As you know, one of our issues with NarxCare is that it has never been externally validated. Bamboo Health/Appriss held a webinar on October 27, 2021. The webinar was named "External Validation of NarxCare as Useful Clinical Decision Support Tool." Here is a recording in case you're interested in watching the 30-minute webinar. They state the aim of the paper the webinar was based on is to "validate the NS metric compared to WHO ASSIST and identify high, moderate and low opioid risk thresholds and to provide actionable data."
Before I go into detail about this webinar, I want to tell you a story about a hypothetical patient named Rachel.
Rachel has had the same prescriber for years, but is moving across the country and has to find a new doctor. According to Rachel's Narcotic Score (NS), Rachel now has two prescribers. This new doctor has a PA in his office that sometimes writes Rachel's prescriptions. Rachel now has 3 prescribers within the last two years. Six months after Rachel started seeing this new doctor, he is raided by the DEA and can no longer prescribe. Rachel finds a new doctor. She now has had 4 prescribers. Rachel's new doctor gives her the first two prescriptions and then transfers her to his NP. This would be Rachel's 5th prescriber. Sadly,18 months after Rachel moved, she is in a terrible car accident and is admitted to the hospital for 4 weeks. Upon discharge, the nurse has Rachel's prescriptions filled at the hospital pharmacy for Rachel so she doesn't have to stop on the way home. This would now be a 6th prescriber. According to the PDMP and NarxCare, she will be flagged as a doctor shopper.
Let's talk about Rachel's pharmacies. Once Rachel moves, she obviously has to get a new pharmacy. Rachel now has 2 pharmacies showing in the PDMP. Rachel's new pharmacy is CVS. CVS starts to give Rachel a hard time about getting her prescriptions filled and her doctor suggests she start going to a small mom and pop pharmacy. Rachel now has 3 pharmacies listed in her PDMP NarxCare score. After a few months, her current pharmacy tells Rachel that due to DEA quota cuts, they can't keep filling her meds consistently and suggests she go to a different, larger pharmacy. Rachel does, and now has 4 pharmacies listed in her PDMP. Remember Rachel's terrible car accident she was in? The pharmacy at the hospital would be Rachel's 5th pharmacy in the past two years. According to her Narcotic Score, she will be flagged as a pharmacy hopper.
After Rachel's nearly fatal car accident, she is given a prescription for sleeping pills because she has a really hard time sleeping. She also is given 10 Ativan because she has PTSD from her accident and gets panic attacks every time she gets in the car to drive to PT. Both of these medications increase her NarxCare Narcotic Score even though they aren't opioids.
Rachel sees her doctor after being discharged from the hospital and he tells her that her Narcotic Score is too high and he can no longer prescribe. Her NS skyrocketed bc she had 6 prescribers, 5 pharmacies, a prescription for sleeping pills and a prescription for Ativan. The only medication her doctor is now willing to prescribe is Suboxone, which means she would now have a diagnosis of Opioid Use Disorder in her EHR. Rachel declines and thankfully, she is able to find a new pain doctor. Her NS now shows she has had a total of 7 prescribers in the last 2 years. Her new doctor receives a warning letter because he is prescribing controlled substances to someone with a very high NS. Rachel's new doctor dismisses her from his office saying he's not willing to risk his license for her. Not only is Rachel unable to find a new pain doctor, she can't even find a regular doctor.
So, was this Narcotic Score helpful? In black and white it looks like Rachel is playing games and has definite signs of OUD. The reality is, she doesn't struggle with addiction, but is now medically abandoned and her only option for pain relief is to go to the streets and hope to get pills that aren't laced with illicit fentanyl. Since Rachel is too afraid to do that, she has to quit her job and file for SSDI since she is no longer able to work due to uncontrolled pain and anxiety.
Now let's discuss this webinar.
At the start of the webinar, they describe the goal and mission of Bamboo Health:
- Goal: "To improve health outcomes and provide more efficient and effective use of resources."
- Mission: "To enable our customers to optimize value-based outcomes by delivering information, actionable insights, and support for the delivery of whole person care through payer and provider collaboration"
He also claims that there is increasing research showing that there has been detected changes in patient behaviors due to PDMP. Notice he doesn’t say anything about patient outcome , just detected changes-that he claims can potentially translate into improved health, or patient safety. All this means is since implementing PDMP, patients are filling fewer prescriptions. As we've showed you before, the statistics actually show since PDMP has been used, overdoses and deaths have increased not decreased.
So, the big question is did they actually externally validate NarxCare? First, let's define what externally validated means. When pertaining to a study, it simply means showing that the conclusions or outcomes of the study can be applied to other settings outside of the study.
There are 4 scores in NarxCare. 1. Narcotic Score (NS) 2. Sedative Score 3. Stimulant Score 4. ORS-Overdose Risk Score The first 3 come directly from PDMP info, and the last one is a combination of PDMP data, EHR (Electronic Health Records)-things like PTSD, anxiety, depression, etc, criminal justice data, along with other things. The ORS is based on a proprietary algorithm, so nobody using NarxCare actually knows what goes into these scores. When Bamboo said they externally validated NarxCare, one would assume they're talking about all of these scores, but specifically the ORS, since that's the score that seems to be used the most against patients. Except, they were only talking about one score, and it wasn't the ORS. The only score they studied was the Narcotic Score (NS).
Since the only score studied was the NS, the rest of this article will be about that, and the claims they make about this score.
The Narcotic score ranges from 0-999. There are four things that go into this score.
- MME (Milligrams of Morphine Equivalent)-As we've shared with you before, MME isn't very scientific, since there are several ways to count this number. Watch Dr. Dasgupta's presentation at a recent FDA meeting showing how MME shouldn't be used as a definite threshold.
- Sedative usage (yes, if you have filled a sedative such as a benzo or sleeping pill, even if you've never filled an opioid, your Narcotic Score will go up)
- Number of overlapping prescriptions
- Number of prescribers over the last two years (which they claim will catch "doctor shoppers")
- Number of pharmacies patient has used to fill a controlled substance over the past two years
Appriss claims there is a strong correlation between Narcotic Score and risk of opioid misuse.
Some basic facts about the study:
- It was funded by NIH (HEAL Iniative) and NIDA
- The study was done by a survey given at pharmacies in Ohio and Indiana.
- Total of 1523 people did survey- of those 1463 individuals had a Narcotic Score
- Participants were 93% white, 62% female, around age 50 and mostly married
- They excluded those who were filling buprenorphine.
- Excluded those with criminal justice involvement
- Included those 18 years and older
- They had to speak English
- 94% had insurance
- Participants were of average health
- Pain was controlled and about average-(this is important because this will often be used to apply to CPP's who don't fit into this category at all
There were some false positives, meaning patients who had a high NS but didn't actually have signs of OUD. These were people who had some sort of work status issue including SSDI, those in poor general health, and those who had increased levels of uncontrolled pain-they are less likely to have substance use involvement. These are people whose pain isn’t adequately managed. This is important because even though it shouldn't be applied to CPP's according to this false positive, it is being applied to CPP's.
There were some false negatives, meaning people had issues with opioid abuse or misuse, but the score didn’t pick it up. A reason they gave for this is the PDMP's algorithm wouldn't pick up on illicit drug use or abuse. Why is this a problem? Well, it means that the patients who are being flagged and denied medication are often patients who don't struggle with OUD, and those who are being missed are often those who do have OUD.
Let's look at some of the main problems.
First, the biggest being there is no room for context. What does this mean? Remember our hypothetical patient, Rachel? All they do is take straight data and don't have anywhere to list reasons or context.
Second, Bamboo confirmed that the NS was the only score that was studied and "validated." What about the other scores? Maybe they should have titled it "Bamboo Health's Narcotic Score was externally validated."
Third, this study didn't look at actual patient outcomes such as overdoses or opioid related adverse events. So, they can look at the outcome of lowered prescribing, which seems to be the only outcome that matters these days. But, in reality, patient outcome is the metric that should matter. As we've seen, it looks like PDMP use is actually associated with increased overdoses and suicides. Seems like our government should use the millions if not billions of dollars that go into PDMP for actual help for those with addiction.