Much of the increase in the prison and jail population in recent years can be attributed to the nationwide public policy of mandatory sentencing for drug offenders; indeed, the proportion of drug offenders in the Federal Bureau of Prisons is expected to increase from 47% in 1991 to 70% by 1995 [5]. This policy has also resulted in a marked rise in the number of inmates in state and local systems. The New York State prison population, for example, rose from 20 000 in 1979 to 59 000 in 1991. A large proportion of inmates are intravenous drug or crack users [5-8].

Inmates are generally poor, undereducated, and over-represented by minorities [6]. Their home communities often have limited medical care access, especially for primary care, prevention, and early detection and treatment. Because of their lifestyles, prisoners present added difficulties in continuity of care and compliance. Given these problems, the period of confinement provides a unique chance to reach an otherwise elusive group, whose risk factors and infection prevalence rates far exceed those of other populations [5, 7, 9-18]. For example, the New York State 1991 release cohort with the highest risk for communicable disease because of drug use had been in prison an average of 16 months.

Confinements of such duration are sufficient for comprehensive primary care interventions that directly serve the persons involved. Such medical care can be justified on both humanitarian and public health grounds. Inmates released to their home communities bring communicable diseases that pose risks to public health that may be reduced by public-health-oriented prevention and treatment.

To underscore the public health risks involved, we briefly review the effects in U.S. prisons and jails of several communicable diseases, including human immunodeficiency virus (HIV) infection, Mycobacterium tuberculosis infection, hepatitis B virus infection, syphilis, gonorrhea, and chlamydial infection (Table 1).

Human Immunodeficiency Virus Infection

The annual aggregate incidence of the acquired immunodeficiency syndrome (AIDS) for all state and federal prison systems was reported to be 202 cases per 100 000 inmates in 1989, which is more than twice the 1988 rate [11]. Moreover, the rate far exceeds the overall estimated incidence of 14.7 cases per 100 000 in 1989 for the U.S. population.

The prevalence of HIV infection among inmates varies nationwide [10, 13, 19-27]. Vlahov and coworkers [10] found HIV-1 seroprevalences of 2.1% to 7.6% for men and 2.7% to 14.7% for women consecutively entering 10 unidentified correctional systems in the United States. In various studies of incoming state inmates, seropositivity rates were less than 1% in Hawaii, Indiana, Michigan, New Mexico, Washington, and Wisconsin, approximately 3% in California, and approximately 7% in Maryland [13]. Higher prevalence rates were found among men (18.0%) and women (26.3%) entering the New York City Correctional System in 1989 [25]. Seroprevalence rates for adolescent male and female inmates were 2.8% and 14.3%, respectively [25]. Use of intravenous heroin was most strongly related to HIV-1 seropositivity. High prevalence rates of HIV-1 infection were also found among New York State prison inmates [26]. Of the 8000 New York State inmates estimated to have HIV infection, 4000 were released in 1991.

Tuberculosis

In 1978, Stead [14] reported that tuberculosis was spread by inmates to their home communities after release from correctional facilities with outbreaks. Since then, the incidence of tuberculosis, in both correctional institutions and urban communities, has risen dramatically because of reactivation of M. tuberculosis due to HIV-related immunosuppression and secondary spread [13, 28-30]. A survey of 29 state health departments revealed that during 1984 and 1985, the incidence of tuberculosis among inmates was more than three times higher than that for nonincarcerated adults 15 to 64 years of age [30]. Since 1985, the national case rate for tuberculosis has increased alarmingly [31]. Incidence-trends analysis shows that approximately 28 000 excess cases occurred from 1985 through 1990. The largest increases were reported in cities with populations over 250 000, such as New York City (+ 975 cases, + 38%), Los Angeles (+ 68 cases, + 8%), and Oakland (+ 62 cases, + 51%) [31]. More than 4000 cases of active tuberculosis were reported in New York City during 1991; in 30% of these cases, patients were resistant to at least one antituberculosis drug. Nosocomial transmission of an M. tuberculosis strain resistant to multiple first-line drugs has been reported in at least three New York City hospitals [32]. These outbreaks were associated with a high case-fatality rate and included inmates in a secure unit at one hospital.

The incidence of active tuberculosis among New Jersey prison inmates was 110 cases per 100 000 during 1987 as opposed to 10 per 100 000 in the general population. In California, these incidence rates were 80 and 13 cases, respectively [13]. The incidence of active tuberculosis among New York State inmates reached 200 cases per 100 000 during 1991, an increase of almost 900% over the period 1977 to 1980. In 1991, 95% of inmates with tuberculosis were co-infected with HIV. Thirty-two percent of 111 isolates tested were multidrug resistant, and some strains were epidemiologically linked by restriction fragment length polymorphism testing done by the Centers for Disease Control (CDC) [33].

The prevalence of induration of more than 10 mm on tuberculin skin tests is high among inmates [6, 34-38]. Skin tests were positive in 12% of Maryland inmates, in 14% of New Mexico inmates, and in 24% of New York State inmates (see Table 1). Because of the high prevalence of anergy among inmates, these figures may represent underestimations [38-40]. The high prevalence of positive tuberculin skin tests among New York State inmates contrasts with a 6% skin-test positivity rate among recently tested correctional employees. This higher prevalence may reflect rates in inmates' home communities: Eighty-five percent of this group previously resided in New York City or in the five surrounding counties. Although tuberculosis outbreaks among New York State inmates [30, 33] may have resulted in some of the increase in the inmate tuberculosis case rate, most of the rise is reflective of an increased community case rate.

Other Communicable Diseases

Inmates have a high prevalence rate of HBV serologic markers [7, 41-43]. The rate of positive markers among male prison inmates was 19% in Wisconsin, 29% in Tennessee, and 47% in New Mexico. Although small subgroups, such as inmates serving unusually long sentences or those using injectable drugs while incarcerated, are at risk for new infection [41], the overall incidence of new HBV infection during incarceration is low [41, 42]. Many inmates are at risk for new infection after release, however, because of resumption of drug use and promiscuous sexual activity [44, 45].

Raba and Obis [16] found that more than 5% of male detainees entering Cook County Jail in Chicago had positive urethral cultures for Neisseria gonorrhoeae, suggesting an annual incidence at least 11 times greater than that in the U.S. population. Most men were asymptomatic and would have posed a risk for transmission after release if they had not been properly screened and treated. Similar rates were demonstrated in male and female entrants to the New York City Jail System in the 1978-1980 period [17]. Lower rates of gonorrhea were detected among U.S. federal prisoners [18], male prisoners in Maryland [46], and female inmates in Pennsylvania [47]. Various studies of incarcerated adolescents [48-50] found an asymptomatic gonorrhea prevalence rate of approximately 2% for boys and 7% to 11% for girls. Approximately 15% of male New York City Jail inmates are currently prescribed empiric treatment for gonococcal and nongonococcal urethritis because of the presence of pyuria (Heyman B. Personal communication). Chlamydial infection was found in 20% of female adolescents in juvenile detention centers in Washington [50] and in 5% of women in Pennsylvania prisons [47]. Women with asymptomatic gonococcal or chlamydial infection, or both, who go unscreened and untreated are more likely to spread infection after release and to suffer ectopic pregnancies, infertility, and chronic pelvic pain.

Rates of true-positive tests for syphilis among inmates have been found to be approximately 1% for boys [48], 0% to 2.5% for girls [47, 48], 1.9% to 3% for men [16, 18, 51], and 5.4% to 20% for women (18; Heyman B. Personal communication). A recent increase in the rate of syphilis has been linked to use of illicit drugs and prostitution or contact with a prostitute [52]. Female prison inmates have significantly higher rates of cervical dysplasia on Papanicolaou smears than are found in local sexually transmitted diseases clinics [47]. Early forms of cervical intraepithelial neoplasia have also been documented among incarcerated girls [53]. The prevalence of human papillomavirus infection among female inmates is unknown; this virus has been associated with cervical carcinoma. Human T lymphotropic virus types I and II are also prevalent among jail and prison inmates [54]. These latter viruses may be spread sexually or through intravenous drug use.

The Correctional Health Care System

A 1972 survey of jail health services conducted by the American Medical Association showed that only 6% of facilities routinely examined all incoming inmates; more than 25% had no regularly scheduled physician visits, and 11% did not have available on-call physicians [55]. A 1983 survey of health services in juvenile detention and correctional facilities conducted by the National Commission on Correctional Health Care showed that almost 20% of such facilities did not provide regular sick call, 39% were not conducting an initial medical screening at admission, and 51% and 42% were not providing on-going mental health and dental services, respectively [56]. Clearly, as Prout and Ross [57] contend in their study of prison medicine, past efforts have not succeeded in guaranteeing the provision of needed medical care to inmates [57]. Substantial legal authority exists, however, for mandating access to medical services for inmates. In Estelle v. Gamble (429 U.S. 97, 104-05 [1976]), the Supreme Court affirmed an inmate's constitutional right to medical care, citing the Eighth Amendment's prohibition of cruel and unusual punishment. Since 1976, this decision has served as the precedent for lower court deliberations and decisions.

Despite the affirmation of the right to access to appropriate care, professional services may be restricted because of the security imperative and coercive nature of the correctional system. It is often impossible to distinguish voluntary from involuntary behavior in the correctional setting because many actions may result from intimidation by guards or inmates, maneuvering for limited goods or political favor [58]. It is often difficult to distinguish between a refusal of care and a denial of care. Rejections of care may be responses to inadequate communication between provider and patient [58]. It may also be difficult to guarantee confidentiality. Privacy is virtually nonexistent, because inmate movement (for example, to specialty clinics) is controlled, and personal effects (for example, vials of zidovudine) are frequently scrutinized as a security precaution. Inmates are required to adhere to general schedules for work, exercise, and diet. They have restricted access to routine over-the-counter medications.

Scarce medical resources are allocated to inmates through screening or triage systems. Even so, access to specialty care and hospitalization is frequently limited by the resistance of community providers to care for people perceived to be dangerous or unclean. Particularly when guarded and physically restrained, the inmate in a community facility is conspicuous and is regarded primarily as a prisoner rather than as a patient. Current federal guidelines for inmate research limit the participation of inmates in clinical trials and epidemiologic studies [59].

More fundamental than these special problems in the care of inmates is the lack of available capital and human resources, which results in a marked variation in organizational structures of prison health services [30]. Health professionals are responsible to the wardens of individual prisons in 10 states (those that are generally less populous). Medical staff in these institutions may be prone to become involved in custodial and supervisory functions such as body cavity searches, the monitoring of health conditions of inmates under discipline, forensic testing for drug and alcohol use, forensic psychiatric evaluations, security restraints, and even execution. Other states have more independent organizational models in which health services are operated solely by a centralized department of corrections, other state agencies, or independent contractors [30].

Few data are available on physician-to-inmate ratios, full-time to part-time physician ratios, use of on-site or off-site consultants, and credentialing practices in jails and prisons. Large variability exists in overall health staffing (nurses, physicians, dentists, and so forth) among states of similar size [22]. Recently reported staffing ratios ranged from 1:29 (Arizona) to 1:49 (Alabama) among the five states with unified medical and mental health services and similar numbers of inmates. Staffing patterns may be influenced by the characteristics of the institution (for example, intake unit, prerelease unit, or a segregated maximum security unit requiring daily rounds by health care personnel), types of inmates (for example, geriatric, physically handicapped, on hemodialysis, or HIV-infected), and the health delivery system available (for example, inpatient service, laboratory, or radiology services).

Recruitment of physicians is often difficult because of salary structures, benefit packages, working environments, and the rural location of many prisons. Strategies for successful recruitment and retention of qualified health professionals to work in correctional institutions have been proposed: These include creating a program such as the National Health Service Corps [14]; improving salary scales; and providing educational subsidies, release time for teaching at an affiliated university, family leave, and child care programs [30]. Availability of employee assistance programs for short-term counseling and referral services for employees with personal problems that affect their job performance may be additional attractive benefits [30]. Other strategies include the offer of flexible work schedules and the wider dissemination of the fact that rates of assaults by prisoners against staff are low. Of the almost 300 000 employees working in state and federal corrections on 1 January 1990, fewer than 1% had been involved in an assault during 1989 that required any medical attention [30]. Appropriate advertising in professional journals, through targeted mailings, and by in-person solicitations are also necessary [30].

Despite these limitations, health services in U.S. jails, prisons, and juvenile confinement facilities have been improving. The National Commission on Correctional Health Care, a not-for-profit organization incorporated in 1983, has provided the framework to improve quality of correctional health services by 1) developing and maintaining standards for correctional health services; 2) offering a voluntary accreditation program for correctional institutions that meet its standards; 3) establishing policy statements relevant to the correctional health care setting; 4) certifying correctional health professionals; 5) offering continuing education programs for correctional health professionals; 6) conducting studies on correctional health topics; and 7) developing and disseminating information about the correctional health field. The Commission currently supports itself through fees for these services and a CDC grant on juvenile health training. These Commission-sponsored interventions have promoted an environment more conducive to public health interventions.

Current Program Needs

In the best jurisdictions, inmates usually undergo an initial examination within 24 hours of incarceration. The medical history is obtained and a physical examination is done. Laboratory studies generally include a leukocyte count, urinalysis, a tuberculin skin test, and a rapid plasma reagin test for syphilis. The availability of voluntary HIV antibody testing varies among facilities [7].

HIV Programs

Various HIV testing and counseling programs have been successful in jail and prison facilities. These programs are being linked increasingly with expanded HIV-related medical care services. The prison setting has been an effective place for medical intervention. The proportion of New York State male prison inmates with AIDS surviving at least 6 months rose significantly, from 42% in 1985 to 64% in 1989 [61]. In 1988, the rate of AIDS-related mortality peaked at 41 deaths per 10 000 inmates [62] and declined to 35 per 10 000 after the introduction of zidovudine therapy and prophylaxis for Pneumocystis carinii pneumonia. Zidovudine use increased from 400 inmates in September 1989 to 1700 inmates in December 1991. Zidovudine is currently estimated to be indicated in 3000 to 4000 New York State inmates with CD4⁺ cell counts below 500 cells/mm³ (unpublished data). Reaching this larger group is limited by voluntary testing (many inmates choose not to be tested) and by poor compliance with drug treatment protocols by inmates.

According to the public health view of correctional health, a proactive stance must be taken to serve a broader community than the inmate's own. With specific attention to communicable disease, prevention and treatment will yield broad social benefits. To achieve success with regard to HIV infection, certain components are necessary, including aggressive voluntary and confidential testing, nondiscrimination, mainstreaming of infected inmates, presymptomatic treatment to decrease morbidity and mortality, and education of both inmates and staff [63]. Maintenance of personal rights and education have been keys to the success of these programs [64-67].

Prison inmates are knowledgeable and are more likely than noninmates to view themselves as being at risk for acquiring HIV infection [68]. Educational materials about HIV can be designed as part of mandatory programs for both staff and inmates. Programs should be in the appropriate language or languages, culturally sensitive, and designed for various media [14, 30, 64-66]. Inmates themselves can be effective health educators [66]. Education to reduce the risk for HIV infection from intravenous drug use and sexual activity should be explicit [14]. Condom distribution should be part of correctional HIV prevention programs [66]. Giving clean drug injection equipment to inmates is not suggested, but inmates should be provided with information they can use to protect themselves from HIV infection after release [30]. Programs should address the special needs of female inmates. Such programs should include education about safer sex techniques, risks for perinatal transmission, and strategies to gain greater control over sexual decision making [14]. Prenatal care and counseling, and, if desired, abortion referrals, should be available to pregnant women [14, 66].

Tuberculosis Control

Although more inmates are now receiving either treatment or prophylaxis for tuberculosis (for example, 5000 New York State inmates in early 1992), containment of tuberculosis reactivation, subsequent intramural spread, and postrelease transmission cannot be addressed solely by prophylaxis. Previous efforts have suggested that control of tuberculosis outbreaks in correctional facilities may result in diminution of subsequent community spread [10]. Many state correctional facilities have instituted yearly tuberculin skin testing among incarcerated persons. Some of these institutions are considering either the institution of preventive therapy with isoniazid for anergic, HIV-infected inmates (after exclusion of active tuberculosis) or routine isoniazid prophylaxis for all HIV-infected inmates [34, 39, 69]. The use of supervised biweekly isoniazid prophylaxis has enhanced the effectiveness of correctional tuberculosis control efforts in New York State prisons.

Other Infection Control Programs

Infection control programs must be aggressively promoted to ensure appropriate utilization of respiratory isolation, restrictions on transfer of ill inmates, rigorous surveillance, and contact tracing [70]. Employee health programs must be promoted to provide postexposure prophylaxis for tuberculosis, hepatitis B virus infection, and HIV infection. Immunization programs administered to staff and inmates provide an immediate and effective preventive intervention [71-73]. In the case of hepatitis B virus, selective screening and vaccination of susceptible inmates with a history of intravenous drug use, previous jaundice, hepatitis, or transfusion may be more cost effective than mass screening and vaccination [43].

Continuity of care after release, especially for tuberculosis control, will help improve compliance with preventive therapy and treatment of active infection, thereby reducing the emergence of drug resistance and secondary transmission in the community. Attention to continuing HIV-related education among drug abusers modifies drug injecting and sexual behavior [74].

Education and Research

Medical schools in New York, Wisconsin, Texas, and other states have relationships with correctional systems; increasing educational opportunities for correctional physicians and for residents in internal medicine and family medicine programs can help improve correctional health programs. The Department of Health and Human Services should modify federal policies on prisoner access to clinical trials, thus promoting the availability of investigational drugs through these academic centers [14, 75]. A National Institute of Corrections inquiry panel concluded that current regulations permit prisoners to participate in therapeutic clinical trials as long as they are not part of a placebo or otherwise untreated control group [75]. Further research on the epidemiology of infectious disease among incarcerated persons (including women and juveniles) will help elucidate improved infection control techniques.

The correctional health care system is positioned to help meet broad public health imperatives through treatment and prevention of highly prevalent diseases. Without such attention, these diseases will pose a risk to the communities to which the inmates return. A public health approach reinforces the need to examine critically the national drug control strategy, which emphasizes incarceration as a method to control the supply of illicit drugs and to treat drug abuse; this policy should be reconsidered, because the human and fiscal cost of imprisonment may be more than the benefit derived for society [76].

Although 83% of intravenous drug users identified through outreach programs have been in jail or prison at some time, 81% and 40%, respectively, have never been in drug treatment [60]. Because the period of incarceration can provide timely access to drug treatment, all prisoners who request such treatment should receive it.

Notwithstanding future changes in drug control policies and in the laws that mandate minimum sentences for drug offenses, correctional health programs must be designed and funded to reflect the personal health care needs of incarcerated persons. Because of the high prevalence of communicable diseases, this is both a public and individual health issue. Budgets should be adequate to address prevention, drug treatment, and medical care for communicable diseases, especially HIV infection, hepatitis B, and sexually transmitted diseases. Special attention should be given to the specific needs of women and mentally ill inmates [76].

Correctional health programs need standards against which their performance can be monitored for continual quality improvement (such as those of the National Commission on Correctional Health Care) [77]. Correctional health programs also need to be brought into the mainstream of clinical medicine and public health; this can be accomplished by fostering ties between correctional health programs and academic and public health departments and by funding research that addresses public policy questions peculiar to inmates and the epidemiology of disease among incarcerated persons.