Table 5Number of studies, strength of evidence domains, magnitude of effect, and strength of evidence (KQ2)

Outcome: All-cause mortality
Number of StudiesTotal NDomains Pertaining to Strength of EvidenceStrength of Evidence
Risk of Bias: Design/ QualityConsistencyDirectnessPrecisionMagnitude and Direction of Effect
Met vs. TZD
4 RCTs4457LowConsistentDirectImpreciseUnclearLow
Met vs. SU
5 RCTs, 8 obs50498MediumInconsistentDirectPreciseSmall, favors metforminLow
Met vs. DPP-4 Inhibitor
0NANANANANANAInsufficient
Met vs. Meg
1 RCT357LowNADirectImpreciseUnclearLow
Met vs. Met + TZD
5 RCTs2554LowConsistentDirectImpreciseUnclearLow
Met vs. Met + SU
4 RCTs, 6 obs30211MediumConsistentDirectPreciseSmall, neither favoredLow
Met vs. Met + DPP-4 Inhibitor
1 RCT190MediumNADirectImpreciseUnclearLow
Met vs. Met + Meg
1 RCT350LowNADirectImpreciseUnclearLow
TZD vs. TZD
0NANANANANANAInsufficient
TZD vs. SU
3 RCTs3986LowConsistentDirectImpreciseUnclearLow
TZD vs. DPP-4 Inhibitor
0NANANANANANAInsufficient
TZD vs. Meg
0NANANANANANAInsufficient
SU vs. DPP-4 Inhibitor
0NANANANANANAInsufficient
SU vs. Meg
1 RCT576MediumNADirectImpreciseUnclearLow
Sitagliptin vs. Meg
0NANANANANANAInsufficient
Met and TZD vs. Met and SU
2 RCTs970LowConsistentDirectImpreciseUnclearLow
Met and SU vs. Met and Meg
2 RCTs, 1 obs4432LowConsistentDirectImpreciseUnclearLow
Met and SU vs Met and DPP-4 Inhibitor
1 RCT1172LowNADirectImpreciseUnclearLow
Met and Meg vs. Met and Another Agent
0NANANANANANAInsufficient
Met and DPP-4 Inhibitor vs. Met and Another Agent
0NANANANANANAInsufficient
Met and Exenatide vs. Met and Another Agent
0NANANANANANAInsufficient
Met and Basal Insulin vs. Met and Another Agent
0NANANANANANAInsufficient
Met and Premixed Insulin vs. Met and SU
2 RCTs819LowConsistentDirectImpreciseUnclearLow
Met and TZD vs. TZD and SU
0NANANANANANAInsufficient
Met and SU vs. TZD and SU
1 RCT639MediumNADirectImpreciseUnclearLow
Met and Meg vs. TZD and Met
1 RCT374MediumNADirectImpreciseUnclearLow
Met and Sitagliptin vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Exenatide vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Basal Insulin vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Premixed Insulin vs. TZD and Another Agent
0NANANANANANAInsufficient
Outcome: Cardiovascular disease mortality
Number of StudiesTotal NDomains Pertaining to Strength of EvidenceStrength of Evidence
Risk of Bias: Design/ QualityConsistencyDirectnessPrecisionMagnitude and Direction of Effect
Met vs. TZD
2 RCTs2950MediumConsistentDirectImpreciseSmall. Neither favored.Low
Met vs. SU
1 RCT 4 obs16788MediumConsistentDirectImpreciseSmall. Met favored.Low
Met vs. DPP-4 inhibitor
0NANANANANANAInsufficient
Met vs. Meg
1 RCT357LowNADirectImpreciseUnclearLow
Met vs. GLP-1 agonist
0NANANANANANAInsufficient
Met vs. Met + TZD
3 RCTs1479LowConsistentDirectImpreciseSmall. Unclear.Low
Met vs. Met + SU
2 obs4968MediumConsistentDirectImpreciseUnclearLow
Met vs. Met + DPP-4 inhibitor
1190LowNADirectImpreciseUnclear.Low
Met vs. Met + Meg
1350LowNADirectImpreciseUnclearLow
TZD vs. TZD
0NANANANANANAInsufficient
TZD vs. SU
12897LowNADirectImpreciseUnclearLow
TZD vs. DPP-4 inhibitor
0NANANANANANAInsufficient
TZD vs. Meg
0NANANANANANAInsufficient
TZD vs. GLP-1 agonist
0NANANANANANAInsufficient
SU vs. DPP-4 inhibitor
0NANANANANANAInsufficient
SU vs. Meg
1576LowNADirectImpreciseUnclearLow
SU vs. GLP-1 agonist
0NANANANANANAInsufficient
DPP-4 inhibitor vs. Meg
0NANANANANANAInsufficient
Met and TZD vs. Met and Another Agent
1 RCT561LowNADirectImpreciseUnclearLow
Met and SU vs. Met and Another Agent
2 RCTs4447LowInconsistentDirectImpreciseUnclearLow
Met and Meg vs. Met and Another Agent
NoneInsufficient
Met and DPP-4 inhibitor vs. Met and Another Agent
1 RCT1172LowNADirectImpreciseUnclearLow
Met and GLP-1 agonist vs. Met and Another Agent
1 RCT1172LowNADirectImpreciseUnclearLow
Met and Basal Insulin vs. Met and Another Agent
1 RCT91LowNADirectImpreciseUnclearLow
Met and Premixed Insulin vs. Met and Another Agent
2438LowConsistentDirectImpreciseUnclearLow
Met and TZD vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and SU vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Meg vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Sitagliptin vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Exenatide vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Basal Insulin vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Premixed Insulin vs. TZD and Another Agent
0NANANANANANAInsufficient
Outcome: Cardiovascular and cerebrovascular disease morbidity
Number of StudiesTotal NDomains Pertaining to Strength of EvidenceStrength of Evidence
Risk of Bias: Design/QualityConsistencyDirectnessPrecisionMagnitude and Direction of effect
Met vs. TZD
4 RCTs
9 obs
640, 910LowInconsistentDirectImpreciseUnclearLow
Met vs. SU
2 RCTs,
5 obs
609, 436LowInconsistentDirectImpreciseUnclear.Low
Met vs. DPP-4 inhibitor
0NANANANANANAInsufficient
Met vs. Meg
1 RCT701LowNAIndirectImpreciseUnclearLow
Met vs. Met + TZD
6 RCTs
1 obs
486,276LowConsistentDirectImpreciseSmall. Met favored.Low
Met vs. Met + SU
1 RCT, 1 obs10449MediumInconsistentDirectImpreciseSmall. Met favored.Low
Met vs. Met + DPP-4 inhibitor
2 RCTs376LowConsistentDirectImpreciseUnclear.Low
Met vs. Met + Meg
1350LowNAIndirectImpreciseUnclearLow
TZD vs. TZD
3 obs585,454HighInconsistentDirectImpreciseUnclearLow
TZD vs. SU
3 RCTs
5 obs
518914LowInconsistentDirectImpreciseUnclear.Low
TZD vs. DPP-4 inhibitor
0NANANANANANAInsufficient
TZD vs. Meg
0NANANANANANAInsufficient
SU vs. DPP-4 inhibitor
0NANANANANANAInsufficient
SU vs. Meg
2 RCTs969LowInconsistentDirectImpreciseUnclearLow
SU vs. GLP-1 agonist
1 RCT411LowNADirectImpreciseUnclearLow
DPP-4 Inhibitor vs. Meg
0NANANANANANAInsufficient
Met and TZD vs. Met and Another Agent
4 RCTs
1 obs
474,644LowInconsistentDirectImpreciseUnclearLow
Met and SU vs. Met and Another Agent
3 RCTs6791LowInconsistentDirectPreciseUnclearLow
Met and Meg vs. Met and Another Agent
0NANANANANANAInsufficient
Met and DPP-4 inhibitor vs. Met and Another Agent
1 RCT665LowNAIndirectImpreciseUnclearLow
Met and GLP-1 agonist vs. Met and Another Agent
1 RCT665LowNAIndirectImpreciseUnclearLow
Met and Basal Insulin vs. Met and Another Agent
1 cross-over105MediumNADirectImpreciseUnclearLow
Met and Premixed Insulin vs. Met and Another Agent
0NANANANANANAInsufficient
Met and TZD vs. TZD and Another Agent
2 RCTs, 2 obs488,535LowInconsistentDirectImpreciseUnclearLow
Met and SU vs. TZD and Another Agent
1 RCT639LowNAIndirectImpreciseUnclearLow
Met and Meg vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and DPP-4 inhibitor vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and GLP-1 agonist vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Basal Insulin vs. TZD and Another Agent
0NANANANANANAInsufficient
Met and Premixed Insulin vs. TZD and Another Agent
0NANANANANANAInsufficient
Outcome: Nephropathy
Number of StudiesTotal NDomains Pertaining to Strength of EvidenceStrength of Evidence
Risk of Bias:ConsistencyDirectnessPrecisionMagnitude and Direction of effect
Met vs. TZD
2 RCT’s1715LowConsistentIndirectPreciseSmall. Favors pioglitazoneModerate
Met vs. SU
1 RCT’s51HighNAIndirectImpreciseUnclearLow
TZD vs. SU
5 RCT’s375HighInconsistentIndirectImpreciseUnclearLow
TZD vs. Meg
1 RCT68LowNAIndirectImpreciseUnclearLow
Met and TZD vs. Met and Another Agent
1 RCT’s389LowNAIndirectImpreciseUnclearLow
Outcome: Neuropathy
Number of StudiesTotal NDomains Pertaining to Strength of EvidenceStrength of Evidence
Risk of Bias:ConsistencyDirectnessPrecisionMagnitude and Direction of effect
Met vs. Met and TZD
1 RCT’s105ModerateNAIndirectImpreciseUnclearLow
Met vs. Met and sitagliptin
1 RCT’s190HighNAIndirectImpreciseUnclearLow
Met and TZD vs. Met and Another Agent
1 RCT’s183HighNAIndirectImpreciseUnclearLow

Meg = meglitinides; Met = metformin; Nateg = nateglinide; Pio = pioglitazone; RCT, randomized controlled trial; Repag = repaglinide; Rosi = rosiglitazone; Sita = sitagliptin; SU = sulfonylurea; TZD = thiazolidinedione

All other comparisons were graded as insufficient since there were no studies of those comparisons.

The strength of the evidence was defined as follows: High = High confidence that the evidence reflects the true effect. Further research is unlikely to change our confidence in the estimate of the effect. Moderate = Moderate confidence that the evidence reflects the true effect. Further research may change our confidence in the estimate of the effect and may change the estimate. Low = Low confidence that the evidence reflects the true effect. Further research is likely to change our confidence in the estimate of the effect and is likely to change the estimate. Insufficient = Evidence is unavailable.

From: Appendix G, Evidence Tables

Cover of Oral Diabetes Medications for Adults With Type 2 Diabetes: An Update
Oral Diabetes Medications for Adults With Type 2 Diabetes: An Update [Internet].
Comparative Effectiveness Reviews, No. 27.
Bennett WL, Wilson LM, Bolen S, et al.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.