Audit of the Federal Bureau of Prisons’ Efforts to Place Inmates Close to Home
Report Number: 25-083 | Date Posted: September 25, 2025 | Division: Audit | Print this page
On December 21, 2018, President Trump signed into law the First Step Act (FSA) to improve criminal justice outcomes and reduce the size of the federal prison population while maintaining public safety. One provision in the FSA amends 18 U.S.C. § 3621(b), requiring the Federal Bureau of Prisons (BOP) to place inmates in facilities as close to their primary residence as possible, and to the extent practicable, within “500 driving miles.” Placing inmates close to home during incarceration reduces recidivism, eases the harm to family members separated from their loved ones, and helps strengthen family ties. The objective of this audit was to assess the BOP’s compliance with the FSA confinement location requirement. Our audit scope covered inmates in BOP custody on September 28, 2024.
BOP’s inmate placement data showed that 36,098, or 33 percent, of the inmates we evaluated were over 500 miles from their release residence on September 28, 2024. However, we found that the BOP used a straight-line, or “as the crow flies” calculation instead of driving miles, as required by the FSA. This resulted in an undercalculation for the inmates evaluated of approximately 8 percent, affecting over 8,600 inmates, as shown in the interactive dashboard below. Read more about the data and our methodology here.
Inmates in BOP Custody in the Contiguous United States on September 28, 2024
The FSA also required that the U.S. Attorney General report to Congress from 2020 to 2025 on the progress made under the FSA, and the reports include the number of inmates housed more than 500 miles from their release residence. However, the information provided to Congress did not reflect driving miles as required by the FSA, because the BOP’s mileage calculations were based on a straight-line distance method.
Additionally, we reviewed a sample of 100 BOP inmates, which included inmates placed both more and less than 500 miles from their residence, to determine if their placements were sufficiently supported. For 26 percent of our sample, we were unable to determine the reason the inmates were placed at their designated facility, particularly when there were comparable facilities closer to the inmate’s residence. Lastly, our review of the BOP’s data identified inaccurate and inconsistent inmate address information, compromising BOP’s ability to calculate distances accurately. Our report contains three recommendations to improve BOP’s inmate placement efforts.
To assess the BOP’s inmate designation efforts, we reviewed applicable laws, regulations, and BOP policies and procedures; interviewed relevant BOP officials; analyzed the September 28, 2024, BOP SENTRY data and inmate documentation; and performed fieldwork at the BOP’s Designation and Sentence Computation Center (DSCC) in Grand Prairie, Texas, to review DSCC’s inmate placement process. Appendix 1 contains additional detail on our audit objective, scope, and methodology.
BOP’s Placement Process
We found that, on September 28, 2024, approximately 41 percent of the BOP inmate population we evaluated were housed more than 500 driving miles from their primary residence.1
Inmates enter, exit, and transfer within BOP continuously, and the BOP must consider multiple factors when placing, or “designating” an inmate to a facility. These factors include bed availability, security level, program needs, faith-based requests, sentencing court recommendations, and other BOP security concerns. The availability of medical and mental health care services at facilities also affects placement. The BOP classifies inmates into four care levels and places them in facilities that meet their medical and mental health needs. Inmates classified at lower care levels have lesser medical and mental health needs than inmates classified at higher care levels.2 As the need for higher levels of care increases, the number of facilities that can meet that need decreases, limiting options closer to inmates’ residences, as reflected in Table 1.
Table 1
Inmate Care Level Data on September 28, 2024
|
|
|||
|---|---|---|---|
| Description | Inmates Evaluated3 | # 500+ Miles Away | % 500+ Miles Away |
| Medical Health | |||
| Care Level 1 | 72,725 | 29,208 | 40% |
| Care Level 2 | 33,268 | 13,590 | 41% |
| Care Level 3 | 2,427 | 1,228 | 51% |
| Care Level 4 | 1,143 | 751 | 66% |
| Mental Health | |||
| Care Level 1 | 104,771 | 42,230 | 40% |
| Care Level 2 | 4,110 | 2,094 | 51% |
| Care Level 3 | 601 | 389 | 65% |
| Care Level 4 | 81 | 64 | 79% |
Source: OIG Analysis of BOP Data
Additionally, the BOP faces fundamental challenges in placing inmates close to their primary residence. As shown in the map below, certain areas of the country lack BOP institutions within 500 miles. Click on the buttons in the top left to see the impact on male or female inmates, and by BOP security level.
Below, the impact is demonstrated by BOP location. The red circles, each of which represents a BOP facility, increase in size with the percentage of inmates more than 500 miles from their release residence. Click on or hover over the map for more details on an institution.
The map below illustrates the effect of the lack of institutions in the northwest and north central parts of the United States. In the map, darker-shaded states indicate a higher risk of inmates from that state being placed more than 500 miles from their release residence. Click on or hover over the map for more details on a state.
The map below provides data at an even more granular level – by congressional district. Both the state map and congressional district map provide valuable information to Congress, primary stakeholders, and decision makers in BOP’s planning and budgetary processes. Click on or hover over the map for more details on a district.
Finally, the dashboard below provides information on inmates by age, ethnicity, security level, race, and sex, which are demographics that the BOP publicly releases on its webpage.4 The data shows, for example, 69 percent of Native American inmates and 51 percent of the BOP’s female population were housed more than 500 miles from their release residence. Contributing factors likely include the lack of facilities in the northwest and north central parts of the United States, where many tribes are located, and the fact that the BOP has fewer female facilities than male facilities. While the OIG does not make recommendations regarding the construction of new facilities, we believe this information is important for the BOP, the Attorney General, and Congress to consider in their planning.
Overall, the interactive dashboards provide different perspectives of the inmate placement data comparing the mileage between the FSA method and the BOP’s straight-line method, providing a comprehensive view of the unique attributes of the inmates in BOP custody on September 28, 2024.
BOP’s Ongoing Efforts to Enhance Placement Decisions
Since the FSA’s enactment, the BOP has worked to house inmates closer to home. For example, the BOP updated its primary designation policy to align with the FSA and identified four facilities for mission changes to create more space near inmates’ release locations.5 Additionally, in May and June 2025, the BOP issued two directives to maximize home confinement availability for those qualifying under the FSA and Second Chance Act.6 Because these actions are ongoing, the impact is not fully reflected in our data, which represents inmates in BOP custody on September 28, 2024. However, in our judgment, these directives could increase the bedspace across BOP facilities, aiding efforts to place inmates closer to home.
Finding 1: The BOP Should Ensure Accurate Calculation of Driving Miles
Both the FSA and BOP policy require the BOP to place an inmate in a facility within 500 driving miles of their primary residence to the extent practicable. The FSA also required that the Attorney General report to Congress from 2020 to 2025 on the progress under the FSA, and the reports included the number of inmates housed more than 500 miles from their release residence. However, the information provided to Congress did not reflect driving miles as required by the FSA because the BOP’s mileage calculations were based on a straightline distance method.7 In January 2025, we informed the BOP of our concern with the BOP’s mileage calculation methodology. BOP officials stated that its current inmate management system, SENTRY, uses an older programming language that cannot integrate driving miles. In September 2021, the BOP awarded a contract to develop and implement a new inmate management system, CICLOPS, to replace SENTRY. BOP officials stated the new system, expected to be functional by December 2026, will use newer technology that would allow the BOP to measure driving miles between BOP institutions and inmate homes. We believe this enhancement is necessary to ensure compliance with the FSA and accurate reporting to Congress. Therefore, we recommend that the BOP ensure its mileage calculation process captures driving miles as required by the FSA.
Finding 2: The BOP Should Properly Document Placement Decisions
To understand the BOP’s inmate placement process and to determine if the BOP placed inmates in accordance with the FSA, we evaluated a judgmental sample of DSCC placements for 100 inmates of the 156,576 inmates in BOP custody on September 28, 2024. The judgmental sample included inmates placed both more and less than 500 miles from their primary residence. For each of the 100 inmates, we observed and verified inmate data in SENTRY and documents in other BOP systems.8 We summarized the results of our testing below.
Table 2
Summary of Inmate DSCC Placement Testing
| Primary FSA Exception | Sufficiently Supported | Insufficiently Supported |
|---|---|---|
| Bed Availability | 9 | 10 |
| Inmate Security Designation | 14 | 2 |
| Program Needs | 5 | 2 |
| Medical and Mental Health Needs | 10 | 1 |
| Recommendations of the Sentencing Court | 13 | 0 |
| Other BOP Security Concerns | 23 | 11 |
| Total | 74 | 26 |
Source: BOP DSCC and OIG analysis
DSCC staff exercise professional judgment and consider various factors when placing inmates. In 74 cases, we found that the inmate’s placement was sufficiently supported, meeting security, medical, and other needs. However, in 26 cases we could not verify that the inmate was placed as close to home as possible. In some cases, we identified multiple options significantly closer to the inmate’s residence that appeared to meet all inmate needs. We asked the BOP designators why, in these cases, the inmates were not placed closer to home. BOP designators suspected placements were likely based on bed availability at the time, or other security or medical concerns, but no documentation supported these decisions in any of the BOP systems or files.
BOP policy requires DSCC staff to note the reason for designation or transfer in the “designator remarks” field in SENTRY. However, DSCC staff explained that SENTRY limits the amount of text that can be entered into that field, and the field can be overwritten, hindering the preservation of placement decisions. A DSCC staff member also stated that once BOP’s new system, CICLOPS, is built, the BOP will explore enhancements to better capture the placement decisions. In our judgment, the BOP should ensure placement decisions are properly documented to demonstrate compliance with the FSA. Therefore, we recommend that the BOP ensure each inmate placement decision is sufficiently supported with proper documentation.
Finding 3: The BOP Should Ensure Address Information Is Accurate and Standardized
To determine the distance between an inmate’s legal residence and their designated facility, the BOP must have the zip codes for both locations.9 BOP’s policy states it is important to have an accurate zip code because it is used to determine the mileage between an inmate’s legal residence and the designated facility. Our review of the SENTRY data found inaccuracies in inmate legal residence addresses, potentially affecting mileage calculations.
First, we identified approximately 8,800 inmate resident zip codes that were either not recognized by the U.S. Postal Service or did not match the inmate’s resident street or city. We discussed this issue with BOP officials, who acknowledged that a process to verify zip codes, which are copied from U.S. Probation Office reports, for accuracy would be beneficial but would require additional time and resources.
We also found over 1,500 instances where the inmate’s legal residence was listed as a government address (e.g., contract detention facilities, BOP institutions, U.S. Marshals Service offices / U.S. Probation Offices, or U.S. Immigration and Customs Enforcement locations). Again, we discussed our concerns with BOP officials, who stated that DSCC previously used addresses reported by the U.S. Probation Office, even if they were government addresses. The BOP believes this practice was not appropriate because the address appears to be a residential address. The BOP stated it now requires DSCC staff to enter either (1) an actual residence address from the Presentence Investigation Report or a U.S. Marshals Service Individual Custody/Detention report, or (2) populate the field as “No Residence” if an actual address cannot be located. During regularly scheduled program review, a Case Manager may change an inmate’s legal address in the Bureau’s Inmate Information System if the inmate provides supporting documentation of a change or correction in legal residence from what was documented in the Presentence Investigation Report.
The BOP’s reliance on information from other agencies, along with SENTRY’s inability to verify or flag nonresidential addresses, increases the risk of inaccurate inmate legal residence addresses. This affects the accuracy of mileage calculations, which rely on the resident address zip code. Therefore, we recommend that the BOP incorporate controls such as address standardization and verification to ensure that each inmate’s legal address is accurate throughout their time in custody and that legal address entries are populated with “No Residence” when appropriate.
Conclusion and Recommendations
While BOP is making efforts to comply with the FSA, we identified areas for improvement. We found that the BOP did not calculate the distance from an inmate’s primary residence to their designated facility using driving miles, which significantly understates the distance inmates are away from their home. Instead, the BOP used a straight-line method to calculate the distance, which also led to inaccuracy in the reports the Attorney General is required to submit to Congress, overstating the reported number of inmates placed within 500 miles of their release residence. Additionally, we could not always verify inmate placements were sufficiently supported. Finally, we identified a need to improve the process for recording the inmate’s address information in SENTRY to promote the accuracy of mileage calculations and the integrity and quality of the data. We provide three recommendations for the BOP to take corrective action and make the necessary improvements.
We recommend that the BOP:
- Ensure its mileage calculation process captures driving miles as required by the FSA.
- Ensure each inmate placement decision is sufficiently supported with proper documentation.
- Incorporate controls such as address standardization and verification to ensure that each inmate’s legal address is accurate throughout their time in custody and that legal address entries are populated with “No Residence” when appropriate.
Appendices
Objective
The objective of this audit was to assess the Federal Bureau of Prisons’ (BOP) efforts to place inmates in a facility within 500 driving miles from their primary residence.
Scope and Methodology
The scope of our audit included inmates in BOP custody on September 28, 2024. To accomplish our objective, we reviewed the First Step Act (FSA) and the BOP’s policies and procedures related to the designation and transfer processes and interviewed BOP officials. We reviewed a judgmental sample of inmate placements and analyzed the SENTRY data for accuracy and reliability. Finally, we calculated the driving miles between the inmate’s primary residence zip code to their designated facility address.
On September 28, 2024, the BOP’s SENTRY data identified 156,576 inmates in its custody. As our audit objective focused on driving miles, we reviewed the mileage data related to inmate residences and BOP facilities that were in the contiguous United States. Therefore, we excluded 47,010 inmates from the 156,576 total inmates in BOP custody.
Table 3
BOP Inmate Population in the Contiguous United States
| Description | Inmate Population |
|---|---|
| Total Inmates in BOP Custody on September 28, 2024 | 156,576 |
| Excluded: inmate residences and BOP facilities outside the contiguous United States (e.g., foreign addresses, Hawaii and Alaska, and U.S. territories), inmates in residential reentry centers (i.e., halfway houses) or on home confinement, inmates not designated by BOP, and inmates who were missing distance data from either the BOP’s or OIG’s mileage calculation.10 | (47,010) |
| BOP Inmate Population in the Contiguous United States | 109,566 |
Source: BOP and OIG
We developed interactive dashboards that illustrate the distance an inmate is away from their primary residence. The dashboards provide this data from various perspectives: (1) inmates’ primary residences by state, (2) inmates’ designated facilities, (3) inmate demographics, (4) facility reach of 500 driving miles, and (5) inmates’ primary residences by congressional district.
Geolocation Analysis, Distance Calculation, and Disclosure Review & Mitigation
The OIG used tools to cleanse and standardize the inmates’ addresses before geocoding. Geocoding is a process which transforms address information to a geographic point on a map. To generate these points, the OIG used the center point of an inmate’s zip code rather than the full address on file because some inmates lacked full address information. The OIG used the standardized and cleansed full address of the BOP facility address to determine its location.
After geocoding, the OIG used software tools to calculate the driving distance from an inmate’s zip code centroid to their assigned facility to measure the movement of cars and find routes that optimize travel time and distance. The model requires driving calculations to follow one-way roads; avoid illegal turns, carpool roads, express lanes, gates, unpaved roads, and roads under construction; and follow other rules that are specific to cars. To enable the model to calculate the distance to some BOP facilities, the OIG also adjusted the settings to enable access to blocked and private access roads.
To ensure the inmate addresses remained confidential, the information in the dashboards was grouped by facility, state or congressional district, and by a distance category.
Statement on Compliance with Generally Accepted Government Auditing Standards
We conducted this performance audit in accordance with generally accepted government auditing standards (GAGAS). Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objective. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective.
Internal Controls
In this audit, we performed testing of internal controls significant within the context of our audit objective. We did not evaluate the internal controls of the BOP to provide assurance on its internal control structure as a whole. BOP management is responsible for the establishment and maintenance of internal controls in accordance with OMB Circular A-123. Because we do not express an opinion on the BOP’s internal control structure as a whole, we offer this statement solely for the information and use of the BOP.11
In planning and performing our audit, we identified internal control components and underlying internal control principles as significant to the audit objective. Specifically, we assessed the design, implementation, and operating effectiveness of internal controls to the extent necessary to address the audit objective. The internal control deficiencies we found are discussed in the BOP’s Placement Process section of this report. However, because our review was limited to those internal control components and underlying principles that we found significant to the objective of this audit, it may not have disclosed all internal control deficiencies that may have existed at the time of this audit.
Compliance with Laws and Regulations
In this audit we also tested, as appropriate given our audit objective and scope, selected records, procedures, and practices, to obtain reasonable assurance that the BOP’s management complied with federal laws and regulations for which non-compliance, in our judgment, could have a material effect on the results of our audit. Our audit included examining, on a test basis, the BOP’s compliance with the following laws and regulations that could have a material effect on the BOP’s operations:
- First Step Act of 2018
This testing included interviewing BOP officials, analyzing SENTRY data, and reviewing supporting documents related to designating an inmate at a BOP facility. As noted in the BOP’s Placement Process section of this report, we found that the BOP’s process to measure the distance from an inmate’s residence to their designated facility did not capture driving miles as required by the FSA.
Sample-Based Testing
To accomplish our audit objective, we performed sample-based testing on the SENTRY records. In this effort, we employed a judgmental sampling design to obtain broad exposure to numerous facets of the areas we reviewed. This non-statistical sample design did not allow projection of the test results to the universe from which the samples were selected.
We requested from the BOP the inmate activity on September 28, 2024. This original file contained 156,576 records. As described above, we excluded inmate addresses and BOP facilities that were not in the contiguous United States. To ensure that inmates were appropriately placed in facilities in accordance with the FSA, we judgmentally sampled 100 inmates and verified documentation that supported the BOP’s decision to place an inmate in a facility.
Computer-Processed Data
During our audit, we obtained information from the BOP’s SENTRY system. We did not test the reliability of the system as a whole, therefore any findings identified involving information from the system were verified with documentation from other sources.
We assessed the reliability of the BOP's SENTRY data by (1) performing electronic testing of select data elements, (2) interviewing BOP officials knowledgeable about the data, and (3) tracing a judgmental sample of data to source documents. We determined that the data is sufficiently reliable for the purpose of responding to our objective.
The Office of the Inspector General (OIG) provided a draft of this audit report to the Federal Bureau of Prisons (BOP). The BOP response is incorporated in Appendix 2 of this final report. In response to our audit report, the BOP concurred with our recommendations and discussed the actions it will implement in response to our findings. As a result, the status of the audit report is resolved. The following provides the OIG analysis of the response and summary of actions necessary to close the report.
Recommendations for the BOP:
- Ensure its mileage calculation process captures driving miles as required by the First Step Act (FSA).
Resolved. The BOP concurred with our recommendation. The BOP stated in its response it is in the process of implementing the Centralized Inmate Case Logistics Operations and Planning System (CICLOPS), which will enable BOP employees to perform similar functions as currently done in SENTRY and is designed to streamline workflows, boost productivity, and simplify tasks. The BOP further stated that when CICLOPS is fully operational, it is expected to capture driving miles in the mileage calculation process. As a result, this recommendation is resolved.
This recommendation can be closed when we receive evidence that the BOP’s mileage calculation process captures driving miles.
- Ensure each inmate placement decision is sufficiently supported with proper documentation.
Resolved. The BOP concurred with our recommendation. The BOP stated in its response that inmate placements are impacted by various factors such as security concerns, medical needs, and bedspace availability and that BOP designators are currently limited by the number of characters that can be entered in the SENTRY remarks data field. Moreover, remarks can be deleted and added by other staff. To implement the recommendation, the BOP intends to have the Correctional Programs Divisions and the Information, Technology and Data Division work together to develop a solution. As a result, this recommendation is resolved.
This recommendation can be closed when we receive evidence that the BOP has ensured it is sufficiently documenting each inmate’s placement decision.
- Incorporate controls such as address standardization and verification to ensure that each inmate’s legal address is accurate throughout their time in custody and that legal address entries are populated with “No Residence” when appropriate.
Resolved. The BOP concurred with our recommendation. The BOP stated in its response that its quarterly refresher training for Designation & Sentence Computation Center (DSCC) staff will include discussion on the verification and address standardization of an inmate’s legal residence. The BOP further stated that the Correctional Programs Branch will meet with the Regional Correctional Programs Administrators to provide institution staff further guidance regarding the review of legal residences at the initial classification and subsequent program review meetings. As a result of these planned actions, through which our expectation is that BOP will identify and implement controls, this recommendation is resolved.
This recommendation can be closed when we receive evidence that the BOP has incorporated controls to ensure an inmate’s: (1) street address and zip code accurately align and, (2) nonresidential address entries are flagged with “No Residence” and tracked appropriately.
1 On September 28, 2024, BOP’s SENTRY data identified 156,576 inmates in its custody. Since the FSA requires the distance to be measured with driving miles, we evaluated the mileage data related to inmate residences and BOP facilities that were in the contiguous United States, which resulted in our evaluating the circumstances of 109,566 inmates. Appendix 1, Table 3 describes the categories of inmates excluded from our evaluation population.
2 Care level 1 inmates are under 70 years of age, generally healthy, and have limited medical needs that can be easily managed by clinician evaluations every 6 to 12 months. Care level 2 inmates are stable outpatients requiring clinician evaluations every 1 to 6 months, with health conditions managed through regular appointments. Care level 3 inmates have complex, chronic health conditions needing frequent clinical contact to maintain stability and prevent complications. Care level 4 inmates require services only available at a BOP medical referral center, which provides 24-hour enhanced medical services and limited inpatient care.
3 Of the 109,566 inmates we evaluated, 3 inmates did not have a medical care level and 3 did not have a mental health care level.
4 BOP, “Inmate Statistics,” September 13, 2025, https://www.bop.gov/about/statistics/statistics_inmate_age.jsp (accessed September 18, 2025).
5 The BOP classifies its facilities into five security levels: minimum, low, medium, high, and administrative. The BOP lowered security levels at FCI Oxford and FCI Memphis from medium to low and, according to the BOP, resulted in 1,070 inmates being placed within 500 miles of the release residence as of February 2025. The BOP also stated it plans to change FCI Estill from medium to low security, but it is undergoing infrastructure renovations expected to be completed by December 2025. Finally, the BOP plans to place female inmates at the FCI Estill satellite camp, which was still not activated as of the conclusion of our fieldwork in July 2025.
6 The Second Chance Act of 2007: Community Safety Through Recidivism Prevention (Pub. L. No. 110-199) aimed to reduce recidivism, increase public safety, and improve the outcome and quality of life of individuals reentering society after incarceration by providing support and resources in four areas: jobs, housing, substance abuse/mental health treatment, and families. The FSA of 2018 reauthorized and expanded the Second Chance Act. 34 U.S.C. § 60501 (2018).
7 The BOP straight-line method uses a spherical distance calculation between the center of the zip codes from an inmate’s residence to the inmate’s designated facility.
8 Other BOP systems include BOP’s internal network drive and the Electronic Inmate Central File, which contained documents such as pre-sentence investigation reports (PSR), program review reports, and Judge and Commitment Orders (J&C).
9 BOP defines the inmate legal residence as the local and state address as reported by the U.S. Probation Office at the time of conviction. However, the inmate’s residence may be updated in SENTRY to the release residence, a verifiable destination where the inmate plans to reside upon release from BOP custody. SENTRY does not differentiate between legal and release residence.
10 For security reasons, the BOP does not publicly release information about an inmate being placed in a residential reentry center or on home confinement; rather, the BOP only publicly identifies the residential reentry management field office overseeing the residential reentry center responsible for housing or monitoring the inmate.
11 This restriction is not intended to limit the distribution of this report, which is a matter of public record.