Do prescription drug monitoring programs encourage prescription – or illicit – opioid abuse?

Abstract

Background

Prescription drug monitoring programs (PDMPs) are tools that states can use to fight prescription opioid misuse within their jurisdiction. However, because PDMPs make prescription opioids more difficult to access, these programs may have the unintended consequence of increasing deaths related to illicit opioids. Methods: This study uses fixed effects models to estimate how PDMP regulatory strength is associated with both prescription opioid- and heroin-related deaths between 1999 and 2016. PDMP regulatory strength is measured by creating a score using multiple correspondence analysis (MCA). Additional models replace the MCA score with a binary indicator for the presence of one particular regulation requiring physicians to access the system before writing opioid prescriptions. Results: Results show that continuous measures of PDMP strength are not generally associated with prescription opioid- or heroin-related death rates. Yet, one model does show that PDMP scores are positively associated with the heroin-related death rate. The models using the binary mandatory access variable show a strong positive association with both prescription opioid and heroin deaths. Conclusions: This study supports the theory that more stringent state PDMPs are associated with higher rates of heroin-related deaths, potentially due to decreases in prescription opioid availability.

Keywords:

• Prescription drug monitoring programs
• opioid epidemic
• heroin
• fixed effects models
• multiple correspondence analysis
• panel data

Notes

i Missouri is the only state without a statewide PDMP program. St. Louis County operates a PDMP which includes 72 jurisdictions within the state as of February, 2019.¹

ii If a regulation began after July 31^st, then it was not considered to be in effect for that year. This coding ensures that the regulations were in place for more than five months out of the year.

iii The ICD-10 codes for prescription opioid deaths that are gathered for this study are: T40.2 (Other opioids), T40.3 (Methadone), and T40.4 (Other synthetic narcotics). The ICD-10 code for heroin deaths is T40.1 (Heroin). Note that prescription opioids obtained through diverted prescriptions would be included in the prescription opioid and not the illicit (heroin) category.

iv Age-adjusted and crude death rates differ since the age-adjusted rates account for differences in state age distributions.³⁰ Crude death rates are simply calculated by taking the number of deaths in a state in a given year and dividing it by the state’s population during that time period, then multiplying the resulting value by 100,000.

v During the data exploration process, discrepancies were discovered between the weights that Pardo assigned to the regulations and the corresponding descriptive statistics. In order to reproduce Pardo’s descriptive statistics, the weights were modified to fit their findings. The weights for the regulations Disclosure, Frequency, and Retention time were adjusted, and are different than the weights Pardo used in his analysis.