Skip to main content

Table 4 Random-effects meta-analysis results of uncontrolled before-and-after studies

From: A systematic review and meta-analysis of the effectiveness of food safety education interventions for consumers in developed countries

Meta-analysis sub-groupa

No. participants/trials/studiesb

No. (%) trials with combined outcome measuresc

Effect measured

Effect estimate average (95 % CI)M/median (range)R,e

I 2

Educational training/courses

 Educators-Behaviours

85/2/2

1 (50)

SMD

0.44 (0.33, 0.54)M

0 %

 Educators-Knowledge

47/3/3

2 (67)

RR

2.86 (1.31, 5.63)R

79 %

 Educators-Attitudes

33/2/2

0 (0)

RR

1.34 (1.06, 1.63)R

73 %

 Adults-Behaviours

11,764/17/16

8 (47)

SMD

0.28 (0.11, 1.49)R

100 %

 Adults-Behaviours

3049/10/10

7 (70)

RR

1.26 (0.95, 2.66)R

97 %

 Adults-Knowledgef

1018/8/7

0 (0)

SMD

0.61 (0.01, 1.04)R

95 %

 Adults-Knowledge

4239/6/6

1 (17)

RR

1.92 (1.18, 3.10)R

99 %

 Adults-Attitudesf

1332/7/7

3 (43)

SMD

0.43 (0.05, 0.95)R

99 %

 Adults-Attitudes

876/4/4

1 (25)

RR

1.09 (1.04, 1.33)R

86 %

 Children/youth-Behaviours

401/3/3

0 (0)

SMD

0.31 (0.14, 1.32)R

99 %

 Children/youth-Behaviours

226/2/2

1 (50)

RR

5.37 (1.04, 9.69)R

94 %

 Children/youth-Knowledge

1028/6/6

1 (17)

SMD

1.06 (0.09, 4.02)R

100 %

 Children/youth-Knowledge

1719/6/6

4 (67)

RR

1.77 (1.11, 5.04)R

98 %

 Children/youth-Attitudes

294/3/3

0 (0)

SMD

0.31 (0.10, 1.32)R

99 %

Media campaigns/other messaging

 Adults-Behaviours

2430/7/6

5 (71)

RR

1.35 (0.90, 2.35)R

93 %

 Adults-Knowledge

1129/7/7

7 (100)

RR

1.58 (1.07, 1.87)R

96 %

 Adults-Attitudesf

1002/3/3

1 (33)

SMD

0.43 (0.13, 0.81)R

99 %

 Adults-Attitudes

2420/6/5

2 (33)

RR

1.10 (1.02, 1.23)R

85 %

 Adults-Stages of changeg

1193/3/2

0 (0)

RR

1.09 (1.07, 1.81)R

70 %

  1. SMD standardized mean difference (Hedges’ g), RR relative risk, CI confidence interval
  2. aSubgroups divided by intervention type, target population, and outcome type. Note that all outcomes in this table had a GRADE rating of very low (the true effect is likely to be substantially different from the measured estimate)
  3. bFor trials with a RR outcome, the number of participants in this column refers to the number in the post intervention group
  4. cRefers to studies that reported multiple measures of the same construct in the same individuals, which were combined post hoc into one overall measure
  5. dNote that all trials in SMD analyses used imputed values for pre-post correlations of 0.81 for knowledge and attitude outcomes from Medeiros et al. (2004) [36] or 0.83 for behaviour outcomes from Kendall et al. (2004) [37]
  6. eSuperscript M indicates that an average estimate of effect and 95 % CI is provided because heterogeneity was low to moderate (I2 = 0–60 %). Superscript R indicates that the median and range of study effect sizes is provided because heterogeneity was high (I2 > 60 %)
  7. fSensitivity analyses for these outcomes revealed that the selection of the imputed correlation value had an appreciable impact on the meta-analysis results (Additional file 12), leading to a downgrading of these findings in the GRADE assessment (Additional file 10)
  8. gRR for this outcome refers to the impact of the intervention to change participants’ stage from contemplation/pre-contemplation/preparation to action/maintenance [32]
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us