LivingGLP1
Network meta-analysis

Diarrhea

Polycystic ovary syndrome

longest follow-upRecorded harmsEach dose separatelyHazard ratio (from event counts)
7
Randomized trials
7
Treatments
380
Participants
59
Events

In plain language

Across 7 trials with 380 participants, 0 of 6 treatments had a clearly better result than placebo on diarrhea, and for 6 the evidence could not tell them apart from placebo. The most favorable estimate was for Metformin: hazard ratio (from event counts) 0.95 (95% credible interval 0.11 to 7.35).

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.
LIRAMETF_SC_1p2-0p85MG_QDLIRAMETF_SC_1p2-1000MG_QDLIRA_SC_1p2MG_QDLIRA_SC_1p8MG_QDLIRA_SC_3p0MG_QDMETFORMINPLACEBO
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

Hazard ratio (from event counts). Lower values favor the treatment.
0.10.20.250.51245102050← favors treatmentfavors reference →MetforminLiraglutide 1.2 mg subcutaneous …Liraglutide and metformin 1.2 to…Liraglutide 1.8 mg subcutaneous …Liraglutide 3.0 mg subcutaneous …Liraglutide plus metformin 1.2 t…
Versus Placebo, log scale. 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
TreatmentBayesian HR (95% CrI)95% prediction intervalFrequentist HR (95% CI)Verdict
Metformin
METFORMIN
0.95 (0.11 to 7.35)0.07 to 11.00Unclear
Liraglutide 1.2 mg subcutaneous once daily
LIRA_SC_1p2MG_QD
1.34 (0.16 to 9.98)0.10 to 14.23Unclear
Liraglutide and metformin 1.2 to 0.85 mg injection once daily
LIRAMETF_SC_1p2-0p85MG_QD
1.79 (0.15 to 17.44)0.10 to 24.83Unclear
Liraglutide 1.8 mg subcutaneous once daily
LIRA_SC_1p8MG_QD
2.40 (0.51 to 14.77)0.29 to 26.13Unclear
Liraglutide 3.0 mg subcutaneous once daily
LIRA_SC_3p0MG_QD
2.94 (0.44 to 24.24)0.29 to 39.66Unclear
Liraglutide plus metformin 1.2 to 1000 mg injection once daily
LIRAMETF_SC_1p2-1000MG_QD
2.96 (0.44 to 22.09)0.27 to 33.81Unclear

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: hazard ratio (from event counts) 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 columnPLACEBOMETFORMINLIRA_SC_1p2MG_QDLIRAMETF_SC_1p2-0p85MG_QDLIRA_SC_1p8MG_QDLIRA_SC_3p0MG_QDLIRAMETF_SC_1p2-1000MG_QD
PLACEBOPLACEBO1.05
0.14 to 8.77		0.75
0.10 to 6.13		0.56
0.06 to 6.62		0.42
0.07 to 1.95		0.34
0.04 to 2.29		0.34
0.05 to 2.29
METFORMIN0.95
0.11 to 7.35		METFORMIN0.71
0.21 to 2.43		0.54
0.11 to 2.92		0.39
0.02 to 4.92		0.32
0.03 to 2.45		0.32
0.06 to 1.61
LIRA_SC_1p2MG_QD1.34
0.16 to 9.98		1.41
0.41 to 4.66		LIRA_SC_1p2MG_QD0.76
0.15 to 3.98		0.55
0.03 to 7.22		0.46
0.05 to 3.26		0.45
0.09 to 2.15
LIRAMETF_SC_1p2-0p85MG_QD1.79
0.15 to 17.44		1.84
0.34 to 9.07		1.31
0.25 to 6.47		LIRAMETF_SC_1p2-0p85MG_QD0.71
0.03 to 10.91		0.59
0.04 to 6.06		0.58
0.07 to 4.48
LIRA_SC_1p8MG_QD2.40
0.51 to 14.77		2.56
0.20 to 41.70		1.83
0.14 to 29.54		1.40
0.09 to 28.75		LIRA_SC_1p8MG_QD0.82
0.06 to 11.53		0.82
0.07 to 11.57
LIRA_SC_3p0MG_QD2.94
0.44 to 24.24		3.10
0.41 to 29.72		2.19
0.31 to 21.48		1.70
0.16 to 22.90		1.22
0.09 to 17.42		LIRA_SC_3p0MG_QD1.00
0.21 to 5.64
LIRAMETF_SC_1p2-1000MG_QD2.96
0.44 to 22.09		3.12
0.62 to 17.16		2.21
0.47 to 11.75		1.71
0.22 to 14.60		1.22
0.09 to 14.97		1.00
0.18 to 4.86		LIRAMETF_SC_1p2-1000MG_QD

Heterogeneity, fit and convergence

Bayesian random effects (primary)

Heterogeneity (tau)
0.475 (95% CrI 0.092 to 1.535), log scale
Residual deviance
19.2 for 16 data points
DIC
32.1 (pD 12.9)
Largest R-hat
1.001
Smallest ESS
bulk 2,405, tail 3,681
Divergent transitions
0
Convergence
All criteria met
Priors
PR02 N(0, 5^2); PR05 N(0, 2.5^2); PR09 log normal(-2.1, 1.58^2) on tau^2 (Adverse events, pharm vs pharm)

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.

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.

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.
  • NoteOutcome assessment, time point: selected time points span 12 to 36 weeks
  • NoteOutcome assessment, analysis set: mixed: completers 2, other 2, randomized 1, safety 2
  • NoteOutcome assessment, definition: mixed: nausea 1, not stated 6
  • NoteRisk of bias, blinding: mixed: double_blind 3, open_label 4
  • NoteRisk of bias, funding: mixed: industry 1, mixed 2, non_industry 4
  • NoteData handling, input conversions: nonserious_only 2, printed_denominator 5
Comparability items per trial
Trialtime weeksanalysis setdefinitioninstrumentadjudicatedproportion methodintercurrent events (discontinuation)intercurrent events (rescue)blindingfundingstart yearEligibility
NCT0290993312completersnot statednot statednot statednot statednot statednot statedopen_labelnon_industry2015women 18 to menopause, PCOS phenotype A by Rotterdam criteria, BMI 30 or higher
SAXAPCOS36safetynot statednot statednot statednot statednot statednot stateddouble_blindindustry2018premenopausal women 18 to 45 years; PCOS by modified NIH 1990; BMI 30 or more; no diabetes
NCT0191146812randomizednot statednot statednot statednot statednot statednot statedopen_labelnon_industry2011premenopausal women over 18; BMI 30 or more; metformin 6 months or longer with under 5 percent weight loss; no diabetes
LIPT26safetynot statednot statednot statednot statednot statednot stateddouble_blindmixed2014Rotterdam PCOS; BMI over 25 or insulin resistance; non hormonal contraception; no diabetes or CVD
NCT0218725012othernot statednot statednot statednot statednot statednot statedopen_labelnon_industry2014women with PCOS (Rotterdam phenotype A), age 18 to menopause, BMI 30 or higher
S00302212completersnauseanot statednot statednot statednot statednot statedopen_labelnon_industry2018women with PCOS (Rotterdam); age over 18; premenopausal; BMI 24 or more
S00310126othernot statednot statednot statednot statednot statednot stateddouble_blindmixed2014premenopausal women 18 or older; Rotterdam PCOS; BMI 25 or more and/or C-peptide above 0.6 nmol/L; no diabetes

Trial data

What each trial contributes. Arm rows give the arm value with its standard error, or events over participants; contrast rows give the hazard ratio (from event counts) on the analysis scale (log) against the trial's first row. Flags mark converted, combined or approximated inputs.
Trial data entering this analysis
TrialArmTreatmentWeekDataValueSEEvents / NFlagsSource
NCT02909933A1LIRA_SC_3p0MG_QD12arm, events5 / 15R0022935:Results
NCT02909933A2LIRAMETF_SC_1p2-1000MG_QD12arm, events6 / 15R0022935:Results
SAXAPCOSA1LIRA_SC_3p0MG_QD36arm, events4 / 55R0022023:Table 4
SAXAPCOSA2PLACEBO36arm, events0 / 27R0022023:Table 4
NCT01911468A1METFORMIN12arm, events1 / 14printed_denominatorR0018103:Results
NCT01911468A2LIRA_SC_1p2MG_QD12arm, events6 / 11printed_denominatorR0018103:Results
NCT01911468A3LIRAMETF_SC_1p2-1000MG_QD12arm, events6 / 11printed_denominatorR0018103:Results
LIPTA1LIRA_SC_1p8MG_QD26arm, events5 / 48nonserious_onlyregistry:adverse_events
LIPTA2PLACEBO26arm, events1 / 24nonserious_onlyregistry:adverse_events
NCT02187250A1METFORMIN12arm, events4 / 14printed_denominatorR0023497:Results
NCT02187250A2LIRA_SC_1p2MG_QD12arm, events1 / 14printed_denominatorR0023497:Results
S003022A1METFORMIN12arm, events4 / 20R0015954:Results
S003022A2LIRA_SC_1p2MG_QD12arm, events4 / 20R0015954:Results
S003022A3LIRAMETF_SC_1p2-0p85MG_QD12arm, events6 / 20R0015954:Results
S003101A1LIRA_SC_1p8MG_QD26arm, events5 / 48R0025661:Table 2
S003101A2PLACEBO26arm, events1 / 24R0025661:Table 2

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