Time restricted eating and exercise training before and during pregnancy for people with increased risk of gestational diabetes: single centre randomised controlled trial (BEFORE THE BEGINNING)

Name: Verified Rigor Record — Time restricted eating and exercise training before and during pregnancy for people with increased risk of gestational diabetes: single centre randomised controlled trial (BEFORE THE BEGINNING)
Item: Time restricted eating and exercise training before and during pregnancy for people with increased risk of gestational diabetes: single centre randomised controlled trial (BEFORE THE BEGINNING)
Rating: 92
Author: Αlpha¹

Sujan MJ, Skarstad HM, Rosvold G, Fougner SL, Follestad T, Salvesen KÅ, Moholdt T.

Paper source

Open DOI

Αlpha¹ Rigor Review

Rigor Review Verified

Critical

High-priority

Action items to resolve · Critical / High

Confidence: highEngine: 7.4.0Source: full text

Not verified in this report

References were not verified against Crossref/OpenAlex.

This Alpha1 Rigor Review uses AI Rigor Reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.

The paper reports a well-designed RCT with adequate randomization, sample-size planning, and exemplary data/code sharing. Key weaknesses are the open-label design (no blinding rationale), missing statements on informed consent and regulatory compliance, and selective outcome reporting (some secondary outcomes deferred).

Evaluated as an interventional trial (RCT). Three independent reviewers assessed all eight dimensions; their ratings converged on 5 dimensions and diverged on 3 (study design, ethical approvals, reporting transparency), resolved by weighing detailed checklist evidence. The statistics verification component re-computed 13 of 15 tests consistently (1 inconsistent, 0 decision errors). No citation concerns were raised.

10 major claims checked against the paper's own evidence: all adequately supported.

supportedAbstract, Conclusions
The combination of time-restricted eating and exercise training started before and continued throughout pregnancy had no significant effect on glycaemic control in late pregnancy.
The primary outcome (two-hour glucose at week 28) showed a non-significant mean difference (0.48 mmol/L, 95% CI -0.05 to 1.01, P=0.08), supporting the claim of no effect.
Evidence: Results, Table 2: mean difference 0.48 mmol/L (95% CI -0.05 to 1.01, P=0.08).
“A combination of time restricted eating and exercise training started before and continued throughout pregnancy had no significant effect on glycaemic control in late pregnancy.”
supportedResults, Secondary glycaemic outcomes in pregnancy
The intervention did not significantly improve secondary glycaemic outcomes.
All secondary glycaemic outcomes reported in Table 2 show non-significant differences, with p-values >0.05.
Evidence: Table 2: all secondary glycaemic outcomes (fasting glucose, HbA1c, fasting insulin, HOMA2-B, HOMA2-IR) have confidence intervals crossing zero and p-values >0.05.
“The intervention did not significantly improve secondary glycaemic outcomes (fasting glucose, fasting insulin, HbA1c, HOMA2-IR, and HOMA2-B) compared with the control group.”
supportedResults, Secondary outcomes
The intervention reduced gestational weight gain and fat mass gain at week 28.
The estimated mean weight gain was 2.0 kg lower (95% CI -3.3 to -0.8, P=0.002) and fat mass gain 1.5 kg lower (95% CI -2.5 to -0.4, P=0.008) in the intervention group, both statistically significant.
Evidence: Results, Secondary outcomes: weight at week 28 difference -2.0 kg (95% CI -3.3 to -0.8, P=0.002); fat mass difference -1.5 kg (95% CI -2.5 to -0.4, P=0.008).
“The estimated mean weight gain in the intervention group at gestational week 28 was 2.0 kg lower (95% confidence interval −3.3 to −0.8, P=0.002), and fat mass gain was 1.5 kg lower (−2.5 to −0.4, P=0.008) than the control group.”
supportedResults, Adherence
Participants adhered to time-restricted eating before pregnancy (average window 9.9 hours/day) with a decline during pregnancy.
The data show an average eating window of 9.9 hours (SD 1.2) in the prepregnancy period, increasing during pregnancy, supporting the claim of good prepregnancy adherence and decline.
Evidence: Results, Adherence: 'The eating window was shorter in the intervention group during the rest of the study period than in the control group (Figure 4). In the prepregnancy period, the intervention group had an average eating window of 9.9 hours (standard deviation 1.2), which increased during pregnancy.'
“In the prepregnancy period, the intervention group had an average eating window of 9.9 hours (standard deviation 1.2), which increased during pregnancy (Figure 4 and supplementary table 2).”
supportedAbstract and Results
The intervention had no significant effect on two hour plasma glucose level at gestational week 28.
The primary analysis shows a non-significant difference (p=0.08).
Evidence: Table 2: mean difference 0.48 mmol/L, 95% CI -0.05 to 1.01, p=0.08
“The intervention had no significant effect on two hour plasma glucose level in an oral glucose tolerance test at gestational week 28 (mean difference 0.48 mmol/L, 95% confidence interval −0.05 to 1.01, P=0.08).”
supportedResults, Secondary outcomes
The intervention reduced body weight and fat mass gain at gestational week 28.
Statistically significant differences in weight and fat mass were observed.
Evidence: Table 2: weight difference -2.0 kg (p=0.002), fat mass difference -1.5 kg (p=0.008)
“The estimated mean weight gain in the intervention group at gestational week 28 was 2.0 kg lower (95% confidence interval −3.3 to −0.8, P=0.002), and fat mass gain was 1.5 kg lower (−2.5 to −0.4, P=0.008).”
supportedResults, Primary outcome and secondary glycaemic outcomes in pregnancy
There was no significant effect on GDM incidence.
Both at week 12 and week 28, the differences were non-significant.
Evidence: Results: at week 12 p=1.00, at week 28 p=0.57
“At gestational week 12, 3/51 participants (5.9%) in each group fulfilled the criteria for GDM diagnosis (P=1.00). The corresponding numbers at gestational week 28 were 8/49 participants (16.3%) in the intervention group and 6/52 participants (11.5%) in the control group (P=0.57).”
supportedResults, Per protocol analyses
The per protocol analysis indicated longer time to pregnancy in the intervention group.
A significant difference was found in per protocol analysis.
Evidence: Per protocol results: difference 48 days, p=0.005
“The results from the per protocol analyses were similar to those from the intention-to-treat analyses ... except for time to pregnancy which was significantly longer in the intervention group (48 days, 95% confidence interval 15 to 81, P=0.005).”
supportedAbstract
The intervention had no significant effect on two hour plasma glucose level in an oral glucose tolerance test at gestational week 28.
The paper reports a mean difference of 0.48 mmol/L (95% CI −0.05 to 1.01, P=0.08), which is not statistically significant, and the conclusion in the abstract and discussion matches this result.
Evidence: Table 2, primary outcome row: estimated effect 0.48 (95% CI −0.05 to 1.01, P=0.08).
“The intervention had no significant effect on two hour plasma glucose level in an oral glucose tolerance test at gestational week 28 (mean difference 0.48 mmol/L, 95% confidence interval −0.05 to 1.01, P=0.08).”
supportedResults, Adherence
Adherence to time-restricted eating was higher than to exercise training, especially in pregnancy.
The data show that 31-44% adhered to the eating window in prepregnancy/pregnancy, while only 15-43% achieved the PAI target; adherence declined more sharply for exercise during pregnancy, supporting the claim.
Evidence: Results, Adherence section: 'In the prepregnancy period, 41/83 participants (49%) in the intervention group adhered to a ≤10 hour eating window.' and 'The proportion of pregnant participants with at least 100 weekly PAI points was ... decreasing to 8/55 (15%) in the third trimester.'
“The proportion of pregnant participants who adhered to a ≤10 hour eating window was 23/55 (42%) in the first trimester, 17/55 (31%) in the second trimester, and 21/55 (38%) in the third trimester of pregnancy.”

Recomputed 13 tests: 13 consistent, 0 inconsistent, 13 via agent-written checks. 1 reported summary statistic mathematically impossible for the stated N (GRIM/GRIMMER).

1 integrity concern flagged (0 high).

Rigor

Eight Dimensions6 pass · 2 warn

10 major claims10 supported

5 copyedit issues flagged: mostly clarity, other, consistency.

Reproducibility

2 data/code links checked; 2 live.

dataZenodoLIVEHTTP 200
https://doi.org/10.5281/zenodo.15675472
Resolves to Zenodo (data repository).
datahttps://clinicaltrials.gov/ct2/show/NCT04585581LIVEHTTP 200
Resolves, but the content could not be matched to the paper.

Registered (1 ID: ClinicalTrials.gov). No reporting guideline cited.

Alpha¹ Rigor Action ItemsFinal pass before journal submission

This is a robustly conducted and transparently reported trial overall. Informed readers should weigh the open-label design and the post-hoc sample size revision as limitations. The missing informed consent statement and deferred secondary outcomes are reporting omissions that warrant a correction or clarification from the authors to meet standard reporting expectations.

1.
HIGHethics
Add an explicit statement about informed consent to the Ethics statements section (e.g., 'All participants provided written informed consent').
Informed consent is a standard reporting requirement for human research; its absence is a concrete transparency gap.
2.
HIGHethics
Add a statement of regulatory compliance (e.g., 'The study was conducted in accordance with the Declaration of Helsinki') to the Ethics statements section.
Explicit compliance statements are expected for human research; the named ethics committee alone is insufficient for complete reporting.
3.
HIGHreporting
Include a CONSORT checklist as a supplementary file or reference adherence to CONSORT guidelines in the Methods section.
This is a registered RCT and a CONSORT checklist is the expected reporting standard; its absence is a transparency limitation.
4.
HIGHreporting
In the Methods/Statistical analysis section, describe how missing data were handled in the primary analysis (e.g., maximum likelihood under MAR assumption for the mixed model).
Missing data handling is a key methodological detail; its absence leaves ambiguity about the robustness of the primary analysis.
5.
MEDIUMreporting
In the Methods/Randomisation and blinding section, add a rationale for the open-label design (why blinding was not feasible or considered unnecessary).
The lack of blinding is the major design limitation; providing a rationale helps readers assess potential bias.
6.
MEDIUMreporting
In the Methods/Sample size section, clarify whether the revised sample size calculation (from 260 to 200 to 167) was pre-specified or post-hoc and whether it was approved by the ethics committee or data monitoring board.
A post-hoc sample size revision without transparency could be seen as a flexibility in design; a clear statement strengthens trust.
7.
MEDIUMreporting
In the Methods, describe what 'standard care' entails for the control group (e.g., any recommendations given about diet or exercise).
Adequate description of the control condition is essential for replication and interpretation of the comparison.
8.
MEDIUMstatistics
In the Methods/Statistical analysis section, describe how outliers were handled in the analysis (or state that no outliers were identified and none were excluded).
Outlier handling is a standard reporting element; its absence is a minor but notable gap.
9.
MEDIUMreporting
In the Results, provide a clear account of which pre-specified secondary outcomes are reported here and which are deferred to separate publications, referencing the trial protocol.
Selective outcome reporting is a transparency concern; clarifying the reporting plan mitigates this.
10.
MEDIUMreporting
In the Methods/Sample size section, provide a power analysis for the final sample size that was actually achieved (167 participants).
The original power analysis was for a different target; a post hoc power analysis contextualizes the null result.
11.
LOWcopyedit
Rephrase the Abstract sentence '31/83 participants (37%) in the intervention group adhered to prespecified criteria, whereas 24/55 participants (44%) in the intervention group who became pregnant fulfilled these criteria' to clarify that the denominators differ.
Clarity issue: the sentence is ambiguous about which denominator (all randomized vs. those who became pregnant) each percentage refers to.
12.
LOWreporting
In the Results, report the exact p-value for the per-protocol analysis of time to pregnancy (P=0.005) alongside the ITT result (P=0.10) and discuss the potential bias from selective removal of non-adherent participants.
The divergence between ITT and per-protocol results for this outcome could be misinterpreted; transparent discussion helps readers weigh the evidence.

Full text (61,144 chars)Analyzed: Jul 4, 2026

Αlpha¹ assesses methodological rigor, not the importance of the findings. See how we calibrate →