Pragmatic methods for reviewing exceptionally large bodies of evidence: systematic mapping review and overview of systematic reviews using lung cancer survival as an exemplar

Table 1 Summary of the review process

Stage	Elements of the review process
Stage 1: initial systematic mapping review of prognostic research	A descriptive map of prognostic research for lung cancer survival based on the assessment of titles and abstracts only. ● Classification of relevant reviews and meta-analysis according to PROGRESS research themes 1–4, lung cancer type, number and type of prognostic factors investigated, and publication type. ● Identification of all prognostic factors, from which a comprehensive coding scheme was developed. ● Identification of prognostic factors deemed to be potentially modifiable (reviewed by two independent public health and clinical stakeholders).
Stage 2: overview of systematic reviews of prognostic factor research	A more in-depth review of a subset of systematic reviews focusing on prognostic factor research (PROGRESS research theme 2) based on the assessment of full text publications. ● Summary of key data from included reviews. ● Coding of reviews according to the prognostic factors they addressed. ● Summary of all prognostic factors investigated by each review, including whether or not they were significantly associated with survival^†, and the direction of the impact.
Stage 3: in-depth evaluation of potentially modifiable factors	A more in-depth evaluation of the results of included reviews reporting modifiable factors. ● Summary of the magnitude of the effect of modifiable prognostic factors (where possible).

^†The term ‘significant’ denotes statistical significance and refers to the results of either the regression or meta-analyses (pooled analysis) or, where no pooled analysis was undertaken, to more than 50% of studies in narrative syntheses. For reviews that reported both pooled analysis and narrative synthesis, this was based on the results of the pooled analysis. Where significant findings were based on only a single study within a review, this was highlighted.