A recent study estimates that more than 275,000 people -- or 4.6 percent of people over the age of 11 in Massachusetts- have opioid use disorder, a figure nearly four times higher than previous estimates based on national data. Led by researchers at Boston Medical Center, the study involved an analysis of several linked public health data sets, allowing for a more accurate picture of the impact of opioids in the state. Published online in the American Journal of Public Health, the results indicate that better surveillance systems need to be deployed locally and nationally to make more accurate opioid use disorder prevalence data available to better address the epidemic.
Traditional methods to estimate prevalence only count those who have interacted with the health care system, which are included in both state and national records. However, for those who don't encounter the system and perhaps overdose, they would not be included in any estimates. Individuals with substance use disorders are less likely to seek medical care or be insured. Additionally, when people are asked about sensitive topics, like drug use, by their provider or in a health survey, they typically don't self-report for many reasons, including stigma.
"There are many people with opioid use disorder who do not encounter the health care system, which we know is a barrier to understanding the true impact of the opioid epidemic," said Joshua Barocas, MD, an infectious disease physician at BMC who led the study. For this study, the researchers examined the Massachusetts Public Health Data Warehouse, a novel database that links data from 16 state agencies for a comprehensive analysis of health care system utilization and fatal and non-fatal opioid overdoses rates. Through a capture-recapture analysis, the researchers were able to identify those known to have an opioid use disorder based on their encounters with the health care system as well as estimate those who have the disorder but aren't included in the data.
According to the results, in 2011 and 2012, the prevalence of opioid use disorder in Massachusetts for those over the age of 11 was 2.72 percent and 2.87 percent, respectively. That increased to 3.87 percent in 2013, and even more, to 4.6 percent in 2015. Surprisingly, those between the ages of 11 and 25 experienced the greatest increase in prevalence of all age groups. The number of "known" persons increased throughout the study period -- from 63,989 in 2011 to 75,431 in 2012, and 93,878 in 2013 to 119,160 in 2015.
The data was also sorted by county, and counties with rural populations had some of the highest prevalence estimates, including: Barnstable (5.03 percent); Berkshire (6.06 percent); Bristol (5.81 percent); Hampden (5.34 percent); and Worcester (4.38 percent). For Suffolk County, which encompasses Boston, the prevalence in 2015 was 3.36 percent.
"Our methodology coupled with the data available from the Department of Public Health provides an innovative approach to more accurately assess opioid use disorder in Massachusetts, which indicates that the issue is more widespread than previous available data was showing," added Barocas, assistant professor of medicine at Boston University School of Medicine. "The information from linked databases can and should be implemented by other states trying to better understand and change the course of the opioid epidemic, which includes deploying resources in areas with the greatest need."
The researchers note that while fentanyl is likely a cause in the surge of overdoses in Massachusetts, the increased prevalence of opioid use disorder is likely also contributing to the continuing increase in overdose deaths. They also highlight that, given the clinical complexity of opioid use disorder, better surveillance can help increase the accuracy of diagnosing and treating the disease. Unlike diseases such as HIV or hepatitis C virus, where a blood test can diagnose a patient, opioid use disorder is based on a clinical diagnosis.
Materials provided by Boston Medical Center. Note: Content may be edited for style and length.
Cite This Page: