User Story: Pharmacy
Janjri Desai, Pharm.D., MBA, BCPS, Administrative Director of Pharmacy at Stanford Health Care
Ordering Provider: Janjri Desai Pharm.D., MBA, BCPS, DPLA
How did you hear about the informatics consult program?
Dr. Topher Sharp, the CMIO at Stanford Health Care, introduced me to Atropos Health. He’d worked with the team previously and recommended testing out the informatics consult service.
What is your primary area of interest?
As the Administrative Director of Pharmacy at Stanford Health Care, I’m most interested in looking at the safety, effectiveness, and cost-effectiveness of medications for different groups of patients. It’s also interesting to see the data Atropos can pull on outcomes of pharmacist intervention.
What question did you ask the Atropos informatics consult service?
I’ve asked a few questions using the informatics consult service recently - to corroborate internal analyses and evaluate drug efficacy. One example concerns a long-acting formulation of bupivacaine, a non-opioid analgesic. It is more expensive than comparable medications and has many potential medication interactions, including with commonly-used anesthetics. The efficacy data from trials isn’t very compelling either. It stayed on the formulary based on anecdotal evidence of it working well, but recently Stanford’s P&T committee decided to conduct a more thorough evaluation of safety and efficacy. My colleague Dr. Ahuja and I asked Atropos questions about the drug to provide additional support for the evaluation process.
Our Questions
Question 1 - For patients who received the drug in question over the past 4 years (total), how many of them required alternative analgesic co-administration, how many died within 30 days, and what was the length of stay? For what procedures was the drug used?
Question 2 - For patients who received the drug in question over the past 4 years for specific procedures, how many of them required alternative analgesic coadministration, how many died within 30 days, and what was the length of stay compared to those who received opioid analgesics?
Atropos Health’s Medical Director met with us to clarify the questions & desired outcomes. Then he used their cohorting technology and analytics pipeline to conduct retrospective observational research, and generate answers based on anonymized EHR data from Stanford Health Care. After clarifications, the process took two days.
The Results
Answer for Question 1 - Based on Atropos’ comprehensive data set, for all patients who received the drug in question (generally, and after specific procedures), the rate of alternate anesthetic vs other opiates within the 96 hours was determined, as well as average length of stay and all-cause mortality. The findings suggested that the drug in question did not provide sufficient pain relief for patients.
Answer for Question 2 Many more patients received opioid analgesics than the drug in question after procedures where such treatment would be expected. The two groups had significant differences in baseline demographic characteristics. Atropos used high-dimensionality propensity score matching to control for confounders so they could compare outcomes across the two groups. Subsequent findings showed that the cohort who received the drug in question had a statistically higher rate of secondary opiate use (expected), a higher rate of alternative local anesthetic use, longer inpatient stay, and a lower rate of death.
Both answers came as Prognostogram real-world evidence reports with detailed figures & tables on methodology, definitions, data span, comorbidities, medications, procedures, and outcomes for both cohorts. They showed that the group on the drug in question had less pain relief & stayed in the hospital longer than the group on opioid analgesics. The P&T committee primarily discussed our internal findings on the safety of the drug in question, but we also found this information useful.
What would you have used to answer this type of question in the past (before the informatics consult service was available)?
Without Atropos, the standard process for answering questions like these would be to pick a PharmD or resident to do manual chart review. The process involves obtaining information from individual patient charts and entering it into a spreadsheet before conducting analyses.
For instance, if I asked a colleague to review the use of Warfarin, they might:
Look in charts to see:
How many patients have elevated INR levels
How much time it takes to get to the target level
How many patients are deviating from the guidelines
Enter all of the information into a spreadsheet
Analyze & share findings
As you can imagine, this process is quite labor intensive and time consuming. So we’re happy to have Atropos Health’s informatics consult service available to answer questions faster & with expanded analytic capabilities. It enables us to ask more questions and takes the burden off of our staff.
In your own words, please describe the Atropos informatics consult service
Overall, I enjoy the experience of submitting a question and getting a credible answer quickly, without the burden on my team. Atropos’s clinical informatics consult service will make it easier for me to learn about areas I’m interested in exploring, like: use of biosimilars, whether guidelines are being followed, cost effectiveness for certain medications, and real world medication efficacy and safety. This evidence can help support regulatory demands & determinations of what’s in the formulary.
What was your overall impression of the service?
It was easy to submit a request. I also found the content in the reports useful (demographics tables, charts) and liked the way it was laid out.
How likely are you to recommend the Atropos informatics consult to your colleagues?
I’d definitely recommend it. It’s already much faster than chart reviews & allows me to ask more questions based on medical datasets.