The reason these reductions had not worked, some experts believed, was because they had not targeted the patients most at risk. About 70% of adults have taken opioids for medical purposes, but only 0.5% have what is officially called “opioid use disorder,” more commonly known as addiction. One study found that even in the most at-risk age group, adolescents and people in their early twenties, only one in 314 privately insured patients prescribed opioids developed problems with it. them.
Researchers have known for years that some patients are at greater risk for addiction than others. Studies have shown, for example, that the more a person has had adverse childhood experiences, such as being abused or neglected or losing a parent, the greater their risk. Another significant risk factor is mental illness, which affects at least 64% of all people with opioid use disorders. But while the experts were aware of these dangers, they had no good way to quantify them.
That started to change as the opioid epidemic escalated and demand increased for a simple tool that could more accurately predict a patient’s risk. One of the first of these measures, the Opioid Risk Tool (ORT), was published in 2005 by Lynn Webster, former president of the American Academy of Pain Medicine, who now works in the pharmaceutical industry. (Webster has also previously received speaking fees from opioid manufacturers.)
To build the ORT, Webster began by researching studies that quantified specific risk factors. Along with the literature on negative childhood experiences, Webster found studies linking risk to personal and family history of addiction, not only to opioids, but to other drugs, including alcohol. . He also found data on the high risk of particular psychiatric disorders, including obsessive-compulsive disorder, bipolar disorder, schizophrenia and major depression.
Putting all of this research together, Webster designed a short questionnaire for patients to determine if a person had any of the known risk factors for addiction. Then he found a way to summarize and weight the responses to generate an overall score.
The ORT, however, was sometimes heavily skewed and limited by its data sources. For example, Webster found a study showing that a history of sexual abuse in girls tripled their risk of addiction. Why only them? Because no similar study had been done on boys. The gender bias this introduced into ORT was particularly strange given that two-thirds of all addictions occur in men.
ORT also ignored the fact that a patient had been prescribed opioids for long periods of time without becoming dependent.
Webster says he did not intend to use his tool to decline treatment for pain, but only to determine who should be watched more closely. As one of the first filters available, however, it quickly won over doctors and hospitals keen to stay on the safe side of the opioid crisis. Today, it has been integrated into multiple electronic health record systems, and is often used by physicians concerned about overprescribing. It is “very, very widely used in the United States and five other countries,” says Webster.
Compared to early opioid risk scanners like ORT, NarxCare is more complex, more powerful, more rooted in law enforcement, and much less transparent.
Appriss began in the 1990s by creating software that automatically alerts victims of crime and other “concerned citizens” when a specific incarcerated person is about to be released. He later switched to health care. After developing a series of databases to monitor prescriptions, Appriss acquired in 2014 what was then the most commonly used algorithm to predict who was most at risk for “controlled substance abuse,” a program developed by the National Association of Boards of Pharmacy, and began to develop and expand it. Like many companies that provide software to track and predict opioid addiction, Appriss is largely funded, directly or indirectly, by the Department of Justice.
NarxCare is one of the many predictive algorithms that have proliferated in several areas of life in recent years. In medical settings, algorithms have been used to predict which patients are most likely to benefit from a particular treatment and to estimate the likelihood that an intensive care patient will deteriorate or die if released.
In theory, the creation of such a tool to indicate when and to whom opioids are prescribed could be useful, perhaps even in addressing medical inequalities. Studies have shown, for example, that black patients are more likely to be denied pain medication and more likely to be seen as drug-seeking. A more objective predictor could, again, in theory, help under-medication patients get the treatment they need.