SEARCH

SEARCH BY CITATION

Keywords:

  • opioids;
  • analgesics;
  • prescription;
  • monitoring;
  • safety;
  • pharmacoepidemiology

ABSTRACT

Purpose

To estimate the association between prescription drug monitoring programs (PDMP) and the probability of analgesic use, overall and by analgesic type.

Methods

This cross-sectional study used 2007 Coordination of Benefits (COB) MarketScan administrative claims data of Medicare eligible and their dependents (n = 2 175 012). The exposure was PDMP status: no PDMP, electronic-only PDMP (ePDMP), or electronic + paper PDMP (e + pPDMP). Outcomes included any analgesic use and, among users, analgesic use by schedule (CII, CIII, CIV, or CV-Rx). Multivariable logistic and multinomial regressions were used to estimate the associations of PDMP status with any analgesic use and schedule of analgesic, respectively, controlling for sociodemographic and clinical factors.

Results

There were 834 489 (38.4%) subjects who received at least one analgesic; of these, 28.9% received one or more opioid analgesics (OAs). Compared to individuals in non-PDMP states, those living in PDMP states had increased odds of receiving any analgesic (ORePDMP = 1.19, 99%CI = 1.19, 1.20; ORe+pPDMP = 1.04, 99%CI = 1.03, 1.05). Among analgesic users, the odds of receiving potent CII analgesics relative to CV-Rx analgesics were lowest for individuals residing in e + pPDMP states (ORe+pPDMP = 0.54, 99%CI = 0.53, 0.55), followed by ePDMP states (ORePDMP = 0.76, 99%CI = 0.75, 0.77) relative to non-PDMP states. The odds of receiving CIII OAs were highest for individuals in PDMP compared to non-PDMP states.

Conclusions

PDMPs are associated with reductions in utilization of targeted prescription OAs and increases in less scrutinized, lower scheduled OAs. Longitudinal studies are needed to determine how PDMPs shift analgesic prescribing and whether such shifts influence clinical care and economic outcomes. Copyright © 2012 John Wiley & Sons, Ltd.