Skip to main content
. 2021 Jul 15;374:n1537. doi: 10.1136/bmj.n1537

Table 2.

GRADE profile for pairwise meta-analyses

No of participants, No of RCTs; mean follow-up Quality of evidence Summary of findings
Risk of bias Inconsistency Indirectness Imprecision Publication bias Overall certainty of evidence Relative effect (OR (95% CI)) Anticipated absolute effect
(event rate per 10 000 people throughout mean follow-up)
Risk with controls Risk difference (95% CI) with statins
Muscle symptoms
65 304, 21; 4.3 years Not serious Not serious Not serious Not serious None ⨁⨁⨁⨁
High
1.06 (1.01 to 1.13) 951 56 (5 to 108)
Muscle disorders
85 740, 25; 4.2 years Not serious Not serious Not serious Serious* None ⨁⨁⨁◯
Moderate
0.88 (0.62 to 1.24) 14 −2 (−5 to 3)
Liver dysfunction
54 803, 21; 3.8 years Not serious Not serious Not serious Not serious None ⨁⨁⨁⨁
High
1.33 (1.12 to 1.58) 92 30 (11 to 53)
Renal insufficiency
32 001, 8; 4.0 years Not serious Serious† Not serious Not serious None ⨁⨁⨁◯
Moderate
1.14 (1.01 to 1.28) 343 45 (3 to 93)
Diabetes
58 629, 9; 4.9 years Not serious Serious‡ Not serious Not serious None ⨁⨁⨁◯
Moderate
1.01 (0.88 to 1.16) 396 4 (−45 to 59)
Eye conditions
25 328, 6; 3.8 years Not serious Serious§ Not serious Not serious None ⨁⨁⨁◯
Moderate
1.23 (1.04 to 1.47) 233 53 (8 to 105)
Myocardial infarction
95 148, 22; 4.4 years Not serious Not serious Not serious Not serious None¶ ⨁⨁⨁⨁
High
0.72 (0.66 to 0.78) 292 −81 (−98 to−63)
Stroke
78 473, 17; 4.7 years Not serious Not serious Not serious Not serious None ⨁⨁⨁⨁
High
0.80 (0.72 to 0.89) 201 −39 (−55 to−22)
Death from cardiovascular disease
95 959, 22; 4.4 years Not serious Not serious Not serious Not serious None ⨁⨁⨁⨁
High
0.83 (0.76 to 0.91) 218 −36 (−52 to−19)

RCT=randomised controlled trial; OR=odds ratio; GRADE=Grading of Recommendations Assessment, Development, and Evaluation.

*

The analysis was underpowered to detect a difference between groups, given the low incidences in both groups.

Four studies reported the presence of proteinuria although the other four studies reported non-specific renal insufficiency.

One study was conducted in patients with impaired glucose tolerance, which resulted in statistically significant (P=0.04) heterogeneity among the included studies.

§

One study reported cataracts, one reported diminished visual acuity, one reported eye inflammation, two reported non-specific eye and adnexa disorders, and one reported a combination of different eye conditions.

Publication bias was detected by examining the asymmetry of the funnel plot for the pairwise meta-analysis of myocardial infarction, but the pooled results did not change after excluding the small studies that caused the detected bias.