Table 40Effect of CHW interventions on health outcomes: strength of evidence

Number of Studies; # of SubjectsRisk of Bias

Design/Quality
ConsistencyDirectnessPrecisionOther Modifying Factors (Intensity, Confounding)ResultsOverall Strength of Evidence
Health promotion and disease prevention: pediatric immunizations
3; 5,40668–72Medium

2 RCTs, 1 cohort/1 good, 1 fair, 1 poor
InconsistentDirectImprecisePresentMixed results: 2 of 3 studies favor CHW intervention vs. control; 1 shows no difference between CHW interventions and no interventionLow
Health promotion and disease prevention: disease prevention
1; 29416,127Medium

1 RCT/fair
Consistency unknown (single study)DirectNot reportedAbsentNo difference between CHW intervention and print intervention (for change in body mass index)Low
Maternal and child health: prenatal care and perinatal outcomes
5; 3,38971,77,79, 83,86,87High

2 RCTs, 3 cohorts/2 fair, 3 poor
InconsistentSome direct, some indirectImprecisePresentMixed results, between CHW and professionals or no intervention or usual careLow
Maternal and child health: child development
4; 1,66475,76,78, 80–83,128Medium

3 RCTs, 1 cohorts/3 fair, 1 poor
InconsistentDirectImprecisePresentMixed results: 2 studies show some benefit for CHW vs. no intervention or health professional, 2 show no difference between CHW and health professionalLow
Maternal and child health: environment conducive to child well-being
7; 2,29967,68,73, 74,78,84,85,128Medium

7 RCTs/4 fair, 3 poor
InconsistentMostly indirectImprecisePresentMixed results: 5 studies show no benefit for CHW over alternatives, 2 show benefit of CHW arm vs. usual care or health professionalLow
Chronic disease management: diabetes mellitus
4; 47927,88–93,124Low

4
RCTs/fair
InconsistentDirectImpreciseAbsentMixed results: 2 of 4 studies found CHW more effective than usual care in decreasing mean HgbA1c, 2 found no differenceLow
Chronic disease management: hypertension
3; 2,82323,94,95,98,99Medium

2 RCTs, 1 cohort/1 fair, 2 poor
ConsistentDirectPrecisePresentNo difference between CHW intervention and CHW in a lesser capacityLow
Chronic disease management: back pain
1; 255114Medium

1 RCT/fair
Consistency unknown (single study)DirectNot reportedAbsentFavors CHW intervention vs. usual care plus a book for Roland score at 6 months and worry score at 12 months; no difference in Roland score at 12 monthsModerate
Chronic disease management: mental health
1; 165120,121High

1
RCT/poor
Consistency unknown (single study)DirectNot reportedPresentNo difference between CHW intervention and usual care (health professionals)Low
Chronic disease management: asthma symptoms
2; 57296,97,100Low

2 RCTs/1 good, 1 fair
InconsistentDirectImpreciseAbsentMixed results: 1 favors CHW vs. delayed intervention; no difference between CHW and less intense interventionLow
Chronic disease management: asthma, caregiver psychosocial outcomes
2; 57296,97,100Low

2 RCTs/1 good, 1 fair
ConsistentDirectPreciseAbsentFavors CHW vs. less intense CHW arm or delayed interventionModerate

CHW, community health worker; RCT, randomized controlled trial.

From: 4, Discussion

Cover of Outcomes of Community Health Worker Interventions
Outcomes of Community Health Worker Interventions.
Evidence Reports/Technology Assessments, No. 181.
Viswanathan M, Kraschnewski J, Nishikawa B, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.