Pub Date : 2025-04-16DOI: 10.1016/j.annepidem.2025.04.012
Raphael E. Cuomo
{"title":"Cannabis use disorder and mortality among patients with colon cancer","authors":"Raphael E. Cuomo","doi":"10.1016/j.annepidem.2025.04.012","DOIUrl":"10.1016/j.annepidem.2025.04.012","url":null,"abstract":"","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"106 ","pages":"Pages 8-10"},"PeriodicalIF":3.3,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-15DOI: 10.1016/j.annepidem.2025.04.011
Tasuku Okui
Purpose
This research examined trends in stroke mortality rates between metropolitan and non-metropolitan areas in Japan.
Methods
Data regarding stroke mortality for individuals aged 40–79 from 1999 to 2023 were sourced from Japan’s Vital Statistics. Age-standardized stroke mortality rate was computed annually for by sex and area. Moreover, an age-period-cohort analysis was performed, and the estimated stroke mortality rate was calculated for each age group, year, and cohort by sex and area.
Results
Throughout the years, the age-standardized mortality rate in non-metropolitan areas consistently exceeded that of metropolitan areas, while the difference in the age-standardized mortality rate diminished over years in men. Furthermore, the difference in estimated mortality rates between these two areas diminished and ultimately vanished from the cohorts born in the 1920s to those born in the late 1930s for men and from cohorts born in the 1920s to those born in the early 1950s for women. Conversely, starting from the cohorts born in the 1950s, the regional difference in the estimated mortality rates began to expand in men.
Conclusions
The results revealed the difference in age-standardized stroke mortality rates between the areas, and the difference in mortality rate varied depending on the birth cohort.
{"title":"Analysis of differences in stroke mortality rates between metropolitan and non-metropolitan areas in Japan from 1999 to 2023","authors":"Tasuku Okui","doi":"10.1016/j.annepidem.2025.04.011","DOIUrl":"10.1016/j.annepidem.2025.04.011","url":null,"abstract":"<div><h3>Purpose</h3><div>This research examined trends in stroke mortality rates between metropolitan and non-metropolitan areas in Japan.</div></div><div><h3>Methods</h3><div>Data regarding stroke mortality for individuals aged 40–79 from 1999 to 2023 were sourced from Japan’s Vital Statistics. Age-standardized stroke mortality rate was computed annually for by sex and area. Moreover, an age-period-cohort analysis was performed, and the estimated stroke mortality rate was calculated for each age group, year, and cohort by sex and area.</div></div><div><h3>Results</h3><div>Throughout the years, the age-standardized mortality rate in non-metropolitan areas consistently exceeded that of metropolitan areas, while the difference in the age-standardized mortality rate diminished over years in men. Furthermore, the difference in estimated mortality rates between these two areas diminished and ultimately vanished from the cohorts born in the 1920s to those born in the late 1930s for men and from cohorts born in the 1920s to those born in the early 1950s for women. Conversely, starting from the cohorts born in the 1950s, the regional difference in the estimated mortality rates began to expand in men.</div></div><div><h3>Conclusions</h3><div>The results revealed the difference in age-standardized stroke mortality rates between the areas, and the difference in mortality rate varied depending on the birth cohort.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"106 ","pages":"Pages 1-7"},"PeriodicalIF":3.3,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-14DOI: 10.1016/j.annepidem.2025.04.009
Tanner Nassau , Zachary Bouck , Seth L. Welles , Alison A. Evans , Shaun Hopkins , Paula Tookey , Dan Werb , Ayden I. Scheim
Purpose
Supervised consumption services (SCS) prevent fatal overdoses, but less is known about their contemporary impact on infectious disease risk.
Methods
We used quantitative bias analyses (QBA) to evaluate the association between SCS use and sharing of injection equipment while accounting for measurement error, selection bias, and unmeasured confounding, drawing on a cross-sectional sample of 695 people who inject drugs in Toronto, Canada surveyed from 2018 to 2020. We estimated the association between self-reported SCS use and equipment sharing using modified Poisson regression. Sensitivity analyses varied SCS use categories and definitions of sharing. QBA estimated the impact of misclassification, selection bias, and unmeasured confounding.
Results
Almost all participants (96.6 %) had recently used a needle and syringe program. Frequent SCS use (≥26 % of injections) was not associated with sharing equipment (adjusted prevalence ratio (aPR): 0.98; 95 % CI: 0.77–1.24). Results of sensitivity analyses did not meaningfully differ. In multiple bias analysis, the median bias-adjusted PR of 1.04 (range: 0.83–1.42) suggested no association between regular SCS use (≥75 % of injections) and syringe sharing.
Conclusions
In summary, SCS use was not associated with equipment sharing in a context of high needle and syringe program coverage. Misclassification, selection bias, and unmeasured confounding did not appear to impact the observed associations.
{"title":"A quantitative bias analysis of the association between supervised consumption service use and sharing of injection equipment among people who inject drugs in Toronto, Canada","authors":"Tanner Nassau , Zachary Bouck , Seth L. Welles , Alison A. Evans , Shaun Hopkins , Paula Tookey , Dan Werb , Ayden I. Scheim","doi":"10.1016/j.annepidem.2025.04.009","DOIUrl":"10.1016/j.annepidem.2025.04.009","url":null,"abstract":"<div><h3>Purpose</h3><div>Supervised consumption services (SCS) prevent fatal overdoses, but less is known about their contemporary impact on infectious disease risk.</div></div><div><h3>Methods</h3><div>We used quantitative bias analyses (QBA) to evaluate the association between SCS use and sharing of injection equipment while accounting for measurement error, selection bias, and unmeasured confounding, drawing on a cross-sectional sample of 695 people who inject drugs in Toronto, Canada surveyed from 2018 to 2020. We estimated the association between self-reported SCS use and equipment sharing using modified Poisson regression. Sensitivity analyses varied SCS use categories and definitions of sharing. QBA estimated the impact of misclassification, selection bias, and unmeasured confounding.</div></div><div><h3>Results</h3><div>Almost all participants (96.6 %) had recently used a needle and syringe program. Frequent SCS use (≥26 % of injections) was not associated with sharing equipment (adjusted prevalence ratio (aPR): 0.98; 95 % CI: 0.77–1.24). Results of sensitivity analyses did not meaningfully differ. In multiple bias analysis, the median bias-adjusted PR of 1.04 (range: 0.83–1.42) suggested no association between regular SCS use (≥75 % of injections) and syringe sharing.</div></div><div><h3>Conclusions</h3><div>In summary, SCS use was not associated with equipment sharing in a context of high needle and syringe program coverage. Misclassification, selection bias, and unmeasured confounding did not appear to impact the observed associations.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"106 ","pages":"Pages 11-16"},"PeriodicalIF":3.3,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-14DOI: 10.1016/j.annepidem.2025.04.010
Hongyan Chen , Xiaotian Chen , Qinyu Yao , Jibin Xin , Yi Zhang , Xiangyuan Huang , Dingmei Wang , Mengru Li , Tiansong Zhang , Taavi Tillmann , Weili Yan , Guoying Huang
Purpose
We performed a 2-sample Mendelian randomization (MR) using maternal MTHFR C677T as the genetic instrument to validate the causal association between maternal red blood cell (RBC) folate and offspring congenital heart disease (CHD) risk.
Methods
We obtained the genetic association for RBC folate through pooling data from 2 genome-wide association studies (the Trinity Student Study [n = 2229]) and Shanghai Preconception sub-cohort [n = 980]). We performed a meta-analysis of genetic studies to obtain the association for CHD (35 studies; 6141 CHDs and 14078 controls) and used the Wald ratio method for the 2-sample MR.
Results
Maternal MTHFR C677T variant was associated with lower RBC folate (-116 nmol/L per risk allele) and higher CHD risk (odds ratio [OR], 1.32 per allele; 95 % CI, 1.18–1.47). Per 100-nmol/L genetically determined higher RBC folate was associated with 21 % lower CHD risk (OR, 0.79 [0.70–0.90]). The association was evident in the Asian populations (0.72 [0.61–0.85]) and regions with low folate status (0.76 [0.65–0.88]) but not in the Caucasian populations (0.96 [0.89–1.04]) or regions with fortification (0.92 [0.79–1.06]).
Conclusions
Our findings support a causal role of maternal folate in offspring CHD risk, mainly confined to Asian populations and regions with low folate status.
{"title":"Appraising the causal relevance of maternal red blood cell folate and congenital heart disease in offspring: 2-sample Mendelian randomization","authors":"Hongyan Chen , Xiaotian Chen , Qinyu Yao , Jibin Xin , Yi Zhang , Xiangyuan Huang , Dingmei Wang , Mengru Li , Tiansong Zhang , Taavi Tillmann , Weili Yan , Guoying Huang","doi":"10.1016/j.annepidem.2025.04.010","DOIUrl":"10.1016/j.annepidem.2025.04.010","url":null,"abstract":"<div><h3>Purpose</h3><div>We performed a 2-sample Mendelian randomization (MR) using maternal <em>MTHFR</em> C677T as the genetic instrument to validate the causal association between maternal red blood cell (RBC) folate and offspring congenital heart disease (CHD) risk.</div></div><div><h3>Methods</h3><div>We obtained the genetic association for RBC folate through pooling data from 2 genome-wide association studies (the Trinity Student Study [<em>n</em> = 2229]) and Shanghai Preconception sub-cohort [<em>n</em> = 980]). We performed a meta-analysis of genetic studies to obtain the association for CHD (35 studies; 6141 CHDs and 14078 controls) and used the Wald ratio method for the 2-sample MR.</div></div><div><h3>Results</h3><div>Maternal <em>MTHFR</em> C677T variant was associated with lower RBC folate (-116 nmol/L per risk allele) and higher CHD risk (odds ratio [OR], 1.32 per allele; 95 % CI, 1.18–1.47). Per 100-nmol/L genetically determined higher RBC folate was associated with 21 % lower CHD risk (OR, 0.79 [0.70–0.90]). The association was evident in the Asian populations (0.72 [0.61–0.85]) and regions with low folate status (0.76 [0.65–0.88]) but not in the Caucasian populations (0.96 [0.89–1.04]) or regions with fortification (0.92 [0.79–1.06]).</div></div><div><h3>Conclusions</h3><div>Our findings support a causal role of maternal folate in offspring CHD risk, mainly confined to Asian populations and regions with low folate status.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"106 ","pages":"Pages 23-29"},"PeriodicalIF":3.3,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143855281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-12DOI: 10.1016/j.annepidem.2025.04.008
Stefan Kuhle , Mary M. Brown , Victoria M. Allen , Jillian Ashley-Martin , Linda Dodds , Christy G. Woolcott
Background
The objective of this study was to examine the association of birth by Caesarean section (CS) with female offspring's risk of CS delivery.
Methods
We used data from the 3G Multigenerational Cohort, which includes women whose births and their own subsequent pregnancies and deliveries were recorded in the population-based Nova Scotia Atlee Perinatal Database. The current analysis was limited to the women's first delivery (n = 23,605). Confounding variables were identified using a directed acyclic graph. The association between birth by CS and later CS delivery was examined with Poisson regression adjusted for confounding variables.
Results
Seventeen percent of women were born via CS, and 23 % delivered by CS. Compared to women born vaginally, women born by CS had an adjusted relative risk (RR) of 1.36 (95 % confidence interval [CI] 1.30, 1.43) for delivering by CS. Restricting the sample to women born to nulliparous mothers did not change the association (RR 1.35), while restriction to women born out of low-risk pregnancies weakened it slightly (RR 1.25).
Conclusions
Birth by CS is associated with a 36 % increased risk of women delivering their first child by CS. This increase is likely due to shared medical and socio-cultural factors.
{"title":"Birth by Caesarean section and female offspring’s risk of Caesarean section delivery","authors":"Stefan Kuhle , Mary M. Brown , Victoria M. Allen , Jillian Ashley-Martin , Linda Dodds , Christy G. Woolcott","doi":"10.1016/j.annepidem.2025.04.008","DOIUrl":"10.1016/j.annepidem.2025.04.008","url":null,"abstract":"<div><h3>Background</h3><div>The objective of this study was to examine the association of birth by Caesarean section (CS) with female offspring's risk of CS delivery.</div></div><div><h3>Methods</h3><div>We used data from the 3G Multigenerational Cohort, which includes women whose births and their own subsequent pregnancies and deliveries were recorded in the population-based Nova Scotia Atlee Perinatal Database. The current analysis was limited to the women's first delivery (n = 23,605). Confounding variables were identified using a directed acyclic graph. The association between birth by CS and later CS delivery was examined with Poisson regression adjusted for confounding variables.</div></div><div><h3>Results</h3><div>Seventeen percent of women were born via CS, and 23 % delivered by CS. Compared to women born vaginally, women born by CS had an adjusted relative risk (RR) of 1.36 (95 % confidence interval [CI] 1.30, 1.43) for delivering by CS. Restricting the sample to women born to nulliparous mothers did not change the association (RR 1.35), while restriction to women born out of low-risk pregnancies weakened it slightly (RR 1.25).</div></div><div><h3>Conclusions</h3><div>Birth by CS is associated with a 36 % increased risk of women delivering their first child by CS. This increase is likely due to shared medical and socio-cultural factors.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"106 ","pages":"Pages 17-22"},"PeriodicalIF":3.3,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143855275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-10DOI: 10.1016/j.annepidem.2025.04.003
David T. Zhu , Joseph Friedman , Suzanne Tamang , Joseph P. Gone
Purpose
To examine epidemiological trends in drug overdose deaths for non-Hispanic American Indian/Alaska Native (AIAN) individuals compared to White individuals.
Methods
We obtained data from the CDC WONDER database to examine crude drug overdose death rates per 100,000 among AIAN and White individuals from 1999–2022. Rates were further stratified by manner of death, sex, age, urbanization, Census region, state, and specific drug types.
Results
From 1999–2022, drug overdose death rates for AIAN individuals increased from 3.17 (95 % CI, 2.48–4.00) to 40.73 (95 % CI, 38.19–43.27), representing a 12-fold increase; in contrast, rates rose by five-fold for White individuals. In 2022, methamphetamine was the leading driver of overdose death rates for AIAN individuals, at 31.39 (95 % CI, 29.16–33.62) per 100,000, followed by fentanyl, at 22.35 (95 % CI, 20.46–24.23) per 100,000. Geographical variations were notable, with the highest rates for AIAN individuals in large central metropolitan regions (53.54 per 100,000) and the Midwest (50.31 per 100,000). Rates were higher for AIAN than White individuals in 20 of the 21 states (95.2 %) included in our analysis.
Conclusions
AIAN individuals are disproportionately impacted by the overdose crisis. Further efforts are needed to expand access to harm reduction-informed, culturally-competent health education, addiction treatment, and social services in the AIAN population.
{"title":"Drug overdose mortality rates among non-Hispanic American Indian/Alaska Native individuals, 1999–2022","authors":"David T. Zhu , Joseph Friedman , Suzanne Tamang , Joseph P. Gone","doi":"10.1016/j.annepidem.2025.04.003","DOIUrl":"10.1016/j.annepidem.2025.04.003","url":null,"abstract":"<div><h3>Purpose</h3><div>To examine epidemiological trends in drug overdose deaths for non-Hispanic American Indian/Alaska Native (AIAN) individuals compared to White individuals.</div></div><div><h3>Methods</h3><div>We obtained data from the CDC WONDER database to examine crude drug overdose death rates per 100,000 among AIAN and White individuals from 1999–2022. Rates were further stratified by manner of death, sex, age, urbanization, Census region, state, and specific drug types.</div></div><div><h3>Results</h3><div>From 1999–2022, drug overdose death rates for AIAN individuals increased from 3.17 (95 % CI, 2.48–4.00) to 40.73 (95 % CI, 38.19–43.27), representing a 12-fold increase; in contrast, rates rose by five-fold for White individuals. In 2022, methamphetamine was the leading driver of overdose death rates for AIAN individuals, at 31.39 (95 % CI, 29.16–33.62) per 100,000, followed by fentanyl, at 22.35 (95 % CI, 20.46–24.23) per 100,000. Geographical variations were notable, with the highest rates for AIAN individuals in large central metropolitan regions (53.54 per 100,000) and the Midwest (50.31 per 100,000). Rates were higher for AIAN than White individuals in 20 of the 21 states (95.2 %) included in our analysis.</div></div><div><h3>Conclusions</h3><div>AIAN individuals are disproportionately impacted by the overdose crisis. Further efforts are needed to expand access to harm reduction-informed, culturally-competent health education, addiction treatment, and social services in the AIAN population.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"105 ","pages":"Pages 80-88"},"PeriodicalIF":3.3,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143833278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-09DOI: 10.1016/j.annepidem.2025.04.006
Miranda Kit-Yi Wong , Marina Mendonça , Nicole Tsalacopoulos , Peter Bartmann , Brian A. Darlow , L. John Horwood , Sarah L. Harris , Eero Kajantie , Chiara Nosarti , Marit S. Indredavik , Kari Anne I. Evensen , Katri Räikkönen , Kati Heinonen , Sylvia van der Pal , Dieter Wolke
Objectives
To assess whether there are differences in fertility between adults born very preterm or at very low birth weight (VP/VLBW) with term-born controls, whether the association of VP/VLBW with fertility differs by sex, and which individual factors are associated with fertility among VP/VLBW adults.
Study design
Prospective longitudinal cohorts with fertility assessed in VP/VLBW and term-born adults were identified from two international consortia: Research on European Children and Adults Born Preterm (RECAP-Preterm), and Adults Born Preterm International Collaboration (APIC). Individual participant data (IPD) on neonatal, medical, sociodemographic, and fertility variables were collected and analyzed using a one-stage approach.
Results
Seven cohorts with 931 VP/VLBW and 1363 term-born young adults (mean ages at assessment ranged from 23 to 30 years) were included. VP/VLBW and term-born young adults did not significantly differ in fertility (i.e., having children) (OR 1.48, 95 % CI 0.99–2.21). No moderation effect of sex could be confirmed (OR 0.87, 95 % CI 0.53–1.42). Among VP/VLBW young adults, higher fertility was significantly associated with female sex, higher age at assessment, being married/cohabiting, the absence of childhood neurosensory impairment, and low levels of maternal and own education.
Conclusions
VP/VLBW is not associated with lower fertility in young adults. Sex does not moderate this association. In addition to childhood neurosensory impairment, mainly sociodemographic factors (partnering, maternal and own education) are associated with fertility in VP/VLBW young adults. The evidence is limited so far to the early reproductive window in the 20 s, further follow-up into established adulthood will be required for definite answers on fertility after VP/VLBW birth.
{"title":"Fertility of young adults born very preterm/very low birth weight: An individual participant data meta-analysis","authors":"Miranda Kit-Yi Wong , Marina Mendonça , Nicole Tsalacopoulos , Peter Bartmann , Brian A. Darlow , L. John Horwood , Sarah L. Harris , Eero Kajantie , Chiara Nosarti , Marit S. Indredavik , Kari Anne I. Evensen , Katri Räikkönen , Kati Heinonen , Sylvia van der Pal , Dieter Wolke","doi":"10.1016/j.annepidem.2025.04.006","DOIUrl":"10.1016/j.annepidem.2025.04.006","url":null,"abstract":"<div><h3>Objectives</h3><div>To assess whether there are differences in fertility between adults born very preterm or at very low birth weight (VP/VLBW) with term-born controls, whether the association of VP/VLBW with fertility differs by sex, and which individual factors are associated with fertility among VP/VLBW adults.</div></div><div><h3>Study design</h3><div>Prospective longitudinal cohorts with fertility assessed in VP/VLBW and term-born adults were identified from two international consortia: Research on European Children and Adults Born Preterm (RECAP-Preterm), and Adults Born Preterm International Collaboration (APIC). Individual participant data (IPD) on neonatal, medical, sociodemographic, and fertility variables were collected and analyzed using a one-stage approach.</div></div><div><h3>Results</h3><div>Seven cohorts with 931 VP/VLBW and 1363 term-born young adults (mean ages at assessment ranged from 23 to 30 years) were included. VP/VLBW and term-born young adults did not significantly differ in fertility (i.e., having children) (OR 1.48, 95 % CI 0.99–2.21). No moderation effect of sex could be confirmed (OR 0.87, 95 % CI 0.53–1.42). Among VP/VLBW young adults, higher fertility was significantly associated with female sex, higher age at assessment, being married/cohabiting, the absence of childhood neurosensory impairment, and low levels of maternal and own education.</div></div><div><h3>Conclusions</h3><div>VP/VLBW is not associated with lower fertility in young adults. Sex does not moderate this association. In addition to childhood neurosensory impairment, mainly sociodemographic factors (partnering, maternal and own education) are associated with fertility in VP/VLBW young adults. The evidence is limited so far to the early reproductive window in the 20 s, further follow-up into established adulthood will be required for definite answers on fertility after VP/VLBW birth.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"106 ","pages":"Pages 30-39"},"PeriodicalIF":3.3,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143855280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-09DOI: 10.1016/j.annepidem.2025.03.011
Gebresilasea Gendisha Ukke , Jacqueline A. Boyle , Rajshree Thapa , Kristie Cocotis , Carli Leishman , Christopher Gilfillan , Ahmed Reja , Wubet Worku Takle , Siew Lim
Purpose
To assess the completion of the type 2 diabetes, heart disease and stroke prevention program (the Life!) among women with a history of gestational diabetes mellitus (GDM) according to participants’ characteristics.
Methods
Data from women with a history of GDM enrolled in the Life! program in Victoria, Australia, between 2014 and 2022 were analysed. Completion rates were assessed using the PROGRESS-Plus (Place of residence, Race/ethnicity/culture/language, Occupation, Gender/sex, Religion, Education, Socioeconomic status, and Social capital Plus age and smoking) framework. Multivariable logistic regression model was fitted.
Results
A total of 2399 women with a history of GDM were enrolled in the program, of which 55 % completed it. Characteristics associated with higher completion rates included being from metropolitan areas (AOR = 1.52, 95 % CI: 1.14–2.01) compared with being from regional areas, having a body mass index in a normal range (AOR = 1.50, 95 % CI: 1.06–2.15) compared with having overweight or obesity, having middle income (AOR = 1.41, 95 % CI: 1.01–1.98) compared with having low- or high-income, and enrolment after 2019 (AOR = 2.3, 95 % CI:1.80–3.06) compared with enrolment in 2019 or earlier. Conversely, having a South or Central Asian background (AOR = 0.65, 95 % CI: 0.46–0.92) is associated with a lower completion rate compared with being from Australia.
Conclusion
Characteristics associated with lower completion rates among women with prior GDM in a cardiometabolic risk reduction program included residing in rural and remote areas, having an elevated BMI (in the overweight or obesity range), low or high income, enrolment in 2019 or earlier and being of South or Central Asian background.
{"title":"An equity audit on program completion among women with a history of gestational diabetes in a state-funded diabetes and cardiovascular risk reduction program","authors":"Gebresilasea Gendisha Ukke , Jacqueline A. Boyle , Rajshree Thapa , Kristie Cocotis , Carli Leishman , Christopher Gilfillan , Ahmed Reja , Wubet Worku Takle , Siew Lim","doi":"10.1016/j.annepidem.2025.03.011","DOIUrl":"10.1016/j.annepidem.2025.03.011","url":null,"abstract":"<div><h3>Purpose</h3><div>To assess the completion of the type 2 diabetes, heart disease and stroke prevention program (the <em>Life!)</em> among women with a history of gestational diabetes mellitus (GDM) according to participants’ characteristics.</div></div><div><h3>Methods</h3><div>Data from women with a history of GDM enrolled in the <em>Life!</em> program in Victoria, Australia, between 2014 and 2022 were analysed. Completion rates were assessed using the PROGRESS-Plus (Place of residence, Race/ethnicity/culture/language, Occupation, Gender/sex, Religion, Education, Socioeconomic status, and Social capital Plus age and smoking) framework. Multivariable logistic regression model was fitted.</div></div><div><h3>Results</h3><div>A total of 2399 women with a history of GDM were enrolled in the program, of which 55 % completed it. Characteristics associated with higher completion rates included being from metropolitan areas (AOR = 1.52, 95 % CI: 1.14–2.01) compared with being from regional areas, having a body mass index in a normal range (AOR = 1.50, 95 % CI: 1.06–2.15) compared with having overweight or obesity, having middle income (AOR = 1.41, 95 % CI: 1.01–1.98) compared with having low- or high-income, and enrolment after 2019 (AOR = 2.3, 95 % CI:1.80–3.06) compared with enrolment in 2019 or earlier. Conversely, having a South or Central Asian background (AOR = 0.65, 95 % CI: 0.46–0.92) is associated with a lower completion rate compared with being from Australia.</div></div><div><h3>Conclusion</h3><div>Characteristics associated with lower completion rates among women with prior GDM in a cardiometabolic risk reduction program included residing in rural and remote areas, having an elevated BMI (in the overweight or obesity range), low or high income, enrolment in 2019 or earlier and being of South or Central Asian background.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"105 ","pages":"Pages 59-65"},"PeriodicalIF":3.3,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143823217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-08DOI: 10.1016/j.annepidem.2025.04.002
Michael C. Sachs , Johan Sebastian Ohlendorff , Adam Brand , Arvid Sjölander , Erin E. Gabriel
Regression standardization is a useful tool for performing causal inference in epidemiology. We have updated the R package stdReg to stdReg2 to be more user-friendly, flexible, and to include two new functionalities, a generalized linear model-based double-robust method and regression standardization for the restricted mean survival. The old package stdReg will continue to function, but new users may find the upgraded version easier to use. We highlight the improvements and implementation in this article. Keywords: average treatment effect, causal inference, regression standardization, statistical software.
{"title":"The next generation of regression standardization with the R package stdReg2","authors":"Michael C. Sachs , Johan Sebastian Ohlendorff , Adam Brand , Arvid Sjölander , Erin E. Gabriel","doi":"10.1016/j.annepidem.2025.04.002","DOIUrl":"10.1016/j.annepidem.2025.04.002","url":null,"abstract":"<div><div>Regression standardization is a useful tool for performing causal inference in epidemiology. We have updated the R package stdReg to stdReg2 to be more user-friendly, flexible, and to include two new functionalities, a generalized linear model-based double-robust method and regression standardization for the restricted mean survival. The old package stdReg will continue to function, but new users may find the upgraded version easier to use. We highlight the improvements and implementation in this article. Keywords: average treatment effect, causal inference, regression standardization, statistical software.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"105 ","pages":"Pages 66-74"},"PeriodicalIF":3.3,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143828414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-08DOI: 10.1016/j.annepidem.2025.04.004
Derek K. Ng , Ankur Patel , George J. Schwartz , Jesse C. Seegmiller , Bradley A. Warady , Susan L. Furth , Christopher Cox , for the CKiD study investigators
Purpose
Clinical management of pediatric chronic kidney disease requires estimation of glomerular filtration rate (eGFR). Currently, eGFR is determined by two endogenous markers measured in blood: serum creatine (SCr) and cystatin C (CysC). Machine learning methods show promise to potentially improve eGFR, but it is unclear if they can outperform regression-based approaches under clinical constraining requiring real time measurement and only two predictors. We constructed a neural network for eGFR (NNeGFR) and compared it to the clinical standard Under 25 (U25eGFR) equations using the same data for training and validation.
Methods
The U25eGFR data comprised 1683 training and 843 validation observations that included iohexol measured GFR (mGFR), SCr and CysC. Sex-stratified feed forward NNs included the same predictors as U25eGFR (i.e., age, height/SCr, CysC) with additional nonlinear transformations. Performance was evaluated by bias (for calibration), proportions within 10 % and 30 % of mGFR (P10 and P30, for accuracy), root mean square error (RMSE, for precision) and R2 (for discrimination).
Results
NNeGFR performed comparably to the U25eGFR equations on all metrics. Biases were minimal, slightly favoring U25eGFR. NNeGFR and U25eGFR had similar P10 (>37 %), P30 (>86 %) and RMSE.
Conclusions
NNeGFR performed as well as established equations to estimate GFR. Without additional biomarkers related to kidney function, which are not currently clinically available in real time, NN methods are unlikely to substantially outperform regression derived GFR estimating equations. Implications for translation of these advanced epidemiologic methods to clinical practice are discussed.
{"title":"A comparison of neural networks and regression-based approaches for estimating kidney function in pediatric chronic kidney disease: Practical predictive epidemiology for clinical management of a progressive disease","authors":"Derek K. Ng , Ankur Patel , George J. Schwartz , Jesse C. Seegmiller , Bradley A. Warady , Susan L. Furth , Christopher Cox , for the CKiD study investigators","doi":"10.1016/j.annepidem.2025.04.004","DOIUrl":"10.1016/j.annepidem.2025.04.004","url":null,"abstract":"<div><h3>Purpose</h3><div>Clinical management of pediatric chronic kidney disease requires estimation of glomerular filtration rate (eGFR). Currently, eGFR is determined by two endogenous markers measured in blood: serum creatine (SCr) and cystatin C (CysC). Machine learning methods show promise to potentially improve eGFR, but it is unclear if they can outperform regression-based approaches under clinical constraining requiring real time measurement and only two predictors. We constructed a neural network for eGFR (NNeGFR) and compared it to the clinical standard Under 25 (U25eGFR) equations using the same data for training and validation.</div></div><div><h3>Methods</h3><div>The U25eGFR data comprised 1683 training and 843 validation observations that included iohexol measured GFR (mGFR), SCr and CysC. Sex-stratified feed forward NNs included the same predictors as U25eGFR (i.e., age, height/SCr, CysC) with additional nonlinear transformations. Performance was evaluated by bias (for calibration), proportions within 10 % and 30 % of mGFR (P<sub>10</sub> and P<sub>30</sub>, for accuracy), root mean square error (RMSE, for precision) and R<sup>2</sup> (for discrimination).</div></div><div><h3>Results</h3><div>NNeGFR performed comparably to the U25eGFR equations on all metrics. Biases were minimal, slightly favoring U25eGFR. NNeGFR and U25eGFR had similar P<sub>10</sub> (>37 %), P<sub>30</sub> (>86 %) and RMSE.</div></div><div><h3>Conclusions</h3><div>NNeGFR performed as well as established equations to estimate GFR. Without additional biomarkers related to kidney function, which are not currently clinically available in real time, NN methods are unlikely to substantially outperform regression derived GFR estimating equations. Implications for translation of these advanced epidemiologic methods to clinical practice are discussed.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"105 ","pages":"Pages 75-79"},"PeriodicalIF":3.3,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143833277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号