Systematic Automated Population-Based Surveillance to Improve Outcomes in Primary Cutaneous Melanoma

Importance: Patients with a primary cutaneous melanoma (PCM) are at increased risk of developing a second primary melanoma (SPM). Earlier stage at diagnosis is associated with better 5-year mortality, yet low compliance with recommended follow-up after treatment for PCM and high rates of patients lost to follow-up are reported in the literature. Strategies to enhance population-based surveillance for SPM have not been well described.

Objective: To determine whether the implementation of a systematic automated population-based program (SAPP) to ensure dermatologist total body skin exams (TBSEs) in the PCM population compared to the pre-automated period and pre-database periods improves compliance with follow-up, detects more SPMs, or affects mortality?

Design: Quality Improvement Study.

Setting: Large integrated health care system with internalized Dermatology services.

Participants: Members with a PCM prior to, or during the study periods of interest, were included in this quality improvement study. There were 6317 eligible individuals with PCMs in 3 study periods: pre-Database (1/1/2005-11/1/2009), pre-Automation (11/1/09-7/31/15), and post-Automation (8/1/15-12/31/2021), respectively.

Intervention(s) or exposure(s): Manual registry for melanoma patients (pre-Automation) and systematic automated population-based program for PCM patients (post-Automation).

Main outcome(s) and measure(s): Mean number of TBSE per person-year. Time Between TBSE. Incidence of SPM detection. Stage of SPM at diagnosis. Melanoma Death and All-Cause Mortality.

Results: In patients with PCM, a SAPP to ensure dermatologist TBSE, reduces median time (months) between TBSEs from 12.3 pre-Database to 8.2 post-Automation (stage 0-2a PCM) and from 10.3 to 6.8 (stage 2b-4 PCM; P < .0001), improves number of TBSEs per person-year with each successive study period (pre-Database mean [SD] 0.56 [1.07]; pre-Automation 0.80 [1.25]; post-Automation 1.78 [4.17]; P < .0001), and improves SPM detection rates per 1000 person-years (pre-Database 20.2 [95% CI 12.2-30.9]; pre-Automation 27.5 [18.2-39.9]; post-Automation 27.5 [26.6-51.8], P < .0001 for trend) at earlier cancer stages. There is an associated annual reduction in all-cause (1.4 per 1000 person-years, P = .0004) and melanoma-cause mortality (0.2 per 1000 person-years, P = .0197).

Conclusion: Implementing a systematic, automated population-based Melanoma program to ensure dermatologist TBSEs improves compliance with follow-up in patients with PCM. More patients received at least annual dermatologist TBSEs and more SPMs were detected at earlier stages. Improvements were sustained and there were significantly lower all-cause and melanoma-cause mortality. The system facilitates prioritization of higher risk patients for intervention and is a unique strategy enhancing patient safety, efficiency, and timeliness of care.