already underserved areas.
Shifting to Medicare rates would carry less risk of hindering access, and providers
would not experience as significant of a reduction in revenues. Medicare rates are more
widely accepted than Medicaid rates among both primary care physicians and specialists, providing a broader pool of physicians for VADOC to contract with to treat inmates. The percentage of physicians who accept Medicare rates is typically about the
same as the percentage that accept commercial insurance, indicating there may not be
a significant change in the number of providers willing to treat inmates under Medicare
rates. The adverse impact on revenues for VCU Health and UVA health system would
also be partly mitigated, with an estimated reduction of about $2 million in FY17
(compared to $4 million using Medicaid rates).
Feasibility and impact of lower rates could be tested through a pilot project

VADOC’s reduced spending on inpatient hospital services from Medicaid expansion
will likely generate some savings in the near term. However, the impacts of also using
lower rates for physician and outpatient services are uncertain, so 2019 may not be the
appropriate time to attempt to reduce rates. It would seem advisable to wait for the
broader Medicaid expansion to take hold and then reassess the feasibility and impact
of using lower rates similar to Medicare rates (lower than current VADOC rates but
higher than Medicaid rates).
To prepare to determine the feasibility and impact of using Medicare rates (or Medicaid rates or any rate lower than the current VADOC rates), the General Assembly

Chapter 3: Strategies to Reduce Spending

should direct VADOC to design a small-scale pilot project. The pilot project’s goal
would be to select certain regions of the state where the private market may be able to
accommodate lower rates for physician and outpatient services, and begin the process
of lowering rates to assess how providers respond. When necessary, VADOC may
need the ability to negotiate higher rates with specific providers if VADOC is unable
to schedule necessary appointments for inmates with certain medical needs. VADOC
could submit the design for the pilot project to the General Assembly, and the project
could be initiated after Medicaid expansion is underway.
RECOMMENDATION 1

The General Assembly may wish to consider including language in the Appropriation
Act directing the Virginia Department of Corrections (VADOC) to design a pilot project that would test the feasibility and assess the impact of using lower rates, potentially
based on Medicare rates, for physician and outpatient services. VADOC should submit
the pilot project design to the House Appropriations and Senate Finance Committees,
and implement the pilot project no later than 2021.
VADOC pays more for certain medications than it would pay through
VCU Health
VADOC spent more than $31 million on prescription drugs in FY17, and prescription
drug spending is growing much faster than overall health care spending at VADOC
and nationwide. VADOC currently buys 60 percent of its prescriptions through a private vendor, while the remaining 40 percent are purchased through VCU Health.
The prices that the vendor negotiates tend to be higher than the prices that VCU
Health gets through the federal 340B drug purchasing program (sidebar). For example,
VADOC pays about $39 for the most common type of inhaler prescribed for inmates
with asthma and other respiratory conditions, while VCU Health purchases the exact
same inhaler for about $8. However, the administrative expenses that the vendor
charges for dispensing the prescriptions are relatively low.
VADOC could substantially reduce the price it pays for certain drugs by
expanding its partnership with VCU Health

VADOC spent more than $8.3 million on medications for auto-immune diseases (biologics), inhalers, and insulin in FY17, and these drugs could be purchased at substantially lower prices if inmates were treated by VCU Health physicians, one of the key
requirements to be eligible for 340B drug prices. The cost of these drugs through the
340B program would have been an estimated $1.7 million (Table 3-1). The estimated
savings include more than a 90 percent reduction in the acquisition price of the drugs,
which would be offset to some extent by higher dispensing fees.

Federal 340B program
requires drug manufacturers to sell drugs at discounted prices to providers that treat a high
number of indigent or
Medicaid covered patients. Providers that
meet these criteria are
able to obtain these advantageous prices for any
patient that they treat in
an outpatient setting.

Chapter 3: Strategies to Reduce Spending

TABLE 3-1
Costs for three high-cost types of drugs would be 80 percent lower if inmates
were treated by VCU Health (FY17)
VADOC
spending

Estimated cost
through VCU Health

Estimated
savings

Inmates
treated

Biologics

$3.49M

$0.16M

$3.33M

150

Inhalers

$2.65

$0.70

$1.47

Insulin

$2.17

$0.80

$1.85

2,400

Total

$8.32

$1.66

$6.65

5,831

3,762

SOURCE: JLARC analysis of data from VADOC and VCU Health, FY17.
NOTE: Estimates are based on comparing actual costs for a selected sample of drugs at a specific point in time. Actual
savings vary by drug over time. “Inmates treated” column does not sum to the total because some inmates are on
more than one medication.

Federal legislation to
limit use of the 340B
program has been introduced for the last several
years. One proposal has
been to make all inmates
ineligible for 340B prices
on prescription drugs, regardless of whether they
are being treated by a
340B-eligible provider.
This raises uncertainty
over the long-term sustainability of cost savings.

VADOC’s current prescription drug vendor
partners with correctional
systems and 340B providers in other states to fill
prescriptions. This reduces the burden on the
pharmacy services of the
340B provider and the
need to fill prescriptions
through different mechanisms.

VADOC recently entered into an agreement with VCU Health to start treating inmates
who need biologic medications. However, VADOC is having difficulty scheduling appointments for some inmates with certain diseases, such as skin conditions, limiting
VADOC’s ability to obtain lower-cost medications. This is due to a limited number of
dermatologists within VCU Health.
The relatively higher number of inmates prescribed inhalers and insulin would necessitate a more extensive partnership and a significant increase in capacity. Inmates in
need of these drugs typically have diabetes, asthma, and chronic obstructive pulmonary disorder (COPD), requiring patients to be seen regularly by a physician in VADOC’s chronic care clinics at each facility.
VADOC should consider other 340B-eligible providers, including the University of
Virginia (UVA) health system, in establishing partnerships that could further reduce
prescription drug spending. In addition to the UVA health system, most other major
hospital systems, local departments of health, and many clinics that serve low-income
populations participate in the 340B program. It is not known which providers may
currently have, or could develop, the capacity to treat inmates in need of high cost
prescription drugs. However, given the capacity limitations at VCU Health and the
challenge of transporting some inmates to Richmond to be treated, seeking providers
closer to VADOC facilities could be advantageous.
RECOMMENDATION 2

Chapter 3: Strategies to Reduce Spending

VADOC pays similar prices for prescription drugs as other public purchasers
that use a multi-state buying group

Other state agencies in Virginia, and many other public health care providers in other
states, purchase prescription drugs through the Minnesota Multistate Contracting Alliance for Pharmacy (MMCAP). MMCAP is able to leverage the volume of drugs purchased across all of its members in negotiating prices with drug manufacturers and
wholesalers. The prices VADOC pays through its current vendor, though, are very
similar to MMCAP’s negotiated prices. Because prescription drug prices fluctuate, VADOC will need to periodically monitor the prices it pays relative to what is available
through other purchasing arrangements. However, any decision to change how it purchases prescription drugs needs to consider the level of service provided by the current vendor, which includes

MMCAP is a buying
group that represents
government purchasers
of prescription drugs
from all 50 states and is
able to leverage that volume into lower prices
with drug manufacturers.

 shipping prescriptions drugs to each VADOC facility in a timely manner;
 accepting the return of unused medications;
 providing an electronic prescription drug record for all inmates;
 managing VADOC’s prescription drug formulary; and
 on demand, customized reporting capabilities.

VADOC effectively employs utilization management
to control spending
Utilization management is a practice that can be used to minimize unnecessary use of
services and thereby control spending. Utilization management typically consists of
(1) a review of service requests to ensure services are medically necessary; (2) approving only those requests for services that are deemed necessary; and (3) providing only
those treatments that have been shown to be cost-effective. Ideally, utilization management principles are applied to both service requests and decisions to prescribe
medication. It is important that VADOC implement utilization management practices
to ensure it pays only for necessary services. Utilization management practices also
protect the state against the risk of litigation when inmates allege they have been denied necessary services.
VADOC review and approval process minimizes unnecessary
procedures
VADOC has procedures in place to minimize the use of unnecessary services. It requires prior approval for nearly all offsite services, which is consistent with industry
practices. When inmates have non-emergency medical issues, they must first be treated
by a physician in the facility. When the physician recommends a specialist, or any other
offsite service, the physician submits a request to the chief physician at the VADOC
central office. The chief physician reviews the request and accompanying notes for

Decisions about medical
necessity in historical
claims data were analyzed by VADOC’s thirdparty administrator using
externally developed criteria. Analysis found few
high-cost services that
would have been subject
to further review and potentially recommended
for alternative treatment
options. Most high-cost
services would have been
approved.

Chapter 3: Strategies to Reduce Spending

medical necessity and either approves the request or recommends alternative treatment. Physicians at contract facilities follow a similar protocol, but the reviewer is a
physician employed by the contractor. This process for reviewing medical necessity is
consistent with other state correctional systems, as well as insurance companies paying
for care in the community.
For decisions about medical necessity, the VADOC chief physician uses internal criteria, supplemented by in-house research on current medical practice. Some other organizations use a more rigorous approach that involves externally developed criteria,
which are comprehensive and continually updated. However, VADOC would probably
not achieve significant cost savings by using external criteria. The external criteria
could help speed up the decisions about medical necessity, but according to analysis
by VADOC’s third-party claims administrator (sidebar), the use of external criteria
would probably not identify many unnecessary services that are not already identified
under VADOC’s current process.

Clinical pharmacy services are a broad group
of activities performed by
a clinical pharmacist. Services can include directly
working with patients and
physicians to review drug
effectiveness, interactions,
and costs, systematic reviews of prescriptions for
lower cost options, and
population-wide analyses
to update formularies
and treatment protocols.

VADOC prescribes drugs approved to be cost-effective but could use
more rigorous approach
VADOC uses a prescription drug formulary as its primary method of managing prescription drug utilization. This approach is consistent with other health care payers.
The formulary provides a list of approved drugs to treat specific conditions and is
developed and regularly updated by a committee of physicians and pharmacists to
identify the most cost-effective medications available. When a physician believes an
inmate needs a medication that is not on the formulary, the physician seeks approval
by the VADOC chief pharmacist, or a similar clinician who works for the comprehensive health services contractor.
It may be possible for VADOC to use a more rigorous approach to prescribing medications. Some organizations use clinical pharmacy services or medication therapy management to maximize the cost-effectiveness of prescribed drugs (sidebar). Research
literature on this practice finds that, in most cases, total medical spending is reduced
while health outcomes are improved. For example, medical research has shown that
 use of clinical pharmacy services for patients with heart disease reduced
the risk of cardiac-related emergency room and hospital visits by 50 percent
and reduced the average cost of these cardiac episodes by 30 percent; and
 use of clinical pharmacy services for patients with asthma reduced the average cost per patient by $725 per year, by reducing emergency room visits
and long-term hospitalizations.
Clinical pharmacy services tend to be most effective when they are targeted to a patient
population with multiple, complex medical conditions that are treated and managed
with many different prescription drugs. Inmate populations often include individuals
with these types of conditions. Faculty at the VCU School of Pharmacy expressed
interest in working with VADOC to determine if a partnership for clinical pharmacy

Chapter 3: Strategies to Reduce Spending

services would be possible. VCU faculty indicated that focusing on patients with complex medical needs and high pharmacy utilization would likely be beneficial, and that
the use of telemedicine for pharmacist consultation would significantly reduce the logistical challenges of providing services. VCU faculty also indicated that it would provide a valuable teaching environment where pharmacy students could learn from
trained clinical pharmacists who are providing services to inmates.
VADOC and the VCU School of Pharmacy should be tasked with implementing a
pilot project to establish the feasibility of clinical pharmacy services for inmates. Establishing a clinical pharmacy program at VADOC would entail identifying an inmate
population for which the services could be most beneficial. It would also entail identifying and hiring clinical pharmacists who are willing to work with VADOC and establishing a protocol to integrate clinical pharmacy services into the current process
for providing care. The purpose of the pilot project would be to evaluate the feasibility,
costs, and benefits of developing and implementing a full-scale, long-term clinical
pharmacy services program.
RECOMMENDATION 3

Virginia has restrictive compassionate release policies,
which contribute to higher spending
One of the most direct ways to reduce spending on health care is to have fewer inmates
in VADOC custody, and compassionate release is one way that an inmate may be released early. Compassionate release can be granted because of a serious and debilitating medical condition, advanced age, or terminal illness. Virginia currently has policies
addressing two types of compassionate release, and judicial discretion allows judges to
reduce sentences based on medical status at sentencing.
Compassionate release policies enable inmates to be considered for conditional release,
but inmates still must meet the same criteria as all other healthy inmates who are eligible for release. Expanding compassionate release would not authorize the release of
any inmate, but would give additional flexibility to the Virginia Parole Board in considering inmates for release in part based on their medical conditions. Those who have
committed particularly serious or violent offenses are unlikely to be released, even with
a serious medical condition, as are those who have behaved poorly while incarcerated.
Seriously ill inmates would not be released unless there is a place in the community for
them to receive appropriate care. If inmates are granted a conditional release, whether
based on illness or time served, they are monitored by probation and parole offices to
ensure they meet the conditions of their release, which for seriously ill inmates would

In Virginia, judges can
reduce an offender’s
sentence at sentencing
because of a serious
medical condition. In
FY17, 162 offenders eligible for a state-responsible sentence received
lesser sentences for medical or mental health reasons, with 68 of these offenders receiving no time
to serve after sentencing.

Chapter 3: Strategies to Reduce Spending

be the same as or similar to those for any other offender. Inmates who violate the
conditions of their release would be re-incarcerated with VADOC.
A small number of very sick inmates account for a large portion of health care spending at VADOC. Even if these inmates were released, many of their medical needs
would still be paid for primarily through Medicaid and Medicare, which would reduce
the financial burden on the general fund.

Virginia’s geriatric conditional release policy
was intended to be a
safety valve for an aging
inmate population following the implementation of truth-in-sentencing in Virginia. In 1995,
Virginia abolished discretionary parole and
codified “truth-in-sentencing” for felonies.
This framework includes
a rule that inmates must
serve at least 85% of
their sentence before
being released.

Spending for inmates with serious medical conditions or near the end
of their lives is substantially higher than for other inmates
In Virginia, some inmates who have serious medical conditions or received a terminal
diagnosis cannot be considered for early release under any of Virginia’s current compassionate release policies. Such inmates do not meet age requirements for geriatric
conditional release (age 60 or 65, depending on years served), and those given a terminal diagnosis have more than three months left to live. Virginia should consider
changes that would reduce costs and bring the policy in line with other states.
VADOC spends much more on inmates near the end of their lives than on inmates
generally (Figure 3-2). VADOC spent $61,000 per inmate in the last twelve months of
life, on average, to provide end-of-life care to 65 inmates who died in FY17. These 65
inmates comprised only 0.2 percent of all inmates, yet their end-of life care (outpatient
and prescription drugs) accounted for 4.7 percent of all VADOC health care spending
on these categories.
FIGURE 3-2
Spending for inmates with terminal illness or complex conditions is far higher
than spending for all other inmates (FY17)
VADOC spending,
average for inmates with

Terminal illness

Complex long-term
conditions

Neither terminal illness
nor complex conditions
(a ll other inmates)

$1,612
' - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - -

10,000

20,000

30,000

40,000

50,000

60,000

$70,000

SOURCE: JLARC analysis of VADOC claims data, FY17.
NOTE: Inmates included in analyses received health care services for at least 3 months while in VADOC custody.
Calculations include all health care spending for in-patient hospital stays, outpatient services, and prescription drugs.
Some inmates with high cost diseases are either not severely impaired, or died suddenly without a terminal diagnosis,
and therefore would not be eligible for less restrictive compassionate release policies.

Chapter 3: Strategies to Reduce Spending

VADOC also spends more on inmates with serious, long-term medical conditions,
which often require high-cost services or medications (Figure 3-2). VADOC spent
more than $20,000, on average, on 810 inmates with serious, long-term medical conditions in FY17. These 810 inmates comprised two percent of all inmates, yet the
offsite services and prescription drugs they required accounted for 19.8 percent of all
VADOC offsite and prescription drug spending in FY17.
CASE STUDY
Inmate has high-cost and serious long-term medical condition, but is not
eligible for release under current VADOC policies
VADOC spent more than $200,000 caring for an inmate with cystic fibrosis in
FY17. This inmate was also among those on whom VADOC spent the most
in FY15 and FY16. Despite his health status and high cost to VADOC, he does
not qualify to be considered for release under any existing compassionate
release policy. He is not terminally ill, so he cannot be considered for release
under the existing medical clemency policy. He is also far too young (47) to
be considered under the only other applicable policy, geriatric conditional
release, which is for inmates older than age 60 or 65.

Collectively, inmates with serious medical conditions or at the end of their lives comprise about three percent of all inmates but about 25 percent of VADOC’s total health
care spending. This is not uncommon; all health systems have similar challenges. It
does, however, present a potential opportunity to reduce VADOC total health care
spending by providing care through other funding streams in the community.
Virginia has more restrictive compassionate release policies than
other states
There are three primary criteria for compassionate release used by other states: terminal illness, serious medical conditions, and advanced age. Many other states have programs similar to Virginia’s for geriatric release, and Virginia’s criteria for this type of
release are in line with those of other states. Virginia’s policy allows inmates who are
at least 65 years old and have served at least 5 years of their sentence, or at least 60
having served 10 years, to be considered for geriatric conditional release. Most other
states have an age requirement of between 55 and 70, some in conjunction with a
requirement for time served that is similar to Virginia’s. The states that do not have a
similar program have other types of medical release programs that may apply to these
offenders. (See Appendix F for additional information on other states’ compassionate
release policies.)
Virginia’s current compassionate release policies allow terminally ill inmates to be considered for release, but the state’s requirement that they have only three months to live

Some inmates would
not be able to be released even if they met
the medical criteria for
compassionate release.
Some offenders’ criminal
history or behavior while
incarcerated would render them ineligible, and
for others, particularly
those convicted of sex
offenses, VADOC would
not be able to find a
nursing or assisted living
facility willing to accept
them. JLARC factored in
these challenges in estimating the number of
inmates potentially able
to be released.

Chapter 3: Strategies to Reduce Spending

is more restrictive than most other states. Moreover, none of Virginia’s current compassionate release policies apply to inmates who are not terminally ill but have serious
medical conditions.
Virginia has the nation’s second-most restrictive compassionate release policy for
inmates with terminal illnesses

Virginia’s medical clemency policy allows an inmate to be considered for release when
he or she has been given a prognosis of three months or less to live. Only Kansas has
a more stringent time requirement than Virginia, at 30 days to live. Twenty-three states
and Washington, D.C., have statutory requirements that a physician provide an estimate of the time an inmate has left to live, typically between 6 and 18 months, in order
for the inmate to be considered for medical release. The remaining states do not specify a time frame, and instead allow flexibility in implementing the policy (Figure 3-4).
As a result of Virginia’s policy, only 15 inmates were released under medical clemency
in the past five years.
Average spending on health care increases as inmates near the end of life. VADOC
spends considerably more during the last six months of an inmate’s life. In FY17,
VADOC spent nearly $32,000 per inmate during his or her last three months of life,
after spending nearly $25,0000 in the preceding three months (Figure 3-3). VADOC
spent nearly $57,000 per inmate in the final six months of an inmate’s life, on average.
The process for inmates to be granted medical clemency takes time to complete, and
some inmates may not live long enough to be granted release even if they are eligible.
Once they receive a prognosis of three months or less to live, there is an investigation
to determine if an inmate may safely be released to the community. This is time-intensive and can take anywhere from four weeks to 10 months. Part of the time involved
includes notifying victims, as required by statute, that an inmate may be released. Cases
that involve terminal illness are expedited so that they are completed more quickly
than discretionary parole cases, but some inmates die before the review is finished.
A policy that allows terminally ill inmates with a prognosis of 12 months or less to live
to be considered for release would help VADOC control inmate health care costs. This
change would also bring Virginia’s policy in line with other states’ policies and would
account for the sometimes lengthy process of reviewing an inmate for release and
finding an appropriate community placement. This policy would be most effectively
implemented if VADOC proactively identified inmates who may be eligible for this
type of release.

Chapter 3: Strategies to Reduce Spending

FIGURE 3-3
Spending on health care increases at the end of an inmate’s life

$31,7S4

$4,413

15 to 18

12 to 15

9 to 12

months

6to 9
months

3 to 6

months

Last 3
months

SOURCE: JLARC analysis of data from VADOC inmate and health care spending data, FY17.
NOTE: Inmates included in analyses received health care services in at least 3 months while in VADOC custody. Calculations include all health care spending for in-patient hospital stays, outpatient services, and prescription drugs.

Unlike other states, Virginia does not allow for compassionate release of inmates
with serious, but not terminal, illnesses

Virginia is the only state (other than Iowa, which does not specify in its policy) that
does not have a policy under which inmates who have a serious medical condition or
are permanently incapacitated may be considered for release (Figure 3-4). Some states
require clemency from the governor for these types of releases, while others have processes similar to Virginia’s parole consideration. In addition, a few states use the cost
and complexity of an inmate’s health care needs as criteria for medical release, in order
to save money on inmate health care and reduce the complexity of health care needs
among the inmate population.

Chapter 3: Strategies to Reduce Spending

FIGURE 3-4
Virginia has more restrictive release policies than nearly all other states

Release policy considers months to live when diagnosed with terminal illness?
■ 53

■

■ 6 to 12 +

■ No ti.n:1e frame
specified

~overed t.hrough .serious
illness or 1ncapac1tation

SERIOUS ILLNESS OR PERMANENT INCAPACITATION ■

••
■■■■11■■11■■■
11■■110■11■■■
11■11■■■ ■ ·■
•••••••
•••••
■

■

Release policy considers serious illness or permanent incapacitation?

■

Yes

■

SOURCE: JLARC analysis of compassionate release statutes and policies in other states, and of Everywhere
and Nowhere: Compassionate Release in the States, Families Against Mandatory Minimums, June 2018.
NOTE: Terminal illness: The District of Columbia also has a policy to allow the release of inmates diagnosed
with a terminal illness. Serious illness or incapacitation: The District of Columbia also has a policy to allow
the release of inmates with serious illness or permanent incapacitation. No information is available about
Iowa’s release policies.

Chapter 3: Strategies to Reduce Spending

Virginia could reduce spending by aligning its compassionate release
policies with other states, but extent of savings would vary
To reduce spending, Virginia could change its compassionate release policies to be
more consistent with policies in other states. Spending reductions would be attributable to lower health care spending and the other costs of housing an inmate (security
and food). The amount of savings likely to accrue would depend on the number of
inmates released, but each release would represent a savings of about $42,000 annually.
Expanding Virginia’s compassionate release policies would only serve to increase the
number of inmates that could be considered for release. Compassionate release does not
authorize the release of any particular inmate, instead making more inmates eligible to
be considered for conditional release based on their medical status. Under these policies, seriously or terminally ill inmates would still have to meet requirements related to
public safety. They would also not be released unless there was appropriate care and a
placement for them in the community. An inmate could be re-incarcerated if he or she
violates the terms of release, just as any inmate released under the current probation
and parole system. In addition, an inmate released under these policies could be sent
back to prison if his or her medical condition resolves.
Expanding compassionate release policies could reduce VADOC spending

States vary widely in how they define serious illness, and the definition Virginia chooses
would have a significant effect on any cost savings. For example,
 South Dakota has a fairly broadly defined compassionate release policy,
which includes inmates who are “seriously ill and not likely to recover” or
require “extensive medical care or significant chronic medical care.” About
800 Virginia inmates could meet this broad definition. About 400 of these
inmates might meet VADOC’s existing public safety and placement criteria.
Releasing these 400 inmates could save about $16.9 million.
 North Carolina has a narrowly defined compassionate release policy, including only inmates who are “permanently and totally disabled.” About 70 Virginia inmates could meet this definition, and 35 of those might meet existing public safety and placement criteria. Releasing these 35 inmates could
save about $1.5 million. (See Appendix B for more information on how
these savings were estimated.)
The number of inmates eligible for this type of release each year would vary. Once
currently eligible inmates are released after being deemed not a public safety threat by
the Virginia Parole Board, savings each year thereafter would likely be less. With an
inmate population of about 30,000, additional inmates will continue to become seriously ill and qualify to be considered for release under a serious illness policy. This
number is not possible to predict, but would result in some savings to VADOC’s
budget each subsequent year.

South Dakota excludes
inmates convicted of
capital offenses from
compassionate release.
North Carolina excludes
inmates convicted of
capital offenses, murder,
offenses related to terrorism, and sex offenses
that require registration
in the sex offender registry from compassionate
release.

Chapter 3: Strategies to Reduce Spending

FIGURE 3-5
Estimated savings to VADOC from compassionate release would depend on
the criteria used
Savings
$20M

More restrictive
policy (e .g. NC)

Less restrictive
policy (e.g. SD)

16
12

8
4
0
0

100

150

200

250

300

350

400

450

Inmat es released

Louisiana recently instituted a medical furlough
program for seriously ill
inmates with permanent
disabilities. Since instituting the policy in early
2018, five inmates have
been released after being deemed not a public
safety threat. If any of
these inmates were to
recover or violate any
other terms of release,
they would likely be reincarcerated. Three inmates’ requests for medical furlough have been
denied this year.

SOURCE: JLARC analysis of prescription drug and off-site outpatient care claims data, FY17.
NOTE: Cost estimates account for the cost of outpatient hospital care, physician services, prescription drugs, housing,
and security. Inpatient hospital costs were not included because most of them will be paid for through the Medicaid
program following Medicaid expansion. To estimate the number of inmates with serious, long-term illnesses, JLARC
identified inmates for which the cost of their offsite and prescription drug services were in the top 10 percent of all
inmates for three consecutive years.

Savings could also result from amending the state’s terminal illness policy. Most other
states define terminal illness as having a prognosis of either six or 12 months, compared to three months in Virginia. Extending Virginia’s terminal diagnosis time frame
to 12 months could result in as many as 30 inmates per year being released, with an
overall savings to VADOC of around $700,000.
There are other financial benefits to expanding Virginia’s compassionate release policies. VADOC has submitted an $80 million capital budget request for approximately
270 infirmary beds to meet the current demand for beds for seriously ill inmates and
those returning from inpatient hospital stays. While expanding Virginia’s compassionate release policies would not totally negate the need for some capital investment, doing so could considerably reduce the number of beds needed and perhaps address the
problem in the near-term. More open infirmary beds could also allow inmates to be
released from inpatient hospitalization sooner, reducing inpatient hospitalization costs.
A less restrictive compassionate release policy would also have non-financial benefits.
Inmates with serious medical conditions usually require more staff attention and time
than other inmates. Releasing some of these inmates would free up VADOC staff
resources to provide care to other inmates, which could improve the efficiency of care
generally or reduce wait times, and reduce employee burnout.

Chapter 3: Strategies to Reduce Spending

Health care costs would be shifted to other public programs, paid for largely
with non-general funds

The costs of care for inmates released under these policies would likely shift to other
public payers like Medicaid and Medicare, and some of these costs would require the
use of general funds. Inmates who are permanently disabled or need nursing home or
assisted living facility services would likely be eligible under existing Medicaid categories, meaning the state would pay for 50 percent of their Medicaid services from general funds. Each Medicaid-eligible inmate with a 50/50 split would result in savings of
about 20 percent (see sidebar). Inmates who are eligible under Medicaid expansion
would have no general fund impact for their Medicaid services.
The savings would also be somewhat offset by the need for a small increase in VADOC
administrative capacity, which could include one or more additional administrative
staff at VADOC central office to supplement existing staff who focus on conditional
release and community re-entry. There could also be additional, but still moderate,
costs if VADOC chose to have a nurse and other staff travel to facilities to conduct
assessments, which may include assessments of whether an inmate is able to perform
activities of daily living. This additional administrative capacity would cost between
$70,000 and $200,000.
Virginia’s compassionate release policies could be broadened through both executive
and legislative action. The governor can extend the prognosis requirement for terminal
illness for medical clemency. The Virginia Constitution gives the governor the power
to grant pardons, including those for medical clemency. The governor establishes the
criteria that inmates must meet to be considered for medical clemency and makes the
final decision if an inmate is granted medical clemency for terminal illness. The Code
of Virginia, though, would need to be amended to establish a new type of conditional
release for seriously ill inmates to be considered for release. Because the General Assembly abolished discretionary parole in 1995 through statutory changes, it was necessary to establish geriatric conditional release by statute. The same would be true of
conditional release for serious illness. In developing such legislation, the General Assembly could consider approaches used by other states to define serious illness and to
evaluate inmates for release.
RECOMMENDATION 4

The governor should extend the life expectancy requirement for terminally ill inmates
to be considered for medical clemency to 12 months.
OPTION 1

The General Assembly could amend Title 53.1 of the Code of Virginia to allow inmates to petition the Virginia Parole Board for conditional release based on serious
illness.

Virginia pays half the
cost of Medicaid services for most recipients.
The estimated savings
from releasing an inmate
under compassionate release policies comes
from health care spending (40%) and other operational costs (60%). If
the general fund pays
for half of the 40% attributable to health care
services, savings would
be about 20% less.

Some states put into
code that an inmate
cannot be a threat to
public safety to be released under compassionate release policies.
Many states, like Texas,
Wisconsin, and Mississippi, exclude inmates
from compassionate release based on their offenses, especially violent
and/or sex offenses.

Chapter 3: Strategies to Reduce Spending

Staffing and Risk Management

SUMMARY All state correctional systems face the financial risk of being found by a court to
be providing inadequate care. Such court findings often result in substantial increases in
spending. Staffing challenges, especially at facilities currently managed through contracts,
make the state vulnerable to these legal and financial risks. Stable, effective health administrative leadership is essential to managing a facility’s health care operations. However, VADOC facilities operated by contractors struggle to maintain a stable group of health administrators; nine of the 12 contract facilities lost their health administrative leadership staff
during FY17 alone. Similarly, key front-line staff, especially registered nurses and licensed
practical nurses, left contract facilities at twice the rate they left non-contract facilities. VADOC’s use of paper-based medical records also hinders its ability to readily assess the overall
quality of care provided by the system, and would impede Virginia’s ability to defend a lawsuit
brought against VADOC’s health system. VADOC does have a monitoring program which
helps manage risk, and has made some recent improvements. The monitoring program,
though, does not comprehensively identify, monitor, and address the full range of risks to
adequate care at facilities (especially at the facilities VADOC manages itself). Issues identified
through monitoring are also not fully tracked to resolution, which makes the state legally
vulnerable. Many of these challenges could be addressed through a more comprehensive
partnership with one of the state’s university hospital systems.

Even with implementing the cost control approaches recommended in Chapter 3,
VADOC—and all prison systems—face the ongoing risk of lawsuits that could substantially increase spending on inmate health care. VADOC, like all correctional systems, is constitutionally required to provide reasonably adequate health care to all inmates. VADOC faces significant financial and legal risks if it fails to effectively deliver
this level of care. Most lawsuits within the last decade against state correctional systems have alleged either lack of access to care for an inmate with a serious medical
need, or lack of sufficient follow-up after treatment or referral. Class-action suits have
arisen when lack of access or follow-up appears to represent systemic problems rather
than isolated incidents.
If a court finds a system to not be providing adequate care, the system (or a just single
facility within a system) can be subject to judicial direction regarding the level and type
of health care provided. When this happens, state spending almost always rises substantially. California provides the most compelling example of these financial consequences. Its entire correctional system has been under a court order to provide improved care, contributing to per-inmate health care costs that are more than three
times what Virginia spends per inmate.

Inmate health care is important for reasons other
than controlling spending. Most inmates eventually return to the community, placing additional
burdens on public health
resources if they are not
provided adequate medical care while incarcerated. Research also indicates that improved
health care can reduce
the likelihood of recidivism.

Chapter 4: Staffing and Risk Management

Recently in Virginia, class action litigation at Fluvanna Correctional Center for Women
has resulted in a substantial amount of state resources required to defend VADOC,
and increasing costs to provide court-mandated care at Fluvanna. The state was also
ordered to pay $1.5 million to date in fees to the plaintiff ’s attorneys.
VADOC can best mitigate the risk of losing a lawsuit by working to ensure the health
care it provides meets the legal standard of care. Having a stable workforce of qualified
health care professionals who adhere to health care standards and procedures is among
the most effective ways to ensure adequate care.

VADOC and its contractors struggle to maintain a
stable health care workforce
Recruiting and retaining qualified medical staff is critical to VADOC’s ability to efficiently deliver health care. Each VADOC facility is led by a person who is the health
authority, or lead nurse administrator, and some larger facilities also have a director of
nursing. The individuals in these health administrator positions are supported by frontline staff who are primarily registered nurses (RN) and licensed practical nurses (LPN).
These staff have a wide range of responsibilities, including treating inmates who get
sick and those with chronic conditions that require ongoing monitoring and treatment.
Medical staff also distribute prescription medications and schedule appointments for
offsite care.
Stability among staff is key, as the correctional setting is different from other clinical
settings. Staff must have a thorough understanding of policies and procedures meant
to ensure the safety of both staff and inmates, and coordinate with security staff on a
daily basis to ensure inmates make their appointments on time, receive their medication, and are transported for offsite care.
Turnover or instability in either health administrators or front-line medical staff makes
an already difficult job even more difficult. Staffing VADOC facilities is inherently
difficult due to a combination of several factors, including a general shortage of nurses
and primary care physicians, the remote locations of many facilities, and a more difficult patient population. There is a national shortage of licensed nurses to meet the
demand from hospitals, private practices, and public providers, and Virginia is experiencing this same challenge. Some of VADOC’s most difficult-to-staff facilities are
those in somewhat rural locations that are still close enough to metropolitan areas that
staff are willing and able to drive to for a job in a hospital system or private practice.
Facilities in the most rural locations are actually better able to recruit and retain staff,
because there are very few other competitor employers. Additionally, the secure nature
of correctional facilities can be less attractive to staff, who must undergo a rigorous
security check each time they enter or exit the facility.
State employees tend to earn lower salaries, but more generous benefits, than they
could earn with other employers. The lower salaries can exacerbate the challenge of
recruiting and retaining a stable workforce in a competitive labor market, and VADOC

Chapter 4: Staffing and Risk Management

experiences this challenge with some of its most critical positions. Average state salaries
for health administrators and qualified mental health professionals were among the least
competitive in the state in 2017 (Total Compensation for State Employees, JLARC 2017).
VADOC’s primary strategy to recruit and retain health administrators and front-line
staff at its most challenging facilities is to contract with a private vendor. This decision
is based on the assumption that contractors can more effectively recruit and retain
staff because they have more flexibility to compensate staff without the restrictions
of the state’s compensation policies.
Contractors have been unable to provide stability in health
administrator positions or front-line staff
Contract facilities have experienced substantial instability in health administrator and
front-line staff positions. All 12 contract facilities experienced turnover in either their
health authority or director of nursing at least once over the last three years. This rate
of turnover is much higher than at non-contract facilities; only three of 15 non-contract facilities (20 percent) experienced turnover in these critical leadership roles between FY15 and FY17. Contract facilities have been similarly unable to provide longterm staffing stability in front-line staff, especially RNs and LPNs. Turnover rates of
staff RNs and LPNs at contract facilities were more than double those of non-contract facilities (Figure 4-1).
FIGURE 4-1
Staff turnover is substantially higher at contract facilities than at VADOCmanaged facilities (FY15-FY17)
Staff turnover
by category

Contract facilities
Health
administrators

RNs

LPNs

SOURCE: JLARC analysis of DHRM personnel data and contractor personnel data, FY15-FY17.
NOTE: Turnover of full-time, salaried staff; excludes temporary staff.

Chapter 4: Staffing and Risk Management

It should be expected that contract facilities would have somewhat higher turnover,
given that VADOC specifically chooses to contract out difficult-to-staff facilities.
However, the substantially higher levels of turnover demonstrate that contractors are
not particularly effective at maintaining stability, especially in the critically important
health administrator role. Contractors themselves identified staffing as the single greatest challenge they face in providing health care. The general labor shortage of nurses
and primary care physicians, coupled with the challenging correctional work environment, is currently too difficult for the contractors to overcome even with greater flexibility in their compensation structure.
Instability in medical staffing hinders operational effectiveness and
increases risk
Staffing instability in key health administrator positions and front-line staff hinders
effective and efficient health care operations. With frequent turnover, health administrators do not develop expertise in correctional health care, limiting their ability to
provide guidance to front-line staff and continuously improve health care operations.
Frequent turnover also makes it very difficult to develop effective working relationships with security staff.
VADOC headquarters staff indicate that continuity in key health administrator (health
authority and director of nursing) positions is critical to running an efficient medical
unit that minimizes VADOC’s legal risk. Health administrators at each facility “set the
tone” for the entire health care operation. Strong health administrators understand the
nuances of delivering care in a correctional setting and can work with new or temporary front-line staff to train them on how to efficiently deliver care. Health administrators also need to develop good working relationships with security staff and the
warden, because all aspects of a facility’s operations need to be well coordinated for
health care to be provided in a timely manner.
Front-line staffing instability can also lead to many problems, particularly if it occurs
at a facility without an experienced health administrator. Some of these problems result in inmate grievances that can escalate to lawsuits. Inmates at contract facilities with
exceptionally high staff turnover filed grievances at nearly twice the rate of inmates at
other contract facilities. Four contract facilities with exceptionally high staff turnover
struggled to meet contract requirements related to delivery of care, resulting in more
than $160,000 in fines (73 percent of all care delivery fines) in FY17.
During interviews and site visits, VADOC staff emphasized the importance of the
role of health administrators for operational effectiveness. Some facilities have longtenured leadership who are effective at working within the correctional environment.
Others, however, are either new to correctional health care, or have ample clinical experience but not enough administrative experience. During JLARC site visits, several
instances of inefficiency or ineffectiveness were observed that likely would have been
identified and remedied by experienced correctional health administrators. For example,

Chapter 4: Staffing and Risk Management

 Front-line medical staff at one facility indicated they were extremely short
on exam room space to treat inmates, hampering their ability to keep up
with sick call and chronic care clinics. However, in the same space, there
was a largely unused dental suite with enough space for four exam chairs,
even though the facility only employed one part-time dentist.
 At one facility, staff restrict access to the medical unit when treating inmates from restricted housing, deferring non-emergency medical care for
most other inmates. In contrast, other facilities send providers to the restricted housing unit, avoiding deferral of non-emergency care for other inmates.
To facilitate operational effectiveness at all facilities, especially those led by new or less
experienced health administrators, VADOC should implement a health administrator
peer review program. Health administrators or other senior front-line staff who have
demonstrated the ability to creatively address operational issues and consistently deliver health care efficiently could visit sites that are having persistent challenges to help
them identify inefficiencies and share potential solutions. This could help improve the
development of less experienced health administrators. VADOC has a similar program
for its security staff. Teams of security staff periodically visit other facilities to review
their security operations, identify challenges, and share ideas for improvement.
RECOMMENDATION 5

The Virginia Department of Corrections (VADOC) should develop and implement a
health administrator peer review program in which experienced leadership or frontline staff review the operations at VADOC facilities—other than the one at which
they work—to identify inefficiencies and share potential solutions.
Inability to provide stable staffing is due in part to misalignment of
contractual incentives
VADOC’s contract requirements place a heavy emphasis on fines for missing shifts,
which incentivizes employing staff to cover open positions rather than on hiring fulltime staff with the knowledge and experience needed to provide efficient and adequate
care. For example, contractors are fined $98 for each hour of missed front-line staff
nursing time (three times the average hourly rate for a nurse in Virginia). This places
an extremely strong emphasis on getting a nurse into the facility to fill a shift, rather
than on hiring and developing stable staffing, which is shown to improve the delivery
of care. Fines were levied for missed hours in every month of 2017, and 10 of the 14
contract facilities paid at least some level of fine.
To provide care and meet the terms of the contract, the contractors often rely on
temporary nursing staff, or “agency nurses,” to fill vacant positions when they experience turnover, just as VADOC does. Interviews indicate that these temporary staff
may lack the necessary knowledge of relevant policies and procedures, such as the

Chapter 4: Staffing and Risk Management

process for logging and scheduling sick call requests, which can lead to a backlog.
Temporary staff also may not receive appropriate onboarding to learn and implement
agency-specific policies. This results in inefficiencies and challenges providing quality
care. Filling vacancies also requires full-time staff to work substantial amounts of overtime, leading to burnout and eventually more turnover.
Contract incentives and structure could be modified to improve staffing stability

VADOC should take steps to improve the stability of full-time staff at contract facilities. Making incremental changes to contract incentives, which can be evaluated over
time before making more sweeping contract changes, is a prudent approach given the
need to maintain continuity of care at each facility. VADOC recently started the first
of three one-year renewal periods under the current contracts. This provides a unique
opportunity to make these changes because the contract terms can be modified in each
of the next two years. Then VADOC can use the information gained from that experience to develop an RFP for its re-procurement of health care services that will better
incentivize staffing stability.
VADOC should first modify its contract requirements to better ensure and incentivize
stable, full-time, health administration and front-line staffing. These changes could include
 retention incentives based on staff tenure, such as bonuses based on the
number of staff with at least one year of service;
 granting VADOC an advisory role in the recruiting and selection process
for health administrators, through which VADOC could help ensure those
chosen have some background or characteristics that makes it more likely
they will stay for a considerable period of time (such as previous employment in a correctional setting); and
 guidelines to formalize and foster relationships between contract health administrators and VADOC wardens and assistant wardens, such as requirements for attending inter-departmental meetings.
It is possible these incremental changes to the contract structure will not materially
improve the stability in staffing at contract facilities. The general health care staffing
shortage and difficulties of working in correctional facilities may be too great to be
overcome through making contractual changes alone. After changing the contracts,
VADOC should assess whether the changes have led to more stable staffing at contract
facilities. This assessment should be based on objective measures such as turnover
rates, the number of compliance findings, and the number of inmate grievances. VADOC should look for measurable improvement from prior years, ideally to the point
where they are operating similarly to non-contract facilities with more stable staffing
and efficient health care delivery. If stability has improved, the changes made to the
contracts to incentive staffing stability should be continued.

Chapter 4: Staffing and Risk Management

This incremental approach to improve stability should be implemented during the renewal periods that are part of the current VADOC contracts. These contracts can be
renegotiated at the end of each contract year, for the next two years.
RECOMMENDATION 6

The Virginia Department of Corrections should evaluate whether the contract modifications have resulted in more stable staffing and efficient care delivery by measuring
turnover rates, compliance findings, and inmate grievances. The results of the evaluation, including a determination of whether staffing stability at contract facilities has
improved from prior years and is similar to non-contract facilities, should be submitted
to the Senate Finance and House Appropriations Committees by the end of 2020.
If stability remains a problem, VADOC should consider a more significant change in
how it contracts for health care. VADOC could make the health administrator position
at some, or even all, contract facilities a classified, state position. This could have several advantages, including giving VADOC direct control over hiring decisions, the ability for VADOC to move health administrators between facilities to meet its needs, and
a more direct reporting relationship between the health administrators and the wardens at each facility. Converting health administrator positions to VADOC employee
positions may necessitate changing the capitated rate model currently used. A contractor may be less willing to still bear the majority of the financial risk for the cost of
care, without having direct control over decisions that affect care. With this change,
the contracts would likely become only staffing contracts for front-line staff, rather
than a comprehensive contract for most health care services.
OPTION 2

Meeting the standard of care is the ultimate objective of having sufficient, stable staffing over the long term. However, current contracts penalize staffing violations more
than meeting the standard of care. No fines are issued during the initial 90 days of
standard of care violations. After that 90-day period, the fine is $2,500 the first month,
and $5,000 each following month. This means that if a problem is corrected within 90
days, but re-emerges as a problem a month later, the 90-day period resets.

Chapter 4: Staffing and Risk Management

In addition, the fine structure for delivery of care applies to a broad range of violations
that vary substantially in their level of seriousness. For example, failure to distribute
medications or treat sick inmates in a timely manner is treated the same as a failure to
properly document care that was delivered.
VADOC should also modify its contracts to more directly emphasize the most critical
aspects of care delivery. VADOC should place a greater priority on certain critical
violations related to standard of care, such as late treatments of sick inmates and
missed medications, by making such violations subject to fines more quickly and at
higher amounts.
RECOMMENDATION 8

VADOC lacks sufficient record-keeping and
monitoring processes needed to manage risk
VADOC bases its health care policies on accepted correctional health care standards
from the American Correctional Association. Though they do not guarantee adequate
care, these standards are looked upon favorably by courts. These standards are one of
two used by state correctional systems. The other set of standards that VADOC does
not use is more specific, but not necessarily more comprehensive, widely accepted, or
generally preferred.
Despite the use of the American Correctional Association standards, VADOC still
faces the risk that the care it provides will be subject to a successful legal action asserting that the care provided is inadequate. An adverse court finding could lead to courtmandated higher levels of spending (as is already occurring at the Fluvanna Correctional Center for Women). To further manage this risk, VADOC needs the ability to
demonstrate that it is providing adequate care and the ability to identify potential adequacy concerns at all facilities. Implementing an electronic medical record and a riskbased monitoring approach would address these needs.
Lack of electronic medical records complicates efforts to demonstrate
adequacy of care, increasing legal and financial risk
VADOC lacks a readily-available and accessible system of recordkeeping that can be
used when needed to answer questions from families, attorneys, or judges about the
level of care being provided to each inmate. Without such a recordkeeping system,
VADOC has difficulty demonstrating, accurately and efficiently, that care is being provided system-wide, at a given facility, or to a single inmate. VADOC’s current paperbased medical record system makes monitoring and risk assessment resource-intensive
and does not allow the agency to make full use of available information. This limits

Chapter 4: Staffing and Risk Management

VADOC’s ability to identify and address problems through monitoring. Inmate health
files contain extensive information, but because of time and resource limitations, only
a small portion of these files can be examined during monitoring.
Reliable and readily-available records are also essential in responding to litigation. Because files are currently on paper at each facility, it is extremely time intensive to review
and provide them to attorneys. Responding to allegations of system-wide problems in
class action lawsuits can require compiling and reviewing thousands of paper medical
records. The Office of the Attorney General indicated that the discovery process as
part of the ongoing litigation at Fluvanna exceeded 100,000 paper documents.
Electronic medical records would facilitate data collection and synthesis across all inmate records. This could help the state demonstrate that allegations of inadequate care
by a single inmate are not a systemic issue with that facility or the entire system.
VADOC has been working for several years to establish a system of electronic medical
records for inmates. In recent years, the General Assembly dedicated nearly $3 million
in funding that VADOC could use for electronic medical records. VADOC worked
with a vendor, in conjunction with VITA, to develop an electronic medical records
system to be integrated with VADOC’s current IT systems. However, the vendor
backed out of negotiations because it was unwilling to meet VITA’s information security requirements. The vendor had proposed using a cloud-based medical records system, but this requires compliance with VITA’s new Enterprise Cloud Oversight Services requirements, which are designed to ensure cloud-based applications comply
with Virginia’s security standards.
The 2018 General Assembly directed a workgroup to evaluate the feasibility of implementing a shared, or at least interoperable, electronic medical records system for
VADOC and other relevant state agencies. Any steps undertaken by VADOC would
need to be done in consultation with this ongoing effort. VADOC recently issued a
request for proposals for an electronic medical records system and is in the process of
reviewing proposals.
Electronic medical records would improve the accessibility and portability of inmate
health records. The state would more easily be able to respond to allegations of inadequate care with readily available information. Staff would be able to quickly access
inmate records at any facility where an inmate might be transferred or transported.
Off-site providers would also be able to receive inmate health information electronically for off-site appointments and hospitalizations, and send information back to VADOC once the inmate is returned to a correctional facility.
There would be up-front administrative challenges to implementing electronic medical
records. Staff would need to be trained on the new system and integrate it into patient
care. Information from paper records for inmates still in VADOC custody may also
need to be entered into the new system, which would involve significant staff time.

Chapter 4: Staffing and Risk Management

VITA should work to help VADOC select a vendor and develop a contract that will
result in a successful, cost-effective system of electronic medical records. VADOC
should work as needed with the Office of Attorney General staff with expertise in
procurement and contract administration. VITA should prioritize its assistance to VADOC to ensure the vendor can accomplish the work and deliver a system consistent
with VITA’s security standards.
RECOMMENDATION 9

The Virginia Information Technologies Agency should collaborate as necessary with
the Virginia Department of Corrections (VADOC) and the Office of the Attorney
General to ensure the selection of a vendor capable of successfully implementing an
electronic medical records system that can meet the specific functional requirements
of the correctional system and be cost-effectively used by all VADOC facilities.
VADOC monitors adequacy of care but does not sufficiently monitor
access and follow-up
VADOC does monitor the care provided to inmates but focuses more on contract facilities than on the facilities managed by VADOC. Contract facilities are more closely
monitored in part because VADOC needs to enforce the terms of the contracts. Contract facilities are monitored each month and a standard set of critical metrics is checked
each time, in addition to other metrics that rotate throughout the year. There are designated VADOC staff who monitor contract facilities; as a result, monitoring is more frequent and addresses a more comprehensive set of topics. VADOC has recently made
improvements to its monitoring, such as standardizing the monitoring form used across
facilities as part of its continuous quality improvement program.
Perhaps because VADOC focuses its monitoring more heavily on contract facilities, it
is finding more problems at contract facilities. VADOC staff say the agency has a
clearer idea of the problems that exist at contract facilities than at non-contract facilities. While there may simply be fewer problems at non-contract facilities because they
tend to have more stable medical staffing, the less rigorous monitoring process does
not provide sufficient assurance that all issues are being identified.
Because VADOC’s medical records system is paper-based, monitoring is time-intensive. Monitoring currently requires extensive review of medical facilities and paper
records. However, implementing electronic medical records would allow VADOC to
better monitor health care delivery and assess risk factors at each facility. Data for all
inmates (rather than a selection of paper files) could be checked electronically (1) for
compliance with policies, and (2) for risk factors such as long waits for inmates who
request care. Some of these tasks could be accomplished by data analysts on a
statewide basis, freeing up monitoring resources for targeted, onsite monitoring. Inmate health outcomes could also be monitored through an electronic records system.

Chapter 4: Staffing and Risk Management

Without risk-based monitoring, VADOC lacks a key mechanism to identify
potential problems that could lead to lawsuits

Correctional health care lawsuits and inmate complaints often involve an inmate alleging he or she did not receive access to care or did not receive prescribed follow-up
treatment. VADOC currently allocates most of its monitoring resources to contract
facilities. Outsourcing health care delivery is an important risk factor that should be
considered, but other facility characteristics and compliance indicators may be better
measures of risk.
VADOC should routinely identify key indicators of potential problems at all facilities
using its existing quality monitoring process and use these indicators as the basis to
govern its monitoring activities. This information could be collected regularly and used
to assess the risk at each facility. Staff could then do in-depth monitoring at those
facilities with the highest risk, similar to what is currently done each month at only
contract facilities. Monitoring would be less frequent but more targeted than what is
currently done for contract facilities.
The key indicators should address issues that can lead to adverse inmate health outcomes, health care complaints, and lawsuits. These indicators would likely include
those related to access to care and follow-up. VADOC should assess critical risk factors
related to these issues on an ongoing basis, such as
 Access – status of nurse and doctor sick call;
 Access – status of chronic care clinic;
 Follow-up – status of recommended specialist care; and
 Follow-up – medication administration.
For the most effective risk-based monitoring approach, this information should be
collected on a regular basis using a standardized format from all facilities. Analysis of
risks at each facility would then be used to determine which facilities need in-depth
monitoring, allowing VADOC to more strategically deploy its limited monitoring resources.
Much of this information to assess risk is not currently available centrally but could
be self-reported by each facility. The status of sick call, chronic care clinics, and specialist care visits are not tracked centrally and would require the medical unit at each
facility to report this information. VADOC does have an electronic record for medication administration, and this could be used to develop indicators of problems with
distributing prescription drugs at each facility.
Tracking of monitoring results to ensure resolution of problems could be
improved, particularly at non-contract facilities

When VADOC’s current monitoring approach does identify problems at facilities, it is
not clear that these problems are resolved in a timely manner because there is no doc-

Sick call is similar to a primary care appointment
or a visit to an urgent
care center for a minor
injury or illness.
Chronic care clinics are
where inmates with
chronic illnesses like diabetes and heart disease
receive their regular care
for those conditions.
Specialist care includes
visits to providers like orthopedists, cardiologists,
and dermatologists, and
usually occurs offsite.
Medication administration refers to inmates receiving their prescribed
medications correctly and
in a timely manner.

Chapter 4: Staffing and Risk Management

Liquidated damages are
fines VADOC imposes on
contractors when certain
contract provisions are
not met.

umentation of follow-up and resolution. VADOC policy directs the quality improvement committee at headquarters to oversee the resolution of problems, but does not
require documentation of the results of corrective actions. At contract facilities, where
monitoring is monthly, there are more assurances that problems are resolved, since
continuous problems result in liquidated damages for the contractors. Nonetheless,
there is no formal documentation of which problems are resolved and which may
require further corrective action. At non-contract facilities, follow-up on problems is
sometimes informal because monitoring only happens quarterly. While unresolved
problems would likely be identified in subsequent monitoring, there is limited documentation of how facilities implement correction action plans.
RECOMMENDATION 10

Pilot project with a public teaching hospital could
improve staffing, risk management, and operations
VADOC could expand its partnership with VCU Health and the UVA health system
by having one of these hospital systems take over the provision of health care for at
least one high-need VADOC facility. The structure of such a partnership could be
similar to VADOC’s current comprehensive health services contracts, but it would
start as a pilot program with a small number of facilities. The university health system
would be responsible for the leadership and staffing for onsite care in the facility and
assume responsibility for the cost of offsite care and prescription drugs.
This type of partnership is most likely to be successful if it is mutually beneficial to
both VADOC and the university health system. Other states have entered into such
partnerships with two main goals: (1) to increase the credibility of the correctional
health care system and (2) to reap long-term cost savings by providing all health care
services through a single system. One strategy to make it beneficial to the university
health system would be to provide a “shared risk” capitated payment for all care, so
that if cost savings are realized, the university is able to share in the financial benefits,
while if expenses are higher than expected, both the state and the university share the
risk. VCU Health recognized that this could be a mutually beneficial partnership when
it submitted a proposal in response to a 2005 RFP to provide all health care at four
VADOC facilities. The proposal was ultimately not selected, because it was not the
lowest cost bid, and because the proposed model would have increased the complexity
of how care was delivered and paid for across facilities. These concerns were valid, but

Chapter 4: Staffing and Risk Management

the potential benefits from such a partnership warrant renewed consideration, given
the positive experiences of other states.
VCU Health is currently experiencing challenges with having sufficient capacity to
treat inmates from VADOC and other correctional populations in Virginia, and any
plan for a pilot partnership should address this issue. VCU Health operates a secure
medical unit for inmates, but 560 inpatient days (10 percent) for VADOC inmates in
FY17 occurred in the general hospital population because the secure unit was at capacity. This places additional burdens on the hospital and VADOC due to the security
needs. VCU Health often has to use a double inpatient room for just one inmate, and
two VADOC correctional officers must be stationed with each inmate that is being
treated in the general hospital population.
Five other states have had extensive partnerships with university health systems, and
the specific financial and administrative responsibilities vary. Texas has the most comprehensive partnership, in which the state department of corrections contracts with
two different university hospitals that are fully responsible for providing and paying
for all inmate health care. Georgia and New Jersey partner with university hospitals to
provide medical staff at correctional facilities, while Michigan partners with a university hospital to provide only medical leadership and administration (Table 4-1). Connecticut’s partnership was recently terminated after the university hospital determined
that it did not benefit sufficiently.
TABLE 4-1
Other states leverage university hospitals in a variety of ways to improve health
care delivery and achieve cost savings
Texas

New Jersey

Georgia

Michigan

provide health services leadership



provide onsite staffing



pay for offsite services and prescription drugs



University hospitals

SOURCE: Interviews with other state correctional staff and other state documents.
NOTE: In New Jersey, the university hospital provides offsite care and prescription drugs, but the state department
of corrections pays for the services.

A pilot program to test a more comprehensive partnership between VADOC and one
of the state’s university health systems could have many potential benefits, including
beginning to address some of VADOC’s challenges noted in this report. Chief among
these are staffing challenges that must be addressed to efficiently delivery health care
in each facility. VADOC would also be able to purchase discounted prescription drugs
through the federal 340B program and develop a way to integrate clinical pharmacy

Chapter 4: Staffing and Risk Management

services as part of onsite care (Chapter 3). If a pilot program were successful and was
then expanded, VADOC could benefit from
 Access to more physicians, particularly those who are willing to treat inmates but do not wish to work for VADOC or a contractor.
 Access to more front-line staff, by exposing students to correctional
health care through their training, making them more likely to choose a career in the field.
VADOC could also benefit from the credibility that would come from being more
closely associated with a highly respected university hospital. This would especially be
the case for the care provided to inmates with very complex or unique health needs—
the type of care a university hospital is known for providing effectively.
If structured properly, a pilot program that was successful and then expanded could
also benefit the participating university hospital. The university would gain an additional clinical setting for its students to observe and learn from other clinicians. The
university health system would have an opportunity to realize financial gains if it could
reduce long-term spending by improving the health of the population and identifying
efficiencies in care delivery to keep inmates out of high-cost, inpatient settings.
Entering into such a comprehensive partnership, even on a pilot basis, would bring
with it substantial administrative complexities. These challenges include
 A substantial amount of staff time and effort required to plan the pilot
partnership. Initial planning would require decisions about which facilities
to include in the pilot, how to initially staff the facilities, how much to pay
for services and how to structure those payments (capitation versus fee-forservice), and how to evaluate the effectiveness of the partnership.
 Neither VCU Health or UVA health system have experience providing onsite care in the correctional setting. This would require substantial orientation and training on the unique aspects of delivering care in this setting.
RECOMMENDATION 11

Appendixes

Appendix A: Study resolution
Rising costs of providing health care for state prison inmates
Authorized by the Commission on September 11, 2017
WHEREAS, the Virginia Department of Corrections (DOC) is granted custody of persons convicted of felonies sentenced to more than one year, and has recently been responsible for about
30,000 offenders; and
WHEREAS, DOC must provide inmates with medical and mental health care and treatment, and
determine how inmates should contribute to the cost of their health care; DOC cannot deny necessary health care services to inmates who cannot afford to pay; and
WHEREAS, inmate health care costs now account for 21 percent ($199 million) of all funds appropriated to operate correctional facilities ($949 million), and the vast majority of appropriations are
state general funds; and
WHEREAS, medical costs per inmate have risen more than 20 percent during the past five years,
outpacing the increase in national health care costs; and
WHEREAS, for about half of inmates, DOC provides health care directly; for the other half, DOC
procures services through contracts; DOC has had some difficulty with the cost and quality of services procured through contracts; and
WHEREAS, one DOC facility, the Fluvanna Correctional Center for Women, is currently under a
federal court order to monitor the quality of inmate health care, and such increased scrutiny may
prompt further increases in inmate health care spending; now, there be it
RESOLVED by the Joint Legislative Audit and Review Commission that staff be directed to review
the rising cost of providing health care for state prisons inmates. In conducting its study, staff shall
(i) compare the cost of providing health care to inmates to the cost of providing health care to other
similar populations in Virginia and other states; (ii) identify the factors contributing to health care
cost increases at facilities managed by DOC and those managed by contractors; (iii) assess whether
DOC efficiently and effectively provides health care to inmates, and procures and administers health
care contracts that leverage purchasing power across facilities; (iv) determine whether DOC sufficiently maintains and adequately uses inmate medical records to make strategic health care decisions;
(v) assess whether DOC adequately partners with community and other resources to provide care;
(vi) determine how well DOC is adapting its facilities and operations to its aging and less health inmate population; (vii) as appropriate, make recommendations; and (viii) as appropriate, research
other issues.
All agencies of the Commonwealth, including the Department of Corrections and all state correctional facilities, Virginia Commonwealth University, the Department of Medical Assistance Services,
and the Virginia Information Technologies Agency shall provide assistance, information, and data to
JLARC for this study, upon request. JLARC staff shall have access to all information in the possession of state agencies pursuant to § 30-59 and § 30-69 of the Code of Virginia including all documents related to disciplinary proceedings or actions of the boards. No provision of the Code of

Appendixes

Virginia shall be interpreted as limiting or restricting the access of JLARC staff to information pursuant to its statutory authority.
Private or for-profit entities that receive state funding to provide health care and other services to
inmates, including through contractual arrangements, are also requested to provide assistance, information, and data to JLARC for this study, upon request. JLARC staff shall, as needed, work with
private entities to develop agreements that sufficiently protect proprietary information during the
course of the study.
JLARC staff shall complete their work and submit a report of its findings and recommendations to
the Commission by December 15, 2018.

Appendixes

Appendix B: Research activities and methods
JLARC staff conducted the following primary research activities as part of its study of the cost of
inmate health care:
 structured interviews with VADOC leadership and employees, other state agencies, state
universities, health care contractors, national experts, and other states;
 site visits to five VADOC facilities and one offsite secure medical facility;
 surveys of VADOC clinical staff and facility health administrators;
 quantitative analysis of demographic, health care encounter, grievance, and staffing data
from VADOC, Department of Human Resource Management, and contractors; and
 review of VADOC documents, documents from other states, and research literature.
Structured interviews
JLARC staff conducted nearly 60 interviews with VADOC leadership and staff at headquarters and
facilities, health care contractors and other stakeholders, national experts on inmate health care, and
eight other states.
Structured interviews with VADOC leadership, staff, and other state agencies

Extensive interviews with VADOC leadership, particularly in Health Services, were conducted as part
of this study. JLARC staff interviewed VADOC leadership outside of health services to hear about
the broad issues surrounding inmate health care, including budgetary considerations and the unique
aspects of delivering health care in a correctional setting. Interviews with health services leadership
and staff at headquarters were conducted throughout the study to understand how health care is
administered and overseen in facilities, the challenges to delivering health care to inmates, utilization
management for offsite care, contract monitoring and administration, and staffing. JLARC staff also
interviewed VADOC information technology leadership to learn about the agency’s efforts to implement electronic medical records.
Other state agencies were also interviewed as part of this project. JLARC staff interviewed staff at
the Department of Medical Assistance Services (DMAS) to understand how billing for VADOC
offsite services compares to Medicaid billing. Staff from the Department of Behavioral Health and
Developmental Services’ (DBHDS) pharmacy operations were interviewed to understand how their
model for purchasing and dispensing prescription drugs compares to VADOC. JLARC staff interviewed leadership at the Compensation Board to learn more about how health care is provided in local
and regional jails, and the Virginia Parole Board to learn about Virginia’s compassionate release policies. JLARC also talked with the Office of the Attorney General about inmate health care lawsuits.
Structured interviews with health care contractors and other stakeholders

JLARC staff conducted interviews with the contractors that work with VADOC to provide health
care and the primary health system that treats inmates, VCU Health. VADOC’s prescription drug
vendor and its health care claims administrator were interviewed to understand available data, pricing
structures under the contracts, and get their perspective on potential strategies to reduce spending.

Appendixes

JLARC staff also interviewed the state’s two comprehensive health services contractors separately for
their perspectives on inmate health care in the VADOC facilities where they operate, and the contracts
they have with VADOC.
JLARC staff also interviewed staff at VCU Medical Center’s secure unit for inmates. These interviews
were designed to get a better idea of how the secure unit is run, who is seen there, and what the
challenges in running the unit are. In addition, VCU Medical Center leadership were interviewed for
their perspectives on VADOC’s partnership with VCU, its history, any ideas for how the partnership
could change, and the advantages and disadvantages of potential changes.
Structured interviews with other states and national experts

Because of their extensive work on inmate health care, JLARC staff interviewed staff at the Pew
Center on the States for background about inmate health care trends throughout the country, innovative ideas other states have implemented to improve care or save money, and what may affect inmate
health care costs. Staff from the Minnesota Multistate Contracting Alliance for Pharmacy (MMCAP)
were interviewed to determine if VADOC may be able to find a more cost-effective way to obtain
prescription drugs.
Eight other states were interviewed for this project: Michigan, Connecticut, Oregon, Maine, New
Jersey, North Carolina, Texas, and Louisiana. These states were chosen either because they had significantly higher or lower per-inmate health care spending than Virginia, or because they utilize partnerships with universities to deliver inmate health care. The goal of these interviews was to understand
how a state’s health care delivery model might affect spending. JLARC staff also sought information
on the advantages and disadvantages of partnering with public universities to deliver care and the
different characteristics of these kinds of partnerships.
Site visits
JLARC staff visited a total of five VADOC facilities for this study.
 Deerfield Correctional Center
 Fluvanna Correctional Center for Women
 Sussex I State Prison
 Sussex II State Prison
 Virginia Correctional Center for Women
In addition, JLARC staff also visited the secure care unit at VCU Medical Center, where VADOC
sends many of its inmates in need of care outside their facility.
The facilities were chosen to ensure a mix of men’s and women’s facilities, facilities with different
security levels, contract and non-contract facilities, and facilities where medical units had different
capabilities.
JLARC staff toured the medical units and other parts of the facilities during each site visit. Interviews
with facility wardens and assistant wardens, health services unit staff, and contract monitors were
conducted to understand how health services units operate in facilities, the day-to-day challenges
health services units face, and possible ideas to make improvements where needed. JLARC staff also

Appendixes

interviewed contract monitors at three facilities to learn more about the process of contract monitoring and quality oversight.
Quantitative analyses
JLARC staff undertook a number of quantitative analyses as part of this study to understand VADOC’s
inmate health care spending and staffing. Table B-1 outlines the data sources used in the analyses.
TABLE B-1
Data used in JLARC quantitative analyses

Data

Health care
spending

Facility
spending &
utilization

Health care
pricing

Staffing

Grievances

Compassionate
release

VADOC data
Expenditures



Offsite and pharmacy claims
Inmate demographic and
health status
Inmate travel records












Facility staffing records
Liquidated damages
to contractor



Inmate grievances
Comprehensive health services contractor data



Offsite and pharmacy claims



Facility staffing records
Other payers’ data
DMAS expenditures
DMAS demographic
and health status







DMAS offsite service prices
MMCAP pharmacy prices
340B pharmacy prices



Appendixes

Analysis of health care spending and utilization across facilities

To understand inmate health care spending trends, JLARC staff compiled FY12-FY17 expenditure
data from the Auditor of Public Accounts (APA) Data Point tool as well as Cardinal extracts provided
by VADOC. JLARC staff separated spending by facility, and categorized spending into five major
categories: contract payments, onsite care, offsite care, prescription drugs, and other. JLARC staff
assessed how spending changed over time overall, by spending categories, and by facility. Appendix C
provides more information on the results of this analysis.
JLARC staff also conducted analysis of offsite and prescription drug spending across facilities for
FY15-FY17. Contract and non-contract facilities can differ significantly in inmate population needs,
so JLARC staff created four regression models to predict prescription drug utilization, prescription
drug spending, offsite service utilization, and offsite service spending. The regression modeling allowed staff to predict utilization and spending based on inmate demographics, health status, and diagnoses, and compare predicted values to actual utilization and spending. (See Appendix E for a more
in-depth discussion of the regression analysis.)
The data included in the regression models came from several different data sources. Data from
VADOC’s CORIS system was used to identify inmate age, gender, race, mental health code, work
code, disease flags, and facility. Claims data from Anthem, Diamond, and VCU Health were used to
calculate service utilization and spending. Anthem claims data were used to sum monthly spending on
offsite services for each inmate. Diamond, VADOC’s prescription drug contractor, provided claims
for most prescription drugs, and claims data on drugs purchased from VCU Health were provided by
VADOC.
Analysis of per-person health care costs for VADOC and DMAS

To calculate a useful comparison between spending on the state’s inmate population and spending on
Medicaid, JLARC staff used a subset of Medicaid recipients with similar characteristics to the state’s
inmate population. JLARC obtained Medicaid expenditure and enrollment data for FY17 from
DMAS. Spending on this subset was compared to VADOC’s spending on inmate health care and
adjusted to account for differences in age, gender, and health status. This was a multi-step process
designed to control for a person’s disability status and whether they have a severe mental illness. The
proportion of people with either a disability or severe mental illness is significantly different between
Medicaid recipients and VADOC inmates, and these individuals incur much higher health care costs.
To determine the percentage of VADOC inmates with a disability, JLARC staff used VADOC work
codes, which vary from 1 to 4, where 1 indicates no restrictions on an inmate’s work ability and a 4
indicates that an inmate cannot work in any capacity. Those inmates with a work code of 4 were
considered disabled for the purposes of this comparison. To determine the percentage of VADOC
inmates with a severe mental illness, JLARC staff relied on VADOC’s mental health codes, which
range from 0 to 4. A mental health code of 2 or higher indicates a serious mental health diagnosis.
The process also entailed adjusting for the differing compositions between the two populations in
terms of gender, as females comprise a much smaller percentage of the total inmate population than
of the Medicaid population. The process also controlled for differences in age, as the vast majority of
inmates are between the ages of 18 and 49, while more than 50 percent of the Medicaid population

Appendixes

is under the age of 17 or over the age of 65. All Medicaid spending on children was excluded from
this comparison.
In order to account for differences in demographics and health status between the two groups, JLARC
staff first used a subset of the Medicaid population that included (1) low-income adults, (2) disabled,
and (3) those with a severe mental illness. Staff used this data to calculate the per-person cost across
all age, gender, and health status categories (example: females age 18-24 with a severe mental illness).
JLARC staff calculated a weighted average cost for Medicaid recipients, using the composition of the
population in VADOC custody.
Analysis of offsite care rates

JLARC staff compared the prices that VADOC pays for offsite care to what Medicare and Medicaid
would pay for the same services. This was done for physician services, outpatient hospital services,
and inpatient hospital services. The methodology for each comparison was different based on the
available data.
Physician reimbursement rates
JLARC staff estimated the savings from using Medicaid rates for physician services by comparing the
rates paid by VADOC in FY17 to Medicaid fee-for-service rates. JLARC staff were able to match 68
percent of all non-inpatient physician claims from FY17 (38,620 of 56,643) using a unique procedure
code for each service, the Current Procedural Terminology (CPT) used by health care providers to
bill for services. These claims accounted for 67 percent of total spending on these services. Major
categories that were not able to be matched were anesthesia claims and the mileage component of
ambulance services, because JLARC staff did not have necessary data on the amount of anesthesia
used or the mileage driven to accurately price each claim using Medicaid rates. The Medicaid fee-forservice rate was compared to the actual amount VADOC paid for each service. JLARC staff compared
the sum of the total VADOC payments and the total Medicaid fee-for-service rates for these claims
to arrive at a percentage difference using Medicaid rates, and then applied that percentage to all noninpatient physician claims in FY17. The savings using Medicare rates were estimated using a DMAS
analysis of the difference between Medicaid and Medicare physician rates. JLARC staff applied this
estimate to the estimated savings using Medicaid rates to arrive at an estimated percentage of VADOC’s costs if Medicare rates were used.
Hospital reimbursement rates
JLARC staff estimated savings of using Medicaid and Medicare rates for hospital services by comparing the percent of hospital charges reimbursed by VADOC, Medicare, and Medicaid for both inpatient
and outpatient services. JLARC staff included emergency room care in outpatient services for the
purposes of this analysis. JLARC staff used eight hospitals in the analysis: VCU and UVA hospitals,
which are both considered “Tier 1” hospitals for the purposes of Medicaid reimbursement, and six
“Tier 2” hospitals. VCU and UVA hospitals account for more than 80 percent of VADOC hospital
spending, and the six Tier 2 hospitals comprise a significant amount of the remaining spending. Medicaid hospital data came from the most recently completed Medicaid hospital cost settlement reports,
which were for 2016. This analysis was done using the following steps (Table B-2):

Appendixes

 Calculated the Medicaid cost-to-charge ratio for inpatient and outpatient services
 Calculated the Medicaid reimbursement-to-charge ratio for inpatient services, and estimated
the reimbursement-to-charge ratio for outpatient services using statewide estimates from
DMAS for Tier 1 hospitals (75%) and Tier 2 hospitals (100% following rate increases in the
2018 Appropriations Act)
 Estimated the reimbursement-to-charge ratio if Medicaid rates were used for VADOC
claims by calculating the weighted average reimbursement-to-charge ratio for inpatient and
outpatient services for both Tier 1 and Tier 2 hospitals, based on the percent of VADOC
hospital spending at each hospital in FY17
 Estimated the Medicare reimbursement-to-charge ratio for inpatient and outpatient services
at Tier 1 and Tier 2 hospitals by using the assumption that Medicare hospital reimbursements
are on average 88 percent of costs
 Applied the estimated Medicare and Medicaid reimbursement-to-charge ratios to the actual
VADOC hospital charges for Tier 1 and Tier 2 hospitals in FY17
TABLE B-2
Example of estimated VADOC savings using Medicaid and Medicare rates for inpatient
services at Tier 1 hospitals
Explanation of step

Calculation

Calculate the total charges for services to VADOC inmates

Calculate the percent of total charges paid by VADOC

Total VADOC inpatient hospital payments to Tier 1 hospitals (A*B)

Calculate the Medicaid reimbursement-to-charge ratio

Calculate the weighted average reimbursement-to-charge ratio
using VADOC payment to each hospital

Estimate total payments using Medicaid rates (A*E)

Calculate the average Medicaid reimbursement-to-cost ratio

69%

Estimate total payments using Medicare rates (.88/F*G)

$12.9

Data source

$46.3M

VADOC claims data

49%

VADOC claims data

$22.7M

VADOC claims data

Varies by hospital
22%
$10.2

Medicaid hospital cost
reports
JLARC calculation
JLARC calculation
Medicaid hospital cost
reports
JLARC calculation

SOURCE: JLARC analysis of VADOC hospital claims data and Medicaid hospital cost reports, FY17.
NOTE: This analysis was repeated for both inpatient and outpatient services at both Tier 1 and Tier 2 hospitals.

Analysis of 340B prescription drug pricing

JLARC staff compared VADOC’s prescription drug costs for a sample of prescription drugs that it
purchases through its prescription fill vendor to the costs if they were purchased through the federal
340B prescription drug program, available through VCU Health. The total cost for prescription drugs
includes the acquisition cost of the drug and the dispensing fee, or administrative cost, paid to the
pharmacy or other vendor that purchases and distributes the drug. To select the sample, JLARC staff

Appendixes

first identified four therapeutic classes of drugs that accounted for 52 percent of all non-340B prescription drug spending in FY17. These were Biologic/Immunologic, Pulmonary, Psychiatric, and Diabetes drugs. JLARC staff selected 21 distinct drugs within these four therapeutic classes that accounted for a significant amount of total spending on that therapeutic class (Table B-3).
TABLE B-3
Prescription drugs selected for comparison accounted for significant amount of total
prescription drug spending (FY17)
Total spending on
therapeutic class
(% of total)

Number of drugs
included in sample

Total spending on drugs
included in sample
(% of therapeutic class)

Biologic/Immunological

$3.49M (18%)

$2.84M (81%)

Pulmonary

$2.65M (14%)

$2.08M (78%)

Diabetes

$2.17M (11%)

$1.82M (84%)

Psychiatric

$1.76M (9%)

$0.20M (11%)

Therapeutic class

SOURCE: JLARC analysis of VADOC prescription drug claims data, FY17.
NOTE: Psychiatric drugs tend to be less expensive for an individual prescription and have many more unique drugs and drug doses,
making it much more difficult to capture a large amount of spending with a small sample.

JLARC staff worked with pharmacy staff from VCU Health to obtain the actual 340B acquisition
cost for each of the 21 unique prescription drugs during FY17. Drugs were matched using their National Drug Code (NDC), a unique identifier for each branded and generic drug, at each specific dose.
JLARC staff used either the average 340B price during the fiscal year or the current 340B price as of
June 2018 if FY17 data was unavailable. For the drugs with available FY17 data, JLARC staff analyzed
the difference between current prices and average price during FY17 and did not find systematic or
substantial differences. JLARC staff used this data to calculate the difference between what VADOC
pays and the 340B price for each type of drug and then multiplied that time the number of prescriptions for VADOC inmates in FY17. JLARC staff then estimated the change in dispensing fees by
comparing the dispensing fees under VADOC’s current contracts with its prescription fill vendor and
VCU Health. These two calculations were combined to reach the net estimated savings.
Analysis of MMCAP prescription drug prices

JLARC staff worked with staff at MMCAP to compare the prices VADOC pays through its prescription fill vendor for non-340B prescription drugs to what MMCAP’s negotiated prices are for the same
drugs. MMCAP staff provided data on what MMCAP would have paid for prescription drugs purchased by VADOC for May 2018. JLARC staff used this data to compare the acquisition cost between
VADOC and MMCAP. The cost of individual drugs varied, sometimes substantially, but on average
there was no significant cost difference between in prices between VADOC and MMCAP.

Appendixes

Analysis of VADOC and contractor turnover data

JLARC worked with VADOC and both of the comprehensive health service contractors to calculate
and compare turnover rates of front-line staff and health care leadership staff at contract and noncontract facilities. To calculate turnover rates, JLARC staff divided the number of turnover events
where an employee left in a given year by the total number of FTEs at each facility. For the purposes
of this analysis, JLARC staff considered Registered Nurses (RNs) and Licensed Practical Nurses
(LPNs) as front-line staff. JLARC staff considered Health Service Administrators (at contract sites)
and Health Authorities (at non-contract sites) as leadership staff. JLARC staff defined a facility as
having either high or low turnover of front-line staff based on whether its total turnover rate was
above the median for all facilities. Because there are so few leadership positions, JLARC staff simply
determined whether or not a facility experienced a change in leadership in a given year, rather than
calculating a turnover rate. JLARC staff then categorized facilities as having high or low turnover of
front-line staff, leadership, or both, resulting in each facility being defined in one of four groups (Table
B-4).
TABLE B-4
Turnover of both leadership and front-line staff was higher at contract facilities (FY17)
Non-contract facilities

Contract facilities

Low turnover of leadership and front-line staff

High turnover of front-line staff only

High turnover of leadership staff only

High turnover of leadership and front-line staff

SOURCE: JLARC analysis of VADOC and contractor staffing data, FY17.
NOTE: This analysis did not include turnover data for front-line staff at two contract facilities, as the data was not made available to
JLARC staff.

Analysis of estimated savings for compassionate release policies

To estimate savings related to potential changes to Virginia’s compassionate release policies, JLARC
staff used actual inmate data for FY17. Demographic and spending data was used to estimate the
number of inmates potentially impacted by expanding compassionate release policies and their associated health care spending. To calculate savings, only outpatient and prescription drug spending was
used because most inpatient services will be covered by Medicaid following Medicaid expansion.
Once JLARC estimated the total number of people who would be potentially eligible for consideration
under each compassionate release policy, staff applied three key assumptions to arrive at the number
who might actually be released, and the associated cost savings:
(1) Approximately two-thirds of inmates who meet the compassionate release criteria could be
considered for release because they would not have committed a violent crime that led to
their current term of incarceration. This assumption was based on the 2017 Virginia Criminal Sentencing Commission annual report, which indicated that two-thirds of offenders

Appendixes

sentenced in FY17 were eligible for the Non-Violent Offender Risk Assessment, meaning
none of the crimes in their current sentencing events were considered violent under Virginia
law.
(2) Approximately 75% of seriously or terminally ill inmates could be placed with family or in
an appropriate community facility given their health needs. This assumption was based on
an estimated placement rate for such inmates from community re-entry staff at VADOC.
(3) The average annual cost of the inmates in these analyses includes VADOC’s average cost
per bed for a state-responsible inmate, not including the cost of health care ($27,201 in
FY17) plus the average cost per inmate for outpatient and prescription drug services in
FY17.
Release for terminal illness
For inmates with terminal illnesses, JLARC staff took spending data for each inmate who passed away
of natural causes in FY17 and calculated average spending in the last 12 months prior to their passing
away. There was one additional assumption for the calculation of savings related to release for inmates
with terminal illnesses:
 Inmates would spend half of the last year of their life still incarcerated, given the time involved in arranging for a terminally ill inmate’s release.
This assumption, along with (1), (2), and (3) above, was used in calculating potential savings if Virginia’s terminal illness policy were expanded to a range of six to 12 months (Table B-5).
TABLE B-5
JLARC calculated possible savings on care for terminally ill inmates based on several
assumptions (FY17)

A Total inmates possibly eligible under new policy

B Percent who committed non-violent crime

66%

C Percent who could likely be placed in the community

75%

D Inmates likely to be released – (A*B)*C

E Portion of last year of life not incarcerated

50%

F Average annual inmate cost

$43,828

G Estimated annual savings – (D*F)*E

$705,083

SOURCE: JLARC analysis of VADOC demographic and outpatient/prescription drug claims data.

Release for serious illness
Inmates possibly eligible for release based on serious illness were identified using spending data from
FY15 through FY17. If an inmate was in the top ten percent of health care spending for all three
years, he or she was identified as an inmate with a serious illness. This group was further narrowed
down by excluding anyone age 60 or older in FY17, since these inmates could be eligible for geriatric

Appendixes

conditional release, depending on their time served. The less restrictive policy option, which mirrors
the policy in South Dakota, included 810 inmates under the age of 60 whose spending was among the
highest in each of the last three years. The more restrictive policy, similar to North Carolina’s, included
68 inmates under the age of 60 whose pending was among the highest in each of the last three years
and who were identified in VADOC data as having a disability that left them totally unable to work in
FY17.
JLARC staff based the calculation for possible savings for seriously ill inmates on assumptions (1),
(2), and (3) above (Table B-6).
TABLE B-6
JLARC staff calculated possible savings on care for seriously ill inmates based on several
assumptions (FY17)
Less restrictive (SD)

More restrictive (NC)

A Total inmates possibly eligible under new policy

810

B Percent who committed non-violent crime

66%

C Percent who could likely be placed in the community

75%

D Inmates likely to be released – (A*B)*C

401

$42,067

$16,866,764

$1,462,224

E Average annual inmate cost
F Estimated annual savings – E*F

SOURCE: JLARC analysis of VADOC demographic and outpatient/prescription drug claims data.

Analysis of inmate grievance data

JLARC staff had access to de-identified inmate health care grievances as part of this study and calculated the average health care grievances per inmate at each facility. The percentage of grievances that
were founded was also calculated, and staff made comparisons across facilities to look for any patterns. JLARC staff also compared the number of health care grievances to total grievances to determine if patterns in health care grievances might be due to systematic issues at the facility.
Review of documents and research literature
JLARC staff reviewed relevant sections of the Code of Virginia, and VADOC policies and procedures
to understand the legal and policy requirements governing inmate health care. Contract documents
and health care oversight documents were also reviewed as part of the study. Additionally, a review of
research literature and inmate health care lawsuits was conducted to identify strategies to reduce spending and the greatest legal risks associated with delivering inmate health care.

Appendixes

Analysis of health care compliance information

JLARC staff examined quality improvement documentation to understand the extent to which VADOC facilities may be complying with health care standards and the specific challenges different facilities and the system as a whole may be facing related to inmate health care. These documents included contract monitoring reports, corrective action plans, Continuous Quality Improvement
Committee (CQI) meeting minutes, and CQI monitoring forms.
Review of research literature

As part of the study, research literature from scholarly journals, government sources, and stakeholder
groups were reviewed. JLARC staff reviewed literature in the following areas:
 inmate health needs,
 electronic medical records,
 clinical pharmacy services and utilization management, and
 compassionate release policies.
The information gathered as part of these literature reviews was used for background research on
each topic, and to identify common approaches and best practices in health care, both in the community and for inmates.
Review of inmate health care lawsuits

JLARC staff partnered with a law student from William and Mary School of Law to find and summarize lawsuits related to inmate health care in the U.S. Fourth Circuit and U.S. Supreme Court. The
student was given a template developed by JLARC staff to gather the necessary information about
each case and used LexisNexis for searching. Using the case summaries, JLARC staff analyzed the
information provided to determine the common reasons for inmate health care lawsuits. This information was used to uncover the most common circumstances under which inmate health care lawsuits
are filed, in order to better understand the risks that the state faces if it provides less than adequate
care to inmates.
JLARC staff also identified class action inmate health care lawsuits nationwide. These cases were useful to identify the characteristics of cases against entire correctional systems and the unique risks to
the state posed by these types of lawsuits.
Surveys
JLARC staff conducted surveys of two groups for this study: (1) clinicians working in VADOC facilities and (2) Health Authorities, the head administrator at each facility.
Survey of clinical staff at VADOC facilities

The survey of clinicians was administered electronically to clinicians—physicians, nurses, dentists,
dental hygienists, dental assistants, psychiatrists, psychologists, and qualified mental health professionals (QMHPs)—working in VADOC facilities. The response rate for this survey was 45 percent, or 334
out of 741 of clinicians. Employees were asked about the following topics:

Appendixes

 what activities clinicians spent their time doing at work;
 how they viewed their workload and the impact of having too much work, where applicable;
 how often they completed tasks that could or should be done by a clinician with more or
less training; and
 what changes would improve efficiency and effectiveness of health care delivery in facilities.
Survey of Health Authorities at VADOC facilities

The survey of Health Authorities was administered electronically to one administrator per facility. The
Health Authority functions as the administrator of the facility’s medical department. At non-contract
facilities, Health Authorities are also clinicians (typically a registered nurse). The response rate for this
survey was 82.5 percent, or 33 out of 40 of Health Authorities. Health Authorities, were asked about
the following topics:
 measures of clinician workload, including the number of sick call, chronic care, and intake
assessment completed;
 what activities clinicians spent their time doing at work;
 how they viewed clinician workload and the impact of having too much work, where applicable;
 how many, if any, additional clinicians would be needed to meet workload demands;
 what changes would improve efficiency and effectiveness of health care delivery in facilities.

Appendixes

Appendix C: Inmate health care spending analysis
JLARC staff analyzed data on facility-level inmate health care spending from FY12 to FY17. This data
was compiled from the Auditor of Public Accounts (APA) Data Point tool as well as Cardinal extracts
provided by VADOC. This appendix provides additional detail on the facilities and services that account for the greatest amount of spending on inmate health care.
Health care spending is concentrated in a few facilities with specialized services
Health care costs at VADOC, like in other health systems, are concentrated in a small number of
inmates with very high health care needs. Seven VADOC facilities have specialized health care services,
and patients with higher health care needs are typically sent to one of these facilities. As a result, the
small number of facilities with specialized capabilities have much higher than average per-inmate
health care spending (Table C-1).

Appendixes

TABLE C-1
Facilities with specialized services have higher per-inmate health care spending (FY17)
Facility
Marion

ADP

Total spending

Per-inmate spending

290

$4,137,229

$14,266

Deep Meadow

1,341

17,983,012

13,410

Deerfield

1,603

18,772,071

11,711

Fluvanna

1,198

13,671,579

11,412

Greensville

3,219

27,931,672

8,677

632

5,102,855

8,074

VCCW
Sussex I

1,144

7,264,369

6,350

Sussex II

1,247

7,910,670

6,344

Lunenburg

950

6,018,457

6,335

Coffeewood

991

4,904,035

4,949

1,147

5,367,363

4,679

Haynesville
Bland

643

2,993,671

4,656

Buckingham

1,273

5,826,581

4,577

Augusta

1,329

6,044,025

4,548

Dillwyn

1,142

5,047,608

4,420

Keen Mountain

705

3,101,648

4,400

Nottoway

1,590

6,796,646

4,275

Indian Creek

1,008

4,136,365

4,104

Green Rock

1,022

4,084,749

3,997

Red Onion
Wallens Ridge
River North

877

3,497,687

3,988

1,081

3,556,438

3,290

969

2,964,425

3,059

St. Brides

1,175

3,534,589

3,008

Pocahontas

1,023

3,003,248

2,936

Major facilities total

27,599

$173,650,991

$6,292

Other spending

1,293

14,336,098

Total spending

28,892

$187,987,089

$6,507

Specialized services










SOURCE: JLARC analysis of Cardinal and Data Point data.
NOTE: Some major facilities have financial responsibility of smaller work centers and facilities. Additional ADP in “other spending” reflects inmates housed at work centers, field units, and detention and diversion centers that are not the financial responsibility of a major
institution. Payments to Lawrenceville and jails are not included.

Most health care spending is for onsite care, offsite care, and prescriptions
JLARC staff categorized spending into five main groups: contract payments, onsite care, offsite care,
prescription drugs, and other. Contract payments include all payments to the comprehensive service
contractors. Onsite care includes spending on state-employed personnel (salaries, benefits) and individually contracted personnel; medical equipment purchased for facilities; and medical supplies purchased for facilities. Offsite care includes payments to community providers and hospitals for inpatient, outpatient, dental, and laboratory services. Spending on prescription drugs includes payments
to VADOC’s prescription fill vendor for most drugs and to VCU Health for drugs purchased through

Appendixes

the federal 340B program. Other expenditures include costs for administrative salaries, travel, and
information technology, among other things.
Health care spending increased by $23.5 million between FY12 and FY17. Over these five years VADOC contracted out the health care at more facilities, increasing contract payments but decreasing
direct spending for onsite care. Offsite care increased due to rising costs but also because VADOC
now pays for the most expensive offsite services, inpatient hospital services. These were included in
comprehensive contracts during FY12. Prescription drug spending has seen substantial increases despite more of this spending being included in comprehensive contracts (Table C-2).
TABLE C-2
Inmate health care spending by type, FY12 and FY17
Spending type
Contract payments
Onsite care
DOC employees

FY12

FY17

$67,699,538

$81,593,950

35,710,587

32,970,069

29,285,475

24,974,165

Individual contracts

4,693,402

5,817,250

Other

1,731,710

2,178,654

41,257,663

40,174,927

Offsite care
Inpatient

22,130,732

27,280,508

Outpatient

17,228,388

10,058,898

1,898,544

2,835,521

13,459,664

20,351,160

Specialty and other
Prescription drugs
Administration and other
Total

6,383,251

12,896,983

$164,510,904

$187,987,089

SOURCE: JLARC analysis of Cardinal and Data Point data.
NOTE: FY12 spending has been adjusted for inflation. This table includes only payments paid by VADOC, not its
contractors. Payments to Lawrenceville and jails are not included.

Contract facilities tend to have the highest total spending because many of them provide care to the
sickest inmates (Table C-3). The onsite, offsite, and prescription drug categories at contract facilities
include only what VADOC pays directly for those services. This includes inpatient hospital stays and
prescription drugs purchased through VCU Health. Contract facilities with significant spending in
these categories, such as Deerfield, have a large number of inpatient stays and inmates in need of
high-cost medications. All services paid for by the contractors are included in the capitation payments
and are not broken out separately in the table.

Appendixes

TABLE C-3
Inmate health care spending by type and facility (FY17)
Facility

Contract
payments

Prescription
drugs

Onsite care

Offsite care

$22,042,842

$983,914

$3,602,360

$1,262,420

$40,136

$27,931,672

12,912,585

502,034

3,585,277

1,739,889

32,013

18,772,071

7,132,532

2,247,160

6,705,882

1,625,545

271,893

17,983,012

Fluvanna

11,355,383

445,858

1,257,407

451,081

161,848

13,671,579

Sussex II

5,686,469

272,613

1,247,468

696,931

7,190

7,910,670

Sussex I

5,590,216

280,912

1,036,701

342,216

14,324

7,264,369

3,038,715

2,525,988

1,147,509

84,434

6,796,646

255,377

1,153,506

676,246

5,036

6,044,025

Greensville
Deerfield
Deep Meadow

Nottoway

Augusta

3,953,860

Lunenburg

3,686,289

Other

Total

315,291

1,567,370

442,858

6,650

6,018,457

Buckingham

1,621,998

3,013,849

1,174,061

16,673

5,826,581

Haynesville

2,041,812

2,045,833

1,260,942

18,776

5,367,363

VCCW

2,806,527

1,420,057

859,365

16,906

5,102,855

Dillwyn

2,032,719

1,863,990

1,121,439

29,460

5,047,608

313,848

548,440

259,935

23,625

4,904,035

2,831,910

696,728

550,608

57,983

4,137,229

225,248

621,552

295,197

5,796

4,136,365

Coffeewood
Marion
Indian Creek

3,758,187
-2,988,572

Green Rock

1,935,485

987,083

628,885

14,562

4,084,749

Wallens Ridge

1,925,907

1,190,850

931,191

27,223

3,556,438

St. Brides

69,956

479,495

475,187

23,209

3,534,589

Red Onion

1,974,224

848,366

629,719

45,377

3,497,687

Keen Mountain

1,449,554

773,525

837,935

40,633

3,101,648

Pocahontas

1,388,411

642,039

944,764

28,035

3,003,248

Bland

1,306,726

902,989

766,285

17,671

2,993,671

River North

1,602,035

519,419

833,676

9,295

2,964,425

Field Units

673,808

600,714

276,865

2,783

1,554,169

Detention and Diversion Centers

482,027

338,040

120,410

7,471

893,947

$32,970,069

$40,174,927

$20,351,160

$1,009,001

$176,099,107

Total

2,486,742

$81,539,950

SOURCE: JLARC analysis of Cardinal and Data Point data, FY17.
NOTE: This table includes only payments paid by VADOC, not its contractors. Totals include only spending attributed to a facility. Payments to Lawrenceville and jails are not included, nor are central administration costs.

Appendixes

Appendix D: Map of VADOC facilities

EE1---..:~--=---__!__-=m=----=-·D
-=m
=---~-

+ Secure medical facilities

■ Non-contract facilities

1 VCU Medical Center

15 Baskerville Correctional Center

2 Southampton Memorial Hospital

16 Bland Correctional Center

17 Buckingham Correctional Center

■ Contract facilities

18 Deep Meadow Correctional Center

19 Dillwyn Correctional Center
3 Augusta Correctional Center

20 Green Rock Correctional Center

4 Coffeewood Correctional Center

21 Haynesville Correctional Center

5 Deerfield Correctional Center

22 Keen Mountain Correctional Center

6 Fluvanna Correctional Center

23 Marion Treatment Center

7 Greensville Correctional Center

24 Nottoway Correctional Center

8 Indian Creek Correctional Center

25 Pocahontas State Correctional Center

9 Lawrenceville Correctional Center

26 Red Onion State Prison

10 Lunenburg Correctional Center

27 River North Correctional Center

11 Powhatan Reception Center

28 Virg inia Correctional Center for Women

12 St. Brides Correctional Center

29 Wallens Ridge State Prison

13 Sussex I State Prison
14 Sussex II State Prison
NOTE: Lawrenceville is a fully privatized prison. Smaller work centers and field units are not included on the map.

Appendixes

Appendix E: Offsite and prescription drug expenditure analysis
JLARC staff used regression analysis to compare health care utilization and spending across VADOC
facilities. A primary goal of these analyses was to determine if contract facilities or non-contract facilities were able to provide health care more cost-efficiently. Because health care spending is highly
dependent on patient characteristics, and inmates with high needs are sent to specific facilities, simply
comparing spending levels was not sufficient. JLARC staff created four regression models to compare
facility spending on offsite services and pharmaceuticals while controlling for differences in inmate
populations at each facility. The models were used to predict
 offsite care utilization,
 offsite care spending,
 pharmaceutical utilization, and
 pharmaceutical spending.
The models used health care claims data for all inmates who that were incarcerated in FY15 through
FY17 at a VADOC facility. This data was aggregated to an analysis file with one observation per
inmate per facility in each year. For example, if an inmate was in one facility for all twelve months in
the year, there was one record for that inmate. If an inmate moved to a new facility during the year,
there would be two records for that inmate. Offsite and prescription drug utilization and spending
was predicted for each inmate based on health and demographic information, and the predicted values
were totaled for each facility. The predicted values were then compared to the actual values at each
facility. Weighted averages were used to account for differences in length of time spent at facilities.
A facility that spent more than predicted is not necessarily inefficient and a facility that spent less than
predicted may not necessarily be more efficient given the relatively small number of inmates at each
facility. One or more inmates with very expensive care can drastically increase the average spending
of an otherwise low-cost facility. However, when considering the results in aggregate, the models
provide evidence of systematic utilization and spending differences between contract and non-contract facilities. The four models were analyzed separately and in combination, for fiscal years 2015,
2016, and 2017.
Data variables and sources
In the models predicting offsite and prescription drug utilization, the dependent variable was whether
the inmate used offsite services or prescription drugs in that fiscal year. In the models predicting
offsite and pharmacy spending, the dependent variable was the average monthly spending on offsite
services or pharmaceuticals. The models included several independent variables, including inmate demographics, health status, and diagnoses (Table E-1).

Appendixes

TABLE E-1
Variables included in JLARC’s offsite and prescription drug expenditure regression models
Variable

Definition

Offsite user (dependent variable)

Whether the inmate had offsite services spending in the period of
observation (0 or 1)

Prescription drug user (dependent variable)

Whether the inmate had pharmaceutical spending in the period of
observation (0 or 1)

Offsite spending (dependent variable)

Total amount charged for inmate offsite care in period of
observation

Prescription drug spending (dependent variable)

Total amount charged for inmate pharmaceuticals in period of
observation

Age

Age in years

Gender

Male or female

Race

Non-white or white

Dummy variables for mental health code

Mental health code (from 0-4) the inmate is assigned by VADOC. A
mental health code of 2 or higher indicates a serious mental health
diagnosis

Work code

Work code (from 1-4) the inmate is assigned by VADOC. A work
code of 1 indicates there are no restrictions on the inmate’s ability to
work while a work code of 4 indicates the inmate cannot work in any
capacity

Cardiovascular

Cardiovascular disease or condition flag (0 or 1)

Respiratory

Respiratory disease or condition flag (0 or 1)

Hepatitis

Hepatitis diagnosis flag (0 or 1)

Orthopedic

Orthopedic disease or condition flag (0 or 1)

Neurologic

Neurological disease or condition flag (0 or 1)

Gastrological

Gastrological disease or condition flag (0 or 1)

Hematologic

Hematologic disease or condition flag (0 or 1)

Diabetes

Diabetes diagnosis flag (0 or 1)

Cancer

Cancer diagnosis flag (0 or 1)

Dialysis need

Whether the inmate needs dialysis (0 or 1)

Infirmary need

Whether the inmate needs to be housed in an infirmary (0 or 1)

Dummy variables for facility

Facility the inmate is housed in for the period (0 or 1 for each facility)

Appendixes

Analysis results
With each model, the predicted utilization or spending was compared to actual utilization or spending
at each facility. When comparing predicted and actual prescription drug utilization, both contract and
non-contract facilities averaged slightly higher actual utilization than predicted. The difference between contract and non-contract facilities, on average, was small in FY17. Non-contract facilities had
two percent higher than predicted utilization, while contract facilities had one percent higher than
predicted utilization (Table E-2).
TABLE E-2
Facility-level actual and predicted prescription drug utilization (FY17)
Facility

Actual
utilization

Predicted
utilization

Difference

Nottoway

0.52

0.64

−0.12

Pocahontas State

0.62

0.66

−0.04

Dillwyn

0.67

0.70

−0.03

Haynesville

0.63

0.65

−0.02

VCCW

0.93

0.95

−0.02

Deep Meadow

0.62

0.63

−0.01

Wallens Ridge

0.64

0.00

Baskerville

0.57

0.01

Red Onion

0.68

0.65

0.03

Green Rock

0.69

0.65

0.04

Keen Mountain

0.73

0.67

0.06

Buckingham

0.71

0.64

0.07

Bland

0.76

0.67

0.09

River North

0.74

0.63

0.10

Marion

0.85

0.73

0.12

Non-contract average

0.69

0.67

0.02

St. Brides

0.56

0.64

−0.07

Coffeewood

0.61

0.65

−0.04

Indian Creek

0.62

0.63

−0.01

Sussex I

0.61

0.62

−0.01

Fluvanna

0.93

0.92

−0.01

Augusta

0.67

0.65

0.02

Sussex II

0.66

0.62

0.04

Deerfield

0.88

0.83

0.05

Lunenburg

0.71

0.65

0.05

Greensville

0.73

0.67

0.06

Contract average

0.70

0.69

0.01

SOURCE: JLARC analysis of VADOC CORIS data and pharmaceutical claims, FY17.
NOTE: The table does not include Powhatan, a contract facility that was a significantly high outlier. Due to rounding, the reported difference and predicted utilization may not total to reported actual utilization.

Appendixes

Compared to predicted and actual offsite service utilization, contract facilities appear to do slightly
better. In FY17, non-contract facilities had three percent higher than predicted utilization and contract
facilities’ actual utilization was equal to the predicted level (Table E-3). This may indicate that contract
facilities are better able to avoid sending inmates offsite for care that can be provided in the facility.
However, this is balanced by higher than predicted offsite spending for inmates who are sent offsite
from contract facilities.
TABLE E-3
Facility-level actual and predicted offsite utilization (FY17)
Facility
Nottoway

Actual
utilization

Predicted
utilization

Difference

0.06

0.15

−0.09

Pocahontas

0.10

0.17

−0.07

Dillwyn

0.14

0.18

−0.04

Baskerville

0.11

0.14

−0.03

River North

0.15

0.00

Green Rock

0.17

0.00

VCCW

0.36

0.33

0.03

Haynesville

0.20

0.17

0.03

Buckingham

0.22

0.16

0.05

Keen Mountain

0.22

0.16

0.06

Red Onion

0.23

0.15

0.08

Wallens Ridge

0.22

0.14

0.08

Bland

0.27

0.17

0.09

Deep Meadow

0.28

0.17

0.11

Marion

0.31

0.20

0.12

Non-contract average

0.20

0.17

0.03

St. Brides

0.08

0.15

−0.07

Coffeewood

0.12

0.17

−0.05

Augusta

0.14

0.16

−0.02

Fluvanna

0.31

0.33

−0.02

Indian Creek

0.15

0.16

−0.01

Sussex I

0.15

−0.01

Lunenburg

0.17

0.01

Greensville

0.21

0.19

0.02

Sussex II

0.18

0.16

0.02

Deerfield

0.41

0.31

0.10

Contract average

0.19

0.20

0.00

SOURCE: JLARC analysis of VADOC CORIS data and offsite claims, FY17.
NOTE: The table does not include Powhatan, a contract facility that was a significantly high outlier. Due to rounding, the reported difference and predicted utilization may not total to reported actual utilization.

Appendixes

When comparing predicted and actual prescription drug spending among prescription drug users,
contract facilities may do slightly better. In FY17, non-contract facilities spent, on average, $10 more
than predicted per person, per month. Contract facilities spent, on average, $8 less than predicted per
person, per month (Table E-4).
TABLE E-4
Facility-level actual and predicted prescription drug spending (FY17)
Facility

Actual
spending

Predicted
spending

Difference

VCCW

$121

$146

−$25

−16

Baskerville
Red Onion

−13

Green Rock

102

−9

−7

109

102

Wallens Ridge
Pocahontas State
Nottoway

River North

111

100

Marion

159

148

Dillwyn

114

Bland

Haynesville

120

Buckingham

126

Keen Mountain

129

Deep Meadow

170

108

$109

$99

$10

141

182

−41

−37

131

157

−26

Non-contract average
Fluvanna
Indian Creek
Greensville
Coffeewood

−17

St. Brides

−13

Augusta

−12

Sussex I

−9

Sussex II

110

113

−3

Lunenburg

109

Deerfield

293

230

$116

$124

−$8

Contract average

SOURCE: JLARC analysis of VADOC CORIS data and pharmaceutical claims, FY17.
NOTE: Spending is reported per person, per month. Averages are only among inmates with prescription drug spending. The table does
not include Powhatan, a contract facility that was a significantly high outlier. Due to rounding, the reported difference and predicted
utilization may not total to reported actual utilization.

Appendixes

When comparing predicted and actual offsite spending among offsite service users, contract facilities
do worse. Non-contract facilities spent $46 more than predicted per person, per month, while contract
facilities spent $147 more than predicted per person, per month in FY17 (Table E-5). This may indicate that while contract facilities are potentially able to send inmates offsite for care less frequently
than non-contract facilities, when they must send inmates to offsite providers, the costs are higher.
TABLE E-5
Facility-level actual and predicted offsite spending (FY17)
Actual
spending

Predicted
spending

Difference

$290

$491

−$201

River North

283

474

−192

Keen Mountain

279

439

−159

Pocahontas

523

670

−147

Facility
Bland

Red Onion

318

451

−133

Wallens Ridge

362

364

−1

Baskerville

384

366

VCCW

487

445

Green Rock

584

541

Deep Meadow

649

595

Haynesville

581

506

Marion

676

510

166

Nottoway

881

546

335

Dillwyn

895

553

341

Buckingham
Non-contract average
Indian Creek

969

525

444

$544

$498

$46

489

529

−40

Fluvanna

779

812

−33

Coffeewood

537

518

St. Brides

531

464

Sussex II

604

537

Deerfield

1.364

1.268

Greensville

1.385

1.198

187

Augusta

795

602

193

Sussex II

1027

568

458

Lunenburg
Contract average

983

524

459

$849

$702

$147

SOURCE: JLARC analysis of VADOC CORIS data and offsite claims, FY17.
NOTE: Spending is reported per person, per month. Averages are only among inmates with offsite spending. The table does not include
Powhatan, a contract facility that was a significantly high outlier. Due to rounding, the reported difference and predicted utilization may
not total to reported actual utilization.

Appendixes

The four models were combined to predict total spending (prescription drug and offsite). For each
facility, the predicted prescription drug utilization rate was multiplied by the facility population to
create a “predicted user” population, and that “predicted user” population was multiplied by the predicted prescription drug spending per person. The same process was completed for offsite service
utilization and spending. This provided a predicted total spending that incorporated all four models,
and could then be compared to actual total spending at each facility. JLARC staff found no evidence
that contract facilities spent systematically more or less on offsite and prescription drug services in
FY17 (Table E-6). The graph in Chapter 3 also presents the results of this combined analysis across
all three years.
TABLE E-6
Facility-level actual and predicted total spending (FY17)
Facility
Pocahontas State
Baskerville
River North

Actual
spending ($)

Predicted
spending ($)

Difference ($)

Difference per
person-month ($)

$1,417,785

$2,151,929

−$734,144

−$62

433,776

495,448

−61,672

−11

1,369,389

1,423,365

−53,976

−5

Red Onion

1,243,404

1,137,673

105,731

Green Rock

2,018,600

1,878,087

140,513

Keen Mountain

1,397,205

1,271,342

125,863

Bland

1,449,319

1,185,102

264,217

Wallens Ridge

1,655,441

1,212,300

443,141

VCCW

1,884,321

1,618,526

265,795

Nottoway

2,851,222

2,088,018

763,204

Haynesville

2,325,786

1,539,215

786,571

Dillwyn

2,675,529

1,774,421

901,108

Buckingham

3,532,890

1,729,037

1,803,853

148

Marion

1,206,420

645,908

560,512

182

Deep Meadow
Non-contract total

2,820,041

1,331,698

1,488,343

189

$28,281,128

$21,482,069

$6,779,059

$47

Fluvanna

4,676,825

5,953,583

−1,276,758

−94

St. Brides

1,359,008

1,809,641

−450,633

−32

Coffeewood

1,460,821

1,685,720

−224,899

−20

Indian Creek

1,554,833

1,779,927

−225,094

−18

Augusta

2,294,988

2,099,330

195,658

Sussex I

2,354,659

1,956,963

397,696

Sussex II

3,037,000

2,244,984

792,016

Greensville

12,881,762

10,931,445

1,950,317

Lunenburg

3,212,200

1,600,859

1,611,341

151

Deerfield
Contract total

9,693,481

7,220,476

2,473,005

200

$42,525,577

$37,282,929

$5,242,648

$49

SOURCE: JLARC analysis of VADOC CORIS data and offsite and pharmaceutical claims, FY17.
NOTE: Spending is reported per person, per month. Does not include Powhatan, a contract facility that was a significantly high outlier.

Appendixes

Appendix F: Compassionate release policies in other states
Compassionate release policies are a tool that states can use to reduce spending on inmate health care
by releasing very ill or aged inmates to appropriate placements in the community. Amending compassionate release is a decision that must be made by state decision-makers in light of policy priorities
related to humanitarian concerns, public safety, appropriate punishment for crimes, and budgetary
pressures. In most cases, release of an inmate is conditional, meaning he or she can be re-incarcerated
should he or she violate the terms of release or medically recover.
Compassionate release process in Virginia
In Virginia, there are two types of compassionate release: one for inmates at least 60 years old having
served a certain portion of their sentence (geriatric conditional release) and one for terminally ill inmates with three months or less to live (medical clemency). Virginia does not have a specific compassionate release statute. Instead, geriatric conditional release is authorized by code, and medical clemency falls under the powers of the governor.
Investigation into suitability of inmate for release

For both types of release, the Virginia Parole Board (VPB) completes an investigation that includes
an assessment of the inmate’s offenses and behavior while incarcerated, his or her potential to be a
public safety risk, his or her medical needs, and whether an appropriate community placement can be
found for seriously/terminally ill inmates. For medical clemency, the inmate must be placed with a
family member or other individual ready and willing to care for the inmate. Inmates released on geriatric conditional release can be placed in community facilities (such as nursing homes or assisted living
facilities) or with family. In addition, VPB is required by law to notify any victims that the inmate is
being considered for release, a process that can take up to 60 days.
Ultimately, VPB makes a recommendation as to whether the inmate should be released. For geriatric
conditional release, they provide this recommendation to VADOC. In the case of medical clemency,
this recommendation is provided to the Secretary of the Commonwealth for a final decision from the
governor.
Geriatric conditional release

An inmate is eligible for geriatric conditional release when he or she is at least age 60 having served
10 years of his or her sentence, or age 65 having served 5 years of his or her sentence. Currently, a
large number of the inmates eligible for geriatric conditional release are also eligible for discretionary
parole because they were sentenced for a crime that was committed before Truth-in-Sentencing abolished discretionary parole in 1995. As a result, they are automatically considered for release when their
age and sentence requirements are met. The VPB estimates there are approximately 200 to 300 inmates
eligible for geriatric conditional release who are not eligible for discretionary parole, and works to
notify these individuals that they may apply for geriatric conditional release. VPB estimates that 10,000
state-responsible inmates who are currently incarcerated with VADOC will be eligible for geriatric
conditional release at some point during their sentences.

Appendixes

Medical clemency

The medical clemency process begins with an application for clemency from the inmate or inmate’s
family to the Secretary of the Commonwealth. Inmates may apply for medical clemency when they
have been given a prognosis of three months or less to live by a physician. The final decision for
clemency rests with the governor.
Compassionate release in other states compared to Virginia
Most states have compassionate release policies related to three situations:
 advanced age,
 serious illness or permanent incapacitation, and
 terminal illness.
The criteria for these types of release vary widely by state (Table F-1). Virginia only considers inmates
of advanced age (geriatric conditional release) and inmates with a terminal illness and three months
or less to live (medical clemency) for release. Virginia is the only state in which there is no way for
seriously ill or permanently incapacitated inmates to be released.
TABLE F-1
Compassionate release criteria vary significantly by state
Serious illness /
permanent
incapacitation
Virginia
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Kansas
Kentucky



















Terminal
illness

Advanced
age

3 months or less

60+ and served at least 10 years or
65+ and served at least 5 years

12 months or less

55+


6 months or less
2 years or less
6 months or less


6 months or less


6 months or less


6 months or less





30 days or less
12 months or less

60+ and at least 10 years served




55+



65+


62+







Appendixes

Serious illness /
permanent
incapacitation
Virginia
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Washington
West Virginia
Wisconsin
Wyoming


































Terminal
illness

Advanced
age

3 months or less

60+ and served at least 10 years or
65+ and served at least 5 years

12 months or less

45+ and at least 25 years served or
60+ and at least 10 years served





60+ and at least 15 years served





18 months or less


12 months or less



60+ and at least 10 years served

6 months or less



6 months or less








12 months or less


6 months or less
6 months or less

65+





6 months or less

65+





6 months or less OR
death imminent

6 months or less OR
death imminent
60+ and at least 10 years or
1/3 of sentence served






12 months or less



18 months or less



2 years or less




70+
65+ and served at least 10 years or
70+ and served at least 30 years



6 months or less

65+











60+ and served at least 10 years or
65+ and served at least 5 years

12 months or less

SOURCE: JLARC staff analysis of compassionate release policies in other states.
NOTE: Iowa has no formal compassionate release policies, and so is not included in this table.



Appendixes

Release for serious illness or permanent incapacitation

Virginia does not consider inmates for release based on serious illness or permanent incapacitation,
and is the only state in the country that does not consider inmates for release on this basis. Fortyseven states and the District of Columbia have policies that allow inmates to be considered for release
based on a serious illness or permanent incapacitation. While definitions of “serious illness” and “permanent incapacitation” vary widely, states’ policies generally fall into one or more of the following
categories:
 an inmate is diagnosed by a physician as having such a severe illness and/or permanent disability that he or she poses no public safety risk,
 an inmate can no longer care for himself or herself in a prison environment or perform
activities of daily living, or
 an inmate has complex health needs that are extremely expensive and better addressed in the
community.
Release for terminal illness

Virginia considers inmates for release if they have a terminal illness and have been given a prognosis
of three months or less to live by a physician. Four states do not have this type of release available,
but all of these states have policies that allow for the release of seriously ill inmates that likely includes
those with terminal illnesses. For release because of a terminal illness, an inmate must have a diagnosis
from a physician that he or she has a terminal illness with a defined period of time left to live.
Virginia has the second most restrictive prognosis requirement for release based on terminal illness,
at three months. In Kansas, inmates must have a prognosis of 30 days or less to be considered for this
type of release. However, Kansas also allows inmates to be considered for release for a serious illness,
providing other terminally ill inmates with an avenue to be considered for compassionate release.
Twenty-three states and the District of Columbia have prognosis requirements for terminal illness
release required by law, with the majority of these being six or 12 months. An additional 21 states will
consider an inmate with a terminal illness for release, but do not have a prognosis requirement under
state law.
Release for advanced age

Twenty-two states, including Virginia, have policies allowing inmates to be released based on their age.
Virginia’s requirement of age 60 with 10 years served or age 65 with five years served is in line with
the majority of other states that have defined ages for release due to advanced age. Six other states
also have time served requirements by state law in addition to age requirements for this type of release.
The states that do not consider inmates for release on this basis all have release based on serious
illness.

Appendixes

Appendix G: Agency responses
As part of an extensive validation process, the state agencies and other entities that are subject to a
JLARC assessment are given the opportunity to comment on an exposure draft of the report. JLARC
staff sent an exposure draft of this report to the Secretary of Public Safety, the Virginia Department
of Corrections, and the Virginia Office of the Attorney General. Relevant excerpts of the report were
also provided to other state agencies and contractors for review and comment. Appropriate corrections resulting from technical and substantive comments are incorporated in this version of the report.
This appendix includes response letters from the Virginia Department of Corrections and the Virginia
Parole Board.

COMMONWEALTH of VIRGINIA
HAROLD W. CLARKE
DIRECTOR

Department of Corrections

P. 0 . BOX 26963
RICHMOND, VIRGINIA 23261
(804) 674-3000

Mr. Hal E. Greer, Director
Joint Legislative Audit and Review Commission
919 East Main Street
Suite2100
Richmond, VA 23219
Dear Mr. Greer:
Thank you for the opportunity to review and comment on the Joint Legislative Audit and Review
Commission's Draft report, Inmate Health Care Spending. The draft report addresses major obstacles faced
by the Virginia Department of Corrections (VADOC) regarding the high cost of healthcare for our offender
population. As you know, offenders frequently enter V ADOC following high risk lifestyles with the
increased health problems that such lifestyles carry. Often these individuals have untreated or undiagnosed
conditions due to sporadic or no healthcare prior to entering V ADOC. Additionally, our offender
population is increasingly aging which magnifies the health care challenges. V ADOC remains committed
to providing quality healthcare to offenders.
We would like to thank the commission for their staff's thorough research, thoughtful recommendations
and the collaborative approach to developing the report. Through our substantive comments recently
addressed by Jeff Lunardi, we believe the report addresses the challenges in both being good stewards of
the Commonwealth's assets and the mandate to continue to provide a level of healthcare required by the
constitution and meet the increasing standards of recent court decisions.
VADOC's long-term vision is for the department to be a progressive and proven innovative leader in the
profession. We feel this report fits this vision and we appreciate the opportunity to have provided input.

Sincerely

ijd~
Director

HWC :srnh

COMMONWEALTH of VIRGINIA
Virginia Parole Board

ADRIANNE L. BENNETT
Chairman

A. LINCOLN JAMES
Member

JEAN CUNNINGHAM
Vice Chairman

SHERMAN P. LEA, SR
Member

October 29, 2018

LINDA L. BRYANT
Member

Hal E. Greer, Director
Joint Legislative Audit and Review Commission
919 East Main Street
Richmond, VA 23219
Dear Mr. Greer:
On behalf of the Virginia Parole Board , I wish to thank you and your staff
for the opportunity to review and comment on the exposure draft report entitled
Spending on Inmate Health Care. The professionalism and thoughtfulness of
your staff was extraordinary.
We are grateful for the consideration given to the challenges associated
with releasing elderly and/or sick offenders. The exposure draft provides insight
into the challenges of finding community placements for elderly and/or sick
offenders whose release is compatible with public safety.
The exposure report is particularly informative as to how the
Commonwealth compares to other states that have compassionate/medical
release policies. We appreciate the clear statement that less restrictive policies
would make more inmates eligible to be considered for release, but not affect the
decision about "who is actually released". Specifically, thank you for recognizing
that all releasing decisions begin first with a comprehensive investigation of
whether an offender's release is compatible with public safety and the potential
impact release would have on victims and the community in general.
We look forward to continuing to work with members of the General
Assembly.
Respectfully,

~~,~~
Chair
6900 Atmore Drive, Richmond, VA 23225 • (804)674-3081 • Fax (804) 674-3284

JLARC.VIRGINIA.GOV
919 East Main St. Suite 2101
Richmond, VA 23219