Examining the Impact of Separate Components of a Multicomponent Intervention Designed to Reduce At-Risk Drinking Among Older Adults: The Project SHARE Study

Background Health promotion interventions often include multiple components and several patient contacts. The objective of this study was to examine how participation within a multicomponent intervention (Project SHARE) is associated with changes in at‐risk drinking among older adults. Methods We analyzed observational data from a cluster‐randomized trial of 31 primary care physicians and their patients aged ≥60 years, at a community‐based practice with 7 clinics. Recruitment occurred between 2005 and 2007. At‐risk drinkers in a particular physician's practice were randomly assigned as a group to usual care (n = 640 patients) versus intervention (n = 546 patients). The intervention included personalized reports, educational materials, drinking diaries, in‐person physician advice, and telephone counseling by health educators (HEs). The primary outcome was at‐risk drinking at follow‐up, defined by scores on the Comorbidity Alcohol Risk Evaluation Tool (CARET). Predictors included whether a physician–patient alcohol risk discussion occurred, HE call occurred, drinking agreement with the HE was made, and patients self‐reported keeping a drinking diary as suggested by the HE. Results At 6 months, there was no association of at‐risk drinking with having had a physician–patient discussion. Compared to having had no HE call, the odds of at‐risk drinking at 6 months were lower if an agreement was made or patients reported keeping a diary (odds ratio [OR] 0.58, 95% confidence interval [CI] 0.37 to 0.90), or if an agreement was made and patients reported keeping a diary (OR 0.52, CI 0.28 to 0.97). At 12 months, a physician–patient discussion (OR 0.61, CI 0.38 to 0.98) or an agreement and reported use of a diary (OR 0.45, CI 0.25) were associated with lower odds of at‐risk drinking. Conclusions Within the Project SHARE intervention, discussing alcohol risk with a physician, making a drinking agreement, and/or self‐reporting the use of a drinking diary were associated with lower odds of at‐risk drinking at follow‐up. Future studies targeting at‐risk drinking among older adults should consider incorporating both intervention components.