Performance Audit Report

Department of Public Safety and Correctional Services
Diminution Credits
Enhancements to Existing Processes are Needed to Help Ensure that Diminution
Credits are Properly Applied
Some Inmate Release Dates Were Incorrect as a Result of Credit Calculation
Errors

December 2004

OFFICE OF LEGISLATIVE AUDITS
DEPARTMENT OF LEGISLATIVE SERVICES
MARYLAND GENERAL ASSEMBLY

•

This report and any related follow-up correspondence are available to the public. Alternate
formats may also be requested by contacting the Office of Legislative Audits as indicated at
the bottom of the next page or through the Maryland Relay Service at 1-800-735-2258.

•

Please address specific inquiries regarding this report to the Audit Manager listed on the
inside back cover by telephone at (410) 946-5900.

•

Electronic copies of our audit reports can be viewed or downloaded from the Internet via
http://www.ola.state.md.us.

•

The Department of Legislative Services – Office of the Executive Director, 90 State Circle,
Annapolis, Maryland 21401 can also assist you in obtaining copies of our reports and related
correspondence. The Department may be contacted by telephone at (410) 946-5400 or (301)
970-5400.

December 10, 2004
Senator Nathaniel J. McFadden, Co-Chair, Joint Audit Committee
Delegate Charles E. Barkley, Vice-Chair, Joint Audit Committee
Members of Joint Audit Committee
Annapolis, Maryland
Ladies and Gentlemen:
We conducted a performance audit to determine if the Department of Public
Safety and Correctional Services (DPSCS)—Division of Correction (DOC) was
properly calculating and accounting for inmate diminution credits. These credits
are earned by inmates assigned to DOC institutions, under certain conditions,
such as for good conduct and for satisfactory performance in work programs.
Diminution credits permit inmates to be released, under a process called
mandatory supervision, prior to serving the full term of their confinements.
Inmates released under mandatory supervision are still under the authority of
DPSCS until the full sentences have expired.
Our audit disclosed that, although DOC had initiated a number of processes,
certain enhancements are needed to help ensure that diminution credits are
properly applied and that inmates are released on correct dates based on credits
earned. We conducted tests at two DOC institutions of a total of 65 inmates
released during calendar year 2003 (selected as a statistical sample). Our tests
disclosed that, because of various errors affecting the application of credits, 22
inmates were not released on the proper dates, but rather, were released from 1 to
112 days early and 3 to 24 days late. The test results were similar for both
institutions. While DOC had established a system of internal controls intended to
provide assurance that recorded credits were accurate, the system did not detect
the errors we identified. We identified several causes for the errors, including the
failure to record changes in work assignments, which affected credit earnings, and
errors in accounting for credits earned while inmates were in the custody of local
jails.
We also noted that, at one of the two institutions, inmates often earned credits for
a specific work assignment even though the institution could not document, as
required, that the inmates had actually performed the applicable work.

As part of our audit we also determined that DOC properly implemented policies
and procedures that reflect State laws, regulations, and court rulings relating to
diminution of credits. Our audit scope, objectives, and methodology are
explained on page 9.
We wish to acknowledge the cooperation extended to us during our audit by
DPSCS and DOC staff and the staff of the institutions selected for review.

Respectfully submitted,

Bruce A. Myers, CPA
Legislative Auditor

2

Table of Contents
Background Information

5

Audit Scope, Objectives, and Methodology

9

Findings and Recommendations

11

Policies and Procedures Governing Diminution Credits
DOC Credit Diminution Policies Are In Accordance with State Law

11

Processes and Controls Over the Calculation of Diminution Credits
Finding 1 – In Some Cases Inmates Were Not Released on the Proper
Dates
Finding 2 – Actual Performance of Work by Inmates to Earn Certain
Diminution Credits at One Institution Was Not Documented as
Required
Finding 3 – DOC Lacked a Formal Policy for Determining the
Reasonableness of Diminution Credit Earning Rates for Specific
Work Assignments
Finding 4 – Access Controls Over OBSCIS Were Not Adequate

Exhibit 1 – Detail of Incorrect Inmate Release Dates Noted in Finding 1
Agency Response

11
13

14

15

17
Appendix

3

4

Background Information
Department of Public Safety and Correctional Services Overview
The Department of Public Safety and Correctional Services—Division of
Correction (DPSCS—DOC) oversees the operations of 23 correctional facilities
in the State. According to DOC records, approximately 23,000 inmates were
incarcerated during fiscal year 2004.

Diminution Credit Overview
All inmates sentenced to DOC institutions are eligible to receive diminution
credits. These credits were created by the State to reduce the costs associated
with holding inmates, by providing a mechanism to release inmates prior to the
expiration of their full sentences and to provide a system of rewards for positive
behavior. Inmates earning credits are eligible for early release, known as
mandatory supervision, which is defined by State law as a conditional release for
inmates who have served the length of the terms less diminution credits earned.
Inmates released early through this process are under the direct supervision of the
DPSCS’ Division of Parole and Probation and come under the authority of the
Maryland Parole Commission, for oversight and monitoring purposes, until the
expiration of their full terms of confinement (that is, the complete length of the
original sentences).
Under State law, the types of diminution credits fall under four broad categories.
Each credit received equals one less day of confinement. State law limits an
inmate’s credits to a maximum of 20 each month.
¾ Good Conduct Credits – These credits are assigned in advance to an
inmate and will reduce the term of confinement if the inmate follows DOC
behavioral guidelines. DOC automatically grants these credits to inmates
(as required by law) on the date of confinement. An inmate may lose
good conduct credits, under a process called revocation, if the inmate
violates specific institutional guidelines for behavior (such as fighting
with other inmates or DOC employees). Inmates can earn up to 10 good
conduct credits monthly during confinement.
¾ Work Task Credits – These credits are awarded to inmates based on
satisfactory performance in approved work programs, such as dietary and
sanitation details. Inmates can receive up to 5 work task credits per
month. Once earned, these credits cannot be subsequently revoked.

5

¾ Education Credits – These credits are awarded to inmates based on
satisfactory progress in vocational, educational, and other training
programs. Inmates can receive up to 5 education credits per month. Once
earned, these credits cannot be subsequently revoked.
¾ Special Project Credits – These credits are awarded to inmates for
satisfactory completion of special tasks or other special programs as
determined by DOC (such as dual cell occupancy in overcrowded
prisons). Inmates can receive up to 10 days per month for special project
credits. These credits may be revoked.

Credit Recordation Process
DOC maintains its diminution credit data in the automated Offender Based State
Correctional Information System (OBSCIS) for all inmates sentenced to its
jurisdiction. The only exception is inmates serving life terms, for which DOC
maintains manual credit records. DPCSC automated the diminution credit
process using OBSCIS in 1996, in part, to minimize human error, which can
occur when manually calculating mandatory supervision release dates.
The diminution process begins when, for each person convicted by the court, the
court generates a commitment record detailing the sentence imposed, its start
date, and any other conditions. DOC’s central commitment office is responsible
for processing information related to diminution credits, using the Judiciary’s
commitment record to initially enter inmate information into OBSCIS, including
length of sentence, good conduct credits assigned, and eligibility for other credits
(which can be dependent on the crime). Based on the initial information entered,
OBSCIS will automatically calculate a mandatory release date, with the
understanding that it could be affected by inmate actions while incarcerated.
Once a commitment record is entered into OBSCIS, the responsibility for
recording subsequent diminution credits and related activity is that of either a
DOC regional commitment office or case management staff at the DOC
institution, depending on the activity. The four regional commitment offices
(organized geographically) are responsible for updating diminution credit
information such as for dual cell occupancies, infractions, and sentence
modifications. Any action taken by a regional office is documented in OBSCIS
and in a hardcopy “commitment file.” An institution’s case management staff
updates OBSCIS for inmate work and educational assignments, for which it
maintains a hardcopy “case management” file. After each credit update, OBSCIS
automatically recalculates an inmate’s mandatory release date. All updates are to
6

be consistent with DOC policies, and are to be based on State law and past legal
interpretations of case law.
Although OBSCIS automated some processes and calculations, the system is still
highly dependent on manual data entry, from the initial establishment of each
inmate’s commitment record, to recording work and educational assignments.
Furthermore, two different hardcopy files are maintained for each inmate: the
regional commitment file and the institution’s case management file. Neither file
includes all the support for the information entered into OBSCIS; therefore, both
files must be considered collectively to reflect a true picture of an inmate’s credit
situation. DOC instituted a number of controls to help ensure the reasonableness
of the impact of actions taken by all parties that affect the frequently-changing
mandatory release dates. Their controls require the following procedures:
¾ An independent review of the initial calculations of the diminution credits
and original mandatory release dates is performed by other commitment
staff.
¾ Once every two years, selected inmate case management files at each
institution are audited by independent case management staff to assess
each institution’s compliance with case management practices and
procedures.
¾ Periodically, commitment staff review OBSCIS generated excess credit
reports. This allows commitment staff to adjust credit records when an
inmate earns more than the legally allowed 20 credits in a month.
¾ A review of an inmate’s commitment and case management files, and the
OBSCIS release date, is performed when an inmate is transferred between
DOC regions (such as Hagerstown and Eastern).
¾ Annual regional commitment office evaluations are performed by
independent commitment staff to assess each office’s compliance with
commitment policies.
¾ DOC regional commitment staff perform desk reviews of all mandatory
supervision releases. OBSCIS generates a monthly report of all inmates
eligible for release within the next 60 days. DOC regional commitment
staff review the commitment files and agree credits to detail OBSCIS
information. Supervisory commitment staff also review the cases.
Completion of each review is documented by a certificate, prepared by the
regional office, authorizing the upcoming release of the inmate.
7

8

Audit Scope, Objectives, and Methodology
Scope
We conducted a performance audit to determine if the Division of Correction
(DOC) is properly assigning and calculating diminution credits. We conducted
the audit under the authority of the State Government Article, Section 2-1221 of
the Annotated Code of Maryland and performed it in accordance with generally
accepted government auditing standards.

Objectives
We had two specific audit objectives:
(1)

To determine whether the policies and procedures implemented by DOC
properly reflect relevant State laws, regulations, and court rulings relating
to diminution credits.

(2)

To assess processes and controls over diminution credits and to determine
if inmate releases are based on properly calculated and recorded credits.

Our audit objectives specifically excluded determining the propriety of credits
earned by inmates while in the custody of local governments’ correctional
institutions; the Patuxent Institution, which releases very few inmates based on
diminution credits; and the DPSCS – Division of Pretrial Detention and Services,
because credits are not earned until an inmate is sentenced at trial.

Methodology
To accomplish our objectives, we reviewed applicable State laws and regulations
as well as policies issued by DOC. We also reviewed the applicability of certain
legal cases with the Office of the Attorney General. In addition, we interviewed
personnel at DOC, including certain DOC institutions. We performed tests to
determine the accuracy of information recorded in the Offender Based State
Correctional Information System (OBSCIS) used by DOC to record offender data,
including diminution credits and release codes.
We also selected a statistical sample of inmates released under mandatory
supervision at two DOC institutions during calendar year 2003 to determine the
accuracy of the release dates based on diminution credits. We projected the
results of these tests to all mandatory supervision releases for these two
institutions for that year.

9

Fieldwork and Agency Responses
We conducted our fieldwork from January 2004 to September 2004. The
Department of Public Safety and Correction Services’ response, on behalf of
DOC, to our findings and recommendations is included as an appendix to this
report. As prescribed in the State Government Article, Section 2-1224 of the
Annotated code of Maryland, we will advise the Department regarding the results
of our review of its response.

10

Findings and Recommendations
Policies and Procedures Governing Diminution Credits
Conclusion
Our audit disclosed that the Division of Correction (DOC) developed and
implemented policies and procedures that properly reflect State laws, regulations,
and court rulings governing the awarding and calculation of diminution credits.
Also, we concluded that DOC disseminated this necessary information to all
appropriate employees.

Processes and Controls Over the Calculation of Credits
Conclusion
Based on the documentation found in both inmate case files (commitment files
and case management files) and DOC’s automated Offender Based State
Correctional Information System (OBSCIS), the audit disclosed that diminution
credits awarded were not always accurate. As a result, some inmates were
released on mandatory supervision on dates other than the dates they were eligible
for release based on actual diminution credits earned. Furthermore, while DOC
had established a system of internal controls intended to provide assurance that
recorded credits were accurate, the control system did not detect certain errors we
noted. The audit also disclosed that, at one institution, inmates often earned
credits for a particular job assignment without proper documentation.
Finding 1
Test results indicate that inmates under mandatory supervision were not
always released on the appropriate dates.
Analysis
In some cases, inmates released by DOC under mandatory supervision were
released on the incorrect dates, based on documented diminution credits earned
and recorded in the inmates’ files and OBSCIS. Specifically, our tests of a
statistical sample of 65 mandatory releases from two DOC institutions during
calendar year 2003 disclosed that the dates for 22 releases were in error. The
incorrect inmate release dates for these 22 errors ranged from 112 days before to
24 days after the dates of eligibility for release (if credits had been properly
applied). Exhibit 1 contains a table providing the detailed results of all 22 errors
including days released early or late and the related sentence served by each
inmate. Early release errors related to 17 of the inmates being released from 1 to

11

112 days early. In these cases, the incorrect releases did not appear to have an
adverse impact on public safety since, according to automated criminal justice
records maintained by DPSCS, none of the inmates released early were arrested
for committing a crime during the period between the erroneous release dates and
the proper release dates.
These errors occurred because incorrect credit data were posted to OBSCIS. For
example, work credits were improperly awarded when inmates were unable to
participate in work, such as when inmates were in segregation status. In addition,
DOC did not always correctly post credits awarded for time served while inmates
were confined in local jurisdictions’ jails prior to transfer to a DOC institution.
DOC did not identify these errors primarily because, prior to an inmate’s release,
DOC procedures did not require a review of all hardcopy documentation in both
the regional commitment file and the institution’s case management file. Rather,
the desk reviews performed included only a review of the regional commitment
files and information recorded in OBSCIS. As a result, documentation that
supports credits awarded was not always reviewed.
Projection of Results
Our statistical sample consisted of tests of a random selection, from OBSCIS, of
calendar year 2003 mandatory releases from two independent DOC institutions.
We found OBSCIS to be a reliable source for test selection purposes since our
tests concluded that OBSCIS properly calculated credits based on data entered.
Our test population consisted of 226 releases from one institution, from which we
selected a sample of 31, and 531 releases from the other institution, from which
we selected a sample of 34 releases. For one institution, based on our test results
of 11 errors, we are 90 percent confident that between 22.4 percent and 48.6
percent of the inmates released on mandatory supervisions in calendar year 2003
were not released on the proper dates. For the other institution, also with 11
errors, we are 90 percent confident that between 19.6 percent and 45.1 percent of
the inmates released in calendar year 2003 were not released on the appropriate
dates. Since these results are based exclusively on calendar year 2003 mandatory
releases for inmates housed at these two DOC institutions, the results cannot be
projected to the remaining DOC institutions or to releases in other periods.
Furthermore, the extent to which errors affect release dates cannot be projected.
Recommendation 1
We recommend that DOC institute a process that, on a test basis, recalculates
the mandatory release dates based on all available documentation currently
maintained by DOC (including an institution’s case management files) prior
to inmate releases. We further recommend that, when applicable, DOC
12

adjust individual inmate release dates based on the test results. We also
recommend that, to the extent the process identifies systemic errors in
calculating diminution credits and mandatory release dates, DOC implement
necessary corrective actions (such as training and changes in policy).

Finding 2
Actual days worked by inmates to earn credits for certain work assignments
at one institution were not documented.
Analysis
One institution’s use of a generic work assignment code to award diminution
credits was not adequately documented. Our test of 25 randomly-selected
inmates that received credits during calendar year 2003 for this work assignment
disclosed that documentation of the actual number of days worked, as required by
DOC, was not on file for any the inmates tested. As a result, we could not
substantiate the propriety of the credits earned by these inmates. According to
DOC’s records, 558 inmates at this institution had this work assignment code
some time during calendar year 2003. Our review of all files maintained by this
institution, relative to this work assignment, disclosed that the documentation
retained for the year supported actual days worked (either partially or in total) for
only 70 of these 558 inmates. According to OBSCIS, these 558 inmates
collectively earned approximately 3,800 days of diminution credits from this
work assignment during calendar year 2003.
DOC established this work assignment code in 2002 to reduce inmate idleness
and to create a non-specialized category of work that would enable those inmates
without regular jobs or duties to earn credit towards earlier mandatory releases.
DOC’s policy requires any institution with an inmate idleness rate of 20 percent
or greater to establish an inmate labor pool. Inmates in this labor pool are to
perform non-specialized tasks, such as snow removal, that are not covered under
existing inmate job classifications, and are eligible to earn up to five credits per
month. However, unlike other work assignments, this policy requires that
institutions document that inmates in the labor pool actually worked at least three
days per week to earn these diminution credits, and that the institutions retain the
related documentation.
According to case management staff at this institution, inmates were primarily
assigned to the labor pool in order to receive diminution credits towards earlier
release dates so as to reduce chronic inmate overcrowding at the institution.

13

Recommendation 2
We recommend that DOC ensure that all its institutions comply with existing
requirements concerning the use of the labor pool job classification and that
diminution credits earned are properly supported by documentation of days
actually worked.

Finding 3
DOC did not have a formal policy for reviewing the reasonableness of the
number of diminution credits to be earned for specific work tasks to ensure
consistency between institutions.
Analysis
DOC did not have a formal policy addressing the amount of diminution credits to
be earned for specific work tasks, which resulted in inconsistent credit earning
rates for similar work at different institutions.
New work tasks and the related credits are recorded in the OBSCIS database by
DOC after the submission by an institution’s case management staff (each
institution’s work task has a unique OBSCIS code). The submissions are to be
approved by supervisory personnel at the institutions and at DOC headquarters;
however, the approval process does not consider either the reasonableness or
consistency of credit earning rates between the new task and existing tasks at
other institutions.
Our test of all work tasks recorded in OBSCIS disclosed 19 tasks where DOC
institutions awarded differing diminution credits. For example, six of eight
institutions with the job task “painter” awarded 10 diminution credits for each
month an inmate satisfactorily completed the assignment; the two remaining
institutions awarded 5 credits each month. Therefore, an inmate performing a
similar job at one institution could receive a greater number of diminution credits
than an inmate at another institution, and, all other things being equal, would earn
an earlier release date under mandatory supervision. While there may be valid
justifications for different credits among institutions, documentation was not
available to explain the differences we noted.

14

Recommendation
We recommend that DOC develop a formal policy for new work tasks that
requires written justifications for the amount of diminution credits to be
awarded, and that requires appropriate supervisory approvals. We also
recommend that such justifications be retained for future verification
purposes. We further recommend that the DOC review existing job tasks
with differing credits, and make appropriate adjustments to promote
consistency, as warranted.

Finding 4
Access controls over critical information contained in OBSCIS were not
adequate.
Analysis
DPSCS did not adequately protect certain critical automated inmate credit data.
A review of OBSCIC system access reports disclosed that a large number of
individuals had transaction screen capabilities that allowed them to modify
critical data files related to inmate diminution credits even though it appeared that
such accesses were not needed for them to perform their job duties. While only
certain central DOC commitment office, regional commitment staff, and
institution case management staff generally need modification access, access also
had been granted to such employees as wardens and correctional officers.
Specifically, we noted the following conditions:
¾ Five of 57 users with access to commitment and diminution credit
screens—used to record information such as inmate eligibility for certain
credits—did not require such access. Generally, these screens should be
limited to DOC central and regional commitment staff employees.
¾ Approximately 900 users with access to case management screens—used
to record individual inmate work task and education assignments—most
likely did not require access. These screens should be limited to case
management staff at the institutions (which we estimated at 240
employees in total). While DOC represented that certain of these 900
users needed access, they had not documented the propriety of assigning
access capabilities to these individuals.
Although we concluded that a significant number of users do not require access,
we were not able to readily determine whether these users included individuals no
longer working for the Department.

15

Recommendation 4
We recommend that only personnel whose job duties require direct
modification access to critical data files be granted access to such files. We
further recommend that DOC perform a review of current users with access
to critical screens and take appropriate actions to remove screen accesses for
those users who do not require it. Finally, we recommend that DOC
periodically assess and document the propriety of employee access
capabilities to critical files.

16

Exhibit 1
Detailed Findings of Incorrect Inmate Release Dates

Release Errors from Statistical Sample (see Finding 1)
Detail of Incorrect Release Date and Related Sentence Served
Early Release
Late Release
Inmate
Sentence
Sentence
Days
Days
Served (years)
Served (years)
1
112
4.6
2
87
3.4
3
69
17.6
4
45
4
5
37
4.8
6
32
20.4
7
29
24.7
8
16
16.3
9
16
5.9
10
14
19.9
11
5
1
12
4
7.2
13
4
1.8
14
3
19.4
15
2
8.4
16
1
6.5
17
1
2.7
3
3
18
19
9
8.6
20
11
5.4
14
1.4
21
22
24
6

17

Department of Public Safety and Correctional Services
Division of Correction
6776 REISTERSTOWN ROAD • SUITE 310 • BALTIMORE, MARYLAND 21215-2342
(410) 585-3300 • FAX (410) 764-4182 • TTY USERS 1-800-735-2258 • www.dpscs.state.md.us
STATE OF MARYLAND
ROBERT L. EHRLICH, JR.
GOVERNOR
MICHAEL S. STEELE
LT. GOVERNOR

December 6, 2004

MARY ANN SAAR
SECRETARY
MARY L. LIVERS, Ph.D.
DEPUTY SECRETARY
FOR OPERATIONS

DIVISION OF CORRECTION

FRANK C. SIZER, JR.
COMMISSIONER

Mary Ann Saar, Secretary
Department of Public Safety and Correctional Services
300 East Joppa Road, Suite 1000
Towson, Maryland 21286-3020
Via

BOBBY SHEARIN
DEPUTY COMMISSIONER

EASTERN
CORRECTIONAL
INSTITUTION
MARYLAND CORRECTIONAL
ADJUSTMENT CENTER

Mary L. Livers, PhD, Deputy Secretary
Department of Public Safety and Correctional Services
300 East Joppa Road, Suite 1000
Towson, Maryland 21286-3020

MARYLAND CORRECTIONAL
INSTITUTION-HAGERSTOWN
MARYLAND CORRECTIONAL
INSTITUTION-JESSUP
MARYLAND CORRECTIONAL
INSTITUTION FOR WOMEN
MARYLAND
CORRECTIONAL
PRE-RELEASE SYSTEM
MARYLAND
CORRECTIONAL
TRAINING CENTER
MARYLAND HOUSE
OF CORRECTION
MARYLAND HOUSE OF
CORRECTION ANNEX
MARYLAND RECEPTION,
DIAGNOSTIC AND
CLASSIFICATION CENTER
METROPOLITAN
TRANSITION CENTER
NORTH BRANCH
CORRECTIONAL
INSTITUTION

Dear Secretary Saar:
The draft performance audit on the Department of Public Safety and Correctional
Services – Division of Correction (DOC) Diminution Credits has been received and
reviewed. The following represents the agency response to the auditor’s findings.

Finding # 1 - Test results indicate that inmates under mandatory supervision were not
always released on the appropriate dates.
Agree: - DOC will institute a process that, on a test basis, recalculates the mandatory
release dates based on all available documentation currently maintained by DOC
(including institution’s case management files) prior to inmate releases. When
applicable, DOC will adjust individual inmate release dates based on the test results. In
addition, to the extent the process identifies systemic errors in calculating diminution
credits and mandatory release dates, DOC will implement necessary corrective actions
(such as training and changes in policy).

ROXBURY
CORRECTIONAL
INSTITUTION
STATE USE
INDUSTRIES
WESTERN
CORRECTIONAL
INSTITUTION

Finding # 2 - Actual days worked by inmates to earn credits for certain work
assignments at one institution were not documented.
Agree: - DOC will ensure that all its institutions comply with existing requirements
concerning the use of the labor pool job classification and that diminution credits
earned are properly supported by documentation of days actually worked.

AUDIT TEAM
Edward L. Shulder, CPA
Audit Manager

Brian K. Roberts, CPA
Senior Auditor

Nathan L. Cooke
Benjamin P. Gill
Joseph E. Walzog
Staff Auditors