Poster Presentation at the XIII International Conference on AIDS in July 2000 in Durban, South Africa. Originally presented at the 9th Annual Canadian Conference on HIV/AIDS Research in April 2000 in Montreal, Quebec.

Defining the High End: Health Care Service Use among 
Young Gay and Bisexual Men Living in Vancouver

Mary Lou Miller, Michael Botnick, Aslam Anis, Keith Chan, Steve Martindale, Kevin JP Craib, Arn Schilder, Martin T. Schechter, Robert S. Hogg

Objective: To characterize the socio-demographic, psychosocial and behavioural determinants of health care service use among young gay and bisexual men (MSM).

Methods: MSM aged 15-30 living in Vancouver who had not previously tested HIV+ were recruited by the Vanguard Project through physicians, clinics and outreach. Medical, out-patient and emergency service use was based on participants who completed a year 4 questionnaire. Comparisons between those who reported low service use (1 to 4 visits), and those who reported high service use (5 or more visits) were carried out using contingency table analysis. Step-wise logistic regression was used to identify independent predictors of high health service use.

Results: This analysis was based on 356 year 4 questionnaires. A total of 321 (90.2%) reported any medical, outpatient and hospital/emergency service use in the previous year. Of these, 118 (36.8%) visited the above services 5 or more times. High users were less likely to be high school graduates (77.8% vs. 88.5%; p = 0.011) and employed (82.2% vs. 95.4%; p < 0.001), and more likely to be HIV+ (8.5% vs. 1.0%; p < 0.001), First Nations (14.8% vs. 5.5%; p = 0.006), an IDU (11.1% vs. 4.5%; p = 0.024), sexually abused (25.0% vs. 15.0%; p =0.028), gay bashed (22.4% vs. 12.5%; p =0.021), a sex worker (14.7% vs. 4.6%; p = 0.002), to live in unstable housing (6.9% vs. 1.5%; p = 0.022), and to have a high depression score (68.1% vs. 39.0%; p <0.001). In multivariate analysis sex trade work (AOR=3.00; 95% CI: 1.20, 7.52) and high depression score (AOR=2.62; 95%CI 1.58, 4.34) were associated with high use of health services.

Conclusion: Our data indicate that there are number of determinants of high use of health services in this population. Further investigation is need to examine prospective predictors of use.

To characterize the socio-demographic, psychosocial and behavioural determinants of health care service use among young gay and bisexual men enrolled in an open cohort.

The Vanguard Project

The Vanguard Project is a prospective study of HIV incidence and risk behaviours, which began in May 1995. Eligible participants are men between the ages of 15 and 30 at baseline, living in the Greater Vancouver area, who have not previously tested HIV-positive. The study is open to all men who have sex with men, regardless of whether they self-identify as gay, bisexual or straight.

Recruitment of participants involves outreach at gay community events, use of print materials (such as posters, brochures, condom packages) and recruitment by health care professionals at medical clinics and local physicians' practices.

Since May 1995, over 850 participants have completed confidential, self-administered questionnaires and undergone HIV testing on an annual basis. Recruitment of street-involved sex-trade workers occurred primarily at an outreach clinic specifically aimed at street youth.

Study Design

Eligible participants for this particular analysis were those who completed the questionnaire in the fourth wave of the study. Questionnaire items included demographics, sexual behaviour with both men and women, and psychosocial scales.

For the purpose of this analysis, self-reported data were examined on medical, outpatient and hospital/emergency service use within the previous twelve months.

Statistical Analysis

Comparisons between those who reported low service use (1 to 4 visits in the previous year), and those who reported high service use (5 or more visits) were carried out using contingency table analysis. Step-wise logistic regression was used to identify independent predictors of high health service use. All reported p-values are two-sided.

Of the 356 men who completed the questionnaire in the fourth wave of the study, 321 (90.2%) reported any medical, outpatient and/or hospital/emergency service use in the previous year. Of these, 118 (36.8%) were categorized as high health service users, having visited the above services five or more times in the previous year.

Univariate analysis (Table 1)

High health service users were less likely to:

• be employed (82.2% vs. 95.4%; p<0.001); and
• have graduated from high school (77.8% vs. 88.5%; p=0.011).

High health service users were also more likely to be:

• HIV-positive (8.5% vs. 1.0%; p<0.001);
• Aboriginal (14.8% vs. 5.5%; p=0.006); and
• living in unstable housing (6.9% vs. 1.5%; p=0.022);

...and more likely to have:

• ever injected drugs (11.1% vs. 4.5%; p=0.024);
• a high depression score (68.1% vs. 39.0%; p<0.001);
• been sexually abused (25.0% vs. 15.0%; p=0.028);
• been queer bashed (22.4% vs. 12.5%; p=0.021);
• been paid for sex in the past year (14.7% vs. 4.6%; p=0.002).

Multivariate analysis (Table 2)

Having been paid for sex (AOR=3.00; 95% CI: 1.20, 7.52) and having a high depression score (AOR=2.62; 95%CI 1.58, 4.34) were independently associated with high use of health services.

Health and medicine exist in a social vortex of institutional and functional paradigms. Over 150 years ago, Rudolph Virchow opined that "medicine is a social science" and that "physicians are the natural attorneys of the poor, and the social problems should largely be solved by them." (1) The associations found in our study confirm the social aspects of Virchow's comment: stress, self-esteem and social relations appear to have a significant bearing on health care usage.

Our data indicate that health service usage among young gay and bisexual men is high, and this usage is related to socio-economic factors, lifestyle issues and the participant's psychological well being.

In our study, psychosocial factors (lifestyle, violence, need for shelter and social support) also weigh heavily on the rationale for the use of hospital services. While this perspective appears to adopt the position of blaming the victim, the person of low socio-economic status who is more likely to be in poor health, or to blame clinicians for not spending sufficient time with their patients, this is not so. We believe that identifying and recognizing the limitations to our health care system should serve to propel health care providers to re-examine service delivery, and rethink intellectual and ethical imperatives to develop supplementary and complementary approaches toward correcting socio-economic disparities in health.

The reductionist biomedical model's drive to amplify in-house medical services to meet expanding demand may not be the most efficacious route to follow. An investment in education, an overall improvement in social conditions, and research on patient self-management may provide greater returns on investment, provided that physicians and administrators support these changes. (2)

For example, Vancouver's St. Paul's Hospital recently announced the effectiveness of a case-management intervention program targeting frequent users of the emergency department. This multi-disciplinary program focused on both the medical and social needs of each individual patient. After individualized care plans were implemented for 24 so-called "frequent flyers," per-patient visits to the ER plummeted from a median of 26.5 in the year prior to the program to a median of 6.5 in the year after intervention. (3)

High end users of the medical system are more likely to have lower education, to be unemployed, to have unstable housing, to be Aboriginal, and in general to be marginalized on the basis of lifestyle (injection drug use, sex work) or to be psycholo-gically damaged (sexually abused, gay bashed, depressed).

The authors gratefully acknowledge the staff, participants and Community Advisory Board of the Vanguard Project; participating doctors and HIV testing clinics; and the National Health Research Development Program, Health Canada, for project funding. Poster designed by Steve Martindale, with assistance from Michael Zarowny.

1. Ackerknecht, LA, Virchow, R:, Wilkinson RG. Income distribution and life expectancy. BMJ 1992; 304: 165-8.
[back to text]
2. Pincus, T. Formal educational level ­ a marker for the importance of behavioral variables in the pathogenesis, morbidity, and mortality of most diseases? [Editorial] J Rheumatol. 1988; 1 5:1457-60.
[back to text]
3. Pope, D. et al. Difficult case management in the emergency department. 2000. Unpublished.
[back to text]