Health System‐Based Unhealthy Alcohol Use Screening and Treatment Comparing Demographically Matched Participants With and Without HIV

Abstract

Background: Unhealthy alcohol use among persons living with HIV (PLWH) is linked to significant morbidity, and use of alcohol services may differ by HIV status. Our objective was to compare unhealthy alcohol use screening and treatment by HIV status in primary care. Methods: Cohort study of adult (≥18 years) PLWH and HIV-uninfected participants frequency matched 20:1 to PLWH by age, sex, and race/ethnicity who were enrolled in a large integrated healthcare system in the United States, with information ascertained from an electronic health record. Outcomes included unhealthy alcohol screening, prevalence, provider-delivered brief interventions, and addiction specialty care visits. Other predictors included age, sex, race/ethnicity, neighborhood deprivation index, depression, smoking, substance use disorders, Charlson comorbidity index, prior outpatient visits, insurance type, and medical facility. Cox proportional hazards models were used to compute hazard ratios (HR) for the outcomes of time to unhealthy alcohol use screening and time to first addiction specialty visit. Poisson regression with robust standard errors was used to compute prevalence ratios (PR) for other outcomes. Results: 11,235 PLWH and 227,320 HIV-uninfected participants were included. By 4.5 years after baseline, most participants were screened for unhealthy alcohol use (85% of PLWH and 93% of HIVuninfected), but with a lower rate among PLWH (adjusted HR 0.84, 95% CI 0.82 to 0.85). PLWH were less likely, compared with HIV-uninfected participants, to report unhealthy drinking among those screened (adjusted PR 0.74, 95% CI 0.69 to 0.79), and among those who screened positive, less likely to receive brief interventions (adjusted PR 0.82, 95% CI 0.75 to 0.90), but more likely (adjusted HR 1.7, 95% CI 1.2 to 2.4) to have an addiction specialty visit within 1 year. Conclusions: Unhealthy alcohol use was lower in PLWH, but the treatment approach by HIV status differed. PLWH reporting unhealthy alcohol use received less brief interventions and more addiction specialty care than HIV-uninfected participants.

Patient and provider factors associated with receipt and delivery of brief interventions for unhealthy alcohol use in primary care

“…The findings are consistent with recent research finding that people with HIV were less likely to receive a BI than a matched sample of HIV-negative patients (Silverberg et al, 2020). Likewise, people with more clinical encounters-visits at outpatient clinics or emergency departments, or hospitalizations-in the preceding year were less likely to receive a BI, perhaps for the same reason.…”

Racial, ethnic, and age disparities in the association of mental health symptoms and polysubstance use among persons in HIV care

“…Communication with patients, including via providers, may help impress the importance of screening measures to improve care quality and their own satisfaction. While a completion level of 31% is similar to that of other primary care-based screening studies [ 55 , 57 ], our findings could have been impacted. If PWH who did not complete a screen were more likely to have mental health and substance use problems, our findings could underestimate their prevalence in this population, particularly among Hispanic PWH, as PWH without a screen were more likely to be Hispanic.…”

The relationship of smoking and unhealthy alcohol use to the HIV care continuum among people with HIV in an integrated health care system

“…Unhealthy alcohol use was defined as either reporting any days of drinking 4+/5+ drinks in the prior 3 months or reporting an average 8+/15+ drinks in a week in the last 90 days, based on clinical guidelines. Lower limits are applied to women and men over the age of 65 and upper limits are applied to men 65 years old or younger (Silverberg et al, 2020; U.S. Department of Health and Human Services and National Institute on Alcohol Abuse and Alcoholism, 2005). Both unhealthy alcohol use and smoking status are collected during routine rooming procedures by medical assistants (MAs) (Mertens et al, 2015).…”

“…Neighborhood deprivation affects multiple health conditions, so an index may have more relevance than single-item measures of SES. The NDI has been used in studies of diabetes (Stoddard et al, 2013) and in our prior work on unhealthy alcohol use screening among people with HIV (Silverberg et al, 2020). Participants' NDI scores were divided into quartiles, with quartile 1 representing the least deprived area and quartile 4 representing the most deprived area.…”

“…We did not include the effects of smoking and AUD treatment in our models to keep the study focused on the potential associations of smoking and unhealthy drinking on the HIV care continuum, and since this would require lengthening the time windows associated with our outcomes. However, prior studies at KPNC have documented that relatively few PWH receive smoking treatment and AUD specialty care (Lam et al, 2020;Silverberg et al, 2020). Thus, the potential impact would likely be low in our sample.…”