Background: Non-melanoma skin cancer, which includes basal cell carcinoma and cutaneous squamous cell carcinoma, is the commonest malignancy worldwide. The mainstay of treatment is surgical excision. Despite this being an exceptionally common procedure, it is not known what the accepted standard is for incomplete excision. Multiple single-centre, regional and national studies have previously reported their incidence of incomplete excision in isolation. Furthermore, is it not known what effect potential risk factors such as the operating group, location of lesions, type of reconstruction, histological components or use of loupe magnification have on the incidence of incomplete excisions. The objective of this study will be to systematically evaluate observational data that present incidence of incomplete surgical excision amongst adult patients with non-melanoma skin cancer worldwide. Methods: We designed and registered a study protocol for a systematic review and meta-analysis of descriptive epidemiology data. A comprehensive literature search will be conducted (from January 2000 onwards) in MEDLINE, EMBASE, Scopus, CINAHL, EMCare and Cochrane Library. Grey literature will be identified through searching Open Grey, dissertation databases (e.g. Open Access Theses and Dissertations) and clinical trial registers (e.g. WHO ICTRP). Observational studies (cohort, cross-sectional, case series and clinical audits) reporting the incidence of incomplete surgical excision and conducted in adult patients with non-melanoma skin cancer will be included. The primary outcome will be the incidence of incomplete surgical excision (defined as residual tumour at either the peripheral or deep margin). Secondary outcomes will be risk factors that may affect incomplete excision (e.g. operating group, location of lesions, types of reconstruction, histological components). Data will not be extracted if the study uses other surgical techniques such as Mohs micrographic surgery, intra-operative frozen section, incision, shave or punch biopsies. Two investigators will independently screen all citations, full-text articles and abstract data. Potential