Background  Case reports and case series are often the first line of clinical evidence in the medical literature and, until stronger evidence becomes available, are the only guidance to new therapies.

Aim  To determine how case reports and case series are reported, what kind of information is provided, and what the authors conclude.

Methods  All case reports and series (n ≤ 10 patients) that described new or innovative therapies were selected from the online version of the Archives of Dermatology published from January 1998 to June 2001. Data were extracted according to a predefined protocol.

Results  Sixty-two case reports and 38 case series were identified. Four of the 100 articles reported treatment failure; 25 reports (40%) and 19 series (50%) concluded that the treatment was efficacious or useful. Systemic treatment was used in 28 reports (45%) and 19 series (50%). Twelve series (32%) reported treatment of consecutive patients. Follow-up could not be assessed in seven case reports (11%). Sixteen reports (26%) and 15 series (39%) reported patients’ experiences; 14 reports (23%) and 10 series (26%) indicated some consent.

Conclusions  This small survey confirms a strong publication bias that favors positive results, exaggerated claims of efficacy and safety, underreporting of patient-centered outcomes, and inadequate reporting of consent. We hope to stimulate a discussion of the quality standards for reporting single cases and case series, and of how information from case reports can be used best to improve patient care.