Seven federal corrections facilities participated in this prospective pilot assessment. One facility that had demonstrated the capacity to implement the evaluation was chosen from each administrative region and one facility volunteered to participate after numerous infected contacts were identified in a large TB contact investigation.

The primary end point for the pilot assessment of INH-RPT was treatment completion defined as receiving at least 11 weekly doses within 16 weeks. The secondary end point was discontinuation of INH-RPT regimen, especially due to signs or symptoms associated with treatment. Participant information was collected during July 2012 to February 2015 and included demographic information, medical and social risk factors, weekly dose and symptom review, adverse events, laboratory monitoring, and final treatment disposition.

Candidates eligible to participate in the pilot assessment included any inmate with a positive result from a tuberculin skin test or an interferon-gamma release assay and no previous treatment of LTBI or TB. Exclusion criteria included inmates living with human immunodeficiency virus (HIV) infection on antiretroviral therapy (due to contraindications for INH-RPT use in combination with antiretroviral therapy); having confirmed or suspected TB disease; being a contact to an index case known to have TB resistant to rifampin or INH, pregnancy; being on warfarin, or on antiepileptic drug therapy; or having hypersensitivity to any of the rifamycins (eg, rifampin, rifabutin). To be included in the assessment, participants needed to have a projected release date beyond the anticipated treatment completion date. The protocol included placing participants on “medical hold” (to defer transfer the inmate to another facility during treatment).

All eligible participants were offered the option of once weekly INH-RPT by DOT for 3 months for treatment of LTBI or 9 months of biweekly INH; only those accepting INH-RPT were enrolled in the pilot assessment. Before starting treatment with INH-RPT, participants underwent a medical assessment including a complete history and a targeted physical examination, laboratory evaluation for hepatitis B virus surface antigen, anti-hepatitis C virus antibody, HIV antibody, liver transaminases, complete blood count with platelets, and a chest radiography unless there was documentation of a negative chest radiograph within the prior 3 months for non–HIV-infected participants and within 1 month for HIV-infected participants.

The INH-RPT regimen was administered weekly by DOT. Monthly liver transaminases were obtained on all participants throughout the treatment phase and once additionally at the completion of treatment. Treatment was discontinued if liver transaminases exceed 3 times the upper limit of normal if associated with symptoms of hepatitis (eg, nausea, vomiting) and 5 times the upper limit of normal if the participant was asymptomatic. Treatment was also discontinued if the participant experienced severe adverse events that may have been a drug hypersensitivity reaction (eg, hypotension).

Statistical analyses were conducted using SAS 9.3 (SAS Institute Inc, Cary, North Carolina). Fisher exact test was used to assess the associations between patient sociodemographic characteristics and symptoms or clinical signs with discontinuation of treatment. A P value of less than .05 was considered statistically significant.

The BOP Research Review Board determined that this was an evaluation project using a recommended LTBI treatment regimen and did not require a full review and approval. The Centers for Disease Control and Prevention determined that the assessment was standard public health practice.

A total of 463 participants started treatment with INH-RPT. The median age was 36 years (range: 20-71 years); 70% were male. The population treated was 12% white, non-Hispanic, 8% black, non-Hispanic, 2% Asian, 1% American Indian, and 77% Hispanic. Two-thirds of participants were foreign-born and 19.9% were contacts to a known TB case (see Supplemental Digital Content Table, available at http://links.lww.com/JPHMP/A481).

Of the 463 inmates treated, 424 (92%) successfully completed treatment. Reasons for discontinuation for 39 (8%) inmates included the following: 17 (44%) signs/symptoms associated with treatment, 9 (23%) transfer or release, 8 (21%) treatment refusal, and 5 (13%) provider error. Five patients who stopped INH-RPT because of symptoms subsequently completed treatment using either INH or rifampin. Asian race was the only characteristic significantly associated with discontinuation of treatment due to an adverse event signs/symptoms related to treatment (Table 1).

Half of the inmates treated reported signs/symptoms while on treatment and 42% of those reported 2 or more signs/symptoms. Overall, the most frequently reported signs/symptoms included fatigue (16%), nausea (13%), and abdominal pain (7%) (Table 2). For those inmates who discontinued treatment because of signs/symptoms, 94% reported experiencing at least 1 sign or symptom, and 71% reported experiencing 2 or more signs/symptoms. The most common signs or symptoms reported among those who discontinued treatment because of an adverse event were nausea (47%), abdominal pain (47%), fever/chills (24%), and sore muscles (24%). Signs/symptoms significantly associated with discontinuation of treatment included abdominal pain (P < .001), appetite loss (P = .02), fever/chills (P = .01), nausea (P = .03), sore muscles (P = .002), and liver transaminases exceeding 5 times the upper limits of normal (P = .03) (Table 3). There was no evidence of cluster effect based on site.

In this pilot assessment of inmates in federal corrections facilities receiving the INH-RPT regimen for LTBI given by DOT, the treatment completion rate was 92%. This completion rate was slightly higher than but comparable with those seen in other studies of the INH-RPT regimen including a community INH-RPT clinical trial (82%), a study in a jail setting (85%), and a large postmarketing nationwide survey (87%).^11,13,14

Although not assessed here, primary factors contributing to the high treatment completion rate for INH-RPT were thought to include the shorter duration of the regimen and reduced number of doses (76 doses for biweekly 9-month INH vs 12 doses for INH-RPT). In addition, the regimen was generally well tolerated with few patients discontinuing therapy because of signs/symptoms associated with treatment. Recent studies have shown fewer adverse events, including hepatotoxicity, in patients treated with the INH-RPT regimen compared with INH.^13,15 In this pilot assessment, approximately half of the inmates reported experiencing signs/symptoms, but only a small number of these inmates subsequently discontinued treatment indicating overall good tolerability of the INH-RPT regimen.

An additional contributing factor to the high completion rate was thought to be implementation of procedures to prevent treatment discontinuation due to transfer or release. Past studies have shown that completion is negatively affected by transfer within or between correctional facilities and release prior to completion.^6–10 In addition, treatment with INH-RPT was not initiated for inmates with projected release dates before the anticipated treatment completion date. The protocol also included placing inmates on “medical hold” during treatment to prevent transfers to other facilities during the treatment. Only 9 inmates in the pilot who discontinued treatment did so because of transfer to another facility or release before completing treatment.

Finally, another factor that may have supported the high treatment completion rate for INH-RPT was a change in the method of medication administration during the pilot assessment that may have improved compliance with treatment. LTBI treatment using INH had traditionally been administered via “pill-line” in which inmates wait in sometimes lengthy lines to obtain treatment. During the pilot, INH-RPT was administered via a special weekly clinic in which participants saw the same health care provider and were placed on a weekly “callout” to keep their appointment. This change in procedure reduced waiting times to receive medication, increased one-on-one interaction between inmates and their health care provider, and included reminders of their appointment using the “callout” system.

Correctional facility populations are potential reservoirs of infection with M tuberculosis that, if left untreated, can lead to TB outbreaks and transmission to communities.^10,16–20 However, jails and prisons are also highly structured environments that provide strategic public health opportunities for containment of infectious diseases, including TB. Screening for TB upon entry into the facility can detect both LTBI and TB disease effectively in incarcerated populations.²¹ Newly incarcerated inmates diagnosed with LTBI can be treated using DOT, an optimal strategy that may be challenging in community-based clinics. The INH-RPT regimen, evaluated in this pilot, offers a very promising alternative to the standard INH treatment regimens for LTBI with its high completion rates and relatively good tolerability. Broader implementation of the INH-RPT regimen for LTBI in US jails and prisons could substantively enhance national TB control efforts.

This assessment had a few limitations. One important limitation was that relatively small numbers of persons did not complete treatment which limited our ability to assess associations. Also, information on behavioral risk factors (eg, drug use) for LTBI was obtained by self-report, which might have resulted in underestimation of the actual prevalence of these behaviors. Another limitation was the difficulty in tracking outcomes for inmates who were transferred to alternate facilities or released from incarceration; this may have led to an underestimation of treatment completion. Finally, facilities were chosen to participate in the assessment on the basis of a demonstrated capacity to implement the protocol and may not be reflective of facilities throughout the federal prison system; this may have resulted in higher completion rates in these facilities than would be seen among other facilities in the system.

We found in the prison setting that the treatment completion rate for LTBI using the INH-RPT regimen was high and, despite the majority of treated participants reporting 1 or more signs/symptoms during treatment, tolerability of the regimen was good. In addition, the regimen has practical advantages to aid in treatment completion including reduced number of DOT doses and a shorter duration of treatment. The INH-RPT regimen is a viable alternative to standard LTBI regimens in the correctional setting.