scorecardresearch Skip to main content

New study suggests opioid addiction in Mass. is much worse than thought

The study estimates that 275,000 Massachusetts residents suffered from opioid use disorder in 2015.SUZANNE KREITER/Globe Staff/File

The crisis of opioid addiction in Massachusetts, long known to be severe, may be even worse than thought — as much as four times worse, according to new research.

A study from Boston Medical Center published Thursday estimates that 275,000 Massachusetts residents, or 4.6 percent of people older than age 11, suffered from opioid use disorder in 2015. Previous estimates based on national surveys pegged the number at just over 1 percent.

More than half the people with opioid addiction haven’t been counted before because they have not obtained health care services related to opioid misuse, the researchers said.


The study, in the American Journal of Public Health, also found that the percentage of people suffering from opioid addiction nearly doubled over the course of four years.

“It’s a good wake-up call,” said Dr. Joshua A. Barocas, an infectious disease physician at Boston Medical Center who led the study. “Our pool of people who are at risk for overdoses is potentially higher than we thought it was.”

The study adds a new dimension to an epidemic until now measured primarily by the number of overdose deaths, said Dr. Michael F. Bierer, president of the Massachusetts Society of Addiction Medicine, who was not involved in the research. It suggests that the rise in addiction may play as an important a role in the high death rate as the prevalence of deadly fentanyl, he said.

“The overdose problem is not just due to a bad batch of street drugs,” said Bierer, a primary care physician at Massachusetts General Hospital. “We’re going to have people sick with opioid use disorder for years and years, and we need to prepare to be able to take care of them.”

But one specialist expressed skepticism about the numbers the Boston group put forth. Dr. Silvia S. Martins, director of the Substance Abuse Epidemiology Unit at the Columbia University Mailman School of Public Health, said the researchers used an overly broad definition of opioid use disorder that might have led to overstating the incidence.


“The analysis could have been done in a more precise way,” she said.

Martins agreed, however, that the prevalence of opioid addiction has been underestimated. “We know most people who have the disorder aren’t even seen by a doctor,” she said.

Delving into a unique Massachusetts data source, the researchers tallied those thought to have an opioid use disorder based on their encounters with the health care system — 119,000 people in 2015, or 2 percent of the population older than 11.

Then they used statistical methods employing the laws of probability to estimate the number of Massachusetts residents who have the disorder but remain undiagnosed — an additional 156,000, or 2.6 percent.

Looking back to 2011, researchers found that opioid addiction increased the most among people ages 11 to 25 (from a third of 1 percent in 2011 to 1.7 percent in 2015).

They also noted geographic differences, with rural areas among the hardest hit.

The proportion of opioid-addicted residents was 6 percent in Berkshire County, 5.8 percent in Bristol County, 5.3 percent in Hampden County, and 5 percent in Barnstable County. For Suffolk County, which includes Boston, the prevalence was 3.4 percent, below the state average.

Funded by the National Institute on Drug Abuse, the National Institute of Allergy and Infectious Diseases, and the Boston University School of Medicine, the study was prompted by a sense among clinicians that the extent of the addiction problem had been underestimated, Barocas said.


Previous prevalence studies were based on interviews, reaching only people who had interacted with the health care system and leaving out those who didn’t admit to using drugs.

To get a better idea of what’s really going on, the researchers examined the Massachusetts Public Health Data Warehouse — a trove established by a 2015 state law that links data from 16 state agencies.

The data contain patient-specific information about such events as hospital and emergency room visits, ambulance runs, insurance claims, opioid-related deaths, and prescriptions. Patients are not identified, but each person has a unique number, so it’s possible to tell, for example, that someone rescued from an overdose was the same person later admitted to a hospital.

The researchers were thus able to count the number of people who had experiences indicating opioid addiction, such as receiving methadone. Then, borrowing a method that wildlife scientists use to estimate animal populations, they extrapolated the number of people who suffer from addiction but have not received health care services or died of an overdose.

Stoddard Davenport, a health care management consultant with Milliman, a large national actuarial and consulting firm, applauded the effort to estimate “unknown” patients.

“It’s great to see some work on that front,” said Davenport, who read the study but wasn’t involved with it.


“There’s a lot more folks suffering with opioid use disorder than the health care system has seen.”

The Boston group’s methodology for estimating “unknowns” has been employed in other public health research, but Davenport said he had never seen it applied to the current opioid epidemic.

He described the results as plausible and noted that they jibe with his own research showing only a fraction of people who use large amounts of opioids have been diagnosed with opioid use disorder.

Felice J. Freyer can be reached at Follow her on Twitter @felicejfreyer