METHODOLOGICAL CHALLENGES IN STUDIES OF PRISONERS' SEXUAL ACTIVITY AND DRUG USE

Nancy Mahon

Prisons and jails are far from ideal places to talk about sex and drug use. Indeed, undertaking a study of prisoners' high-risk behaviours invites many methodological, logistical and ethical challenges. These challenges stem primarily from three aspects of prisoners' lives: correctional facilities are by nature coercive environments; sex and drug use violate correctional regulations; and, sexual behavior involves identity issues that often spur shame and a fear of homophobic violence from other inmates. Not surprisingly, studies of prisoners' high-risk acts are relatively small in number. They are also concentrated in a few countries, particularly England, Australia, Canada, and, most recently, the United States of America. This article briefly outlines and discusses the methodological challenges of performing research on prisoners' sexual and drug-related activities and the limitations that these hurdles may place on the gathered data. At first blush, research methodology may seem to be of little significance (and interest) to harm reduction advocates. Yet, in a field such this, where resources - be they funds or the number of knowledgeable and committed researchers - are so scarce, ensuring the maximum veracity and reliability of research results as well as their accessibility to advocates is critically important. For, advocates must thoroughly understand the nature and limitations of research in order to effectively employ it to advocate for programmes.

THE NATURE AND USES OF QUALITATIVE AND QUANTITATIVE RESEARCH

Researchers have conducted both qualitative and quantitative studies on prisoners'sexual activity and drug use, and each kind of research is useful for different aspects of the design and implementation of harm reduction programmes. Qualitative studies, consisting of individual interviews and/or focus groups, have been conducted in the USA and England to learn about the nature of prisoners' high-risk behaviours including the practices and patterns of their sexual activity and drug use (Mahon, 1996; Turnbull et al., 1994). The type of information that can be gleaned from such research includes the locations and circumstances under which prisoners engage in high-risk activities. For instance, Turnbull and colleagues found t at male prisoners often have oral and anal sex in the shower area (Turnbull et al., 1994). On the issue of drug use, this author found that syringes, let alone sterile syringes, are so scarce in New York State correctional facilities that prisoners use dirty needles picked form the hospital clinic garbage, and make-shift syringes from basketball pumps and the filaments of light bulbs to shoot drugs intravenously (Mahon, 1996).

Taken as a whole, the qualitative data indicate that highrisk activity in prisons and jails is generally riskier than similar behaviour in the larger community because of the time pressures, lack of privacy, economy of scarcity and the coercive environment of correctional facilities. Also, participants' narrative accounts of how, when and where they had sex and used drugs in prisons and jails is of particular use and interest to harm reduction service providers. For clearly, as in the larger community, sex and drug use in correctional facilities is often spontaneous and situational. Similarly, simply because a condom, dental dam or clean needle is available, does not necessarily mean that an individual in the community or in a correctional facility will chose to use a harm reduction device. Qualitative data help service providers increase the likelihood that the availability of harm reduction devices will lead to use their use both by lending insight into cultures and subcultures of prisons and jails and allowing for the accessibility of harm reduction tools in the locations whe~e sex and drug use often occur.

In contrast to qualitative research, quantitative studies are used to assess the prevalence and inc idence of prisoners'high- risk activities. Quantitative studies on this subject are more numerous than qualitative studies in large part because most research in this area has been performed to explore, and, often, justify the need for harm reduction programmes . Simply stated, to be convinced of the need foraharm reduction programme, correctional officials want to see numbers. Thus, when they are told that harm reduction programmes are necessary because prisoners engage in risky behaviours, government officials ask how many prisoners engage in sex and shoot drugs, and how often do they do it?

SELECTION OF A STUDY SAMPLE:  PRISONERS VERSUS NON-PRISONERS

In designing a study, employing current or former prisoners as participants presents different methodological challenges and, in turn, places different limitations on the results. Attesting to the difficulty of performing research on prisoners' high-risk behaviours, of the dozen or so studies on the topic worldwide, only four have used current prisoners as subjects. Of these, only two have successfully collected data, and each of these studies used a convenience sample. The first study, which was performed in the USA, used a preexisting AIDS/HIV programme for inmates as a sample (Mahon, 1996). In the second study, which was performed in Australia, self-identified injecting drug users (IDUs) were recruited for the study by word of mouth by a peer worker (Crofts er al., 1996). As these studies illustrate, it is impossible for researchers to perform unfettered outreach to a prison or jail population, and studies must consequently recruit subjects in a manner that will inevitably lead to some sampling bias.

In addition to sampling bias, use of prisoners as research subjects presents numerous other hurdles, including: (1) obtaining permission from correctional officials and, in some instances, public health officials, to perform the study; (2) gaining access to the individual prison or jail facility to collect data with research tools, such as, pens and tape recorders which are considered contraband in correctional facilities; (3) assuring confidentiality to the prisoner/participants given the necessity of having correctional officers within close physical proximity to the researchers and participants; (4) ensuring the maximum veracity and reliability of the data given the prohibited nature of sex and drug use in prisons and jails, the high degree of homophobia in the population, and the large amount of peer pressure on prisoners not to tell civilians about prisoners' sexual behaviour; (5) obtaining truly informed consent despite the oppressive nature of prisons and jails and the necessity of using a convenience sample; and lastly, (6) the low literacy rate of prisoners and the limitations this places on the use of surveys to collect data.

Significantly, two early studies that attempted to collect quantitative data from prisoners on their highrisk behaviour failed to do so. In an English study, which focused on drug dependence in prisoners, inmates refused to answer a question about drug use in prison because they were'suspicious of its purpose' (Madden et al., 1991). On the other hand, a Canadian researcher focusing on HIV preven rion in prisons sought to ask similar questions of Canadian prisoners, but the Ministry of Correctional Services barred her from doing so on the dubious grounds that prisoners'sexual and drug-related behaviours did not relate to HIV prevention in prisons (Toepell, 1992). Together these studies illustrate two recurrent themes in studies with current prisoners: researchers' difficulty in gaining the trust of inmates and correctional officials' reluctance to admit the existence or importance of inmates'sexual activity and drug use.

In the face of such obstacles, it is not surprising that the vast majority of studies on prisoners' highrisk activities have used samples of former, rather than current, prisoners. Studying former prisoners eliminates many obstacles that a sample of current prisoners present and exchanges them for two relatively minor limits: participants'potential problems with recalling events that they witnessed or participated in while they were incarcerated (Mahon, 1996); and the difficulty of performing outreach to a population offormer inmates, which isgeographically scattered, reluctant to identify as formerly incarcerated, and frequently does not appellr for scheduled appointments (Dolan et al., 199 1).

DIFFICULTY OF GENERALISING DATA FROM PRISONS AND JAILS STUDIES

There is one additional limitation that is common to all studies of prisoners' behaviours regardless of the nature of the sample: the difficulty of generalising findings from one particular correctional facility or group of prisoners to entire correctional systems or prisoner populations. By nature, correctional facilities, particularly prisons (in which prisoners serve sentences of one year or longer), are separate, largely self-contained fiefdoms. Thus the cultures between and among staff and inmates, the degree and nature of coercion, official tolerance for consensual and non-consensual sex among inmates and between inmates and staff, and the availability of drugs, may vary drastically from one facility to the next. Studies that collect data from prisoners and/or former prisoners who have been incarcerated in a range of individual facilities within a correctional system consequently will yield data that is far more generaliseable than a study which collects data at any one facility.

CONCLUSION

Methodological barriers that researchers and, by extension, advocates face in learning about prisoners'lives could been seen only as hurdles that must be surpassed in order to attain the information sought. Yet, the obstacles that researchers face also shape prisoners'daily lives. Thus, to truly understand prisoners'behaviours and environment, and the circumstances under which harm reduction providers may in the future deliver services, researchers, advocates, and service providers alike must understand the barriers to gathering valid information on prisoners' sexual activity and drug use.

Nancy Mahon, J. D., Center on Crime, Communities and Culture, Open Society Institute, 888 Seventh Avenue, New York, NY 10 106, USA

REFERENCES

Crofts N, Thompson S, Wale E, Hernberger F (1996). Risk behaviours for blood-borne viruses in a Victorian prison. In press.

Madden A, Swinton M, Gunn J (1991). Drug dependence in prisoners. BritishMedicaljournal302:800.

Mahon N (1996). New York inmates HIV risk behaviours: The implications for prevention policy and programmes. Americanjournal of Public Health 86:1211-1215.

Toepell AR (1992). Prisoners and AIDS: Knowledge, Attitude and Behavior. Toronto: John Howard Society of Metropolitan Toronto; December: No 6.

Turnbull R Stimson GV, Stilwell 0 (1994). Drug Use in Prison. West Sussex, England: The Centre for Research on Drugs and Health Behaviour.