Behavioral Health, Mental Health, Substance Abuse
Predictors of the Propensity of Marylanders to Ever Talk with Their Physicians about the Risks of Taking Prescription Opioids
(School of Pharmacy (UMB) PHSR Faculty)
Introduction: Opioid misuse is a major public health concern. In 2014, 206,000 Marylanders reported non-medical use of prescription opioids (POs), which contributed to 329 PO related fatalities. Healthcare providers can educate the public about the risks of use and misuse of PO(s). Our study examines the predictors for the propensity of Marylanders to discuss the risks of PO use with their physicians. Methods: Demographics, PO use, awareness and perceptions related to misuse in the community were gathered from a 52 question Maryland Public Opinion Survey on Opioids over three weeks between February and March of 2015, using web-based methods to reach statewide audience. The propensity of Marylanders to discuss the risks of using PO(s) with their physicians was predicted by the multivariable logistic regression model Results: A total of 6623 Marylanders took the survey (63.14% females; 37.90% > 45 years-old; 71.46% white; 57.21 % with minimum college education; 35.94 % from Southern Maryland). The question of discussing the risks of PO use with the health care provider was responded by 3259 people (17% with physicians and 6.39% with pharmacists). Those who have ever used PO (OR=1.4491, CI (1.243, 1.789)) and heroin (OR=2.206; CI (1.675, 2.906)), not perceive PO misuse (OR=4.123, CI (2.258, 7.530)) and/or with less than a college education (OR=1.191; CI (1.003, 1.415)) are significantly more likely to discuss the risks of PO use with physicians than the others. The 18-25 age-group (OR=0.736; CI (0.546, 0.993)) are less likely talk with their physicians about PO risks than the older group. Conclusion: Higher educational level, ever use of PO(s) or heroin and lack of perception to PO misuse seem to be good predictors to discuss with health care providers the risks and dangers of PO use, controlling for other variables such as ethnicity, gender and regions.