Positive predictive value for presumptive diagnoses of skin cancer and compliance with follow-up among patients attending a community screening program

Abstract

Background: There is an ongoing need for evaluating the effectiveness of skin cancer screening programs and barriers to obtaining definitive diagnoses and treatments among screen-positive patients.

Methods: We studied 132 patients who were given presumptive diagnoses of skin cancer while attending a free skin cancer screening program during May 1994 in San Diego, California (464 screened). We assessed compliance with recommendations to see a physician for definitive diagnostic workup. Logistic regression models included potential determinants of compliance: age, gender, health insurance, educational level, anatomic site, number of suspicious skin lesions, personal and family history of skin cancer, propensity to sunburn, and history of recent change in moles. The positive predictive value (PPV) of skin cancer diagnoses was defined conservatively as a range by inclusion then exclusion of patients not biopsied.

Results: Follow-up information was obtained for 115 patients, and 88 (77%) of these reported obtaining a definitive workup. The final regression model controlling for age showed that patients without a family history of skin cancer were less compliant than those with a history [OR for noncompliance 3.29, 95% confidence interval (CI) 1.10-9.83], and patients with only one presumptive skin cancer lesion were less compliant than those with additional cancer lesions (OR 5.31, 95% CI 1.07-26.43). Older patients were more compliant than younger. Among 82 patients under 65 years of age, the above factors were important only among men. Also, noncompliance was significantly higher among younger men without versus with health insurance (OR 5.14, 95% CI 1.23-20.68). Two melanomas, 1 squamous cell carcinoma, and 23 basal cell carcinomas were histopathologically confirmed. The PPV for melanoma was 6-15%, for squamous cell carcinoma 4-12%, and for basal cell carcinoma 32-85%.

Conclusions: This study confirms the ability of skin cancer screening programs to successfully identify individuals with cutaneous malignancies. The significant association of noncompliance to seek recommended follow-up care with lack of health insurance for men too young to participate in Medicare suggests that approaches to definitive workup for uninsured patients be developed alongside free screening programs.