Grant: Travel Awards - Institute Community Support

Administrative databases depicting healthcare utilization have become an important data source in HIV epidemiological research, providing a unique opportunity to implement rapid, inexpensive measurements crucial to public health. With people living with HIV (PLWH) living longer than ever before, large-scale administrative data are often used to estimate the prevalence and incidence of chronic age-associated comorbidities, which further inform intervention investment and prioritization that best addresses the healthcare needs of PLWH. However, to date, only a handful of methodological studies on administrative data have evaluated the influence of study design choices, such as the number of years of data examined, in the estimates of prevalence and incidence. As such, we conducted a data-driven exploration to determine optimal look-back windows (LBWs) for measuring prevalence and incidence of nine chronic comorbidities in PLWH, compared to HIV-negative controls, using administrative data. To do so, we applied multiple LBWs, from zero to 16 years, to estimate the prevalence and incidence of cardiovascular, kidney, lung, liver, and neurocognitive diseases, cancers, diabetes, hypertension, and osteoarthritis in 2012. The presence of these diseases in both populations was ascertained using BC-wide hospitalization, physician billing and drug dispensation databases. We found that a shorter LBW underestimated prevalence and overestimated incidence. On the contrary, extending LBWs resulted in a higher estimated prevalence, but had a relatively small effect on the incidence rates. The patterns of LBW greatly varied across comorbidities and between PLWH and HIV-negative controls. Overall, this study highlights the underacknowledged concern for widely varying incidence and prevalence estimates, with implications for treatment planning and intervention development in aging PLWH and beyond.