PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 7129742

Simultaneous inference in epidemiological studies.

D R Jones·L Rushton
View original source ↗Export metadata ↓

Abstract

Some difficulties encountered in using and interpreting significance tests in both exploratory and hypothesis testing epidemiological studies are discussed. Special consideration is given to the problems of simultaneous statistical inference--how are inferences to be modified when many significance tests are performed on the same set of data? Although some partial solutions are available, greater emphasis on estimation methods and less use of and reliance on significance testing in epidemiological studies is more appropriate.

Explore related subjects

EpidemiologyProbabilityStatistics as TopicJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

D R Jones, L Rushton. 1982. Simultaneous inference in epidemiological studies.. https://doi.org/10.1093/ije%2F11.3.276

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2026

Incidence patterns and genetic validation of primary glaucoma subtypes among 1 million adults in China and the UK.

BACKGROUND/AIMS: Primary open-angle glaucoma (POAG) and primary angle-closure glaucoma (PACG) are distinct diseases, yet many glaucoma cases in population-based datasets lack subtype specification. We assessed incidence patterns of glaucoma subtypes in China and the UK and used genetic evidence to infer the likely subtype composition of cases recorded as unspecified glaucoma. METHODS: Incident primary glaucoma was identified from linked inpatient records in the prospective China Kadoorie Biobank (CKB; n=512 504) and UK Biobank (UKB; n=492 329) studies. Cohort-specific phenotyping algorithms defined POAG, PACG and unspecified glaucoma. Adjusted incidence rates were estimated by direct standardisation. To support subtype inference, polygenic risk scores (PRSs) were constructed using ancestry-specific genome-wide association studies, including a new East Asian PACG meta-analysis, and tested for association with glaucoma phenotypes using multivariable logistic regression. RESULTS: Over 12 years of follow-up, 1658 primary glaucoma cases were identified in CKB and 7643 in UKB. Most (>68%) cases lacked subtype specification. Incidence increased with age and was twofold higher among women for PACG in both cohorts and for unspecified glaucoma in CKB. In UKB, POAG incidence was fivefold higher among Black than White participants, with a similar but attenuated pattern for unspecified glaucoma. PRS analyses indicated that unspecified glaucoma closely aligned with PACG in CKB but was more heterogeneous in UKB. CONCLUSION: Healthcare-recorded incidence patterns for POAG and PACG were consistent with established demographic risk factors, whereas unspecified glaucoma showed differences in subtype composition between populations. Integrating epidemiological and genetic evidence improves interpretation of glaucoma phenotypes when detailed clinical information is unavailable.

Epidemiology↗
PubMed2026

Neonatal outcomes following antenatal corticosteroid administration before planned caesarean birth at late preterm or early term (36-38 weeks' gestation): a retrospective cohort study.

OBJECTIVE: To investigate the benefits and harms associated with antenatal corticosteroids (ACS) prior to planned caesarean birth at late preterm or early term. DESIGN: Retrospective cohort study. SETTING: South Australia, Australia, 2005-2018. PATIENTS: Women with singleton pregnancies undergoing planned caesarean birth between 36+0 and 38+6 weeks' gestation. METHODS: We used routinely collected electronic maternity and neonatal data. Adjusted risk differences (aRD) with 95% CIs were calculated for neonatal and maternal outcomes. OUTCOME MEASURES: Neonatal outcomes included respiratory distress requiring oxygen, hypoglycaemia requiring glucose, neonatal unit (NNU) admission and prolonged NNU admission (≥48 hours). RESULTS: Among the cohort of 4049 women, the proportion receiving ACS was 54.8%, 48.4% and 5.6% at 36, 37 and 38 completed weeks' gestation, respectively. ACS administration was associated with a significant reduction in respiratory distress across all gestations, with the greatest reduction seen at 36 weeks' gestation (aRD -12.5%; 95% CI -21.4% to -3.6%). At 38 weeks' gestation, ACS administration was associated with an increased risk of neonatal hypoglycaemia (aRD 5.7%; 95% CI 1.1% to 10.2%), NNU admission (aRD 5.4%; 95% CI 0.1% to 10.7%) and prolonged NNU admission (aRD 4.6%; 95% CI 0.2% to 8.9%). CONCLUSIONS: ACS administration prior to late preterm and early term planned caesarean birth was associated with reduced risk of respiratory distress. The net benefit of ACS administration before planned caesarean birth remains unclear; however, there was evidence of potential neonatal harm when birth occurred after 37 weeks' gestation.

Epidemiology↗