Objectives The USPSTF recommends lung cancer screening to adults aged 55 to 80 years who are current or former heavy smokers. Guidelines also recommend a thorough patient understanding of harms, including overdiagnosis, prior to screening. Many US medical centers, both academic and community, have screening programs and publish information about screening on their websites. Because websites could influence patients’ perceptions of screening benefits and harms, and subsequent decisions, some experts have expressed concern about persuasive or overly optimistic portrayals of screening in patient-facing websites. We evaluated the reporting of overdiagnosis on these websites in relation to the reporting of benefits, other harms, and suggested next steps for patients.

Method Our study sample included lung cancer screening program websites found through a standardized web search from two types of institutions: 1) academic medical centers associated with all (n=152) US allopathic medical schools; 2) a state-matched random sample of community (non-academic) lung cancer screening programs drawn from a list of all (n=1779) American College of Radiology designated lung cancer screening centers. Two investigators independently evaluated each website for the reporting of overdiagnosis, benefits, other harms, and recommendations for next steps. Discrepancies were resolved by consensus. Analysis included bivariate comparisons between reporting rates for overdiagnosis and other harms and benefits.

Results We identified 81 academic centers and 81 community centers (total n=162) with screening websites. Overdiagnosis was reported on 11 (6.7%) websites and more frequently on academic sites (13.6% vs 0.0%, p<0.01). Of sites reporting overdiagnosis, 7 (63.6%) used the term overdiagnosis, with the others only explaining the concept. Of all websites, benefits were noted more frequently than overdiagnosis (98.1% vs 6.7%, p<0.01) and overdiagnosis was the least reported harm: false positives 44.4% (p<0.01), radiation 34.0% (p<0.01), false negatives 16.1% (p=0.08), incidental findings 8.6% (p=1.00). Centers reporting overdiagnosis more often reported the presence of screening harms (100% vs 44.4%, p<0.01). These centers also more frequently recommended patients weigh the harms and benefits of screening (100% vs 22.0%, p<0.01) and explicitly suggested that patients discuss harms and benefits with their primary care provider (72.7% vs 13.9%, p<0.01).

Conclusions Overdiagnosis was uncommonly reported by academic medical center lung cancer screening websites and was absent from community-based screening program websites. It was the least commonly reported harm and was noted significantly less than benefits of screening. Websites noting overdiagnosis were more likely to report potential harms of screening and to recommend patients weigh the harms and benefits of screening themselves and in consultation with their primary care provider. Research is needed to determine how portrayal of screening-related harms and benefits in this setting may contribute to imbalanced perceptions of harms and benefits and how these perceptions influence patient screening decisions.