Carla Shoff1, Tse-Chuan Yang2, Seulki Kim2. 1. Office of Enterprise Data and Analytics, Centers for Medicare & Medicaid Services, Baltimore, Maryland. 2. Department of Sociology, University at Albany, Albany, New York.
Abstract
PURPOSE: While research has been done comparing rural/urban differences in opioid prescribing to the disabled Medicare Part D population, research on opioid prescribing among the aged Medicare Part D population is lacking. This study aims to fill this gap by exploring the predictors of opioid prescribing to aged Medicare Part D beneficiaries and investigating whether these predictors vary across rural and urban areas. METHODS: This is an analysis of ZIP Codes in the continental United States (18,126 ZIP Codes) utilizing 2017 data from Centers for Medicare & Medicaid Services. The analytic approach includes aspatial descriptive analysis, exploratory spatial analysis with geographically weighted regression, and explanatory analysis with spatial error regime modeling. FINDINGS: Both beneficiary and prescriber characteristics play an important role in determining opioid prescribing rates in urban ZIP Codes, but most of them fail to explain the opioid prescribing rates in rural ZIP Codes. CONCLUSION: We identify potential spatial nonstationarity in opioid prescribing rates, indicating the complex nature of opioid-related issues. This means that the same stimulus may not lead to the same change in opioid prescribing rates, because the change may be place specific. By understanding the rural/urban differences in the predictors of opioid prescribing, place-specific policies can be developed that can guide more informed opioid prescribing practices and necessary interventions.
PURPOSE: While research has been done comparing rural/urban differences in opioid prescribing to the disabled Medicare Part D population, research on opioid prescribing among the aged Medicare Part D population is lacking. This study aims to fill this gap by exploring the predictors of opioid prescribing to aged Medicare Part D beneficiaries and investigating whether these predictors vary across rural and urban areas. METHODS: This is an analysis of ZIP Codes in the continental United States (18,126 ZIP Codes) utilizing 2017 data from Centers for Medicare & Medicaid Services. The analytic approach includes aspatial descriptive analysis, exploratory spatial analysis with geographically weighted regression, and explanatory analysis with spatial error regime modeling. FINDINGS: Both beneficiary and prescriber characteristics play an important role in determining opioid prescribing rates in urban ZIP Codes, but most of them fail to explain the opioid prescribing rates in rural ZIP Codes. CONCLUSION: We identify potential spatial nonstationarity in opioid prescribing rates, indicating the complex nature of opioid-related issues. This means that the same stimulus may not lead to the same change in opioid prescribing rates, because the change may be place specific. By understanding the rural/urban differences in the predictors of opioid prescribing, place-specific policies can be developed that can guide more informed opioid prescribing practices and necessary interventions.
Authors: Douglas R Oyler; Svetla Slavova; Patricia R Freeman; Zhengyan Huang; Jeffery Talbert; Sharon L Walsh; Philip M Westgate Journal: J Rural Health Date: 2022-02-27 Impact factor: 5.667