A Predictive Model to Identify Skilled Nursing Facility Residents for Pharmacist Intervention

Area of Science:

• Pharmacoeconomics
• Health Services Research
• Geriatric Pharmacy

Background:

• Skilled nursing facilities (SNFs) have complex medication regimens.
• Identifying patients who benefit from clinical pharmacist interventions is crucial for optimizing care.
• Current methods for identifying these patients may not be efficient.

Purpose of the Study:

• To develop and validate a predictive model to identify patients in SNFs requiring clinical pharmacist intervention.
• To utilize administrative data for predicting the need for medication reviews.

Main Methods:

• A retrospective, cross-sectional study was conducted in nine SNFs.
• Multivariable logistic regression was used to identify predictive factors from administrative data.
• A model was developed in a derivation cohort and validated in a separate cohort.

Main Results:

• The final model included 22 factors and demonstrated good predictive performance (AUC = 0.79).
• Key predictors for requiring intervention included clopidogrel dispensing, falls, and vertebral fracture diagnoses within 180 days.
• The model showed consistent performance in the validation cohort (AUC = 0.79).

Conclusions:

• Administrative data can effectively predict which SNF patients require clinical pharmacist intervention.
• Real-time application of this model can lead to time savings for pharmacists.
• This can enhance pharmacy services through more focused patient care and resource allocation.