Table 1 Study level characteristics of 22 included studies
Study ID | Included Jurisdiction with a PMP | Study Population | Study Design | Outcome Groups | Outcome Data Source, Years |
|---|---|---|---|---|---|
Ali 2017 [24] | All states except MO and DC | general population | Pooled Cross-sectional Logit and generalized linear models | illicit opioid usea, opioid dependencea | National Survey on Drug Use and Health (NSDUH), 2004–2014 |
Birk 2017 [43] b | All states that established PMPs between 1998 and 2012 | general population, opioid-related deaths | Controlled interrupted time series | opioid-related care outcomes, opioid-related adverse events | Treatment Episode Data Set (TEDS), 1998–2012; National Vital Statistics System (NVSS), 1999–2012 |
Branham 2018 [27] b | AL, AZ, CO, CT, IN, IA, LA, ME, MN, MS, NV, NM, NC, ND, OH, SC, TN, UT, VT, VA, WV, WY | people who use drugs admitted to treatment | Controlled interrupted time series | opioid-related care outcomesa | TEDS, 1992–2012 |
Dave 2017 [44] b | AK, AZ, AR, CO, CT, DE, FL, GA, IA, KS, LA, ME, MD, MN, MS, MT, NE, NH, NJ, NM, NC, ND, OH, OR, SC, SD, TN, WA, WI, WY, VT, VA | general population | Controlled interrupted time series | opioid-related care outcomes | TEDS, 2003–2014 |
Delcher 2015 [50]c | FL | drug-related deaths | Controlled interrupted time series | opioid-related adverse eventsa | Florida Medical Examiners Commission, 2003–2012 |
Kim 2013 [51] | AL, AZ, CA, CO, CT, IL, IN, LA, ME, MI, MS, NM, NC, ND, OH, OK, SC, TN, VA, WV, NY, HI, ID, OA, RI, TX, MA, NV, UT, WY, KY | general population | Interrupted time series | opioid-related adverse events | CDC Wide-Ranging Online Data for Epidemiologic Research (WONDER); 1999–2008 |
Kinsell 2015 [40] c | NY, FL | general population | Controlled before and after/ Difference-in-difference approach | opioid-related adverse eventsa, opioid-related care outcomesa | CDC WONDER & Florida Medical Examiner Drug Related Death Data & Florida Agency for Health Care Administration & New York State Inpatient Databases (SID)/ State Emergency Department Databases (SEDD), 2009–2012 |
Li 2014 [52] | MI, VA, NY, ME, MS, NC, CA, MA, TX, AZ, SC, TN, IL, RI, CO, CT, OH, AL, LA, OK, HI, ID, NM, KY, IN, UT, PA, WV, WY, ND, NV | general population | Controlled interrupted time series | opioid-related adverse events | NVSS, 1999–2008 |
Mallatt 2017 [55] c | NSDUH: 49 states (excluding MO) ARCOS: AK, AL, AR, AZ, CA, CO, CT, DE, FL, GA, HI, IA, ID, IL, IN, KS, KY, LA, MA, MD, ME, MI, MN, MS, MT, NC, ND, NE, NH, NJ, NM, NV, NY, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VA, VT, WA, WI, West VA, WY NIBRS: 26 states | general population | Repeated cross-sectional (NSDUH); interrupted time series (NIBRS & ARCOS) | opioid-related legal and criminal outcomesa | National Incident-Based Reporting System (NIBRS), 2004–2014 |
Maughan 2015 [25] c | Detroit, Phoenix, San Francisco, Denver, Chicago, Boston, Miami, Minneapolis, New York City | general population | Generalized Estimate Equations with repeated measures | opioid-related care outcomes | eDrug Abuse Warning Network (DAWN), 2004–2011 |
McLaughlin 2016 [28] | AL, AZ, CA, CO, CT, IL, IN, IA, LA, ME, MI, MN, MS, NE, NM, NC, ND, OH, OK, SC, TN, VT, VA, WV, AK, NY, HI, ID, PA, RI, TX, MA, NJ, DE, NV, OR, UT, WA, WY, KS, SD, WI, FL, KY, MT | general population | Repeated cross sectional | illicit opioid usea | NSDUH, 2004–2012 |
Meara 2016 [48] | AL, AZ, CA, CO, CT, IL, IN, IA, LA, ME, MI, MN, MS, NE, NM, NC, ND, OH, OK, SC, TN, VT, VA, WV, AK, NY, HI, ID, PA, RI, TX, MA, NJ, DE, NV, OR, UT, WA, WY, WY, SD, WI, FL, KY, MT | insured population | Cohort; before and after | opioid-related adverse events | Medicare Provider Analysis and Review (MEDPAR), Outpatient, Carrier, and Medicare Beneficiary Summary files, 2006–2012 |
Meinhofer 2017 [53] | AL, AK, AZ, AR, CO, CT, DE, FL, GA, IA, WY, LA, ME, MD, MN, MS, MT, NE, NJ, NM, NC, ND, OH, OR, SC, SD, TN, VT, VA, WA, WI WY | general population | Interrupted time series | opioid-related adverse eventsa | NVSS, 2000–2013 |
Nam 2017 [45] c | AL, AZ, CO, CT, IA, LA, ME, MN, MS, NC, ND, NM, OH, SC, TN, VA, VT, WV, WY | general population | Interrupted time series | opioid-related adverse eventsa | CDC WONDER, 1999–2014 |
Patrick 2016 [47] b | WA, OR, AZ, NM, CO, WY, MT, SD, NE, WY, MN, IA, AR, LA, MS, AL, GA, FL, SC, NC, TN, VA, WV, OH, MD, DE, NJ, CT, VT, ME, AK, ND, WI, KY | general population | Controlled interrupted time series | opioid-related adverse events | CDC WONDER, 1999–2013 |
Paulozzi 2011 [46] | CA, HI, ID, IL, IN, KY, ME, MA, MI, NV, NM, NY, OK, PA, TX, UT, VA, WV, WY | general population | Controlled interrupted time series | opioid-related adverse events | CDC WONDER, 1999–2005 |
Pauly 2018 [49] | 48 states (all but MO and PA), and DC | insured population | Interrupted time series | opioid-related adverse events | Truven Health Marketscan claims database, 2004–2014 |
Radakrishnan 2015 [39] b | All that fall into 1992–2011 | general population | Interrupted time series/ Repeated cross section | opioid-related care outcomesa, opioid-related adverse eventsa | TEDS, 1992–2010; NVSS, 1999–2010 |
Reifler 2012 [42] c | CT, MA, ME, NY, PA, RI, AZ, CA, CO, HI, ID, NV, UT, WY, IA, IL, IN, MI, ND, OH, AL, KY, MS, NC, OK, SC, TN, TX, VA, WV, WA | general population | Controlled interrupted time series | opioid-related adverse events, opioid-related care outcomesa | Researched Abuse Diversion and Addiction-Related Surveillance System (RADARS), 2003–2009 |
Reisman 2009 [19] | CA, HI, IL, IN, MA, MI, NY, OK, TX, ID, KY, NV, RI, UT | people who use drugs admitted to treatment | Cross-sectional (with interrupted time series) | opioid-related care outcomes | TEDS, 2003 |
Simeone 2006 [41] | 20 states | general population | Prospective cohort study | opioid-related care outcomes | TEDS, 1997 & 2003 |
Surratt 2014 [26] c | FL | general population | Interrupted time series | opioid-related legal and criminal outcomes | RADARS, 2009–2012 |