Skip to main content
. 2015 Jan 13;3(1):E35–E46. doi: 10.9778/cmajo.20140047

Table 5: Subgroup analyses for harms.

Outcome, subgroup Effect
Statistical heterogeneity (within-group),
Test for between-group differences,
No. of participants No. of studies GRADE quality of evidence*
Risk ratio (95% CI) Absolute risk increase, % No. needed to harm (95% CI) p value I2 value, % p value I2 value, %
Any adverse events
Behavioural
Not estimable†


NA
NA
482
3
Moderate
Pharmacological + behavioural
1.03
(0.99 to 1.08)


NA
533
1
Low
Serious adverse events
Behavioural
0.51
(0.09 to 2.73)


NA
0.37
(0)
322
1
Moderate
Pharmacological + behavioural
1.25
(0.46 to 3.35)


0.56
0
573
2
Low
Gastrointestinal events
Pharmacological + behavioural
3.77
(2.56 to 5.55)
36.7
3
(2 to 5)
NA
NA
533
1
Moderate
Study withdrawal due to adverse events
Pharmacological + behavioural 2.49 (0.79 to 7.87) 0.45 0 NA 573 2 Low

Note: Note: GRADE =Grading of Recommendations Assessment, Development and Evaluation, NA = not applicable.
* Rating reflects confidence in the estimate of effect assessed through 5 domains of the evidence (risk of bias, indirectness, imprecision, inconsistency and reporting bias) as follows: moderate = downgraded for risk of bias; low = downgraded for risk of bias and imprecision.
†No events reported in both groups in all studies.