Factors mediating and moderating the relationship between gender and utilization of health care among Puerto Rican drug users in New York

Methods for prospectively incorporating gender into health sciences research

Numerous studies have demonstrated that sex (a biological variable) and gender (a psychosocial construct) impact health and have discussed the mechanisms that may explain these relationships. Funding agencies have called for all health researchers to incorporate sex and gender into their studies; however, the way forward has been unclear to many, particularly due to the varied definition of gender. We argue that just as there is no standardized definition of gender, there can be no standardized measurement thereof. However, numerous measurable gender-related variables may influence individual or population-level health through various pathways. The initial question should guide the selection of specific gender-related variables based on their relevance to the study, to prospectively incorporate gender into research. We outline various methods to provide clarification on how to incorporate gender into the design of prospective clinical and epidemiological studies as well as methods for statistical analysis.

Association between mental health service utilisation and sharing of injection material among people who inject drugs in Montreal, Canada

BACKGROUND

High-risk injection behaviors are associated with high prevalence of mental health problems among people who inject drugs (PWID). However, whether the use of mental health services is associated with lower risk of sharing injection material remains undetermined. This study aims to examine the association between mental health service utilisation and receptive sharing risk, and determine the potential modifying effect of psychological distress on this association.

METHODS

Participants answered an interviewer-administered questionnaire at 3-month intervals gathering information on sociodemographic characteristics, substance use and related behaviors, services utilisation and significant mental health markers. Relationship between the use of mental health services and receptive sharing was modeled using the generalized estimating equation (GEE), controlling for age at baseline, gender, and other potential confounders. Psychological distress was estimated using the Kessler Psychological Distress Scale (K10). Effect modification was investigated by adding an interaction term in the univariate GEE analysis.

RESULTS

358 participants contributed to 2537 visits (median age 40.3, 82% male). Mental health service utilisation was reported in 631 visits (25%), receptive sharing in 321 visits (13%) and severe psychological distress in 359 visits (14%). In multivariate GEE analyses, a significant association was identified between receptive sharing and the use of mental health services (aOR = 0.69; 95% CI = 0.50-0.94). We found no evidence of effect modification by psychological distress.

CONCLUSION

Among PWID, mental health service utilisation was associated with lower likelihood of receptive sharing, regardless of level of psychological distress. These findings should be taken into account when designing harm reduction strategies for this population.

Longitudinal changes in psychological distress in a cohort of people who inject drugs in Melbourne, Australia

BACKGROUND

Previous research into psychological distress among people who inject drugs (PWID) is predominantly cross-sectional; we determined longitudinal predictors of change in psychological distress among a cohort of PWID.

METHOD

We examined Kessler Psychological Distress Scale (K10) scores from 564 PWID (66% male) enrolled in the Melbourne Injecting Drug User Cohort Study. Gender-stratified linear models with fixed effects for each participant were used to examine correlates of change in individual K10 scores. Further linear regressions of adjusted K10 scores were used to measure correlations between demographic variables.

RESULTS

Participants reported higher K10 scores (higher psychological distress) than the general Australian population (mean K10 scores 23.4 (95%CI 22.6-24.2) and 14.5 (95%CI 14.3-14.7) respectively). The cohort's mean K10 score did not significantly differ over time, but individual variations were common. Women reported higher K10 scores than men (mean baseline K10 scores 25.2 (95%CI 23.9-26.6) and 22.4 (95%CI 21.5-23.3) respectively), however no significant differences remained after controlling for temporal factors. Key predictors of increases in K10 scores were being the victim of an assault in the past six months (P<0.001 for women and men) and intentionally overdosing in the past 12 months (P=.010 for women and P<0.001 for men).

CONCLUSIONS

PWID experience higher levels of psychological distress than the general population. Temporal rather than individual factors may account for the higher levels of psychological distress reported among women. Interventions to reduce rates of assault and/or intentional overdose should be explored to reduce high levels of psychological distress among PWID.

Prevalence of depressive symptoms and associated factors among people who inject drugs in China

BACKGROUND

People who inject drugs (PWID) have a high prevalence of mental health problems (e.g., depression) which compromise the effects of HIV prevention. We investigated the prevalence of depression and associated factors among PWID in Dazhou, China.

METHODS

Anonymous face-to-face interviews were administered to eligible participants by well-trained doctors. The Center of Epidemiological Studies Depression Scale (CES-D) was used to measure depression, using hierarchical linear regression models.

RESULTS

Among the 257 participants, 7.0%, 11.3%, and 75.1% were probable cases of mild (16 ≤ CES-D < 21), moderate (21 ≤ CES-D < 25), and severe (CES-D ≥ 25) depression, respectively. Hopefulness (standardized β = -0.34, p < 0.001) and emotional family support (β = -0.21, p < 0.001) were negatively associated, whereas severity of drug dependence (β = 0.12, p = 0.034) and unmet service needs (β = 0.20, p < 0.001) were positively associated, respectively, with CES-D scores. The adjusted analysis showed that resilience was negatively associated with CES-D, but it was not selected by the stepwise model containing hopefulness and emotional family support. Protective factors (i.e., hopefulness and emotional family support) and risk factors (i.e., severity of drug dependence and unmet service needs) remained significant in the same regression model.

CONCLUSION

Protective and risk factors were independently associated with depressive symptoms. Integrated interventions covering basic and psychosocial needs for PWID are warranted. Such interventions should strengthen emotional family support and cultivate positive factors such as hopefulness.

Social network members' roles and use of mental health services among drug users in New York City

Depression is more common among drug users (15-63 %) than the general population (5-16 %). Lack of social support network members may be associated with low mental health service (MHS) use rates observed among drug users. We investigated the relationship between social network members' roles and MHS use among frequent drug users using Social Ties Associated with Risk of Transition into Injection Drug Use data (NYC 2006-2009). Surveys assessed depression, MHS use, demographics, drug use and treatment, and social network members' roles. Participants reporting lifetime depressive episode with start/end dates and information on social/risk network members were included (n = 152). Adjusting for emotional support and HIV status, having one or more informational support network members remained associated with MHS use at last depressive episode (adjusted odds ratio (AOR) 3.37, 95 % confidence interval (CI) 1.38-8.19), as did history of drug treatment (AOR 2.75, 95 % CI 1.02-7.41) and no legal income (AOR 0.23, 95 % CI 0.08-0.64). These data suggest that informational support is associated with MHS utilization among depressed drug users.