Network meta-analysis

HbA1c change

Type 2 diabetes (adults); background therapy: no glucose lowering drug (lifestyle only)

weeks 36 to 64 (target week 52)Not stated by the trialsDoses of a drug pooledMean difference

Randomized trials

Treatments

1,790

Participants

Trial arms

In plain language

Across 5 trials with 1,790 participants, 1 of 6 treatments had a clearly better result than placebo on HbA1c change, and for 5 the evidence could not tell them apart from placebo. The most favorable estimate was for Dulaglutide injection: mean difference -1.41 percentage points (95% credible interval -2.36 to 0.35).

Lower values favor the treatment. "Clearly" means the 95% credible interval excludes no difference. These are averages across trials; they do not say how any one person will respond. Not medical advice.

Network of trials

Each circle is a treatment; each line joins treatments that were randomized against each other in at least one trial.

Node area shows participants; line width shows trials comparing two treatments directly. Blue: GLP-1 receptor agonist containing treatments; grey: comparators. Hover over or tab to a node or line for details.

Effects compared with Placebo

Mean difference in percentage points. Lower values favor the treatment.

Versus Placebo. Blue: Bayesian random effects posterior median with 95% credible interval (thick) and 95% prediction interval (dashed). Amber diamond: frequentist estimate with 95% confidence interval. Hover over or tab to a row for its values.

Effects of each treatment versus Placebo
Treatment	Bayesian MD (95% CrI)	95% prediction interval	Frequentist MD (95% CI)	Verdict
Dulaglutide injection DULA_SC	-1.41 (-2.36 to 0.35)	-2.72 to 0.67	-1.60 (-2.03 to -1.17)	Unclear
Semaglutide tablet SEMA_PO	-1.32 (-2.10 to -0.09)	-2.55 to 0.40	-1.41 (-1.69 to -1.13)	Better
Albiglutide injection ALBI_SC	-1.26 (-2.26 to 0.51)	-2.63 to 0.92	-1.43 (-1.86 to -1.00)	Unclear
Liraglutide injection LIRA_SC	-1.23 (-1.95 to 0.06)	-2.32 to 0.58	-1.40 (-1.79 to -1.01)	Unclear
Efpe sc EFPE_SC	-0.84 (-1.91 to 0.35)	-2.32 to 0.83	-0.87 (-1.50 to -0.24)	Unclear
Glibenclamide GLIBENCLAMIDE	-0.71 (-1.75 to 1.07)	-2.04 to 1.52	-0.90 (-1.38 to -0.42)	Unclear

Ranking

How likely each treatment is to be among the best. Rankings ignore the size of differences and their certainty, so read them with the effects above.

League table

Every treatment against every other: mean difference of the row treatment versus the column treatment, posterior median and 95% credible interval. Green: row clearly better; red: row clearly worse.

Every treatment in a row compared with every treatment in a column
Row versus column	PLACEBO	DULA_SC	SEMA_PO	ALBI_SC	LIRA_SC	EFPE_SC	GLIBENCLAMIDE
PLACEBO	PLACEBO	1.41 -0.35 to 2.36	1.32 0.09 to 2.10	1.26 -0.51 to 2.26	1.23 -0.06 to 1.95	0.84 -0.35 to 1.91	0.71 -1.07 to 1.75
DULA_SC	-1.41 -2.36 to 0.35	DULA_SC	-0.11 -1.33 to 1.53	-0.16 -1.56 to 1.33	-0.18 -1.13 to 0.90	-0.57 -2.03 to 1.48	-0.71 -2.16 to 0.75
SEMA_PO	-1.32 -2.10 to -0.09	0.11 -1.53 to 1.33	SEMA_PO	-0.06 -1.65 to 1.21	-0.08 -1.23 to 0.85	-0.47 -1.84 to 1.15	-0.60 -2.30 to 0.68
ALBI_SC	-1.26 -2.26 to 0.51	0.16 -1.33 to 1.56	0.06 -1.21 to 1.65	ALBI_SC	-0.02 -0.97 to 1.03	-0.40 -1.87 to 1.62	-0.55 -2.06 to 0.87
LIRA_SC	-1.23 -1.95 to 0.06	0.18 -0.90 to 1.13	0.08 -0.85 to 1.23	0.02 -1.03 to 0.97	LIRA_SC	-0.38 -1.65 to 1.33	-0.53 -1.70 to 0.45
EFPE_SC	-0.84 -1.91 to 0.35	0.57 -1.48 to 2.03	0.47 -1.15 to 1.84	0.40 -1.62 to 1.87	0.38 -1.33 to 1.65	EFPE_SC	-0.15 -2.29 to 1.35
GLIBENCLAMIDE	-0.71 -1.75 to 1.07	0.71 -0.75 to 2.16	0.60 -0.68 to 2.30	0.55 -0.87 to 2.06	0.53 -0.45 to 1.70	0.15 -1.35 to 2.29	GLIBENCLAMIDE

Heterogeneity, fit and convergence

Bayesian random effects (primary)

Heterogeneity (tau): 0.334 (95% CrI 0.016 to 1.069)
Residual deviance: 11.0 for 11 data points
DIC: 22.0 (pD 11.0)
Largest R-hat: 1.002
Smallest ESS: bulk 1,368, tail 2,483
Divergent transitions: 0
Convergence: All criteria met
Priors: PR01 N(0, 5^2); PR03 N(0, 2.5^2); PR07 half normal(0, 0.5^2) on tau

Frequentist check (netmeta)

Method: inverse variance, REML, MD
Tau: 0.000
Q (total): 0.00 on 0 df
Q (between designs): 0.00

Inconsistency

Do trials comparing two treatments directly agree with the indirect evidence through other treatments?

No comparison here has both direct and indirect evidence, so inconsistency cannot be checked by node splitting; the between designs Q above is the global check.

Small study effects

A comparison adjusted funnel plot: each point is one trial comparison, centered on its own network estimate. Asymmetry can signal that small trials with unfavorable results are missing.

Fewer than ten trial comparisons, too few to judge funnel plot asymmetry.

Baseline characteristics of the trials

The baseline value of each effect modifier or prognostic factor that applies here, per trial: mean or percent over arms, weighted by the number randomized. A note under a value names a stand in (a median for a mean, 100 minus the male share, Asian subgroup rows, an all Asian site list); "not reported" means no source of the trial prints it.

Baseline effect modifiers and prognostic factors per trial
Trial	Baseline HbA1c (percent) effect modifier, prognostic factor; 5 of 5	Background glucose lowering therapy (stratum S0 to S4) effect modifier, prognostic factor; 5 of 5	Diabetes duration (years) effect modifier, prognostic factor; 5 of 5	Asian ancestry or region (percent Asian) effect modifier, prognostic factor; 3 of 5	Age (years) prognostic factor; 5 of 5	Treatment duration (weeks) effect modifier; 5 of 5
NCT00393718 S000330	9.27	S0	8.24	100 region: all sites in Asia	58.3	52
NCT01558271 S000837	8.14	S0	6.59	98.98	57.47	52
NCT01733758 S001075	8.11	S0	6.65	not reported	58.25	52
PIONEER 9 S001943	8.2	S0	7.6	100	59.4	52
AMPLITUDE-M S002081	8.07	S0	5.1	not reported	58.51	56

Comparability of the trials

Checked across designs (the set of treatments each trial compares) for eligibility, outcome definition and assessment, and the baseline factors above. Flags point to possible intransitivity; notes describe variation that is not tied to design.

FlagBaseline, Baseline HbA1c: design means range from 8.1 (EFPE_SC vs PLACEBO) to 9.3 (GLIBENCLAMIDE vs LIRA_SC) percent; threshold 0.5
FlagBaseline, Diabetes duration: design means range from 5.1 (EFPE_SC vs PLACEBO) to 8.2 (GLIBENCLAMIDE vs LIRA_SC) years; threshold 3
FlagBaseline, Asian ancestry or region: not reported by 2 of 5 trials
FlagHeterogeneity, tau above threshold: tau 0.33 exceeds the outcome threshold 0.3; the flags of this network are candidate explanations
NoteOutcome assessment, time point: selected time points span 52 to 56 weeks
NoteOutcome assessment, analysis set: mixed: completers 1, full_analysis 3, randomized 1
NoteData handling, input conversions: arms_combined 3

Comparability items per trial
Trial	time weeks	analysis set	definition	instrument	adjudicated	proportion method	intercurrent events (discontinuation)	intercurrent events (rescue)	blinding	funding	start year	Eligibility
NCT00393718	52	full_analysis	not stated	not stated	not stated	not stated	not stated	not stated	double_blind	industry	2006	age 20 or older; HbA1c 7.4 to 10.4 percent NGSP; with or without prior oral agents
NCT01558271	52	full_analysis	not stated	not stated	not stated	not stated	not stated	not stated	double_blind	industry	2012	Japanese; age 20 or more; BMI 18.5 to 35; HbA1c 7.0 to 10.0 percent; OAM naive or OAM stopped
NCT01733758	52	completers	not stated	not stated	not stated	not stated	not stated	not stated	double_blind	industry	2013	Japanese adults; diet and exercise with or without one oral drug; HbA1c 7 to 10 percent after washout
PIONEER 9	52	full_analysis	not stated	not stated	not stated	not stated	not stated	not stated	double_blind	industry	2017	Japanese adults 20 y or older; HbA1c 6.5 to 9.5 on one oral drug washed out, or 7.0 to 10.0 on diet and exercise
AMPLITUDE-M	56	randomized	not stated	not stated	not stated	not stated	not stated	not stated	double_blind	industry	2017	type 2 diabetes on diet and exercise, HbA1c 7 to 10 percent, no glucose lowering drug in prior 3 months

Trial data

What each trial contributes. Arm rows give the arm value with its standard error, or events over participants; contrast rows give the mean difference on the analysis scale against the trial's first row. Flags mark converted, combined or approximated inputs.

Trial data entering this analysis
Trial	Arm	Treatment	Week	Data	Value	SE	Events / N	Flags	Source
NCT00393718	A1	LIRA_SC	52	arm, final	7.700	0.100	263		R0022766:Table 2
NCT00393718	A2	GLIBENCLAMIDE	52	arm, final	8.200	0.100	130		R0022766:Table 2
NCT01558271	A1	DULA_SC	52	arm, change	-1.390	0.060	280		R0011754:Results
NCT01558271	A2	LIRA_SC	52	arm, change	-1.190	0.080	136		R0011754:Results
NCT01733758	A2+A3	ALBI_SC	52	arm, change	-1.199	0.047	283	arms_combined	registry:outcomes
NCT01733758	A4	LIRA_SC	52	arm, change	-1.170	0.080	95		registry:outcomes
PIONEER 9	A1+A2+A3	SEMA_PO	52	arm, change	-1.309	0.080	122	arms_combined	registry:outcomes
PIONEER 9	A4	LIRA_SC	52	arm, change	-1.300	0.156	41		registry:outcomes
PIONEER 9	A5	PLACEBO	52	arm, change	0.100	0.120	34		registry:outcomes
AMPLITUDE-M	A1+A2+A3	EFPE_SC	56	arm, change	-1.267	0.115	304	arms_combined	R0007136:Results
AMPLITUDE-M	A4	PLACEBO	56	arm, change	-0.400	0.300	102		R0007136:Results

Rerun this analysis yourself

Remove treatments or trials and refit the frequentist and Bayesian models in your browser, with the same methods as the published analysis.

Open in the NMA lab