Skip to main content

Table 3 Summary of recommendations from 93 publication by key item

From: Methods to systematically review and meta-analyse observational studies: a systematic scoping review of recommendations

Key item No of articles providing recommendation Topic of recommendation N articles addressing area (%)a
Protocol development 16 Need for protocol to be written in advance 12 (75%)
   Items to be included in protocol 11 (69%)
Research question 20 Scope of research question 20 (100%)
Search strategy 33 General methods for conducting searches in context of observational studies 22 (67%)
   Specific challenges in searching for observational studies 12 (36%)
Study eligibility 22 Specifying eligibility criteria 22 (100%)
   Assessment of eligibility 6 (27%)
Data extraction 9 Methods for data extraction 9 (100%)
Dealing with different study designs 25 Inclusion of different study designs in a single review 10 (40%)
   Combining results from different study designs in a single meta-analysis 15 (60%)
Risk of bias assessment 39 Methods to assess the risk of bias in individual studies 39 (100%)
Publication bias 20 Inclusion of unpublished studies 5 (25%)
   Methods to assess publication bias 7 (35%)
Heterogeneity 39 Measurement of heterogeneity 39 (100%)
   Exploring potential causes of heterogeneity 16 (41%)
Statistical analysis 52 Deciding to combine results in a single effect estimate 20 (38%)
   Choosing fixed or random effects meta-analysis 16 (31%)
  1. aPercentages do not add up to 100% because articles can contribute recommendations to more than one topic and only the most frequent areas of recommendation for each key item are listed