Menu Back to Poster-Presentations-Details

P124: Real-World Data in Comparative Effectiveness Research: Impacts of Pharmacotherapy and Social Determinants of Health

Poster Presenter

Brittin Wagner

Senior Research Scientist
PointClickCare
Canada

Objectives

Objective is to address comparative effectiveness of pharmacotherapy for AUD by examining Activities of Daily Living (ADL) outcomes across 9 medications for 16,814 treatment-seeking adults in skilled nursing facilities in the United States undergoing > 12 weeks of pharmacotherapy for AUD.?

Method

Criteria from a recently published comparative effectiveness review protocol for AUD was applied to deidentified EHR data from 265,479 US adults diagnosed with AUD. Of these, 16,814 were treatment-seeking adults who remained on one or more of 9 AUD medications for at least twelve weeks.

Results

We utilized real-world data providing clinical and demographic characterization of adults in our study, including social determinants of health (SDoH) measures, status as taking one or more of nine medications prescribed for AUD, and outcomes including activities of daily living at baseline and at 12-24 weeks post treatment initiation. We found that the average number of AUD medications prescribed to persons 18–64 is 1.3 times greater than 65+ year olds.? Persons 65+ compared to 18-64 years old are shown to have a 0.2-point significantly higher mean ADL.? However, when controlling for concurrent conditions, polypharmacy count and SDoH measure indicating barriers to care, ICD-10 Z Codes, age category is not a significant predictor of ADL.? Persons taking Naltrexone are shown to have a 1.0-point significantly lower mean ADL than those not taking it. ?They show the 2nd lowest ICD-10 Z code presence, lowest days in facility and ADL increase of AUD drugs.? Persons taking Baclofen are shown to have a 1.5-point significantly higher mean ADL than those not taking it. ?They show the 2nd highest presence of ICD-10 Z Code, highest days in facility and ADL increase.? The most prevalent condition was muscle weakness for Baclofen versus hypertension for all other AUD medications.? Findings are consistent with APA guidelines indicating that first line, FDA-approved drugs Naltrexone and Acamprosate are associated with better follow-up ADLs.? Persons with certain conditions (muscle weakness, chronic pain, cancer etc.) and/or taking multiple medications may experience different outcomes:? The two most common co-prescribed medications for this population are Gabapentin and Ondansetron. Our findings show they have opposite effects on ADLs (Table 2).?

Conclusion

Comparative effectiveness research on outcomes such as Quality of Life (QoL) is rare, as are studies that include measures of social determinants of health (SDoH). Our study utilizing large EHR, "real-world data" simulates a head-to-head clinical trial using retrospective data, and generates empirically-based treatment targets (e.g. those most likely to benefit from prescription treatment for AUD) and confirm populations unlikely to benefit from novel medication indications (e.g. those with conditions such as muscle weakness, which require other therapies). While this is a study on AUD, it is important to understand patient lifestyle priorities and conditions other than AUD that contribute to the findings. ?Our findings related to co-therapies and co-occurring conditions supports the use of meaningful, patient-centered measures for comparative effectiveness research. We especially find value of measures such as these for identifying ways to improve care for people with complex health conditions including substance use disorders like AUD.