iPLEDGE Weaknesses: Is It Time to Address the Flaws?

A majority of patients had one documented course of isotretinoin treatment (82.1%, N=293), however, additional trials of isotretinoin therapy were documented in 17.9% (N=64). Of the patients who had previously received isotretinoin therapy, chart information regarding prior courses was available in 60.9% of cases (N=39). For known courses, four months of therapy was the most common prior course duration (15.4%, range: 1-14 months).

Regarding the primary treatment course, congruence between chart documentation and iPLEDGE data was observed in 73.1% of cases (N=261; Einstein=69.8%, Austin=82.8%, Downstate=47.1%; Table II). Average course duration for congruent charts was 4.3 months, range: 1-12 months (Figure 1). The number of discrepant doses between the chart and iPLEDGE is highlighted in Figure 2. Of the 96 discrepant charts, the majority (81.4%) was due to a missed dose(s), with an addition(s) in the remainder of cases. However, as many individual charts had multiple discrepancies, the total number of discrepant doses was 249, again with misses representing the majority (90%). An iPLEDGE restart was required for 8.7% (N=31) of patients. The number of isotretinoin doses administered before the restart is demonstrated in Figure 3.

Per FDA Drug Safety and Risk Management and Dermatologic and Ophthalmic Drug Advisory Committee goals, iPLEDGE should monitor the prescribing, dispensing, and distribution of all isotretinoin brands. Given the large percentage of discordant data found in our study, we suggest that iPLEDGE is falling short of its goal to monitor every dose of isotretinoin. Nevertheless, several factors could potentially account for the discordance observed in this study.

A "restart," an easily modifiable error, accounted for 8.7% of all charts in this study. Restarts occur when a patient is "lost to follow-up," and locked out of the iPLEDGE system until re-registration is complete. Per iPLEDGE rules, a patient may become lost to follow-up in four ways: 1) The patient has been in a "Registered" status for greater than 90 days and the prescriber/designee has not interacted with the patient's record; 2) Patient has been in a "Requires Confirmation" status for greater than 53 days and the prescriber/designee has not interacted with the patient's record; 3) Females of childbearing potential (FCBP) who are missing their pregnancy test at date of last dose; 4) FCBPs who are missing their final pregnancy test 30 days post treatment. Should a patient become "lost to follow-up" prior to the end of the planned treatment course, a physician may re-register that patient, but the practitioner is required to contact the iPLEDGE program and receive an "override code." Once patients are re-registered, they are eligible for isotretinoin according to the same rules as any other registered patient. However, in our experience, the online data readily available to the practitioner prior to the re-registration is lost. This phenomenon is coded in our study as a "restart." Though the data actually remains stored within the iPLEDGE system, it is not readily accessible to providers. Long-term maintenance and storage of information is not mentioned in any iPLEDGE supporting materials, and was discovered only when the primary investigators called to inquire about missing data. It is also interesting to note that while patients may be locked out of the system as soon as 30 days post treatment, three of our patients were found to have greater than 90 days between dispensed doses with no lost data.

An "addition" occurs when isotretinoin is noted to be dispensed through iPLEDGE without a corresponding office visit or other documented means of a patient obtaining an isotretinoin prescription. While revised iPLEDGE rules allow patients to receive more than one dispensed prescription of isotretinoin each month (exact dates and prescription window periods vary based on childbearing status of the patient), this requires a new prescription each time accompanied by corresponding chart documentation. Possible explanations for the observed additions