A Commentary on Age Segregation for Older Prisoners

Abstract

The growing number of older prisoners in state and federal prisons has fostered an important discussion in literature regarding the potential benefits of age-segregated living arrangements for older inmates. This article begins with a brief review of the reasons for America's aging prison population. Thereafter, it uses a multidisciplinary literature review to clarify a 4-point rationale for age-segregated prisons: (a) cost savings via centralized health care for older prisoners; (b) the reduction of civil liabilities for correctional systems that centralize disability services as per requirements of the Americans with Disabilities Act of 1990; (c) the advancement of prisoner safety for older inmates; and (d) the promotion of rehabilitation by advancing treatment opportunities with a group that is most likely to desist from future criminal activity (in part) due to age-related desistance from crime. Conclusions focus on age segregation within the historical context of segregation in prison based on sociodemographic characteristics.

On June 30, 2006, federal and state correctional facilities housed 1,556,518 prisoners in the United States, yielding a rate of incarceration at 497 prisoners per 100,000 U.S. residents (Sabol, Minton, & Harrison, 2007). Although the vast majority of these inmates are relatively young, the number of older prisoners, inmates 50 years of age and above (Morton, 1992), in state and federal correctional facilities is growing exponentially in the United States (Williams et al., 2006). The number of prisoners above the age of 50 more than doubled between 1994 and 2001 (from 50,478 to 113,358 inmates). During this period, the proportion of all older prisoners also expanded from 5.9% in 1994 to 7.9% in 2001 (Camp & Camp, 1994–1995, 1996–1999, 2000–2001). Since 1995, the total population of older prisoners has grown by about 10,000 per year (Aday, 2003); it can be safely estimated that state and federal prisons housed about 180,000 inmates who were 50 years or older (about 11.6% of all prisoners) on January 1, 2008.

In the year 2030, about one third of the U.S. prison population will be 55 years of age and older (Enders, Paterniti, & Meyers, 2005; Williams et al., 2006). This rapid increase in the proportion of older inmates is due, in part, to the prevailing use of federal and state sentencing statutes that emphasize the advancement of longer sentences for both younger and older inmates and the reduction of options for early release via restricted or eliminated access to parole and/or good time. Examples of such statutes include mandatory minimum sentencing strategies, determinate sentencing strategies, sentencing guidelines, truth in sentencing laws that mandate convicted offenders to typically serve 85% of their sentences, and three-strikes sentencing strategies that require life sentences (often without access to parole) for recidivists (Benekos & Merlo, 1995; Ditton & Wilson, 1999; Turner, Sundt, Applegate, & Cullen, 1995).

Even though it is uncertain whether these sentencing strategies have acted as general deterrents and/or led to real reductions in crime rates, it is clear that these sentencing schemes have increased the general rate and length of incarceration (Austin & Irwin, 2001; Benekos & Merlo, 1995; Petersilia, 1992; Steffensmeier & Harer, 1993; Turner et al., 1995; Visher, 1987; Zimring & Hawkins, 1991). In addition, such sentencing strategies have created a “stacking effect,” whereby cohorts of younger inmates are often held without any access to parole or early-release options (Zimbardo, 1994, p. 3). Increases in the rate and length of incarceration have, therefore, led to two issues regarding a growing population of older inmates: (a) the need to provide developmentally appropriate environmental conditions and service delivery systems and (b) the need to explore alternatives to mainstreaming older prisoners given the consequences of placing older inmates in facilities designed to hold a younger, more aggressive, more physically robust population (Aday, 2003; Anderson & McGehee, 1991, 1994; Kerbs & Jolley, 2007; Lemieux, Dyeson, & Castiglione, 2002; Morton, 1993, 1994). Age segregation is a possible alternative to the status quo of mainstreaming and can potentially improve the safety of, and service provision for, older prisoners. This article aims to provide a rationale for age segregation that is qualified by a summary of considerations that are often raised against age segregation.

Cost Savings With Centralized Health Care

As compared to younger prisoners, older prisoners tend to be in poor health (Maruschak & Beck, 2001). Based on data from the 1997 Survey of Inmates in State and Federal Correctional Facilities (Maruschak & Beck, 2001), the percentage of state prisoners who reported any health condition (i.e., learning, speech, hearing, vision, physical, or mental impairment) increased as one moved up the age brackets from ages 24 or younger (23.8%) to 25–34 years of age (26.8%), to 35–44 years of age (34.0%), and finally to 45 or older (47.6%). For federal prisoners, the same pattern emerged as one moved up age brackets from ages 24 or younger (13.8%) to 25–34 years of age (16.9%), to 35–44 years of age (22.1%), and finally to 45 or older (38.6%). Some academics (Rubenstein, 1984) have suggested that the health of older prisoners tends to deteriorate quickly during incarceration. This purported rapid decline in health may be due to a constellation of factors that include unhealthy lifestyles prior to entering prison (substance abuse, poor diet, lack of exercise), unhealthy lifestyles within prison, and the harsh and stressful experiences within prison that can “aggravate” and “accelerate” the aging process (Fattah & Sacco, 1989). Thus, a “50-year old inmate may have a physiological age that is 10 to 15 years older” than their chronological age (Mitka, 2004, p. 423). For this and other reasons, most of the correctional literature views age 50 as the line of demarcation for older prisoners (Morton, 1992).

Older prisoners average three serious health problems, which may include chronic health conditions and substance abuse or dependence problems (Chaiklin & Fultz, 1985; Corrections Today, 1990; Kerbs, 2000a; Marquart, Merianos, & Doucet, 2000; Wilson & Vito, 1986). In a recently published integrative review of health-related research on older prisoners, Loeb and AbuDagga (2006) noted that the most commonly reported health problems included arthritis, back problems, cardiovascular diseases, endocrine disorders, psychiatric conditions, respiratory diseases, sensory deficits (vision and hearing problems), and substance abuse problems. In addition, there may be gender differences in the reporting of health problems by older inmates. Research by Kratcoski and Babb (1990) found that older female prisoners were two times as likely as older male prisoners to report serious health problems such as cardiac, degenerative, and respiratory illnesses.

Although a significant proportion of older inmates require and are legally entitled to medical treatment, they are not necessarily receiving adequate medical care. The best available research (Dabney & Vaughn, 2000; Lundstrom, 1994; Robbins, 1999; Vaughn & Collins, 2004) suggests that prisoners of all ages are medically and psychiatrically neglected, despite a legal mandate to treat serious illnesses as per Estelle v. Gamble (1976). In this case, the U.S. Supreme Court noted that “deliberate indifference to serious medical needs of prisoners constitutes the `unnecessary and wanton infliction of pain'… proscribed by the Eighth Amendment” (Estelle v. Gamble, 1976, p. 104). Moreover, prisons were directed to provide a community standard of care for serious medical problems to include acute care, dental care, and general medical, nutritional, psychiatric, and long-term care services (Kratcoski & Pownall, 1989). Despite a legal directive to provide inmates with a community standard of medical care, 90% of all prison hospitals in the United States failed to meet the basic standards of care as established by the medical profession (Lundstrom, 1994). Furthermore, this failure to meet professionally designated standards of care occurred in reference to prison hospitals and prison-based medical services that are oriented toward treating the acute care needs of younger inmates, a population with proportionally fewer chronic and serious medical conditions as compared to older inmates (Alston, 1986). Thus, it appears as if older prisoners are at particular risk of medical neglect given the complexities associated with chronic health conditions and aging.

Although many states and the federal government have increased their correctional health care budgets, prison systems are still failing to meet inmates' basic medical needs (Dabney & Vaughn, 2000; Robbins, 1999; Vaughn & Collins, 2004). This is problematic given that correctional health care systems across the nation collectively spent approximately US$3.87 billion in fiscal year 2002 (Camp, 2003). Since then, prison health care budgets have increased rapidly; for example, the California Department of Corrections and Rehabilitation increased its health care spending by 263% since 2000 to US$2.1 billion a year for the fiscal year that ends in July of 2008 (Rau, 2007). Despite this infusion of money, the scholarly literature indicates that prison systems routinely hire physicians who have restrictions on their medical licenses and can practice medicine only in prisons due to prior findings of medical negligence and malpractice in the free society; hence, correctional physicians are often substandard and lacking in basic competence (Dabney & Vaughn, 2000), which can necessitate the use of expensive community-based outsourcing. When prison hospitals and correctional medical facilities and staff are not adequate to the task of treating complex conditions, older prisoners must be taken to community-based hospitals and treated by specialists with guards providing continuous and expensive supervision (Kerbs, 2000b; Morton, 2001; Rikard & Rosenberg, 2007). Such supervision and outsourcing of medical care has been shown to increase the average cost of incarceration per year per inmate. Whereas the average younger prisoner costs around US$22,000 per year to imprison, the average older prisoner costs three times that amount (between US$60,000 and US$69,000 per year), in part due to the augmented costs of treating chronic health conditions (Aday, 2003; Kerbs, 2000b).

Age segregation could help ameliorate many of the health care delivery problems experienced by older prisoners. Age segregation would allow correctional authorities to relocate older prisoners from (a) facilities that are largely oriented toward the care and needs of younger inmates to (b) facilities designed to care for the specific and specialized health care needs of older prisoners. This may help avoid costly litigation associated with law suits alleging medical neglect, malpractice, and Eighth Amendment violations as per Estelle v. Gamble (1976; Dabney & Vaughn, 2000; Robbins, 1999; Vaughn & Collins, 2004). If staffed properly, age-segregated facilities would also help avoid costly outsourcing to community-based hospitals and associated costs for continuous and expensive supervision that is required while the prisoner is in community-based hospitals (Kerbs, 2000b). Furthermore, most prison hospitals are typically located in expensive medium or maximum security settings so that inmates of all classifications can be served, but older inmates “generally pose less of a security threat than do younger violent offenders, and a large number can be safely housed in a minimum-security setting if medical care is available” (Morton, 2001, p. 83). Indeed, most research suggests that, as compared to younger prisoners, older prisoners are easier to supervise because they are less likely to escape, violate prison rules, or receive disciplinary reports (Goetting, 1984; Rubenstein, 1982, 1984; Wilson & Vito, 1986). Although there is a small percentage (9.5%) of older inmates who present serious disciplinary problems (McShane & Williams, 1990), these inmates may be better served in currently existing higher-security prison hospitals and correctional facilities, albeit with the need for medically competent treatment designed to meet the needs of geriatric inmates. For the other 90.5% of older prisoners, funds could be saved by centralizing chronic-care services in less expensive, age-segregated minimum-security settings.

Reducing Civil Liabilities With Centralized Disability Services

In 1998, the U.S. Supreme Court ruled in Pennsylvania Department of Corrections v. Yeskey that the Americans With Disabilities Act (ADA) of 1990 is unambiguously applicable to federal and state prisoners. In legal terms, this court case requires that the physical structures of prisons be adapted (or accommodations be made) to provide disabled prisoners with unobstructed access to all facilities; disabled prisoners (young and old) must also have programmatic access to services to include educational, legal, medical, recreational, religious, and social services (Atlas & Witke, 2000; Burke, 1999; Lester, 2003; Russell & Stewart, 2001).

Although many federal and state prisoners are not disabled, a large proportion of prisoners report some kind of physical or emotional condition that impairs their ability to function; in addition, advanced age is associated with increases in the proportion of prisoners reporting physical and emotional conditions. For example, Maruschak and Beck (2001) examined age-specific breakdowns of the proportion of federal and state prisoners reporting physical impairments and mental conditions in 1997. As compared to the proportion of younger federal and state prisoners below age 45, a larger proportion of older prisoners ages 45 and above reported hearing, vision, physical impairments and mental conditions. In addition, Maruschak and Beck (2001, p. 4) noted that “48% of State inmates and 39% of Federal inmates age 45 or older said they had a physical impairment or mental condition compared to 24% of State prisoners and 14% of Federal prisoners age 24 or younger.

Although most states mainstream the vast majority of older prisoners (including those who are disabled) into the general prison population, many states have arranged for a limited amount of housing with special accommodations for prisoners (both young and old) who are disabled or infirm. Aday (2003) noted that about half of all states commingle younger and older prisoners who are infirm and/or disabled in units and facilities that are “often described as `aged/infirm,' `medical/geriatric,' `disabled,' or simply `geriatric'” (p. 153).

Although older prisoners in such facilities may have greater access to specialized programming (rehabilitative, recreational, vocational, and work-related programming), the majority of older prisoners in the general prison population do not necessarily have such access due to a multiplicity of problems. Most correctional programs have been primarily developed for younger prisoners (Aday, 1994a, 2003); for example, General Educational Development programs and vocational training may have utility in the eyes of someone who is under the age of retirement, but older prisoners may not see these programs as having any relevance to their life status. Even if older prisoners did want to access programs developed for younger prisoners, there appears to be a bias against placing older inmates into such programs. Goetting (1983) noted that prison staff were averse to the idea of placing older prisoners in educational and vocational programs, and other studies and experts suggest that aging inmates rarely access counseling, educational, and/or vocational prison programs (Sabath & Cowles, 1988; Wiegand & Burger, 1979; Wilson & Vito, 1986). Although it is of concern that these programs are of little use to aging inmates due to their structure and content (Aday, 1994a, 1994b; Wiegand & Burger, 1979), it is even more disconcerting to hear that prison staff either actively discourage or openly deny older prisoners access to programs (Goetting, 1983; Wiegand & Burger, 1979). Wiegand and Burger (1979), for example, noted that some correctional staff members felt that “you can't teach an old dog new tricks,” and “since the number of available openings in the educational program was limited, staff were unwilling to fill those spots with older offenders. The preference was to offer the courses to the younger offender” (p. 50). Fattah and Sacco (1989) argued that older prisoners receive reduced access to rehabilitative services because “rehabilitation, as a correctional objective, is often dismissed … [for older prisoners] on the grounds that it is neither feasible nor desirable” (p. 123).

Since the affirmation of the ADA of 1990 as per Pennsylvania Department of Corrections v. Yeskey (1998), correctional facilities have become more responsive to the needs of older prisoners (Aday, 2003), but they still have a long way to go as evidenced by the legal literature (Burke, 1999; Lester, 2003; Russell & Stewart, 2001). The failure to safeguard and guarantee older prisoners' right of access to facilities and programs in the face of potential disabilities and age-based discrimination suggests that the rights of older and disabled prisoners appear to be largely illusory and rhetorical (Kerbs, 2000a). Not surprisingly, the continued failure to comply with such rights has led to and will continue to foster civil liabilities as individual and class-action law suits are filed against prison facilities and correctional personnel (Atlas & Witke, 2000; Booth, 1989; Burke, 1999; Dugger, 1988; Goetting, 1985; Lester, 2003; Lundstrom, 1994; Morton, 1992; Russell & Stewart, 2001). Given the preponderance of physical impairments and emotional conditions reported by older prisoners, correctional agencies could reduce the civil liabilities related to the ADA of 1990 if they centralize disability services in age-segregated facilities that fully comply with ADA requirements. Because many facilities in the United States were built prior to the passage of the ADA of 1990, the potential costs of structural and programmatic changes across all facilities in America appears prohibitive. Nonetheless, those inmates who need ADA-compliant facilities should be given access to such facilities, and age segregation may provide a sound first step toward helping older prisoners who need fully compliant facilities because of physical impairments or mental conditions.

Advancing Inmate Safety for Older Prisoners

Another significant reason for age segregation stems from studies concerning inmate-on-inmate victimization. Of course, inmates of all ages can experience the vagaries of victimization, and the best available research using inmate self-report data suggest that inmate-on-inmate victimization is a daily event in the context of prison life. In a recent statewide study, 25% of all prisoners (252 inmates per 1,000) reported physical victimization by other inmates (Wolff, Blitz, Shi, Seigel, & Bachman, 2007). However, many studies rely on official statistics to document victimization. For example, (Camp & Camp, 1999) documented 59 inmate deaths by homicide and a total of 27,169 inmate-on-inmate attacks in federal and state correctional facilities in 1998. In 2006, federal, state and local correctional authorities reported an estimated 6,528 allegations of sexual violence involving incarcerated men and women (Beck, Harrison, & Adams, 2007). Although some may consider official statistics to be accurate, most experts suggest that they grossly underestimate the frequency of victimization for two key reasons (Bowker, 1982; Cohen, Cole, & Bailey, 1976). First, such official statistics often measure only incidents requiring medical attention, but many minor incidents are unreported. Second, many inmates avoid reporting incidents because of their fear of retaliation and humiliation. McCorkle (1993b) argued that official statistics also fail to include the “more minor, surreptitious acts of physical abuse, sexual aggression, and extortion that are considered `normal' parts of prison life” (p. 74). Not surprisingly, self-report studies find much higher victimization rates than official statistics.

Although inmate-on-inmate victimization appears to be a common occurrence that undermines prison safety and inmate rehabilitation (Bowker, 1980; Byrne & Stowell, 2007; DeLisi, 2003; Gibbons & Katzenbach, 2006; Gilligan & Lee, 2004; Hochstetler, Murphy, & Simons, 2004; McCorkle, 1992), correctional officials have a duty to protect prisoners from such victimization (Buchanan, 2007; Byrne & Hummer, 2007; Ross, 2005). Failure to protect prisoners can create a civil liability for correctional officers under the Eighth Amendment (see Farmer v. Brennan, 1994) if prisoners who are victimized by other prisoners “can show that officials knew of substantial risk of harm [victimization] and recklessly disregarded that risk” (Ross, 2005, p. 160). Recent quantitative studies have suggested that bisexuals, homosexuals, mentally ill inmates, and White prisoners are at increased risk of experiencing sexual inmate-on-inmate victimization (Hensley, Tewksbury, & Castle, 2003; Struckman-Johnson & Struckman-Johnson, 2000; Wolff, Blitz, & Shi, 2007); in addition, quantitative studies have shown gender differences for self-reported experiences with sexual coercion in prison for both inmate-on-inmate and guard-on-inmate victimization (Struckman-Johnson & Struckman-Johnson, 2006; Struckman-Johnson, Struckman-Johnson, Rucker, Bumby, & Donaldson, 1996; Wolff, Blitz, Shi, Bachman, & Siegel, 2006). One question that remains unanswered by legal scholars and case law that is pertinent to age-segregation debates is whether or not older prisoners represent a class of prisoners who are generally endangered by one or more types of inmate-on-inmate victimization due to placement in the general prison population.

A number of anecdotal and speculative reports suggest that older prisoners may be at risk of victimization. Based upon an extensive literature review on prison-based victimization, Bowker (1980) argued that older inmates can become “easy prey for prison `wolves,' `gorillas,' and other `rip-off artists'” (p. 159). Since then, Chaneles (1987) used anecdotal evidence to show how younger prisoners become depressed about their long sentences and turn their subsequent anger and rage into “exploitative behavior” toward ageing prisoners. Moreover, studies by Krajick (1979), Weigand and Burger (1979), and Vito and Wilson (1985) reported similar anecdotal evidence. For example, Vito and Wilson (1985) noted that “victimization and fear of victimization by younger, stronger inmates is a serious problem for elderly inmates” (p. 18).

The sociocultural environment in prison and current shifts in this environment may offer additional support for the notion that older prisoners are at risk of inmate-on-inmate victimization. Goetting (1985) argued that older prisoners were previously afforded a higher status (respect) because of their generally extensive criminal experience; more recently, Hunt, Riegel, Morales, and Waldorf (1993) quoted one prisoner as saying that younger prisoners “don't have respect for the old timers. They disrespect the old men now” (p. 406). DeLuca (1998) noted that “like the elderly in society, older inmates no longer get the respect once accorded to them, and also similar to their counterparts in free society, are more likely to be victimized by younger, more aggressive inmates” (p. 211).

Although their overall social status may be deteriorating, there are reasons to believe that older prisoners may confront varying levels of disrespect and victimization in prison based on their background characteristics and how younger prisoners behave in response to inmates with differing backgrounds. For example, prisoners with sex-offense histories (e.g., pedophiles and rapists) appear to be at high risk of victimization (Bowker, 1982), perhaps because they are among the most despised by the general prison population and they have a very low status in the inmate hierarchy (Dumond, 1992). The stigma associated with sex offenders suggests that older inmates might present high levels of victimization given that one third of all older prisoners are sex offenders (Flynn, 1998, 2000), and a majority of prisoners convicted of rape and sexual assault are self-proclaimed pedophiles (U.S. Department of Justice, 1997).

Quantitative studies have consistently demonstrated that younger inmates are more likely to experience various forms of victimization (Cooley, 1993; Maitland & Sluder, 1998; Porporino, Doherty, & Sawatsky 1987; Wolff, Blitz, Shi, Siegel, & Bachman, 2007; Wolf et al., 2006; Woolredge, 1998; Wright, 1991). This would lead one to believe that older prisoners, as compared to younger prisoners, are at a relatively lower risk of victimization. Although this may be true, a possible reduction in the risk of victimization does not justify the placement of older inmates in an environment where (a) they are being consistently targeted by younger inmates and (b) as a class, their victimization rates are high, regardless of differences between younger and older prisoners.

With regard to the former issue, Kerbs and Jolley (2007) randomly sampled 65 male prisoners (50 year of age and older) from a statewide Department of Corrections and completed face-to-face surveys regarding their victimization experiences over the last 12 months prior to the interview. Qualitative analyses from this study indicated that younger prisoners were the primary perpetrators of psychological, property, physical, and sexual inmate-on-inmate victimization against older prisoners, which is why the majority of older prisoners in this study supported age-segregated living arrangements.

With regard to the latter issue, the study by Kerbs and Jolley (2007) found high rates of psychological and property victimization with lower rates of physical and sexual victimization. Psychologically, significant proportions of older prisoners reported that other inmates had verbally threatened them (16.9%), threatened them with fake punches (24.6%), insulted them (40%), and/or cut in front of them in lines for food and/or other services (84.6%). While such acts seem minor, they can escalate into more serious forms of inmate-on-inmate victimization. Interaction among inmates occurs in such a manner that “All of the forms of prison victimization are related so that each becomes a causal factor in the other, forming an insane feedback system through which prison victimization rates are under constant pressure to increase” (Bowker, 1980, p. 31). Thus, being victimized within one category of offense increases the likelihood of being victimized in other categories (Edgar & O'Donnell, 1998).

Older inmates in the study by Kerbs and Jolley (2007) also reported that other inmates had victimized them economically. In total, about 28% had things in their cell stolen by another inmate, and 29% reported being cheated or conned out of money by another inmate in the past year. Although rates of physical and sexual victimization were lower, these forms of victimization nonetheless occurred. In total, 10.8% of all inmates reported physical attacks and assaults without weapons, 1.5% reported such attacks and assaults with weapons, and 6.2% reported being robbed. Finally, 10.8% reported being sexually harassed and 1.5% reported being raped.

In sum, there are some key findings from studies of older prisoners: (a) They are victimized by younger inmates; (b) they feel vulnerable to attack by younger prisoners; (c) they prefer to live with inmates in their own age bracket; and (d) they live at times in age-segregated protective-custody units (Aday, 1994a, 1994b, 2003; Aday & Webster, 1979; Kerbs & Jolley, 2007; Krajick, 1979; Walsh, 1989). So, why do prison officials mainstream older inmates with younger inmates? The published scholarly literature from the 1970s onward noted that correctional officials had a longstanding and deliberate tradition of mainstreaming because older prisoners (as compared to younger prisoners) were relatively well behaved and their good behavior was recognized by prison administrators as having a stabilizing, calming, or quieting effect on younger and more aggressive inmates in the general prison population (DeLuca, 1998; Wiegand & Burger, 1979). In short, a heterogeneous prison population of young and old inmates together was seen as optimal because older prisoners purportedly reduced the risk of violence and riots initiated by younger inmates (Krajick, 1979; Vito & Wilson, 1985).

Although there appears to be an empirical basis for such beliefs (see, for example, Mabli, Holley, Patrick, & Walls, 1979), studies to this effect do not negate the fact that mainstreaming may be helping younger prisoners' behavior at the expense of the older prisoners' safety. Although correctional managers were relying on older inmates to calm and maintain control over aggressive and violent younger prisoners, researchers (see, for example, Cooley, 1993; Ekland-Olson, Barrick, & Cohen, 1983; Wooldredge, 1998; Wright, 1991) did not adequately examine the consequences of mainstreaming because most prior studies examined the safety of inmates of all ages with few (if any) older inmates represented in samples given their limited representation in the overall prison population. To properly clarify the collateral consequences of mainstreaming this numerical minority, studies have had to oversample older prisoners or restrict samples to older prisoners (Douglas, 1991; Kerbs & Jolley, 2007). When studies sample older prisoners, findings suggest that they do have high rates of victimization and that younger prisoners are the perpetrators; hence, it could be argued that older prisoners represent an endangered class of inmates who deserve special living accommodations to enhance their safety while decreasing civil liabilities related to mainstreaming this at-risk population.

Promoting Rehabilitation Services for Older Prisoners

One of the most common measures of rehabilitation for prisoners is their rate of recidivism to include their rate of rearrest, reconviction, and/or reincarceration after release. Unfortunately, national data suggest that about 7 in 10 released prisoners will be rearrested within 3 years, and about half of all released prisoners will return to prison in this same period (Visher & Travis, 2003). Given that there were more than 600,000 prisoners released from federal and state facilities in 2002, the gravity of such recidivism rates is clear for the communities that receive released inmates (Visher & Travis, 2003).

In terms of the predictors of such recidivism, age plays a big role in the rate of rearrest, reconviction, and return to prison. One of the most consistent findings across the criminological literature pertains to the fact that age is negatively associated with recidivism. As age increases, recidivism rates decrease, which means that older prisoners are most likely to desist from criminal activity and benefit from rehabilitation services as evidenced by their low rates of recidivism. As reproduced in Table 1, data from a study published by Langan and Levin (2002) clearly depict the negative association between age and recidivism. As compared to the proportion of younger state prisoners (ages 14 to 44) who were released in 1994, a smaller proportion of older state prisoners (45 and older) were rearrested, reconvicted, or reincarcerated within 3 years.

Table 1.

Rate of Recidivism for State Prisoners Within 3 Years of Release in 1994

Age at Release	Percentage Rearrested	Percentage Reconvicted	Percentage Reincarcerated With New Prison Sentence	Percentage Reincarcerated With or Without a New Prison Sentence
14–17	82.1	55.7	38.6	56.6
18–24	75.4	52.0	30.2	52.0
25–29	70.5	50.1	26.9	52.5
30–34	68.8	48.8	25.9	54.8
35–39	66.2	46.3	24.0	52.0
40–44	58.4	38.0	18.3	50.0
45 or older	45.3	29.7	16.9	40.9

Note: Data reproduced from Langan and Levin (2002).

Research by Langan, Schmitt, and Durose (2003) also found a negative relationship between age and recidivism for sex offenders. As reproduced in Table 2, the data suggested that a smaller proportion of older sex offenders (ages 45 and above) released from prison in 1994 were rearrested for any type of crime within 3 years, a finding that held true for rapists, sexual assaulters, child molesters, and statutory rapists. Hence, despite the common stereotype of once a sex offender always a sex offender, it appears that even sex offenders can desist from crime after release. This is an important finding given that, as discussed earlier, one third of all older prisoners are sex offenders (Flynn, 1998, 2000), and a majority of prisoners convicted of rape and sexual assault are self-proclaimed pedophiles (U.S. Department of Justice, 1997).

Table 2.

Percent of Sex Offenders Rearrested for Any Crime Within 3 Years of Release From Prison in 1994

Age at Release	Percentage of Rearrested Rapists	Percentage of Rearrested Sexual Assaulters	Percentage of Rearrested Child Molesters	Percentage of Rearrested Statutory Rapists
18–24	58.6	60.2	59.6	70.0
25–29	53.8	54.3	51.4	56.4
30–34	52.6	46.7	46.5	47.7
35–39	46.1	38.9	38.0	37.9
40–44	41.2	31.6	28.0	44.4
45 or older	23.0	23.7	23.8	23.8

Note: Table reproduced from Langan, Schmitt, and Durose (2003).

Not surprisingly, antisocial and criminal behavior in prison often predict the same behavior in the community, which is why it is so important to promote prosocial behavior and rehabilitation behind bars. Indeed, one of the stronger predictors of recidivism in the community is offender misconduct in correctional facilities (Gendreau, Little, & Goggin, 1996). Again, the inverse age–crime curve holds true for studies documenting misconduct in prison; as age increases, prison misconduct decreases (Cooper & Werner, 1990; Craddock, 1996; DeLisi, 2003; Flanagan, 1980, 1983; Goetting, 1984; Goetting & Howsen, 1986; Ireland, 2000; Light, 1991; McShane & Williams, 1990; Wooldredge, 1991, 1994; Wooldredge, Griffin, & Pratt, 2001). In fact, DeLisi (2003) wrote, “Younger inmates are disproportionately volatile, difficult to manage, violent and prone to misconduct” (p. 655). Although all studies find an inverse relationship between age and prison misconduct, studies vary with respect to the identified ages at which the rates of prison misconduct begin to decrease; some studies find that decreases begin in the late teens (Kuanliang, Sorensen, & Cunningham, in press) and early twenties (Cunningham, Reidy, & Sorensen, 2008), whereas other studies show declines in the late twenties (DeLisi, 2003) and even the midthirties (Cunningham & Sorensen, 2007). To some extent, these figures also correspond to trajectories of crime among delinquent boys who were followed to age 70 in Sampson and Laub's (2003) examination of life-course trajectories; the mean age of desistance for all crimes reported for boys in this study was 37.5 years, but there were differences for the mean age of desistance across crimes including violent (31.3), property (26.2), alcohol or drug (36.8), and other crimes (32.7).

Issues of desistance are most evident when one examines the precautionary measures used by younger versus older prisoners who try to safeguard themselves against violence in prison. McCorkle (1992) found a positive association between age and the use of passive precautionary behaviors such as keeping more to self, avoiding certain areas of the prison, spending more time in one's cell, and avoiding activities. As age increased, there was an increasing reliance on passive precautionary behaviors. In contrast, age was negatively associated with aggressive precautionary behaviors such as having to get tough with other inmates, keeping weapons nearby, and lifting weights. Hence, as age decreased, younger prisoners were more reliant on aggressive precautionary behaviors than older prisoners. In addition, McCorkle (1992) noted that younger inmates who were heavily engaged in aggressive behavioral strategies were also “more oriented toward the [violent] prison culture … as a source of social, psychological, and material gratification” (p. 170).

In sum, older prisoners use passive behaviors to avoid victimization (McCorkle, 1992) while distancing themselves from an inmate culture that provides high status and rank in the social order to inmates who use violence, drug activity, financial schemes, and general predatory behavior to exploit and dominate other inmates and resources (Bowker, 1980; Byrne & Hummer, 2007; Byrne & Stowell, 2007; DeLisi, 2003; Dumond, 1992; Edgar & O'Donnell, 1998; Forst, Fagan, & Vivona, 1989; Gibbons & Katzenbach, 2006; Gilligan & Lee, 2004). Passive behavior and a lack of participation within the inmate culture identify older prisoners in two distinct ways.

First, these behaviors potentially compound the vulnerability of older inmates and can increase their likelihood of victimization because they are not in the process of building or maintaining a high social status within the prison community. The ability to be safe in prison involves more than just avoiding conflict and quietly living on the margins of an aggressive and violent culture (Wooldredge, 1998); in fact, the ability to survive in a hostile environment is influenced by the inmate's capacity to demonstrate a potential for doing harm to others. An inmate's capacity to demonstrate such potential is based on other inmates' perceptions of how well the prisoner embodies the traits of a true warrior: a person who exudes the highest level of social fitness, and therefore the most pronounced ability to thrive, because the inmate is capable of using physical strength, street smarts, and courage to gain control over resources and power over others (Bowker, 1980; Dumond, 1992; Edgar & O'Donnell, 1998; Gilligan & Lee, 2004; McCorkle, 1992). An older prisoner who quietly avoids conflict, does not take advantage of others, fails to engage in violent behavior, and displays an average of three serious medical conditions is certainly not viewed as a capable warrior. Within a prison's social order, the older inmate is not only an easy target for victimization, but he or she is also an obvious and socially acceptable target for victimization.

Second, older inmates'practice of avoidance behaviors and distancing from the inmate culture locate these prisoners within an age-graded life-course framework for desistance from criminal behavior (Laub & Sampson, 2001; Sampson & Laub, 2005). Laub and Sampson (2001) have created a framework that examines offender behavior across the life-course. One of the most salient features of their perspective is the focus on the important role social context plays in an offender's ability to desist from crime. Offenders can move from criminal identity, behaviors, and associations to a prosocial identity, behaviors, and associations by interacting with people, events, and the consequences of their choices within the social environment. The authors stress the fact that desistance is an ongoing process that is marked by the following correlates: a good marriage, stable work, a transformation of identity, and aging. Additional indicators include separation from deviant peers and unstructured time replaced by participation in a routine set of structured prosocial activities.

An examination of older inmates and their use of avoidance behaviors and distancing from the inmate culture within a framework of desistance support an argument for the segregation of older inmates. First, because older inmates are generally well behaved and even passive in their approach to prison life, they may be vulnerable targets for victimization as discussed earlier. In addition, with an average of three serious medical conditions, should older inmates need to defend themselves, they may be compromised in their ability to do so. Second, older inmates may be engaged in a meaningful process of desistance (Laub & Sampson, 2001; Sampson & Laub, 2005; Shover & Thompson, 1992). Their ability to make substantial changes in their lives is directly tied to their social environment, and given the nature of a social order that is built on a high regard for ongoing criminal activity, older inmates' participation in a process of desistance is likely to be dangerous and unsuccessful. If older inmates are potentially poised for rehabilitation, then would it not be more beneficial to provide an age-segregated, ADA-compliant living environment with centralized health care and programming to support inmates who may have already begun a process of desistance?

A similar argument has been applied to the case of juveniles who were tried, convicted, and sentenced in adult courts. Researchers decried the placement of juveniles in adult facilities, fearing for their physical safety due to their small size (Schindler & Arditti, 2001; Ziedenberg & Schiraldi, 1998) and fearing for their identity development and social trajectories based on their immersion in a violent inmate culture (Forst, Fagan, & Vivona, 1998; Redding, 1999). However, this same argument applies just as well to a call for segregating older inmates. In fact, empirically and theoretically, it would appear that older inmates are in a unique position to benefit from the same kind of age-segregated living arrangements that have been supported for juveniles: environments designed to provide a less violent, age appropriate context suitable for rehabilitation. Segregation for older inmates would support rehabilitation through age-appropriate programming and through the provision of an environment where basic survival is not the foremost task of the day. According to Styve, MacKenzie, Gover, and Mitchell (2000), “To make a positive impact on inmate adjustment and reduce criminal activity, correctional environments at a minimum must provide an environment that is perceived as safe” (p. 298). Furthermore, victimization and the fear of victimization are strong predictors of symptoms associated with psychopathology (Hochstetler, Murphy, & Simons, 2004; McCorkle, 1993a; Wooldredge, 1999). A prisoner's sense of general well-being as measured by indicators of anxiety, depression, and psychophysiological symptoms has implications for his or her ability to adjust to prison life, benefit from rehabilitation programming, and avoid recidivism (French & Gendreau, 2006; Hochstetler, Murphy, & Simons, 2004; Silver, 2006; Wooldredge, 1999).

In closing, if the goal of incarceration is rehabilitation and preparation for successful reentry into free society, then it makes sense to provide an environment that is theoretically and empirically associated with the correlates of desistance (i.e., vocational training, work programs, improved contact with families, and cognitive-behavioral training to support identity transformation) and the factors that would support successful desistance (i.e., improved physical health, improved mental health, reduced victimization, and reduced fear of victimization).

Conclusions

As noted by Rikard and Rosenberg (2007), there is an ongoing debate among academics and policy makers around the pros and cons of age segregation for older prisoners. Even though this academic debate will continue in the literature for years to come, the pragmatic demands of 180,000 older prisoners have led some states to open facilities that largely provide geromedical and other specialized services to their aging prisoners (Rikard & Rosenberg, 2007). To date, at least 26 states have run geriatric facilities or units and another 18 states have operated hospice programs for end-of-life care (Aday, 2003). Unfortunately, the extant research has not critically examined other issues surrounding the emergence, development, and functioning of such programs; most research has focused primarily on programmatic descriptions and not outcome evaluations (Aday, 1994a, 2003; Rikard & Rosenberg, 2007). In sum, age segregation and end-of-life programming appear to be de facto phenomena that will probably grow along with the number and proportion of older prisoners in the United States, but additional outcome evaluations are needed to determine if and how these geriatric programs meet older prisoners' needs.

Of course, this is not the first time in history that age has been used as a defining demographic characteristic for correctional reform. Age-segregated housing for juveniles began on January 1st, 1825, when the New York House of Refuge admitted six girls and three boys (Pisciotta, 1985). Since then, we have seen an evolution of age-segregated correctional institutions to include cottage houses, juvenile reformatories, and today's age-segregated detention centers and training schools. For those juveniles who enter today's adult correctional facilities, many states have opened age-segregated prisons that separate younger (juvenile) offenders from adult prisoners (Redding, 1999). States have moved in the direction of age segregation for juveniles because their placement with adults in prisons significantly increases the juveniles' chances of physical and sexual assaults by adult inmates and staff; not surprisingly, juveniles in adult prisons are also more likely to commit suicide than juveniles in age-segregated juvenile facilities (Beyer, 1997; Dumond, 1992; Forst, Fagan, & Vivona, 1989; Kerbs, 1999; Redding, 1999).

Analogous to the de facto segregation of inmates based on age for younger and older prisoners, inmates have also been segregated based on other sociodemographic characteristics such as gender and race. The biggest similarity among all calls for segregation, regardless of the sociodemographic group in question, is related to the advancement of prisoner safety. For example, the call for gender-segregated living arrangements for women in prison stems from a long history of male prisoners, guards, and wardens sexually abusing women in prison throughout the ages (Buchanan, 2007; Pollack, 2002). Indeed, Davis (1998) noted that the sexual abuse of women in prison is so pervasive that it “has become an institutionalized component of punishment behind prison walls” (Davis, 1998, p. 350). In part, such abuses continue to this day because many states fail to aggressively prosecute guards for custodial misconduct, despite the fact that Congress and 44 states have criminalized sexual activity between guards and inmates (Buchanan, 2005, 2007; Parker, 2002; Pollack, 2002). Racial segregation in prison is also rooted in the belief that such segregation abates violence by decreasing intergroup contact in racially divided prison settings with a propensity for intergroup violence. Moreover, de facto segregation exists, in part, because the courts have specifically supported racial segregation to abate prison violence and advance security (Robertson, 2006).

Hence, there appears to be precedence for segregation based on sociodemographic characteristics. Given the rapid growth in the number of older prisoners, the time might be right for states to more aggressively advance the development of age-segregated living arrangements. As discussed, older inmates support the use of age-segregated housing, and prison systems seem to be developing such facilities. If these facilities were infused with adequate funds to enhance programs and services that are developmentally appropriate for an aging prison population, federal and state prison systems might reap significant savings. Financially, centralized health care for older prisoners with chronic health problems could help circumvent costly outsourcing to community-based hospitals and potential law suits stemming from medical neglect of serious medical problems as per Estelle v. Gamble (1976). If age-segregated facilities were developed to comply with ADA guidelines, federal and state prisons can also circumvent costly law suits associated with precedence set by Pennsylvania Department of Corrections v. Yeskey (1998). On a more humanitarian note, the safety of older prisoners would be significantly enhanced by age-segregated facilities. However, prison systems should be careful to avoid replacing one form of victimization (inmate-on-inmate victimization) with another (guard-on-inmate victimization and/or pro-grammatic neglect) given that there have been examples of age-segregated facilities that perpetrated institutional maltreatment (see Stewart v. Rhodes, 1979).

Beyond institutional maltreatment, there are additional iatrogenic problems cited by scholars who oppose the advancement of age segregation. Such consequences could include the placement of older prisoners in poorly resourced facilities in remote (rural) locations that do not have adequate access to programming and competent medical staff. While such deprivations certainly plague inmates of all ages in general, these concerns are particularly problematic for older prisoners with complex bio-psycho-social problems. This is why Morton (2001), for example, advocated for mainstreaming older inmates in an effort to “help insure that older inmates are not denied access to programs and services available to other inmates” (p. 83).

Granted, given the status quo in relation to a dearth of programming for all inmates, age segregation might lead to even fewer options for a specialized cohort of older inmates. However, we are not proposing the creation of age-segregated facilities that cater to the status quo; instead, we propose the creation of cost-effective and humane facilities that operate according to the principles of rehabilitation with age-appropriate and developmentally specific programs and services. Although we agree with Morton that older inmates should have access to appropriate programs and services, we believe that mainstreaming does little more than compromise their safety and subsequent rehabilitation. Pragmatically, requiring these systems to provide specialized care for older inmates necessitates a careful analysis of available resources and the strategic delivery of such resources to maximize benefits to inmates of all ages. As a compromise to age-segregated facilities, it might be wise to consider the benefits of housing older inmates within separate wings or units of larger facilities that already include competently staffed medical centers for inmates of all ages. Philosophically, it is important that we do not use pragmatic considerations (e.g., economic and political constraints) as justifications for sacrificing progress toward creating safe environments that support the rehabilitation of older prisoners.

Furthermore, in light of the issues raised thus far, the very act of mainstreaming raises serious questions about America's commitment to the rehabilitation of older prisoners. Do we as a society believe in rehabilitation for older prisoners? Do they deserve it? If yes, policy makers must take into account the impact a violent inmate culture has not only on the physical safety and psychological health of the inmate but also on the inmate's ability to engage in a meaningful process of desistance that will prepare the prisoner for successful reentry into free society. Forst, Fagan, and Vivona (1989), when discussing concerns about juveniles in adult prisons, noted that such mainstreaming teaches juveniles “little else other than the institutional world … [and] the reciprocal cycle of violence and retaliation” (p. 11). The authors question the placement of juveniles in what is known to be a violent inmate culture; yet the very existence of such a culture and its impact on inmates of all ages has not been fully taken into consideration. If the existence of the social order in prison is dangerous and counterproductive to rehabilitation, why place any individual, young or old, in such an environment?

Therefore, an argument in favor of segregation for older inmates finally comes to rest on two important premises. First, the purpose of incarceration should be one framed by a philosophy of rehabilitation for all inmates that is operationalized according to the theoretical and empirical correlates of a successful participation in desistance. Second, older inmates should be viewed as equally deserving of rehabilitative correctional environments. Scholars and policy makers have advocated for the removal of juveniles and women from the general prison population on the basis of their demonstrated vulnerability to violence and abuse. Implicit in this argument is the underlying worth of these inmates in that they are harmed to a greater degree by prison violence and abuse, and they therefore require and deserve separate facilities. Research shows that older inmates are also vulnerable: They have little social status within a violent prison order coupled with a diminishing physical fortitude. Even so, older inmates are expected to survive in facilities dominated by younger, stronger, and more aggressive inmates. Are older prisoners, like other vulnerable inmates, worthy of accommodations that promote their safety and rehabilitation in a more appropriate setting or are they simply expected to fend for themselves?

Biographies

John J. Kerbs is an assistant professor in the Department of Criminal Justice at East Carolina University. His current research focuses on the measurement and prevention of violence in prisons and schools. His articles have appeared in the American Journal of Criminal Justice, Annals of the American Academy of Political and Social Science, Crime & Delinquency, Criminal Justice Review, Federal Probation, Hamline Journal of Public Law and Policy, Perspectives: The Journal of the American Probation and Parole Association, and Professional Psychology—Research and Practice.

Jennifer M. Jolley, MSW, is a doctoral student and a National Institute of Mental Health predoctoral fellow in the George Warren Brown School of Social Work at Washington University (St. Louis, Missouri) and a police officer and grants specialist with the Winterville Police Department (Winterville, NC). She specializes in the design and implementation of community, correctional, and forensic programs that address interpersonal violence. Her articles have appeared in the American Journal of Criminal Justice, Crime & Delinquency and Criminal Justice Review.