Methods | Data from studies | Public health guidance | |||
---|---|---|---|---|---|
 | Preventive and promotive interventions (n = 50) Systematic reviews: 12 Systematic review + meta: 36 Overview: 2 | Therapeutic nutrition interventions (n = 34) Systematic reviews: 10 Systematic review + meta: 24 Overview: 0 | Fortification (folic acid) (n = 2) Systematic reviews: 0 Systematic review + meta: 2 Overview: 0 | Delivery platforms (CCT) (n = 4) Systematic reviews: 3 Systematic review + meta: 0 Overview: 1 |  |
Study designs (primary studies) | RCTs: 25 reviews Mixed (RCTs, quasi and other designs): 25 reviews | RCTs: 23 reviews Mixed (RCTs, quasi and other designs): 11 reviews | RCTs: 0 reviews Mixed (RCTs, quasi and other designs): 2 reviews Overview: 0 reviews | RCTs: 0 reviews Mixed (RCTs, quasi and other designs): 4 reviews | RCTs may have limited generalizability or not be feasible or ethical. Non-randomized trials or cluster RCTs may be more appropriate for the question |
Searching non-health databases, number seen out of total (range) | 6 (1–5) | 11 (1–2) | 0 | 3 (3–5) | Public health may require range of databases and detailed text words because terminology is imprecise |
Assessing risk of bias | Tools used: Cochrane risk of bias: 17 GRADE: 5 Jadad: 1 CONSORT: 1 Other tools: 18 Not stated: 8 | Tools used: Cochrane risk of bias: 14 GRADE: 0 Jadad: 0 CONSORT: 0 Other tools: 19 Not stated: 1 | Tools used: Cochrane risk of bias: 0 GRADE: 1 Jadad: 0 CONSORT: 0 Other tools: 1 | Tools used: Cochrane risk of bias: 2 GRADE: 0 Jadad: 0 CONSORT: 0 Other tools: 2 | Appraisal criteria depend on the type of study in the review. Authors should use Cochrane risk of bias tool for RCTs |
Ethics and inequalities | HIC: 9 LMIC: 21 Both: 20 | HIC: 8 LMIC: 15 Both: 11 | HIC: 1 LMIC: 0 Both: 1 | HIC: 0 LMIC: 2 Both: 2 | Public health interventions may unintentionally increase health inequalities if the more advantaged benefit more than the worst off. Authors should consider whether populations are likely to be disadvantaged across PROGRESS+ acronym factors (place of residence, race or ethnicity, occupation, gender, religion, education, socioeconomic position (SES) and social capital |
Context (explore context as an effect modifier) | 29 | 13 | 1 | 0 | Public health interventions depend on social, economic and political context in which they are developed and implemented. Context should be explored as an effect modifier (e.g. host organization, target population) |
Sustainability (sustainability was assessed or described) | 11 | 6 | 0 | 4 | Consider the extent to which the intervention’s effects on outcomes are sustained over time, e.g. by assessing economic and political variables, strength of implementing institutions, integration of services into existing services, training component, community involvement |
Applicability (conducted subgroup analysis to assess applicability to different settings) | 29 | 13 | 1 | 0 | Authors are well-positioned to collect data needed to make judgments about applicability to different settings, including political environment, social norms, cultural preferences, resources required, educational level of target population, organizational capacity to deliver the intervention, structural barriers to intervention, health care providers have capacity to deliver this intervention |