reports

Diabetes, gender and neighbourhood ethnicity: inequalities in healthcare resource utilization

Canadian Journal of Diabetes 33:281.

Provincial Health Services Authority, Vancouver, BC, Canada

AIMS: Lower utilization of medical resources by immigrants are often observed in self-reported surveys and usually interpreted as the so called healthy immigrant effect. However, whether the lower utilization rate by immigrants is solely a reflection of better health or is indicative of barriers to accessing the health system is not clear. In this study, we examine the gender-specific differences in healthcare resource utilization between neighbourhoods with varied immigrant density based on patients’ medical records.

METHODS: For healthcare planning purposes, the entire population of Richmond, BC, is geographically grouped into five neighbourhoods with similar population size: South Arm, Richmond Centre, Blundell, Steveston and East Richmond. The expected health resource dependency for each resident was assessed using the Johns Hopkins University Adjusted Clinical Group (ACG) case-mix system that accounts for his/ her morbidity, complexity and costs of care with the overall Richmond rate as baseline. For each neighbourhood we aggregated the observed hospital bed-days and primary care physician use across all ACG categories and compared them with the corresponding expected values obtained by applying Richmond’s ACG-specific rates. The resulting differences are used to signal if resources are under or over used for that neighbourhood. Healthcare use due to pregnancy, obstetrics and relevant complications were excluded as they preclude gender based comparisons.

RESULTS: Overall, diabetic women were more likely to visit their primary care physicians and less likely to stay in hospitals compared to diabetic men. No obvious neighbourhood variations in the use of primary care physicians both among diabetic women and among diabetic men were observed. Taking the entire Richmond female population as baseline, however, we found great neighbourhood differences in hospital use by diabetic women. Diabetic women from the two neighbourhoods with the lowest proportion of immigrants had the greatest use of hospital resources (18.5% higher in Blundell and 5.2% higher in Steveston) whereas diabetic women from the neighbourhoods with the highest proportion of immigrants used less hospital resources (18.6% lower in East Richmond and 10.5% lower in Richmond Centre) than expected. No such effect was observed for diabetic men.

CONCLUSIONS: In this study we demonstrated that diabetic women from neighbourhoods with a high density of immigrants used less hospital acute care. This could indicate the existence of potential barriers of gender, language, and ethno-culture to immigrant diabetic women. The results can be used to help health authorities and policymakers respond to the needs of immigrant women by addressing potential barriers to the access of health care.