Choice of measurement approach for area-level social determinants of health and risk prediction model performance.

OBJECTIVE: The objective of this paper is to provide empirical guidance by comparing the performance of six different area-level SDoH measurement approaches in predicting patient referral to a social worker and hospital admission after a primary care visit. METHODS: We compared the performance of six area-level SDoH measurement approaches in predicting patient referral to a social worker and hospital admission after a primary care visit using random forest classification algorithm. Data came from 209,605 patient encounters at a federally qualified health center. Models with each area-based measurement approach were compared against the patient-level data only model using area under the curve, sensitivity, specificity, and precision. RESULTS: Addition of area-level features to patient-level data improved the overall performance of models predicting need for a social worker referral. Entering area-level measures as individual features resulted in highest model performance. CONCLUSION: Researchers seeking to include area-level SDoH measures in risk prediction may be able to forego more complex measurement approaches.

Evaluating expected outcomes of acid remediation in an intensively mined Appalachian watershed.

Assessments of watershed-based restoration efforts are rare but are essential for the science of stream restoration to advance. We conducted a watershed scale assessment of Abram Creek before and after implementation of a watershed-based plan designed to maximize ecological recovery from acid mine drainage (AMD) impairment. We surveyed water chemistry, physical habitat, benthic macroinvertebrates, and fish community structure in three stream types: AMD-impacted (14 streams), AMD-treated (13 streams), and unimpaired reference (4 streams). We used in-stream measurements to quantify ecological loss from AMD, the amount of ecological recovery expected through remediation, and the observed degree of post-treatment recovery. Sites impaired by AMD improved in water quality with AMD treatment. Dissolved metals and acidity declined significantly in treated streams, but sulfate and specific conductance did not. Likewise, sites impaired by AMD improved in bio-condition scores with AMD treatment. EPT genera increased significantly but were lower compared to unimpaired streams. We found fish at nine treated sites that had none before treatment. Community-level analyses indicated improved but altered assemblages with AMD treatment. Analysis of pre-treatment conditions indicated that only 30% of the historic fishery remained. Remediation was expected to recover 66% of the historic fishery value, and assessment of post-treatment conditions indicates that 52% of the historic fishery has been recovered after 3 years. Developing expected endpoints for restoration outcomes provides a tool to objectively evaluate successes and can guide adaptive management strategies.