TY - JOUR T1 - Patterns of opioid use (codeine, morphine or meperidine) in the Canadian population over time: analysis of the Longitudinal National Population Health Survey 1994–2011 JF - BMJ Open JO - BMJ Open DO - 10.1136/bmjopen-2019-029613 VL - 9 IS - 7 SP - e029613 AU - Mayilee Canizares AU - J Denise Power AU - Y Raja Rampersaud AU - Elizabeth M Badley Y1 - 2019/07/01 UR - http://bmjopen.bmj.com/content/9/7/e029613.abstract N2 - Objective This study aimed to investigate cohort effects in selected opioids use and determine whether cohort differences were associated with changes in risk factors for use over time.Design This study presents secondary analyses of a longitudinal survey panel of the general population that collected data biannually.Setting Data from the Canadian Longitudinal National Population Health Survey 1994–2011.Population This study included 12 542 participants from the following birth cohorts: post-World War I (born 1915–1924), pre-World War II (born 1925–1934), World War II (born 1935–1944), Older Baby Boom (born 1945–1954), Younger Baby Boom (born 1955–1964), Older Generation X (born 1965–1974) and Younger Generation X (born 1975–1984).Main outcome Responses to a single question asking about the use of codeine, morphine or meperidine in the past month (yes/no) were examined.Results Over and above age and period effects, there were significant cohort differences in selected opioids use: each succeeding recent cohort had greater use than their predecessors (eg, Gen Xers had greater use than younger baby boomers). Selected opioids use increased significantly from 1994 to 2002, plateauing between 2002 and 2006 and then declining until 2011. After accounting for cohort and period effects, there was a decline in use of these opioids with increasing age. Although pain was significantly associated with greater selected opioids use (OR=3.63, 95% CI 3.39 to 3.94), pain did not explain cohort differences. Cohort and period effects were no longer significant after adjusting for the number of chronic conditions. Cohort differences in selected opioids use mirrored cohort differences in multimorbidity. Use of these opioids was significantly associated with taking antidepressants or tranquillisers (OR=2.52, 95% CI 2.27 to 2.81 and OR=1.60, 95% CI 1.46 to 1.75, respectively).Conclusions The findings underscore the need to consider multimorbidity including possible psychological disorders and associated medications when prescribing opioids (codeine, morphine, meperidine), particularly for recent birth cohorts. Continued efforts to monitor prescription patterns and develop specific opioid use guidelines for multimorbidity appear warranted. ER -