Article Figures & Data
Tables
Organization Tools Used (Technical) Data Lead (Responsible for Tracking and Monitoring) 1 Study registry EHR query reports MA who oversaw all refills; no other clinical care responsibilities 2 Study registry EHR query reports Excel MA with clinical care responsibilities 3 Excel-based registry IT Project CoordinatorCare coordinator (position changed mid-study) 4 Study registry Proprietary software Population health data analyst 5 EHR integrated registry Programmer and quality improvement coordinator 6 Study registryExcel MA with clinical care responsibilities EHR, electronic health record; IT, information technology; MA, medical assistant.
- Table 2.
Organizational Approaches Used to Overcome Challenges in Identifying Patients on LtOT
Approach Strengths Weaknesses Program using prescriptions to pull patients into an EHR-linked registry Can use multiple EHR variables to produce the registry list and update it efficiently Required significant IT resources.Difficult to develop an LtOT definition that was searchable.EHR prescription data were sometimes identified as inaccurate when vetted. Query EHR for patients with opioid treatment agreement Organizations frequently prioritized getting opioid treatment agreements signed as an early step in improving care Required developing a custom, searchable data field for the opioid treatment agreement if not already present in the EHR, or doing a chart review.Sometimes included patients taking other non-opioid controlled substances.Missed patients without a signed opioid treatment agreement. Query EHR for patients with documentation of MED Organizations frequently prioritized calculating MED as an early step in improving care Required developing a custom, searchable data field for MED if not already present in the EHR, or doing a chart review.MED often not calculated or inconsistently calculated.If MED not updated to 0 after cessation of LtOT, over-counted patients.Assumed MED is calculated only for patients using LtOT. Query EHR for patients with prescription for opioid medication Directly uses the primary element of interest, opioid prescriptions Search complicated due to many different types of opioids, each with many brand names.Opioid medication lists require updating as new opioid medications become available.Required significant manual cleaning time to target only patients who were “currently” receiving LtOT and who met the definition of LtOT rather than acute opioid therapy. Query EHR for patients with a designated diagnosis used to code for LtOT Clinician-led cohort identification increased accuracy of diagnosis Clinicians resisted applying a designated diagnosis.At the time of the study, there was no clear diagnosis for patients on LtOT.Relied on care teams knowing how to consistently apply the diagnosis. Pull provider reports from the state prescription monitoring database Useful cross-check of internal data Organizations thought the state drug database lists were inaccurate.Was not possible to run a clinic-wide report, required running individual provider reports.Required manual cleaning time to identify only those patients who met the definition of LtOT rather than acute opioid therapy. EHR, electronic health record; IT, information technology; LtOT, long-term opioid therapy; MED, morphine-equivalent dose.
Data Element Prescriber Date of last appointment Date of next appointment Diagnosis to identify patients on LtOT MED Co-prescription of opioids and sedatives Date opioid treatment agreement signed Function assessment (e.g., PEG) Risk assessment (e.g., ORT) Depression assessment (e.g., PHQ) Date of last state prescription monitoring database check Result of last state prescription monitoring database check Date of last urine drug test Result of last urine drug test Sleep apnea assessment (e.g., STOPBang) PTSD assessment Anxiety assessment (e.g., GAD-7) GAD-7, General Anxiety Disorder - 7; MED, morphine-equivalent dose; ORT, opioid risk tool; PEG, Pain, Enjoyment, and General activity pain assessment tool; PHQ, patient health questionnaire; PTSD, post-traumatic stress disorder.
Strategy Strengths Weaknesses Quote EHR-linked registry Data extracted directly from the EHR into the registry.Easy to access detailed reports. Required significant resources to build (time, skills, funding).Clinicians had to click out of the EHR to reach the registry. “Once the data is there and the structure is built, the work is just finding the data. A lot of the work was collaborative with the folks who work with the EHR. Digging and finding where the template data was stored.” Excel spreadsheet Easy to control (change variables, edit entries, track elements for updating).Inexpensive.Provided an interim system until an EHR-integrated system was possible Required manual chart review or data entry by clinical personnel (e.g., medical assistant) to populate with data.Hand-entered data from chart reviews onto excel spreadsheet (errors more likely, time consuming).Needed a cue to know when there were new data to enter. Cumbersome to keep historical data, therefore difficult to track trends.Required IT support to make more usable (e.g., turning the font red when patient overdue for a urine drug test).Not integrated with the EHR for use in patient care. Excel is a “quick and easy reference.”“If I was gone or something, I would miss getting the flags from the nurses saying that hey, we refilled this medication. So I never really took the time to go and backtrack, I just went forward from there.” Proprietary software Data from the EHR extracted with proprietary software into a report.Simple to use, others could step in with minimal training. Not all proprietary software in use at organizations was nimble enough to easily create LtOT reports.Proprietary software still needed a list of patients to query, which required maintenance. “They’re adding COT module, but they haven’t done that yet; we’re already married up to them and we like it, but they aren’t there yet.” EHR query Data extracted directly from the EHR.Did not require additional system or proprietary software. Required translation of query into a tracking report.Required double-checking via chart review or provider consultation as reports often included errors.Difficult to troubleshoot why errors occurred.Difficult to limit to current patients on LtOT.Exporting from EHR to Excel produced a report that took hours to make readable. “Inquiries in [EHR] are pretty primal; created several of our own, but they’re fraught with problems; we never get the same list of patients.”“Exporting is a pain… It does not produce a usable spreadsheet—it takes hours to go through it to make it usable.” COT, chronic opioid therapy; EHR, electronic health record; IT, information technology; LtOT, long-term opioid therapy.
