2. Outcome category: health‐related knowledge.
| Study ID | Health topic | Measure | No. of participants | Time point(s) |
Intervention arm(s) Mean (SD)* |
Control arm(s) Mean (SD)* |
Notes |
| 1 Culturally and literacy adapted self‐management programme vs no health literacy intervention | |||||||
| Bloom 2014 | Breast health/breast cancer | Not reported | N: 230 | 6 months post‐intervention | — | — | MD 0.5 (P < 0.0001) Cluster‐RCT; "GEE were used to account for clustering (sample and analysis)" (Bloom 2014) Increased knowledge did not increase mammography |
| Koniak‐Griffin 2015 | Cardiovascular disease | Heart knowledge questionnaire, adapted from a previous survey by Mosca et al (2004) (10 items, true/false format, 0 to 10, higher score is better) |
IG: 98 CG: 95 |
6 months after randomisation (immediately post‐intervention) |
7.9 (2.6) | Not reported | — |
| IG: 100 CG: 94 |
9 months after randomisation (at 3‐month follow‐up) |
9.4 (1.9) | |||||
| Rosal 2011 | Type 2 diabetes | ADKnowl, adapted version (23 item‐sets (104 items), 0 to 104, higher score is better) |
IG: 124 CG: 128 |
12 months after randomisation (immediately post‐intervention) | 0.089 (range ‐0.065 to 0.113) | 0.033 (range 0.009 to 0.057) | Intervention effect 0.056 (0.022 to 0.090) P = 0.001 |
| van Servellen 2005 | HIV | (1) HIV Illness and Treatment Knowledge and Misconceptions Scale (17 items, 0 to 17, higher score is better) (2) Knowledge of risk of getting sicker 1 item, 1 = very high risk to 4 = nonexistent risk, lower score is better |
IG: 34 CG: 35 |
6 months after randomisation (immediately post‐intervention) | (1) 1.20 (3.19) (2) ‐0.24 (0.78) |
(1) 1.40 (2.59) (2) 0.09 (0.67) |
Change scores are reported To improve the interpretation of results, the original scale has been transformed into a positive scale with higher values indicating better performance (see Analysis 1.4) |
| 2 Culturally and literacy adapted self‐management programme vs written information on the same topic | |||||||
| Han 2017 | Cervical/breast cancer | Breast Cancer Knowledge Test (0 to 18, higher score is better) |
IG: 278 CG: 282 |
6 months after randomisation (immediately post‐intervention) | 11.0 (3.9) | 10.4 (3.8) | Cluster‐RCT; data have been re‐analysed for meta‐analyses using the appropriate unit of analysis with the use of the ICC reported by Han 2017. In addition, combined scores for breast cancer knowledge and cervical cancer knowledge were calculated (see Analysis 2.10; Analysis 2.11). Estimated MD 0.7 (95% CI ‐0.1 to 1.6) MD estimated from linear mixed‐effects models adjusted for baseline knowledge, age, insurance, English proficiency, years of US residence, years of education, employment and family history of breast cancer. |
| Cervical Cancer Knowledge Test (0 to 20, higher score is better) |
5.6 (2.4) | 5.3 (2.6) | Estimated MD –0.1 (95% CI –0.3 to 0.1) | ||||
| Kaur 2019 | Oral health | Questionnaire on oral self‐care knowledge and oral self‐care behaviour (0 to 15, higher score is better) |
IG: 70 CG: 70 |
3 months after randomisation | 4.389 (2.15) | 0.82 (2.013) (95% CI 0.34 to 1.31) | MD 3.57 (2.88 to 4.26) Group x time P < 0.0001 Mean (SD) was calculated from reported linear mixed model analysis |
| Kim 2009 | Type 2 diabetes | DKT (14 items, 0 to 14 (general test, knowledge I), 9 items insulin subscale (knowledge II)1, higher score is better) |
IG: 40 CG: 39 |
30 weeks after randomisation | Knowledge (I) 2.4 (2.3) Knowledge (II) 0.3 (3.7)1 |
Knowledge (I) 0.7 (2.4) Knowledge (II) 0.4 (0.8)1 |
Change scores are reported Knowledge (I) P = 0.00 Knowledge (II) P = 0.27 |
| Kim 2014 | High blood pressure | HBP knowledge questionnaire (0 to 26, higher score is better) |
IG: 184 CG: 185 |
12 months after randomisation | 20.8 (2.7) | 19.3 (3.7) | Cluster‐RCT; data have been re‐analysed for meta‐analysis using the appropriate unit of analysis with the use of the ICC reported by Han 2017. Group x time P = 0.001 (see Analysis 2.10; Analysis 2.11; Analysis 2.14; Analysis 2.13) |
| 18 months after randomisation (6‐month follow‐up) | 20.8 (2.8) | 20.1 (3.2) | |||||
| Kim 2020 | Type 2 diabetes | DKT (14 items, 0 to 14 (general test), 9 items insulin subscale (results not reported), higher score is better) |
IG: 105 CG: 104 |
12 months after randomisation | 10.3 (SE 0.2) | 8.3 (SE 0.3) | Group P < 0.001 |
| Rosal 2005 | Type 2 diabetes | ADKnowl, adapted version (23 item‐sets (104 items), 0 to 104), higher score is better |
IG: 15 CG: 10 |
3 months after randomisation (immediately post‐intervention) | 0.05 (0.15) | ‐0.02 (0.11) | Change scores are reported Group x time P = 0.27 |
| 6 months after randomisation (4.5 months post‐intervention) | 0.05 (0.13) | ‐0.03 (0.08) | |||||
| 3 Culturally adapted health literacy skills building course vs no/unrelated health literacy intervention | |||||||
| Elder 1998 | Nutrition/cardiovascular health | Nutrition knowledge test (0 to 12, higher score is better) |
IG: 134 CG: 157 |
3 months after randomisation (immediately post‐intervention) | 6.76 | 6.04 | Cluster‐RCT; unadjusted values are reported Group x time P ≤ 0.001 |
| At 6‐month follow‐up | 6.90 | 6.11 | |||||
| Otilingam 2015 | Nutrition/heart and brain health | US Department of Agriculture's Diet and Health Knowledge Survey (0 to 9, higher score is better) |
IG 1: 32 IG 2: 33 CG 1: 16 CG 2: 18 |
Immediately post‐intervention | IG 1: 6.86 (1.27) IG 2: 7.03 (0.91) Combined: 6.95 (1.10) |
CG 1: 5.94 (1.12) CG 2: 6.22 (0.94) Combined: 6.09 (1.02) |
Group x time P = 0.0293 (combined IGs vs CG 1) Both IGs and CGs were combined for meta‐analyses (see Analysis 3.3) CG 2 was assessed post‐test only |
| IG 1: 29 IG 2: 29 CG 1: 16 CG 2: 18 |
At 1‐month follow‐up | IG 1: 6.72 (1.33) IG 2: 6.66 (1.11) IG 1, 2*: 6.69 (1.21) |
CG 1: 5.56 (1.71) | ||||
| Taylor 2011 | Hepatitis B prevention, no specific health problem of participants reported | Questionnaire (0 to 5, higher score is better) |
IG: 80 CG: 100 |
At 6‐month follow‐up | 3.68 (1.12) | 2.87 (1.38) | Cluster‐RCT; data have been re‐analysed for meta‐analysis using the appropriate unit of analysis with the use of the ICC reported by Han 2017. |
| Immigrants are more likely
to be infected with HBV
AOR 2.12 (1.12 to 4.03) HBV can be spread during childbirth AOR 2.10 (0.96 to 4.62) HBV can be spread during sexual intercourse AOR 2.58 (1.29 to 5.15) HBV can be spread by sharing razors AOR 5.42 (1.91 to 15.39) HBV infection can cause liver cancer AOR 2.08 (1.08 to 4.02 |
AOR estimated through GEE models were used to account for clustering; adjusted for ESL organisation, class time, country of origin, years since immigration, gender, age group, years of education and marital status | ||||||
| Tong 2017 | Colorectal cancer | Questionnaire (0 to 5, higher score is better) |
IG: 161 CG: 168 |
6 months after first session (at 3‐month follow‐up) | Knowledge of colon polyps: 23.6% to 78.3%, MD 54.7% Screening start age at 50 years: 14.3% to 36.0%, MD 21.7% FOBT yearly: 10.6% to 38.5%, MD 27.9% Sigmoidoscopy every 5 years: 3.7% to 24.2%, MD 20.5% Colonoscopy every 10 years: 2.5% to 20.5%, MD 18% |
Knowledge of colon polyps: 19.6% to 37.5%, MD 17.9% Screening start age at 50 years: 11.9% to 14.3%, MD 2.4% FOBT yearly: 11.9% to 17.3%, 5.4% Sigmoidoscopy every 5 years: 1.2% to 4.2%, MD 3% Colonoscopy every 10 years: 3.6% to 6.5%, MD 2.9% |
MD 36.8%, P < 0.0001 MD 19.3%, P = 0.0056 MD 22.5%, P = 0.0001 MD 17.5%, P < 0.0001 MD 15.1%, P = 0.012 Cluster‐RCT. No composite score reported. Authors state that GEE models were used to account for clustering. "For every point increase on the knowledge score (0‐5), the odds of ever‐screening and being up to date with screening were significantly increased, supporting knowledge as a mediator of the intervention effect." (Tong 2017 |
| Wong 2020 | Mental health (depression) | CBT‐Q (0 to 9, higher score is better) |
IG: 18 CG: 19 |
Immediately post‐intervention | 5.06 (0.10) | 4.33 (1.24) | P = 0.07 |
| At 2‐month follow‐up | — | ||||||
| 4 Culturally and literacy adapted telephone education vs unrelated culturally and literacy adapted telephone education | |||||||
| Lepore 2012 | Prostate cancer screening | Questionnaire (0 to 14, higher score is better) |
IG: 215 CG: 216 |
Approx. 7 months post‐intervention | 61.6 (SE 0.009) | 54.7 (SE 0.009) | P < 0.001 Adjusted for education, any PSA claim prior to pretest, and percent correct on knowledge index at pretest |
| 5 Culturally and literacy adapted audio‐/visual education without personal feedback vs no health literacy intervention | |||||||
| DeCamp 2020 | Child health | Questionnaire (0 to 5, higher score is better) |
IG: 72 CG: 63 |
10 to 13 months after randomisation (immediately to 3 months post‐intervention) | 0.67 (0.15) | 0.52 (0.15) | Change scores are reported P = 0.52 |
| Hernandez 2013 | Depression | Depression Knowledge Scale (0 to 17, higher score is better) | IG: 72 CG: 64 |
Immediately post‐intervention | 2.44 (2.24) | 0.02 (1.79) | Change scores are reported |
| Thompson 2012 | Child nutrition and feeding | Questionnaire (0 to 19, higher score is better) |
IG: 80 CG: 78 |
Immediately post‐intervention | 17.25 (1.7) | 13.7 (2.1) | P < 0.001 |
| 90.8 (9) | 72.3 (11.2) | ||||||
| 6 Culturally and literacy adapted audio‐/visual education without personal feedback vs written information on the same topic | |||||||
| Gwede 2019 | Colorectal cancer | Awareness of colorectal cancer and screening tests (Questionnaire based on NCI’s Health Information National Trends Survey and on literature, 0 to 11, higher score is better) |
IG: 32 CG: 27 |
At 3‐month follow‐up | 7.9 (2.0) | 6.4 (2.2) | — |
| Payán 2020 | Breast cancer | Questionnaire (0 to 16, higher score is better) |
IG 1: 79 (Cuidarse brochure) IG 2: 79 (Cuidarse brochure, CHW delivered) CG: 82 (standard brochure) |
Immediately post‐intervention | IG 1: 11.7 (2.7) IG 2: 11.5 (2.6) IG 1, 2: 11.6 (2.64) |
CG: 11.5 (3.0) | 10 to 13 months after randomisation; and IGs were combined for meta‐analysis (see, Analysis 6.6; Analysis 6.7; Analysis 6.8; Analysis 6.9) |
| IG 1: 67 IG 2: 61 CG: 65 |
At 3‐month follow‐up | IG 1: 10.3 (3.1) IG 2: 10.2 (2.8) IG 1, 2: 10.25 (2.95) |
CG: 10.7 (2.7) | ||||
| Poureslami 2016a | Asthma | Functional knowledge of asthma symptoms, triggers and factors that could make asthma worse (5‐point Likert scale, range not reported, higher score is better) |
Group 1: 22 Group 2: 21 Group 3: 20 Group 4: 22 |
At 3‐month follow‐up | Knowledge of asthma symptoms Group 1: ‐0.19, 95% CI ‐0.78 to 0.40 Group 2: 0.33, 95% CI ‐0.30 to 0.97 Group 3: 0.88, 95% CI ‐0.02 to 1.79 Knowledge of asthma triggers Group 1: 0.50, 95% CI ‐0.62 to 1.62 Group 2: 1.29, 95% CI ‐0.03 to 2.54) Group 3: 0.29, 95% CI ‐0.99 to 1.58 Knowledge of triggers that could make asthma worse Group 1: ‐0.18, 95% CI ‐2.37 to 2.01 Group 2: 0.86, 95% CI ‐0.51 to 2.22 Group 3: 0.35, 95% CI ‐1.12 to 1.94 |
Knowledge of asthma symptoms Group 4: 0.17, 95% CI ‐0.62 to 0.95 Knowledge of asthma triggers Group 4: 1.22, 95% CI 0.38 to 2.07 Knowledge of triggers that could make asthma worse Group 4: 0.45, 95% CI ‐1.41 to 2.31 |
6‐month assessment not reported No composite score reported, data were not combined as no score range was reported; the scale could not be standardised on a scale ranging from 0 to 100 Results reported are adjusted for age, gender, educational level and ethnicity Data have been extracted from the secondary reference (see Poureslami 2016a for all trial reports related to this study) |
| Poureslami 2016b | COPD | "Some" questions of BCKQ | — | ||||
| Unger 2013 | Depression | Depression Knowledge Scale (0 to 17, higher score is better) | IG: 69 CG: 70 |
Immediately post‐intervention | 2.37 (SE 0.32) | 0.86 (SE=0.27) | |
| 1‐month follow‐up | t = 5.09, P < 0.05 | t = 2.64, P < 0.05 | "[T]he data collectors reported that several students shared their photonovel with students in the text pamphlet group after the posttest." (Unger 2013, p. 405) | ||||
| Valdez 2015 | Cervical cancer | Questionnaire (0 to 12, higher score is better) |
IG: 290 CG: 318 |
At 1‐month follow‐up | 8.9 (1.6) | 7.1 (2.0) | P < 0.0001 |
| Valdez 2018 | Cervical Cancer | Questionnaire (0 to 5, higher score is better) |
IG: 383 CG: 344 |
At 6‐month follow‐up | 3.7 (1.6) | 3.1 (1.4) | P < 0.0001 |
| 7 Culturally and literacy adapted audio‐/visual education without personal feedback vs another culturally and literacy adapted audio‐/visual education without personal feedback | |||||||
| Ochoa 2020 | Cervical cancer | Questionnaire (0 to 8, higher score is better) |
IG: 61 CG: 48 |
At 2‐week follow‐up | 5.10 (1.45) | 4.44 (1.15) | P = 0.011 |
| At 6‐month follow‐up | 5.38 (1.27) | 5.29 (1.17) | P = 0.718 | ||||
| Poureslami 2016a | Asthma | Functional knowledge of asthma symptoms, triggers, and factors that could make asthma worse (5‐point Likert scale, range not reported, higher score is better) |
Group 1 (physician‐led knowledge video): 22 Group 2 (narrative, peer‐led video): 21 |
At 3‐month follow‐up | Knowledge of asthma symptoms Group 1: ‐0.19, 95% CI ‐0.78 to 0.40 Knowledge of asthma triggers Group 1: 0.50, 95% CI ‐0.62 to 1.62 Knowledge of triggers that could make asthma worse Group 1: ‐0.18, 95% CI ‐2.37 to 2.01 |
Knowledge of asthma symptoms Group 2: 0.33, 95% CI ‐0.30 to 0.97 Knowledge of asthma triggers Group 2: 1.29, 95% CI ‐0.03 to 2.54) Knowledge of triggers that could make asthma worse Group 2: 0.86, 95% CI ‐0.51 to 2.22 |
6‐month assessment not reported No composite score reported Results are adjusted for age, gender, educational level and ethnicity |
| Poureslami 2016b | COPD | "Some" questions from BCKQ | A 3‐month follow‐up | — | |||
*Unadjusted mean (SD) if not otherwise reported.
1 Assessed only for those injecting insulin (intervention, n = 5; control, n = 7). Data were not included in the meta‐analyses.
ADKnowl: Audit of Diabetes Knowledge; AOR: adjusted odds ratio; BCKQ: Bristol COPD Knowledge Questionnaire; CBT: cognitive behavioural therapy; CBT‐Q: Knowledge of CBT questionnaire; CG: control group; CI: confidence interval; COPD: chronic obstructive pulmonary disease; DKT: Diabetes Knowledge Test; ESL: English as a second language; GEE: generalised estimating equations; HBP: high blood pressure; HBV: hepatitis B virus; IG: intervention group; NCI: National Cancer Institute; OR: odds ratio; PSA: prostate‐specific antigen; SD: standard deviation; SE: standard error