Purpose: The COVID-19 pandemic has led to the publication of numerous primary studies and meta-analyses; however, conclusive evidence on whether HIV infection influences COVID-19 outcomes among people living with HIV (PLHIV) is still lacking. This research uses a novel technique, the exit meta-analysis, to conclusively update the evidence of HIV’s impact on COVID-19-related mortality, hospitalization, and need for Intensive Care Unit (ICU) admission in severe disease. Methods: A search of PubMed, EMBASE, Cochrane Reviews (CDSR), SCOPUS, CINAHL reviews and Google Scholar databases was conducted up to the 18 January 2024 for meta-analyses and observational studies that reported adjusted associations for the effect of HIV on COVID-19 related mortality, hospitalization, and ICU admission. Evidence from existing meta-analyses was summarized narratively, and an updated meta-analysis was carried out using a bias-adjusted inverse variance heterogeneity model. Subgroup analysis was carried out for age groups and geographical regions. Results: Of 3153 records identified, 20 meta-analyses and 56 primary studies, with a total of 27,936,428 participants, including 655,882 PLHIV, were included. A review of the meta-analyses showed conflicting results for all outcomes. In the updated synthesis, HIV was associated with higher odds of mortality (aOR 1.43, 95% CI: 1.01–1.86, I² = 90.7%) and ICU admission (aOR 1.49, 95% CI: 0.67–2.30, I² = 88.8%), but not hospitalization (aOR 1.11, 95% CI: 0.78–1.48, I² = 97.5%). The results for both ICU admission and hospitalization include the null value, leading to lower certainty. The exit meta-analysis suggested conclusive results for mortality (DAts score = −0.012) and hospitalization (DAts score = −0.014), but not for ICU admission. Conclusions: This exit meta-analysis provides conclusive evidence that HIV increases mortality in people with COVID-19; however, more studies may be required to address ICU admission and hospitalization.

Background: Gestational diabetes mellitus (GDM), defined as hyperglycaemia first identified during pregnancy, is associated with increased risks of adverse maternal and neonatal outcomes. Standard screening is typically performed at 24–28 weeks’ gestation; however, earlier screening before 20 weeks has been proposed to improve outcomes. Evidence supporting early screening or diagnosis remains inconsistent. We aimed to evaluate whether early screening or early diagnosis of GDM (before 20 weeks) influences maternal and neonatal outcomes compared with standard screening or diagnosis at 24–28 weeks.Methods: We conducted a systematic review and meta-analysis of analytical studies comparing early versus standard timing of GDM screening or diagnosis. A systematic search of PubMed, Embase, Web of Science, and Cochrane CENTRAL was performed from inception to 2025. Eligible studies included those comparing early versus usual screening in unselected pregnant populations and/or early versus usual diagnosis among women with GDM; these comparisons were carefully predefined to minimize potential bias. Two reviewers independently screened studies, extracted data, and assessed methodological quality using the MASTER scale. Pooled effects were generated using the quality-effects model. Publication bias was assessed using Doi plots. For binary outcomes, odds ratios were calculated and converted to absolute measures (risk differences and numbers needed to treat or harm) across a range of baseline risks to enhance interpretability. Outcomes assessed included gestational weight gain, preterm delivery, preeclampsia or eclampsia, macrosomia, neonatal hypoglycemia, neonatal intensive care unit admission, large-for-gestational-age infants, and caesarean delivery.Finding:14 studies met the inclusion criteria. Early diagnosis was not associated with a statistically divergent difference from usual diagnosis in terms of pregnancy-weight gain (WMD -1.08 kg), preterm delivery (OR 1.09), preeclampsia/eclampsia (OR 1.81), macrosomia (OR 0.97), PIH (OR 1.06), neonatal hypoglycemia (OR 1.61), NICU admission (OR 1.27), LGA (OR 0.98) and cesarean section (OR 1.18). Across the general obstetric population, early screening did not demonstrate a statistically divergent difference in any maternal or neonatal outcome including preterm delivery (OR 1.54), preeclampsia/eclampsia (OR 1.24), neonatal hypoglycemia (OR 1.19), Cesarean section (OR 0.87), macrosomia (OR 0.96) and LGA (OR 1.15). Translation of relative effects into absolute measures indicated no absolute clinical benefit for either early diagnosis or early screening. For early diagnosis, absolute risk differences were generally small and baseline-dependent, with directionally harmful effects for only few outcomes, particularly preeclampsia/eclampsia and neonatal hypoglycemia (NNH = 7 to 20), whereas most other outcomes demonstrated risk differences close to zero. Additionally, for early screening, absolute risk differences were largely neutral, with only preterm birth showing a small increase in absolute risk and all other outcomes indicating negligible differences.Interpretation: Early diagnosis of GDM does not identify a group with substantially different maternal and neonatal risks, likely reflecting a population for which intervention is unnecessary. Early screening, when applied to the general population of pregnant women, also does not improve outcomes and offers no absolute interventional benefit. Current evidence does

Background: There is ongoing debate about the ideal blood glucose thresholds for the diagnosis of gestational diabetes mellitus (GDM), as individual randomized controlled trials (RCTs) comparing the lower International Association of Diabetes and Pregnancy Study Groups (IADPSG) blood glucose to criteria which use higher blood glucose thresholds have not been conclusive.Objectives: This meta-analysis compared lower to higher blood glucose thresholds for GDM diagnosis in terms of GDM prevalence and adverse maternal and neonatal health outcomes.Methods: We searched Cochrane CENTRAL, PubMed, Scopus, CINAHL, and preprint archives up to May 30, 2025, for RCTs that compared lower to higher glucose thresholds in the diagnosis of GDM in pregnancy. Endpoints were GDM prevalence and adverse pregnancy outcomes. After study quality assessment, meta-analysis was done using an inverse variance heterogeneity model, heterogeneity was assessed using I2, and publication bias by funnel plots. The quality effects model was used to synthesize evidence that was presented using forest plots.Results: Ten RCTs, comparing IADPSG to higher blood glucose thresholds, with a total of 88,311 pregnant women, were included. After meta-analysis, the lower glycemic criteria compared to the higher glycemic criteria was associated with almost double the prevalence of GDM diagnosis (13.7% vs 8.04%, respectively) and the odds of GDM diagnosis (OR = 1.87, 95%CI 1.47-2.37, I2 = 93.4%), with moderate certainty GRADE evidence. For adverse pregnancy outcomes, the lower glycemic criteria compared to the higher glycemic criteria were associated with a trend towards lower odds for gestational hypertension (OR 0.57, 95%CI 0.30-1.07), and neonatal hypoglycemia (OR 0.35, 95%CI 0.06-1.93, I2=88.2%). However, there were no differences between the glycemic thresholds for cesarean section (OR 0.94, 95% 0.78-1.13, I2= 0%). For all the outcomes, there was low certainty GRADE evidence.Conclusion: While the lower glycemic threshold results in close to double the prevalence of GDM and potentially higher workload for the healthcare system, this meta-analysis suggests that there may not be a difference in maternal and fetal outcomes when using lower vs higher thresholds in diagnosing GDM. However, the very low to low GRADE certainty suggests that further RCTs are needed to improve the certainty of evidence.