NarxCare, a product of a company called Bamboo Health, is a proprietary data analytics program. It uses up to 70 data points (that only Bamboo Health knows) and mixes them with your PDMP (prescription history) to assign four 3-digit scores letting your doctor or pharmacy know if you have a high risk of abuse or overdose. There is a Narcotic Score, a Stimulant Score, and a Sedative Score, and an Overdose Risk Score (ORS). According to Bamboo Health, "NarxCare aids care teams in clinical decision making, provides support to help prevent or manage substance use disorder, and empowers states with the comprehensive platform they need to take the next step in the battle against prescription drug addiction." Essentially, NarxCare pulls data from multiple state registries looking for red flags of drug seeking behavior.
The three categories of prescription medication it looks at are narcotics (opioids), stimulants (ADHD meds) and sedatives (benzos, sleeping meds, etc.). "The NarxCare report identifies risk factors with interactive visualizations, as well as an Rx Graph, and a set of scores that numerically correspond to the patient’s PDMP data." Although Appriss states on their website that their product shouldn't be used by itself to make medical decisions, that's not what's actually happening. As shown in this NarxCare article, patients are being denied medication or even being dismissed from a medical practice based on a NarxCare score alone.
Some of the risk factors used in the proprietary algorithm are:
- The number of prescribers a patient has had in a two-year period.
- The number of pharmacies a patient used in a two-year period.
- The dosage (MME-Milligram Morphine Equivalent).
- Amount of other medications that may increase potency of other medications.
- Number of times prescriptions overlap with prescriptions from other providers
- Any Mental Health Diagnosis
- Distance from patient to doctor
Nobody knows (except Bamboo health) exactly what goes into the ORS (Overdose Risk Score). Some of the factors we've found listed are criminal history, diagnoses in your EHR (Electronic Health Record) including mental health diagnosis (such as depression or PTSD), distance patient drives to the doctor, and diagnosis of sleep apnea.
What are some of the problems with NarxCare? Why isn't it a good thing?
- There is no room for context. What does this mean? As mentioned above, if someone has more than one prescriber within a two-year period, that will increase the ORS. But, there isn't a way to include an explanation such as if a patient moves, if a patient has multiple surgeries within that time period, or if a patient's doctor retires causing the need for a new physician.
- A patient's pet's prescription history is included. This means if a patient has surgery and if her pet has surgery, it appears as though the patient is getting prescriptions from more than one doctor.
- The NarxCare algorithm has never been externally validated. What does this mean? Essentially, it means the results of the study the algorithm is based on haven't been proven to apply to people outside of the study. Yet, that's exactly what they are doing. They are calling it evidence-based even though it is not.
- If someone is prescribed 90 MME (Morphine Milligram Equivalent) or more, the NarxCare score is automatically at the to 1% of all Overdose Risk Scores. Why is this bad? Well, the whole concept of MME and arbitrary thresholds such as 90 MME are flawed and not based in science.
- The NarxCare algorithm has inherent gender and racial bias. How? Since they use criminal history, it is racially biased. Since more it women tend to have mental health diagnoses such as PTSD due to trauma, it automatically is biased against women. We have heard stories of women who have been denied prescription opioids due to having been a survivor of sexual abuse.
As summarized in this Wired article, the entire concept of NarxCare is flawed and is flagging patients who shouldn't be flagged. "According to one study, 20 percent of the patients who are most likely to be flagged as doctor-shoppers actually have cancer, which often requires seeing multiple specialists. And many of the official red flags that increase a person's risk scores are simply attributes of the most vulnerable and medically complex patients, sometimes causing those groups to be denied opioid pain treatment."
Many people are fighting against the use of NarxCare. Almost every state uses Aprriss' PDMP platform and roughly 25 states in USA use NarxCare. Contact your local Department of Health to find out if your state uses it. For more information about Narxcare, we've included links to a few articles.