Review of inverse probability weighting for dealing with missing data

Association of solid fuel use with cognitive function and the modifying role of lifestyle: A nationwide cohort study in China

BACKGROUND

As opposed to a healthy lifestyle, indoor air pollution from solid fuel use may be harmful for cognitive function. However, the extent to which lifestyle modifies the association between solid fuel use and cognitive function remains unknown.

METHODS

A total of 21,008 individuals aged 16 to 92 were enrolled in 2010 and followed up to 2014 in the China Family Panel Studies (CFPS). Cognitive function was assessed using standardized math and word tests in two waves. Solid fuel use at baseline was assessed by self-reporting of firewood, straw, or coal used for cooking. Lifestyle profile was classified into two groups (favorable vs. unfavorable) based on five modifiable lifestyle factors including alcohol drinking, smoking, body mass index, diet, and physical activity. Linear mixed-effects models were employed to assess the association of solid fuel use and lifestyle with cognitive function. The effect modification of lifestyle was analyzed.

RESULTS

A total of 49.7% of the study population used solid fuels for cooking and 17.4% had a favorable lifestyle. Solid fuel use was associated with a significant decrease in cognitive function (β = -0.29, 95% CI: -0.39, -0.19 for math test; β = -0.62, 95% CI: -0.84, -0.41 for word test). Lifestyle significantly modified this association (p-interaction: 0.006 for math test; 0.016 for word test), with the corresponding association being less pronounced among participants adhering to a favorable lifestyle compared to those with an unfavorable lifestyle.

CONCLUSION

A favorable lifestyle may attenuate the adverse association between solid fuel use and cognitive function. Adopting a favorable lifestyle has the potential to mitigate the adverse neurological effects due to indoor air pollution.

Is social disadvantage a chronic stressor? Socioeconomic position and HPA axis activity among older adults living in England

INTRODUCTION

Living in socioeconomic disadvantage has been conceptualised as a chronic stressor, although this contradicts evidence from studies using hair cortisol and cortisone as a measure of hypothalamus-pituitary-adrenal (HPA)1 axis activity. These studies used complete case analyses, ignoring the impact of missing data for inference, despite the high proportion of missing biomarker data. The methodological limitations of studies investigating the association between socioeconomic position (SEP)2 defined as education, wealth, and social class and hair cortisol and cortisone are considered in this study by comparing three common methods to deal with missing data: (1) Complete Case Analysis (CCA),3 (2) Inverse Probability Weighting (IPW) 4and (3) weighted Multiple Imputation (MI).5 This study examines if socioeconomic disadvantage is associated with higher levels of HPA axis activity as measured by hair cortisol and cortisone among older adults using three approaches for compensating for missing data.

METHOD

Cortisol and cortisone levels in hair samples from 4573 participants in the 6th wave (2012-2013) of the English Longitudinal Study of Ageing (ELSA)6 were examined, in relation to education, wealth, and social class. We compared linear regression models with CCA, weighted and multiple imputed weighted linear regression models.

RESULTS

Social groups with certain characteristics (i.e., ethnic minorities, in routine and manual occupations, physically inactive, with poorer health, and smokers) were less likely to have hair cortisol and hair cortisone data compared to the most advantaged groups. We found a consistent pattern of higher levels of hair cortisol and cortisone among the most socioeconomically disadvantaged groups compared to the most advantaged groups. Complete case approaches to missing data underestimated the levels of hair cortisol in education and social class and the levels of hair cortisone in education, wealth, and social class in the most disadvantaged groups.

CONCLUSION

This study demonstrates that social disadvantage as measured by disadvantaged SEP is associated with increased HPA axis activity. The conceptualisation of social disadvantage as a chronic stressor may be valid and previous studies reporting no associations between SEP and hair cortisol may be biased due to their lack of consideration of missing data cases which showed the underrepresentation of disadvantaged social groups in the analyses. Future analyses using biosocial data may need to consider and adjust for missing data.

Characteristics and predictors of Long Covid in children: a 3-year prospective cohort study

Background

Children can develop Long Covid, however long term outcomes and their predictors are poorly described in these patients. The primary aim is to describe characteristics and predictors of Long Covid in children assessed in-clinics up to 36 months post-SARS-CoV-2 infection, as well as investigate the role of vaccines in preventing Long Covid, risk of reinfections and development of autoimmune diseases.

Methods

Children aged 0-18 years old with confirmed SARS-CoV-2 infection were invited for a prospective follow-up assessment at a peadiatric post-covid clinic in Rome, Italy, at serial intervals (3-, 6-, 12-, 18-, 24- and 36-months post-infection onset, between 01/02/2020 and 28/02/2024). Long Covid was defined as persistence of otherwise unexplained symptoms for at least three months after initial infection.

Findings

1319 patients were initially included, 1296 reached the 3 months follow-up or more. Of the patients who underwent multiple follow-ups, 23.2% (301), 169 (13.2%), 89 (7.9%), 67 (6.1%), 47 (7.1%) were diagnosed with Long Covid at 3-6-12-18-24 months, respectively For the primary outcome of Long Covid at three months, age >12 years (P < 0.001, OR 11.33, 95% CI 4.2; 15.15), comorbidities (P = 0.008, OR 1.83, 95% CI 1.06; 2.44), being infected with original variants (P < 0.001, OR 4.77, 95% CI 2.46; 14.47), female sex (P < 0.001, OR 1.62, 95% CI 1.02; 1.89) were statistically significant risk factors. Age >12 years (P = 0.002, OR 9.37, 95% CI 1.58; 8.64), and infection with original (P = 0.012, OR 3.52, 95% CI 1.32; 8.64) and alfa (P < 0.001, OR 4.09, 95% CI 2.01; 8.3) SARS-CoV-2 variants remained statistically significant risk factors for Long Covid duration for at least 18 months. Vaccination was associated with a lower risk of long covid at 3, 6 and 12 months for older children and a lower risk of reinfections. Being infected with the original SARS-CoV-2 variant was associated with a higher risk of new-onset autoimmune diseases ((P = 0.035, 95% CI 1.12; 2.4). One patient was diagnosed with Long Covid after a re-infection.

Interpretation

This is the longest follow-up study of children with SARS-CoV-2 infection, showing a significant and long-lasting burden of Long Covid in the pediatric population. Our findings highlight the urgent need of investing in pediatric Long Covid in order to find effective diagnostic and therapeutic approaches, as well can inform preventive strategies in case of future pandemics.

Funding

This study has been funde by Pfizer non-competitive grant, granted to DB (#65925795).

Hearing Loss and Cardiovascular Disease Risk Profiles: Data from the Hispanic Community Health Study/Study of Latinos

BACKGROUND

 Individual cardiovascular disease (CVD) risk factors (RFs) have been associated with hearing loss (HL). The relationship to aggregate risk is poorly understood and has not been explored in the Hispanic/Latino population.

PURPOSE

 The aim of this study was to characterize the association between aggregate CVD RF burden and hearing among Hispanics/Latinos.

RESEARCH DESIGN

 Cross-sectional examination.

STUDY SAMPLE

 Participants (18-74 years; n = 12,766) in the Hispanic Community Health Study/Study of Latinos.

DATA COLLECTION AND ANALYSIS

 Thresholds (0.5-8 kHz) were obtained, and HL was defined dichotomously as pure-tone average (PTA0.5,1, 2,4) > 25 dB HL. Optimal CVD risk burden was defined as follows: systolic blood pressure (SBP) < 120 mm Hg and diastolic blood pressure (DBP) < 80 mm Hg; total cholesterol < 180 mg/dL; nonsmoking; and no diabetes. Major CVD RFs were diabetes, currently smoking, SBP >160 or DBP > 100 mm Hg (or antihypertensives), and total cholesterol > 240 mg/dL (or statins). Thresholds were estimated by age (18-44 and ≥45 years) and sex using linear regression. The association between CVD risk burden and HL was assessed using multivariable logistic regression. Models were adjusted for age, sex, Hispanic/Latino background, center, education, income, alcohol use, body mass index, and noise exposure.

RESULTS

 In the target population, 53.03% were female and 18.81% and 8.52% had all RFs optimal and ≥2 major RFs, respectively. Elevated BP (SBP 120-139 mm Hg or DBP 80-89 mm Hg) was associated with HL in females < 45 years (odds ratio [OR], 2.18; 95% confidence interval [CI], 1.14-4.16). Diabetes (OR, 1.37; 95% CI, 1.01-1.84) and tobacco smoking (OR, 1.44; 95% CI, 1.03-2.01) were associated with HL in females ≥ 45 years. The odds of HL were higher for females ≥ 45 years with ≥2 RFs versus those with all RFs optimal (OR, 1.99; 95% CI, 1.12-3.53). Elevated BP (SBP 140-159 mm Hg or DBP 90-99 mm Hg), but not aggregate risk burden, was associated with HL in males ≥ 45 years (OR, 1.49; 95% CI, 1.02-2.19). No relationships with major CVD RFs were significant in males < 45 years.

CONCLUSIONS

 HL is associated with elevated BP in females < 45 years, with diabetes and hypertension in males ≥ 45 years, and with diabetes, smoking, and having ≥2 major CVD RFs in females ≥ 45 years. Future studies are needed to examine if these factors are associated with incident HL.

Financial Hardship Among Patients With Early-Stage Colorectal Cancer

Importance

The degree of cancer patients' financial hardship is dynamic and can change over time.

Objective

To assess longitudinal changes in financial hardship among patients with early-stage colorectal cancer.

Design, Setting, and Participants

In this prospective longitudinal cohort study, English-speaking adult patients with a new diagnosis of stage I to III colorectal cancer being treated with curative intent at National Cancer Institute (NCI) Community Oncology Research Program (NCORP) practices between May 2018 and July 2020 and who had not started chemotherapy and/or radiation were included. Data analysis was conducted from March to December 2023.

Main Outcomes and Measures

Patients completed surveys at baseline as well as at 3, 6, 12, and 24 months after enrollment. Cost-related care nonadherence and material hardship, as adopted by Medical Expenditure Panel Survey, were measured. Factors associated with financial hardship were assessed using longitudinal multivariable logistic regression models with time interaction.

Results

A total of 451 patients completed baseline questions, with 217 (48.1%) completing the 24-month follow-up. Mean (SD) age was 61.0 (12.0) years (210 [46.6%] female; 33 [7.3%] Black, 380 [84.3%] White, and 33 [7.3%] American Indian or Alaska Native, Asian, multiracial, or Native Hawaiian or Other Pacific Islander individuals or those who did not report race or who had unknown race). Among 217 patients with data at baseline and 24 months, 19 (8.8%) reported cost-related care nonadherence at baseline vs 20 (9.2%) at 24 months (P = .84), and 125 (57.6%) reported material hardship at baseline vs 76 (35.0%) at 24 months (P < .001). In multivariable analysis, lower financial worry (odds ratio [OR], 0.90; 95% CI, 0.87-0.93), higher education (OR, 0.34; 95% CI, 0.15-0.77), and older age (OR, 0.94; 95% CI, 0.91-0.98) were associated with lower nonadherence. Receipt of chemotherapy was associated with higher material hardship (OR, 2.68; 95% CI, 1.15-6.29), while lower financial worry was associated with lower material hardship (OR, 0.83; 95% CI, 0.80-0.96). Over 24 months, female sex was associated with lower nonadherence (OR, 0.90; 95% CI, 0.85-0.96), while higher education was associated with higher nonadherence (OR, 1.09; 95% CI, 1.03-1.17). Being employed was associated with lower material hardship (OR, 0.85; 95% CI, 0.78-0.93), while receipt of care at safety-net hospitals was associated with higher hardship (OR, 1.09; 95% CI, 1.01-1.17).

Conclusions and Relevance

In patients with early-stage colorectal cancer, material hardship was more common than cost-related cancer care nonadherence and decreased over time, while nonadherence remained unchanged. Early and longitudinal financial screening and referral to intervention are recommended to mitigate financial hardship.

Parental Binge Eating and Child Binge Eating and Weight-Control Behaviors: Cross-Sectional and Longitudinal Findings From the EAT 2010-2018 Study

OBJECTIVE

In cross-sectional and retrospective research, parental binge eating is associated with their children's eating psychopathology. The current study extended the evidence by cross-sectionally and longitudinally examining the relation between parental binge eating and binge eating and weight-control behaviors in the next generation of their adolescent children and young adult children in a population-based sample.

METHODS

Adolescents (Time 1: M = 14.5, SD = 2.0 years) (n = 2367), followed into adulthood (Time 2: M = 22.1, SD = 2.0 years), and their parents (n = 3664) were enrolled in EAT 2010-2018 and Project F-EAT 2010. The current study examined parental binge eating, and child binge eating and weight-control behaviors. Adjusted models covaried for child gender, age, and race/ethnicity.

RESULTS

Approximately 7% of adolescents at Time 1 had at least one parent who reported binge eating with no differences by child's age, gender, or race/ethnicity. Having at least one parent experiencing binge eating at Time 1 (vs. not) was associated cross-sectionally with adolescent children's use of extreme weight-control behaviors (9.6% vs. 4.8%; Risk Difference [RD] = 4.9%) and associated longitudinally with binge eating during young adulthood (21.1% vs. 11.6%; RD = 9.5%). Other associations did not reach statistical significance.

CONCLUSIONS

Children of parents with binge eating appear to have elevated risk of extreme weight-control behaviors during adolescence and binge eating in young adulthood. Clinicians should assess whether eating psychopathology extends to other family members, and offer additional support to parents with binge eating. Further research is needed to identify risk factors in the children of parents with binge eating and to assess strategies for prevention.

A comprehensive assessment of care competence and maternal experience of first antenatal care visits in Mexico: Insights from the baseline survey of an observational cohort study

BACKGROUND

Comprehensive antenatal care (ANC) must prioritize competent, evidence-based medical attention to ensure a positive experience and value for its users. Unfortunately, there is scarce evidence of implementing this holistic approach to ANC in low- and middle-income countries, leading to gaps in quality and accountability. This study assessed care competence, women's experiences during the first ANC visit, and the factors associated with these care attributes.

METHODS AND FINDINGS

The study analyzed cross-sectional baseline data from the maternal eCohort study conducted in Mexico from August to December 2023. The study adapted the Quality Evidence for Health System Transformation (QuEST) network questionnaires to the Mexican context and validated them through expert group and cognitive interviews with women. Pregnant women aged 18 to 49 who had their first ANC visit with a family physician were enrolled in 48 primary clinics of the Instituto Mexicano del Seguro Social across 8 states. Care competence and women's experiences with care were the primary outcomes. The statistical analysis comprised descriptive statistics, multivariable linear and Poisson regressions. A total of 1,390 pregnant women were included in the study. During their first ANC visit, women received only 67.7% of necessary clinical actions on average, and 52% rated their ANC experience as fair or poor. Women with previous pregnancies (adjusted regression coefficient [aCoef.] -3.55; (95% confidence intervals [95% CIs]): -4.88, -2.22, p < 0.001), at risk of depression (aCoef. -3.02; 95% CIs: -5.61, -0.43, p = 0.023), those with warning signs (aCoef. -2.84; 95% CIs: -4.65, -1.03, p = 0.003), common pregnancy discomforts (aCoef. -1.91; 95% CIs: -3.81, -0.02, p = 0.048), or those who had a visit duration of less than 20 minutes (<15 minutes: aCoef. -7.58; 95% CIs: -10.21, -4.95, p < 0.001 and 15 to 19 minutes: aCoef. -2.73; 95% CIs: -4.79, -0.67, p = 0.010) and received ANC in the West and Southeast regions (aCoef. -5.15; 95% CIs: -7.64, -2.66, p < 0.001 and aCoef. -5.33; 95% CIs: -7.85, -2.82, p < 0.001, respectively) had a higher probability of experiencing poorer care competence. Higher care competence (adjusted prevalence ratio [aPR] 1.004; 95% CIs:1.002, 1.005, p < 0.001) and receiving care in a small clinic (aPR 1.19; 95% CIs: 1.06, 1.34, p = 0.003) compared to a medium-sized clinic were associated with a better first ANC visit experience, while common pregnancy discomforts (aPR 0.94; 95% CIs: 0.89, 0.98, p = 0.005) and shorter visit length (aPR 0.94; 95% CIs: 0.88, 0.99, p = 0.039) were associated with lower women's experience. The primary limitation of the study is that participants' responses may be influenced by social desirability bias, leading them to provide socially acceptable responses.

CONCLUSIONS

We found important gaps in adherence to ANC standards and that care competence during the first ANC visit is an important predictor of positive user experience. To inform quality improvement efforts, IMSS should institutionalize the routine monitoring of ANC competencies and ANC user experience. This will help identify poorly performing facilities and providers and address gaps in the provision of evidence-based and women-centered care.

Fear of COVID-19 predicts increases in anxiety, depressive symptoms, health anxiety, psychosocial distress, and loneliness: Findings from a prospective two-year follow-up study

The role of fear of COVID-19 in prospectively predicting changes in psychopathological symptoms during the COVID-19 pandemic remains unclear. The present data were obtained from a longitudinal non-probability sample in Germany, initially assessed at the beginning of the COVID-19 pandemic in Germany (April-May 2020) and reassessed after two years (n = 846; 83% female; mean age: 44.59 years, SD = 12.32; response rate: 19.5%). Multiple linear regressions were used to examine associations of fear of COVID-19 at baseline with depressive symptoms, anxiety, health anxiety, psychosocial distress, and loneliness controlling for (a) the respective symptom measure, and (b) all psychopathological symptoms at baseline. The data were weighted to minimize attrition and representativeness biases. Overall, loneliness decreased from the beginning of the COVID-19 pandemic until the two-year follow-up, whereas all other symptoms did not change. Fear of COVID-19 at the beginning of the pandemic predicted an increase in anxiety symptoms, health anxiety, psychosocial distress, and loneliness two years later. In addition, fear of COVID-19 predicted higher health anxiety, depressive symptoms, psychosocial distress, and loneliness, but not anxiety symptoms when controlling for all baseline symptom measures at once. Fear of COVID-19 seems to play a central role in predicting negative mental health outcomes, emphasizing the necessity of indicated prevention and intervention to decrease worry and manage anxiety, thereby reducing the negative impact on mental health caused by fear during future pandemics.

Effect of Starting Dialysis Versus Continuing Medical Management on Survival and Home Time in Older Adults With Kidney Failure : A Target Trial Emulation Study

BACKGROUND

For older adults with kidney failure who are not referred for transplant, medical management is an alternative to dialysis.

OBJECTIVE

To compare survival and home time between older adults who started dialysis at an estimated glomerular filtration rate (eGFR) less than 12 mL/min/1.73 m2 and those who continued medical management.

DESIGN

Observational cohort study using target trial emulation.

SETTING

U.S. Department of Veterans Affairs, 2010 to 2018.

PARTICIPANTS

Adults aged 65 years or older with chronic kidney failure and eGFR below 12 mL/min/1.73 m2 who were not referred for transplant.

INTERVENTION

Starting dialysis within 30 days versus continuing medical management.

MEASUREMENTS

Mean survival and number of days at home.

RESULTS

Among 20 440 adults (mean age, 77.9 years [SD, 8.8]), the median time to dialysis start was 8.0 days in the group starting dialysis and 3.0 years in the group continuing medical management. Over a 3-year horizon, the group starting dialysis survived 770 days and the group continuing medical management survived 761 days (difference, 9.3 days [95% CI, -17.4 to 30.1 days]). Compared with the group continuing medical management, the group starting dialysis had 13.6 fewer days at home (CI, 7.7 to 20.5 fewer days at home). Compared with the group continuing medical management and forgoing dialysis completely, the group starting dialysis had longer survival by 77.6 days (CI, 62.8 to 91.1 days) and 14.7 fewer days at home (CI, 11.2 to 16.5 fewer days at home).

LIMITATION

Potential for unmeasured confounding due to lack of symptom assessments at eligibility; limited generalizability to women and nonveterans.

CONCLUSION

Older adults starting dialysis when their eGFR fell below 12 mL/min/1.73 m2 who were not referred for transplant had modest gains in life expectancy and less time at home.

PRIMARY FUNDING SOURCE

U.S. Department of Veterans Affairs and National Institutes of Health.

Do parents engage in weight- and health-focused conversations with their emerging adult children and are there cross-sectional associations with weight and well-being outcomes?

BACKGROUND

Research indicates harmful associations between parental weight-focused conversations and markers of pediatric health and well-being. However, little is known about the prevalence and consequences of parent conversations focused on weight or health behaviors (i.e., physical activity or nutrition) with emerging adult children.

METHODS

Data are from the 2018 follow-up survey of the population-based EAT 2010-2018 (Eating and Activity over Time) in cohort from Minneapolis-St. Paul, MN. Participants were emerging adults at follow-up with ages 18-26. Regression models adjusted for sociodemographic characteristics were conducted.

RESULTS

Over two-thirds (68%) of mothers and 44% of fathers engaged in weight-focused conversations with their emerging adult children; 25% of both parents reported engaging in conversations focused only on health behaviors; and 8% of mothers and 26% of fathers reported not engaging in either type of conversation. Health-focused conversations by both parents were associated with lower body mass index (BMI) and disordered eating behaviors, higher intake of fruit and vegetables, and psychosocial well-being in emerging adult children. Weight-focused conversations with both parents were associated with higher BMI and disordered eating behaviors in emerging adults. There were gender moderated associations of paternal conversations about weight and health with vegetable intake, binge eating, and depressive symptoms.

DISCUSSION

The high prevalence and negative health outcomes associated with weight-focused conversations coupled with the low prevalence and positive health outcomes associated with health-focused conversations by parents suggests the need for public health messaging and intervention development aimed at reducing parental weight talk with emerging adult children.

Arts and cultural engagement and subsequent social deficits among older adults: A three-year longitudinal study using the Japan Gerontological Evaluation Study

Arts and cultural engagement has the potential to reduce social deficits such as loneliness and social isolation. However, as most evidence is from Western countries, less is known whether the protective association of engagement with social deficits can also be seen in different cultural settings such as Asia. We explored the associations of arts and cultural engagement, focusing on engagement continuity and type, with loneliness and social isolation among older adults in Japan, one of the fastest-ageing countries. This three-year longitudinal study involved 4,383 individuals (mean age = 74.3 years; 51.3% women) from the Japan Gerontological Evaluative Study 2019; 2022 waves. To assess engagement continuity, respondents were categorised into four groups: none, decreasing, increasing and sustained engagement. A latent class analysis identified four classes of engagement type: low, receptive, creative and diverse engagement. Loneliness and social isolation were measured using the University of California Los Angeles (UCLA) 3-Item Loneliness Scale (range: 3-9) and the Social Isolation Index (range: 0-5), respectively. We applied ordinary least squares regressions to investigate the associations between engagement and the outcomes. Regarding engagement continuity, those who increased or sustained their engagement across waves reported lower loneliness compared with those who did not engage (increased: coef. = -0.22, 95% confidential interval [CI] = -0.41, -0.04; sustained: coef. = -0.26, 95% CI = -0.36, -0.16). Individuals who sustained their engagement also reported lower social isolation (coef. = -0.18, 95% CI = -0.27, -0.09). Regarding engagement type, engaging in diverse activities was associated with lower loneliness compared to low engagement (coef. = -0.34, 95% CI = -0.59, -0.10), while creative and diverse engagement were associated with lower social isolation (creative: coef. = -0.13, 95% CI = -0.22, -0.04; diverse: coef. = -0.33, 95% CI = -0.54, -0.12). These findings suggest that offering a diversity of creative arts and cultural activities and supporting sustainable engagement of older adults may help alleviate their social deficits.

Trauma exposure among young children in rural Pakistan: Associations with gender, mental health, and cognitive skills

BACKGROUND

The burden of childhood trauma (violence, injury/illness, loss) in low-resource settings is high, although the effect of trauma on children's mental and cognitive health is under-researched. Child gender may moderate the association between trauma and outcomes; boys are more likely to experience trauma, but girls are more likely to show distress following trauma.

METHODS

We draw on data from the Bachpan cohort (n = 888), a sample of mother-child dyads in rural Pakistan, to investigate these associations among 6-year-old children in a South Asian, low-resource setting. Mothers reported on children's lifetime exposure to 15 possible traumas and their current mental health. In addition, children were assessed for their verbal skills, working memory, and inhibitory control. We estimated trauma prevalence and used generalized estimating equations to test the association between number of traumatic events and child mental health and cognitive skills in the overall sample and by gender.

RESULTS

90.5% of children experienced at least one trauma. The most common traumas were death of a loved one (47%) and hearing about war/terrorism on the TV/radio (48%). On average, boys experienced more traumas (M = 3.00) than girls (M = 2.67). Specifically, boys were more likely than girls to experience an injury or hospitalization (30% vs. 21%, p < .05) and to hear about war/terrorism on the TV/radio (52% vs. 43%, p < .05). Trauma was associated with increased mental health difficulties and increased anxiety scores. There was little evidence that trauma exposure was associated with child cognitive skills. The strength of association between trauma and outcomes was similar for boys and girls.

CONCLUSIONS

Findings show that childhood trauma exposure is common in this setting and associated with worse mental health problems among young boys and girls. Results suggest that trauma-informed interventions are important for supporting child mental health in South Asia.

Effect of retirement on loneliness: a longitudinal comparative analysis across Australia, China and the USA

BACKGROUND

There is evidence that the transition to retirement can bring social challenges and may increase loneliness. Few studies have examined the impact of retirement on loneliness; most have been conducted in Western countries. It is important to examine the differences in loneliness postretirement across countries to identify patterns and risk factors that might influence the health and well-being of older adults. We aimed to examine the effect of retirement on loneliness among older adults in Australia, China and the USA.

METHODS

Longitudinal analysis of data from population-based samples of Australian, Chinese and American adults over 50. Lagged and fully lagged adjusted models were applied. Social engagement was examined as an effect modifier and a sensitivity analysis was conducted among urban participants.

RESULTS

Retirees had a higher predicted prevalence of loneliness than non-retirees in Australia (19.4% (95% CI 18.0% to 20.9%) vs 17.0% (95% CI 15.7% to 18.4%)) and in the USA (19.3% (95% CI 17.5% to 21.1%) vs 15.7% (95% CI 14.3% to 17.3%)). These differences were significant only in the USA. In China, loneliness was significantly lower in those who had retired (10.0% (95% CI 7.9% to 12.5%) vs 17.1% (95% CI 15.7% to 18.5%)). In Australia and the USA, voluntary retirees had the lowest loneliness and involuntary retirees had the highest. Social engagement did not modify the association between retirement and loneliness.

CONCLUSIONS

Our findings imply that the effect of retirement should be considered within a cultural context to inform suitable and effective strategies to alleviate loneliness.

Blood metabolites mediate effects of breakfast skipping on heart failure via Mendelian randomization analysis

Numerous observational studies have suggested a potential causal relationship between skipping breakfast and cardiovascular diseases, including heart failure (HF). However, these studies are susceptible to inherent confounders and the challenge of reverse causation, and the underlying metabolic factors are not yet clear. Therefore, our aim is to assess the causal impact of breakfast skipping on HF and the role of potential mediating metabolic products from a genetic perspective, by conducting Mendelian Randomization (MR) studies and mediation analysis. We leveraged summary data from the most extensive genome-wide association studies to date on breakfast skipping (with 193,860 participants), blood metabolites (with 118,461 participants), and HF (involving 47,309 cases and 930,014 controls). To explore the causal relationship between breakfast skipping and HF, as well as the role of 249 potential blood metabolite mediators, we conducted bidirectional MR and mediation MR analyses. We primarily employed the Inverse Variance Weighted (IVW) method, complemented by various other techniques to ensure the comprehensiveness and reliability of our analysis. Our research confirms a causal association between breakfast skipping and an increased risk of HF (odds ratio [OR]: 1.378, 95% confidence interval [CI]: 1.047-1.813; p = 0.022). Furthermore, our research findings demonstrate that breakfast skipping is positively correlated with 6 blood metabolites and negatively correlated with 2 others. Notably, our mediation MR analysis further reveals that three blood metabolites act as mediators in the relationship between breakfast skipping and the risk of HF. Specifically, the mediating effects are attributed to the ratio of docosahexaenoic acid (DHA) to total fatty acids (proportion mediated = 9.41%, 95% CI: 2.10-28.61%), glucose (proportion mediated = 6.17%, 95% CI: 0.97-28.53%), and glycoprotein acetyls (GlycA) (proportion mediated = 5.68%, 95% CI: 0.94-21.62%). The combined mediating effects of these three factors total 20.53% (95%CI: 8.59-91.06%). Our research confirms the causal relationship between genetically instrumented breakfast skipping and HF, underscoring the potential mediating roles played by three key blood metabolites: ratio of DHA to total fatty acids, glucose and GlycA. This discovery offers valuable perspectives for clinical strategies targeting HF.

Balancing efficacy and computational burden: weighted mean, multiple imputation, and inverse probability weighting methods for item non-response in reliable scales

IMPORTANCE

Scales often arise from multi-item questionnaires, yet commonly face item non-response. Traditional solutions use weighted mean (WMean) from available responses, but potentially overlook missing data intricacies. Advanced methods like multiple imputation (MI) address broader missing data, but demand increased computational resources. Researchers frequently use survey data in the All of Us Research Program (All of Us), and it is imperative to determine if the increased computational burden of employing MI to handle non-response is justifiable.

OBJECTIVES

Using the 5-item Physical Activity Neighborhood Environment Scale (PANES) in All of Us, this study assessed the tradeoff between efficacy and computational demands of WMean, MI, and inverse probability weighting (IPW) when dealing with item non-response.

MATERIALS AND METHODS

Synthetic missingness, allowing 1 or more item non-response, was introduced into PANES across 3 missing mechanisms and various missing percentages (10%-50%). Each scenario compared WMean of complete questions, MI, and IPW on bias, variability, coverage probability, and computation time.

RESULTS

All methods showed minimal biases (all <5.5%) for good internal consistency, with WMean suffered most with poor consistency. IPW showed considerable variability with increasing missing percentage. MI required significantly more computational resources, taking >8000 and >100 times longer than WMean and IPW in full data analysis, respectively.

DISCUSSION AND CONCLUSION

The marginal performance advantages of MI for item non-response in highly reliable scales do not warrant its escalated cloud computational burden in All of Us, particularly when coupled with computationally demanding post-imputation analyses. Researchers using survey scales with low missingness could utilize WMean to reduce computing burden.

Effectiveness of early rehabilitation interventions in patients with traumatic brain injury using a large database

BACKGROUND

Rehabilitation is important for patients with moderate-to-severe traumatic brain injury (TBI). However, the timing of early rehabilitation initiation is ambiguous, and its safety and effectiveness are unknown.

OBJECTIVE

To examine the effectiveness and safety of early rehabilitation in patients with moderate-to-severe TBI using propensity score analysis and a large database.

DESIGN

Retrospective cohort study.

SETTING

A large medical database (JMDC database) of tertiary care facilities was used to compare outcomes of early and delayed rehabilitation.

PATIENTS

Patients aged between 20 and 90 years who were diagnosed with TBI were admitted to acute care hospitals. Inclusion criteria were patients undergoing rehabilitation within 7 days of admission with a Glasgow Coma Scale score of 3 to 12 on admission. This study included 3074 patients with moderate-to-severe TBI.

INTERVENTIONS

Patients were classified into an early rehabilitation group (within 2 days of admission) or a delayed rehabilitation group (3 to 7 days postadmission), depending on when rehabilitation started after TBI. Rehabilitation was defined as any type or intensity of intervention provided by a physical, occupational, and/or speech/language therapist. Interventions were not controlled.

MAIN OUTCOME MEASURE(S)

The primary outcome was Barthel Index (BI) efficiency (BI gain/length of stay). Secondary outcomes included BI gain (discharge BI - admission BI), incidence of aspiration pneumonia complications during hospitalization, discharge to home, mortality, and length of stay.

RESULTS

After applying inverse probability weighting with propensity scores, the total was 6152 patients. 3074 (50.0%) patients received early rehabilitation. The early rehabilitation group showed no difference in inpatient mortality (p = .438), improved BI efficiency (β = 0.86, p < .001), and shorter length of stay (β = -5.00, p = .018).

CONCLUSIONS

Early rehabilitation in patients with moderate-to-severe TBI is associated with more efficient functional improvement and reduced hospital stays without an increase in inpatient mortality.

An Empirical Dietary Pattern Associated with the Gut Microbial Features in Relation to Colorectal Cancer Risk

BACKGROUND & AIMS

Epidemiologic evidence for dietary influence on colorectal cancer (CRC) risk through the gut microbiome remains limited.

METHODS

Leveraging 307 men and 212 women with stool metagenomes and dietary data, we characterized and validated a sex-specific dietary pattern associated with the CRC-related gut microbial signature (CRC Microbial Dietary Score [CMDS]). We evaluated the associations of CMDS with CRC risk according to Fusobacterium nucleatum, pks+Escherichia coli, and enterotoxigenic Bacteroides fragilis status in tumor tissue using Cox proportional hazards regression in the Health Professionals Follow-Up Study (1986-2018), Nurses' Health Study (1984-2020), and Nurses' Health Study II (1991-2019).

RESULTS

The CMDS was characterized by high industrially processed food and low unprocessed fiber-rich food intakes. In 259,200 participants, we documented 3854 incident CRC cases over 6,467,378 person-years of follow-up. CMDS was associated with a higher risk of CRC (Ptrend < .001), with a multivariable hazard ratio (HRQ5 vs Q1) of 1.25 (95% CI, 1.13-1.39). The association remained after adjusting for previously established dietary patterns, for example, the Western and prudent diets. Notably, the association was stronger for tumoral F nucleatum-positive (HRQ5 vs Q1, 2.51; 95% CI, 1.68-3.75; Ptrend < .001; Pheterogeneity = .03, positivity vs negativity), pks+E coli-positive (HRQ5 vs Q1, 1.68; 95% CI, 0.84-3.38; Ptrend = .005; Pheterogeneity = .01, positivity vs negativity), and enterotoxigenic Bacteroides fragilis-positive CRC (HRQ5 vs Q1, 2.06; 95% CI, 1.10-3.88; Ptrend = .016; Pheterogeneity = .06, positivity vs negativity), compared with their negative counterparts.

CONCLUSIONS

CMDS was associated with increased CRC risk, especially for tumors with detectable F nucleatum, pks+E coli, and enterotoxigenic Bacteroides fragilis in tissue. Our findings support a potential role of the gut microbiome underlying the dietary effects on CRC.

Characterisation and validation of lactation information from structured electronic health records for use in pharmacoepidemiological studies

BACKGROUND

Breastfeeding information stored within electronic health records (EHR) has recently been used for pharmacoepidemiological research, however the data are primarily collected for clinical care.

OBJECTIVES

To characterise breastfeeding information recorded in structured fields in EHR during infant and postpartum health care visits, and to assess the validity of lactation status based on EHR data versus maternal report at research study visits.

METHODS

We assessed breastfeeding information recorded in structured fields in EHR from one health system for a subset of 211 patients who were also enrolled in a study on breast milk composition between 2014 and 2017 that required participants to exclusively breastfeed their infants until at least 1 month of age. We assessed the frequency of breastfeeding information in EHR during the first 12 months of age and compared lactation status based on EHR with maternal report at 1 and 6-month study visits (reference standard).

RESULTS

The median number of breastfeeding records in the EHR per infant was six (interquartile range 3) with most observations clustering in the first few weeks of life and around well-infant visits. At the 6-month study visit, 93.8% of participants were breastfeeding and 80.1% were exclusively breastfeeding according to maternal report. Sensitivity of EHR data for identifying ever breastfeeding was at or near 100%, and sensitivity for identifying ever exclusive breastfeeding was 98.0% (95% CI: 95.0%, 99.2%). Sensitivities were 97.3% (95% CI: 93.9%, 98.9%) for identifying any breastfeeding and 94.4% (95% CI: 89.7%, 97.0%) for exclusive breastfeeding, and positive predictive values were 99.5% (95% CI: 97.0%, 99.9%) for any breastfeeding and 95.0% (95% CI: 90.4%, 97.4%) for exclusive breastfeeding.

CONCLUSIONS

Breastfeeding information in structured EHR fields have the potential to accurately classify lactation status. The validity of these data should be assessed in populations with a lower breastfeeding prevalence.

Psoriasis and mental health in adolescents: A cross-sectional study within the Danish National Birth Cohort

BACKGROUND

Psoriasis is a chronic skin disease associated with lower quality of life and higher risk of anxiety and depression in adults. We investigate whether adolescents with psoriasis also experience poorer mental health than their peers.

METHODS

In this cross-sectional study, we included questionnaire data on psoriasis and mental health from the 18-year follow-up of the Danish National Birth Cohort. We estimated odds ratios (OR) and 95 % confidence intervals (CI) using a logistic regression with inverse probability weighting to account for potential selection bias, adjusted for potential confounders identified a priori. We estimated associations between self-reported psoriasis and multiple aspects of mental health (self-rated health, life satisfaction, mental well-being, loneliness, overall and internalizing behavioral difficulties, depressive symptoms, and anxiety symptoms). In sensitivity analyses, we examined doctor-diagnosed psoriasis and psoriasis with and without joint pain.

RESULTS

Of the 44,838 included in this study, 1147 (2.6 %) reported psoriasis. Adolescents with psoriasis had a higher risk of nearly all outcomes, including depressive symptoms (OR 1.38; 1.19-1.58) and panic/agoraphobia among both males (OR 1.72; 1.33-2.19) and females (OR 1.60; 1.33-1.92). Associations attenuated when restricted to doctor-diagnosed psoriasis. Associations with poor mental health were mainly observed for adolescents with psoriasis also reporting joint pain.

LIMITATIONS

We could not establish temporality and lacked data on joint pain in referents.

CONCLUSION

Psoriasis is associated with poor mental health in adolescents. This appears to be driven by adolescents with psoriasis also reporting joint pain and is less evident in those with a doctor-confirmed diagnosis.

Sociodemographic Trends and Perinatal Outcomes in Fathers 50 Years and Older

Importance

The age of fathers at childbirth is rising, with an increasing number of births attributed to older fathers. While the impact of advanced paternal age has been documented, sociodemographic data about fathers aged 50 years and older remain scarce.

Objectives

To explore sociodemographic and temporal trends among the oldest US fathers (age ≥50 years) and their associations with perinatal outcomes.

Design, Setting, and Participants

This retrospective cross-sectional study included data from all US births from 2011 to 2022 using the National Vital Statistics System. Data were analyzed from August 2023 and May 2024.

Exposures

Reported paternal age at childbirth.

Main Outcomes and Measures

Outcomes of interest were sociodemographic factors, temporal trends in older fatherhood, and perinatal outcomes, including preterm birth, low birth weight, gestational diabetes, gestational hypertension, assisted reproductive technology (ART), rates of maternal primiparity, and the infant sex ratio.

Results

From 2011 to 2022, the US recorded 46 195 453 births, with an overall mean (SD) paternal age of 31.5 (6.8) years and 484 507 (1.1%) involving fathers aged 50 years or older, 47 785 (0.1%) aged 60 years or older, and 3777 (0.008%) aged 70 years or older. Births to fathers aged 50 years or older increased from 1.1% in 2011 to 1.3% in 2022 (P for trend < .001). Fathers aged 50 years or older were more diverse, with variations in educational achievement and race and ethnicity. Marital status and maternal racial and ethnic and educational backgrounds also varied by paternal age and race. Despite controlling for maternal age and other sociodemographic and perinatal factors, every 10-year increase in paternal age was consistently associated with greater use of ART (eg, age 50-59 years: adjusted odds ratio [aOR], 2.23; 95% CI, 2.19-2.27), higher likelihood of first maternal birth (eg, age 50-59 years: aOR, 1.16; 95% CI, 1.15-1.17), and increased risks of preterm birth (eg, age 50-59 years: aOR, 1.16; 95% CI, 1.15-1.18) and low birth weight (eg, age 50-59 years: aOR, 1.14; 95% CI, 1.13-1.15) compared with fathers aged 30 to 39 years. No significant changes in the infant sex ratio were observed, except among fathers aged 70 years or older (aOR, 0.92; 95% CI, 0.86-0.99) and 75 years or older (aOR, 0.84; 95% CI, 0.73-0.97), who showed a decreased likelihood of having male offspring.

Conclusions and Relevance

In this cross-sectional study of all US births from 2011 to 2022, the percentage attributed to older fathers, while small, increased. Notable variations in paternal and maternal race and education were identified. Older fatherhood was associated with increased ART use, first-time maternal births, adverse perinatal outcomes, and altered sex ratio. Further research of this population is crucial for improving patient counseling and family planning.

Childhood and Adolescent Depression Symptoms and Young Adult Mental Health and Psychosocial Outcomes

Importance

Depression is a leading cause of disability. The timing and persistence of depression may be differentially associated with long-term mental health and psychosocial outcomes.

Objective

To examine if depression symptoms during early and middle childhood and adolescence and persistent depression symptoms are associated with impaired young adult outcomes independent of early risk factors.

Design, Setting, and Participants

Data for this prospective, longitudinal cohort study were from the Québec Longitudinal Study of Child Development, a representative population-based Canadian birth cohort. The cohort consists of infants born from October 1, 1997, to July 31, 1998. This is an ongoing study; data are collected annually or every 2 years and include those ages 5 months to 21 years. The end date for the data in this study was June 30, 2019, and data analyses were performed from October 4, 2022, to January 3, 2024.

Exposures

Depression symptoms were assessed using maternal reports in early childhood (ages 1.5 to 6 years) from 1999 to 2004, teacher reports in middle childhood (ages 7 to 12 years) from 2005 to 2010, and self-reports in adolescence (ages 13 to 17 years) from 2011 to 2015.

Main Outcomes and Measures

The primary outcome was depression symptoms at age 20 years, and secondary outcomes were indicators of psychosocial functioning (binge drinking; perceived stress; not being in education, employment, or training; social support; and experiencing online harrasment) at age 21 years. All outcomes were self-reported. Adult outcomes were reported by participants at ages 20 and 21 years from 2017 to 2019. Risk factors assessed when children were aged 5 months old were considered as covariates to assess the independent associations of childhood and adolescent depression symptoms with adult outcomes.

Results

The cohort consisted of 2120 infants. The analytic sample size varied from 1118 to 1254 participants across outcomes (56.85% to 57.96% female). Concerning the primary outcome, adjusting for early risk factors and multiple testing, depression symptoms during adolescence were associated with higher levels of depression symptoms (β, 1.08 [95% CI, 0.84-1.32]; P < .001 unadjusted and Bonferroni adjusted) in young adulthood. Concerning the secondary outcomes, depression symptoms in adolescence were only associated with perceived stress (β, 3.63 [95% CI, 2.66-4.60]; P < .001 unadjusted and Bonferroni adjusted), while both middle-childhood (β, -1.58 [95% CI, -2.65 to -0.51]; P = .003 unadjusted and P < .001 Bonferroni adjusted) and adolescent (β, -1.97 [95% CI, -2.53 to -1.41]; P < .001 unadjusted and Bonferroni adjusted) depression symptoms were associated with lower levels of social support. There were no associations for binge drinking; not being in education, employment, or training; or experiencing online harrasment.

Conclusions and Relevance

In this cohort study of Canadian children and adolescents, childhood and adolescent depression symptoms were associated with impaired adult psychosocial functioning. Interventions should aim to screen and monitor children and adolescents for depression to inform policymaking regarding young adult mental health and psychosocial outcomes.

A machine learning personalized treatment rule to optimize assignment to psychotherapies for grief among veterans

BACKGROUND

Complex grief patterns are associated with significant suffering, functional impairments, health and mental health problems, and increased healthcare use. This burden may be even more pronounced among veterans. Behavioral Activation and Therapeutic Exposure (BATE-G) and Cognitive Therapy for Grief (CT-G) are two evidence-based interventions for grief. The goal of this study was to use a precision medicine approach to develop a personalized treatment rule to optimize assignment among these psychotherapies.

METHODS

We analyzed data (N = 155) from a randomized clinical trial comparing BATE-G and CT-G. Outcome weighted learning was used to estimate an optimal personalized treatment rule. Baseline characteristics including demographics, social support, variables related to the death, and psychopathology dimensions were used as prescriptive factors of treatment assignment.

RESULTS

The estimated rule assigned 72 veterans to CT-G and 56 to BATE-G. Assigning participants according to this rule was estimated to lead to markedly lower mean grief level following 6 months from treatment compared to assigning everyone to either BATE-G (Vdopt - VBATE-G = -18.57 [95 % CI: -29.41, -7.72]) or CT-G (Vdopt - VBATE-G = -20.89 [95 % CI: -30.7, -11.07]) regardless of their characteristics.

LIMITATIONS

Participants were primarily male veterans, and identified with Black or White race. The estimated rule was not externally validated.

CONCLUSION

The estimated rule used relatively simple, easily accessible, client characteristics to personalize assignment to treatment using a precision medicine approach based on machine learning and causal inference. Upon further validation, such a rule can be easily implemented in clinical practice to prescriptively maximize treatment benefits.

Cognitive Trajectories in Older Adults Diagnosed With Hematologic Malignant Neoplasms

Importance

More people are surviving long-term after diagnosis with hematologic malignant neoplasm (HMN), yet there are limited data on cancer-related cognitive impairment in people with HMN. Better understanding cognitive outcomes after HMN in older adults is important for patient counseling and management.

Objective

To model cognitive trajectories and rates of cognitive decline before and after HMN diagnosis in older adults compared with a matched noncancer cohort.

Design, Setting, and Participants

In this population-based cohort study, older adults from the Health and Retirement Study (HRS) diagnosed with HMN between 1998 and 2016 after age 65 years were matched 1:3 to participants without cancer from the same HRS wave using propensity scores incorporating variables relevant to cognition. Cognitive trajectories were modeled with piecewise linear splines, and rates of cognitive decline before, during, and after diagnosis were compared in the 2 groups. Data were analyzed from April 2022 to April 2024.

Exposures

HMN diagnosis by Medicare diagnosis codes.

Main Outcomes and Measures

Cognitive function was assessed by the Langa-Weir cognitive summary score from 1992 to 2020. Sociodemographic and health-related variables relevant to cognition were incorporated into propensity scores.

Results

At baseline, there were 668 participants in the HMN cohort (mean [SD] age, 76.8 [7.6] years; 343 [51.3%] male; 72 [10.8%] Black, 33 [4.9%] Hispanic, and 585 [87.6%] White) and 1994 participants in the control cohort (mean [SD] age, 76.5 [7.3] years; 1020 [51.2%] male; 226 [11.3%] Black, 91 [4.6%] Hispanic, and 1726 [86.6%] White). The HMN cohort consisted predominantly of more indolent diagnoses, and only 96 patients (14.4%) received chemotherapy. Before and in the 2 years around the time of diagnosis, the HMN and control cohorts had similar rates of cognitive decline. At 1 year postdiagnosis and beyond, the rate of cognitive decline was slower in the HMN cohort (-0.18; 95% CI, -0.23 to -0.14) than in the control group (-0.24; 95% CI, -0.26 to -0.23) (P = .02), but this difference was no longer significant after accounting for the competing risk of death (HMN group, -0.27; 95% CI, -0.34 to -0.19; control group, -0.30; 95% CI, -0.33 to -0.27; P = .48).

Conclusions and Relevance

In this cohort study of older adults, the HMN and matched noncancer control cohorts had similar rates of cognitive decline before, during, and after diagnosis after accounting for the competing risk of death.

Using instruments for selection to adjust for selection bias in Mendelian randomization

Selection bias is a common concern in epidemiologic studies. In the literature, selection bias is often viewed as a missing data problem. Popular approaches to adjust for bias due to missing data, such as inverse probability weighting, rely on the assumption that data are missing at random and can yield biased results if this assumption is violated. In observational studies with outcome data missing not at random, Heckman's sample selection model can be used to adjust for bias due to missing data. In this paper, we review Heckman's method and a similar approach proposed by Tchetgen Tchetgen and Wirth (2017). We then discuss how to apply these methods to Mendelian randomization analyses using individual-level data, with missing data for either the exposure or outcome or both. We explore whether genetic variants associated with participation can be used as instruments for selection. We then describe how to obtain missingness-adjusted Wald ratio, two-stage least squares and inverse variance weighted estimates. The two methods are evaluated and compared in simulations, with results suggesting that they can both mitigate selection bias but may yield parameter estimates with large standard errors in some settings. In an illustrative real-data application, we investigate the effects of body mass index on smoking using data from the Avon Longitudinal Study of Parents and Children.

What role do socioeconomic and clinical factors play in disease activity states in rheumatoid arthritis? Data from a large UK early inflammatory arthritis audit

BACKGROUND

Persistently active rheumatoid arthritis (pactiveRA) may be due to the interplay between biological and non-biological factors. The role of socioeconomic factors remains unclear.

OBJECTIVES

To explore which biological and non-biological factors associate with pactiveRA.

METHODS

Adults with early RA in the National Early Inflammatory Arthritis Audit, recruited from May 2018 to October 2022, were included if having pactiveRA or persistently low RA (plowRA). The pactiveRA was defined as three consecutive Disease Activity Score-28 joints (DAS28) of >3.2 at baseline, 3 and 12 months. The plowRA was defined as DAS28 ≤3.2 at 3 and 12 months. Stepwise forward logistic regression was used to explore associations with pactiveRA (outcome). Age and gender were included a priori, with socioeconomic factors and comorbidities as exposure variables.

RESULTS

682 patients with pactiveRA and 1026 plowRA were included. Compared with plowRA, patients with pactiveRA were younger (58, IQR: 49-67) versus (62, IQR: 52-72), and included more women (69% vs 59%). The pactiveRA was associated with worse scores in patient-reported outcomes at baseline, and anxiety and depression screens. Overall, there was clear social patterning in pactiveRA, with age-by-gender interaction. Logistic regression indicated age, gender, social deprivation and previous or current smoking, were independently associated with pactiveRA, after controlling for disease severity markers (seropositivity). Depression, lung disease, gastric ulcers and baseline corticosteroid use, were also associated with pactiveRA (p<0.05 for all).

CONCLUSION

Socioeconomic factors and deprivation were associated with pactiveRA, independent of clinical and disease characteristics. Identifying 'adverse' socioeconomic drivers of pactiveRA can help tailor interventions according to individual need.

Terror catastrophizing: association with anxiety, depression, and transgenerational effects

Background & Objectives: Terror catastrophizing, defined as an ongoing fear of future terrorist attacks, is associated with a higher incidence of anxiety disorders, among other psychological impacts. However, previous studies examining terror catastrophizing's relationship to other mental health disorders are limited. The current study sought to determine if patients diagnosed with anxiety and depression would experience increased terror catastrophizing. Additionally, this study aimed to investigate whether parental terror catastrophizing increases children's internalizing symptoms.Design & Methods: Individuals were randomly drawn from the Danish Civil Registration System and invited to complete a series of questionnaires to measure terror catastrophizing tendency, lifetime parental trauma, and children's internalizing symptoms. In total, n = 4,175 invitees completed the survey of which 933 reported on a child between 6 and 18 years. Responses were analyzed using a generalized linear regression model.Results: Participants diagnosed with anxiety alone or comorbid with depression were more likely to experience symptoms of terror catastrophizing than undiagnosed participants (β = 0.10, p < .001; β = 0.07, p = .012). Furthermore, the parental tendency to catastrophize terror was associated with higher internalizing symptoms in children (β = 0.09, p = .006), even after taking parental diagnoses, as well as lifetime and childhood trauma into account.Conclusion: The results can inform clinical practices to account for a patient's potential to exhibit increased terror catastrophizing tendencies or be more affected by traumatic events. Additionally, they can offer insights for designing novel preventative interventions for the whole family, due to the relation between parental tendencies for terror catastrophizing and the internalizing symptoms observed in children.

Body mass index and weight gain in pregnancy and cardiovascular health in middle age: A cohort study

OBJECTIVE

To examine associations between body mass index (BMI) in early pregnancy and gestational weight gain (GWG) with cardiovascular health in middle age using the 'Life's Essential 8' (LE8) concept of the American Heart Association (AHA).

DESIGN

Population-based cohort study.

SETTING

Swedish CardioPulmonary bioImage Study (SCAPIS).

POPULATION

A total of 8871 women from SCAPIS were included.

METHODS

Information on cardiovascular health in middle age was collected from SCAPIS and linked to pregnancy weight data obtained from the Swedish Medical Birth Register, with an average follow-up time of 24.5 years. An LE8 score between 0 and 100 was determined, where a score under 60 points was defined as poor cardiovascular health. Binary logistic regression and restricted cubic splines were used.

MAIN OUTCOME MEASURES

Cardiovascular health according to LE8 in middle age.

RESULTS

The odds of having poor cardiovascular health in middle age were significantly higher in women who had overweight (adjusted odds ratio, aOR 3.30, 95% CI 2.82-3.88) or obesity (aOR 7.63, 95% CI 5.86-9.94), compared with women classified as being of normal weight in pregnancy. Higher odds were also found for excessive GWG (aOR 1.31, 95% CI 1.09-1.57), compared with women who gained weight within the recommendations.

CONCLUSIONS

A high BMI in early pregnancy and excessive GWG were associated with greater odds of poor cardiovascular health in middle age. Although further studies are needed, our results highlight pregnancy as an important period to support long-term cardiovascular health.

Association between changes in cardiovascular health and the risk of multimorbidity: community-based cohort studies in the UK and Finland

Background

Better cardiovascular health is associated with lower risk of various chronic diseases, but its association with multimorbidity is poorly understood. We aimed to examine whether change in cardiovascular health is associated with multimorbidity risk.

Methods

The primary analysis was conducted in the Whitehall II multiwave prospective cohort study (UK) and the validation analysis in the Finnish Public Sector cohort study (Finland). Change in cardiovascular health was assessed using the American Heart Association Life's Simple 7 (LS7) and Life's Essential 8 (LE8) at baseline and re-assessments, using objective measures in Whitehall II and self-reports and pharmacy claims in the Finnish Public Sector cohort study, respectively. Multimorbidity was defined as the presence of two or more of 12 chronic diseases during follow-up. We estimated hazard ratios (HR) and 95% confidence intervals (CI) using Cox's proportional hazard models with age as time scale, adjusting for sex, education, occupation, marital status, and ethnicity.

Findings

In the primary analysis among 9715 participants, mean age was 44.8 (standard deviation 6.0) years and 67.6% participants were men at baseline. During the median follow-up of 31.4 (interquartile range 26.8-32.3) years, 2751 participants developed multimorbidity. The hazard of multimorbidity decreased by 8% (HR 0.92, 95% CI 0.88-0.96) per ideal LS7 metric increment over 5 years and by 14% (HR 0.86, 95% CI 0.80-0.93) per ten points increase in LE8 score over 10 years. These findings were replicated in the validation analysis among 75,377 participants in terms of 4-year change in cardiovascular health.

Interpretation

Improvement in cardiovascular health was associated with lower multimorbidity risk in two community-based cohort studies. Interventions improving cardiovascular health of the community may contribute to multimorbidity prevention.

Funding

None.

Comparative effectiveness of an individualized model of hemodialysis vs conventional hemodialysis: a study protocol for a multicenter randomized controlled trial (the TwoPlus trial)

BACKGROUND

Most patients starting chronic in-center hemodialysis (HD) receive conventional hemodialysis (CHD) with three sessions per week targeting specific biochemical clearance. Observational studies suggest that patients with residual kidney function can safely be treated with incremental prescriptions of HD, starting with less frequent sessions and later adjusting to thrice-weekly HD. This trial aims to show objectively that clinically matched incremental HD (CMIHD) is non-inferior to CHD in eligible patients.

METHODS

An unblinded, parallel-group, randomized controlled trial will be conducted across diverse healthcare systems and dialysis organizations in the USA. Adult patients initiating chronic hemodialysis (HD) at participating centers will be screened. Eligibility criteria include receipt of fewer than 18 treatments of HD and residual kidney function defined as kidney urea clearance ≥3.5 mL/min/1.73 m2 and urine output ≥500 mL/24 h. The 1:1 randomization, stratified by site and dialysis vascular access type, assigns patients to either CMIHD (intervention group) or CHD (control group). The CMIHD group will be treated with twice-weekly HD and adjuvant pharmacologic therapy (i.e., oral loop diuretics, sodium bicarbonate, and potassium binders). The CHD group will receive thrice-weekly HD according to usual care. Throughout the study, patients undergo timed urine collection and fill out questionnaires. CMIHD will progress to thrice-weekly HD based on clinical manifestations or changes in residual kidney function. Caregivers of enrolled patients are invited to complete semi-annual questionnaires. The primary outcome is a composite of patients' all-cause death, hospitalizations, or emergency department visits at 2 years. Secondary outcomes include patient- and caregiver-reported outcomes. We aim to enroll 350 patients, which provides ≥85% power to detect an incidence rate ratio (IRR) of 0.9 between CMIHD and CHD with an IRR non-inferiority of 1.20 (α = 0.025, one-tailed test, 20% dropout rate, average of 2.06 years of HD per patient participant), and 150 caregiver participants (of enrolled patients).

DISCUSSION

Our proposal challenges the status quo of HD care delivery. Our overarching hypothesis posits that CMIHD is non-inferior to CHD. If successful, the results will positively impact one of the highest-burdened patient populations and their caregivers.

TRIAL REGISTRATION

Clinicaltrials.gov NCT05828823. Registered on 25 April 2023.

Addressing label noise for electronic health records: insights from computer vision for tabular data

The analysis of extensive electronic health records (EHR) datasets often calls for automated solutions, with machine learning (ML) techniques, including deep learning (DL), taking a lead role. One common task involves categorizing EHR data into predefined groups. However, the vulnerability of EHRs to noise and errors stemming from data collection processes, as well as potential human labeling errors, poses a significant risk. This risk is particularly prominent during the training of DL models, where the possibility of overfitting to noisy labels can have serious repercussions in healthcare. Despite the well-documented existence of label noise in EHR data, few studies have tackled this challenge within the EHR domain. Our work addresses this gap by adapting computer vision (CV) algorithms to mitigate the impact of label noise in DL models trained on EHR data. Notably, it remains uncertain whether CV methods, when applied to the EHR domain, will prove effective, given the substantial divergence between the two domains. We present empirical evidence demonstrating that these methods, whether used individually or in combination, can substantially enhance model performance when applied to EHR data, especially in the presence of noisy/incorrect labels. We validate our methods and underscore their practical utility in real-world EHR data, specifically in the context of COVID-19 diagnosis. Our study highlights the effectiveness of CV methods in the EHR domain, making a valuable contribution to the advancement of healthcare analytics and research.

Participation and engagement in online cognitive testing

Web-based testing of cognitive abilities allows for large-scale assessments without geographical constraints. Yet, the extent to which it can reach populations beyond the typical demographic groups recruited for cognitive studies is unclear. This study focused on comparing the characteristics of individuals from a general population study (HUNT4) who chose to participate in a cognitive study (HUNT4-Hjernetrim) with those who did not. Additionally, we investigated participants' engagement and user experience. We obtained data on socio-demographics, health conditions (both physical and mental), self-reported cognitive or learning difficulties, and lifestyle factors of Hjernetrim participants and non-participants from the HUNT4 database. Hjernetrim involved 13 cognitive tests, administered through the online platform Memoro. We used logistic regressions to assess participation biases and linear regressions to assess participants' engagement and user experience. Of 65,851 HUNT4 participants invited via regular mail to Hjernetrim, 5634 (9.4%, aged 13-97, 54% women) participated. The best represented in the sample were 50-79-year-olds, women, tertiary educated, living alone, from urban areas, not occupationally active, and reporting memory complaints. Individuals who were aged 80+, had motor or vision impairments, and teenagers with learning disabilities, were underrepresented. Participants were more likely to have mental health problems, have or survived cancer and less likely to have cardiovascular disease. Participants logged on mainly during weekdays, the preferred time of day varied by age. On average, participants used 42 min and completed 78% of the tasks. Using PCs provided the most complete data. In terms of user experiences, 65% were positive while 14% were negative or reported technical difficulties. Overall, the study demonstrated that web-based methodology allowed for a relatively well-represented sample that included groups typically difficult to reach. The presence of somatic and mental diseases had a variable influence on participation. Participants finished most tests and reported positive experiences overall.

The iceberg model of suicidal ideation and behaviour in Danish adolescents: integration of national registry and self-reported data within a national birth cohort

Hospital records are used to identify suicide attempts in many countries but not all individuals present to hospital after a suicide attempt i.e., suggesting a 'hidden number'. Our aim was to present the prevalence of suicide ideation, plans, attempts, and suicides among Danish adolescents, including attempts not resulting in hospital contact. The study population consisted of participants in the Danish National Birth Cohort participating in an 18-year follow-up, with individual-level linkage to national register data. Prevalence was estimated with a variable with mutually exclusive categories ranging from no suicidality to self-reported suicide ideation, -plans, -attempt and hospital-recorded suicide attempt and stratified on sex and parental income. The 'hidden number' was estimated as the ratio between suicide attempts with and without hospital contact. Among 47 858 participants, all aged 18-years, 36% girls and 28% boys reported suicide ideation at least once in their life. In addition, 6% girls and 3% boys had either reported or been recorded with a suicide attempt. For every attempt recorded in the hospital setting, two girls (ratio, 1:2) and six boys (ratio, 1:6) reported having attempted suicide without hospital contact. The prevalence of any suicide attempt was 8% and 3% in the lowest and highest income group, respectively. Before age 18, 0·011% girls and 0·016% boys had died by suicide. In conclusion, suicidal ideation and behaviour are common in adolescents and there is a substantial 'hidden number' of adolescents with suicide attempt. These results emphasize the need for early age suicide preventive interventions in community-settings e.g., school environments.

Users' perception of quality as a driver of private healthcare use in Mexico: Insights from the People's Voice Survey

OBJECTIVE

The Mexican government has pursued multiple initiatives to improve healthcare coverage and financial protection. Yet, out-of-pocket health spending and use of private sector providers in Mexico remains high. In this paper, we sought to describe the characteristics of public and private healthcare users, describe recent visit quality across provider types, and to assess whether perceiving the public healthcare sector as poor quality is associated with private health sector use.

METHODS AND FINDINGS

We analyzed the cross-sectional People's Voice Survey conducted from December 2022 to January 2023. We used Chi-square tests to compare contextual, individual, and need-for-care factors and ratings of most recent visits between users of public (social security and other public providers) and private sector providers (stand-alone private providers and providers adjacent to pharmacies). We used a multivariable Poisson regression model to assess associations between low ratings of public healthcare sources and the use of private care. Among the 811 respondents with a healthcare visit in the past year, 31.2% used private sources. Private healthcare users were more educated and had higher incomes than public healthcare users. Quality of most recent visit was rated more highly in private providers (70.2% rating the visit as excellent or very good for stand-alone private providers and 54.3% for pharmacy-adjacent doctors) compared to social security (41.6%) and other public providers (46.6%). Those who perceived public health institutions as low quality had a higher probability of seeking private healthcare.

CONCLUSION

Users rated public care visits poorly relative to private care; at the population level, perceptions of poor quality care may drive private care use and hence out-of-pocket costs. Improving public healthcare quality is necessary to ensure universal health coverage.

Post-diagnostic multivitamin supplement use and colorectal cancer survival: A prospective cohort study

BACKGROUND

Use of multivitamin supplements has been associated with lower incidence of colorectal cancer (CRC). However, its influence on CRC survival remains unknown.

METHODS

Among 2424 patients with stage I-III CRC who provided detailed information about multivitamin supplements in the Nurses' Health Study and Health Professionals Follow-up Study, the authors calculated multivariable hazard ratios (HRs) of multivitamin supplements for all-cause and CRC-specific mortality according to post-diagnostic use and dose of multivitamin supplements.

RESULTS

During a median follow-up of 11 years, the authors documented 1512 deaths, among which 343 were of CRC. Compared to non-users, post-diagnostic users of multivitamin supplements at a dose of 3-5 tablets/week had lower CRC-specific mortality (HR, 0.55; 95% confidence interval [CI], 0.36-0.83, p = .005), and post-diagnostic users at doses of 3-5 and 6-9 tablets/week had lower all-cause mortality (HR, 0.81; 95% CI, 0.67-0.99, p = .04; HR, 0.79; 95% CI, 0.70-0.88), p < .001). The dose-response analysis showed a curvilinear relationship for both CRC-specific (pnonlinearity < .001) and all-cause mortality (pnonlinearity = .004), with the maximum risk reduction observed at 3-5 tablets/week and no further reduction at higher doses. Compared to non-users in both pre- and post-diagnosis periods, new post-diagnostic users at dose of <10 tablets/week had a lower all-cause mortality (HR, 0.81; 95% CI, 0.71-0.94, p = .005), whereas new users at a dose of ≥10 tablets/week (HR, 1.58; 95% CI, 1.07-2.33) and discontinued users (HR, 1.35; 95% CI, 1.14-1.59) had a higher risk of mortality.

CONCLUSIONS

Use of multivitamin supplements at a moderate dose after a diagnosis of nonmetastatic CRC is associated with lower CRC-specific and overall mortality, whereas a high dose (≥10 tablets/week) use is associated with higher CRC-specific mortality.

Sex differences in the association of pretransfusion haemoglobin and cognition in preterm infants

OBJECTIVES

To assess sex-specific differences in the association between pre-transfusion haemoglobin values and early neurodevelopmental function.

DESIGN

Observational follow-up of infants with birth weights <1000 g and gestational ages 22-28 weeks who were enrolled in the NICHD Neonatal Research Network Transfusion of Prematures (TOP) Trial at 19 U.S. sites, 2012-2017.

MAIN OUTCOME MEASURES

Pretransfusion haemoglobin values were obtained longitudinally through 36 weeks' postmenstrual age. The infant's mean pretransfusion haemoglobin was used as a marker of degree of anaemia (n=1655 measures). Measures of brain function were obtained at 22-26 months' corrected age using the Bayley Scales of Infant & Toddler Development, third edition (BSID-III) (n=1290 BSID-III scores). Sex-specific estimates for the linear relation between pretransfusion haemoglobin and BSID-III scores were obtained from repeated-measures regression analysis, adjusted for gestational age, birth weight, study site, clinical characteristics, and demographic covariates.

RESULTS

The relation of pretransfusion haemoglobin with 24-month BSID-III scores showed significant, independent interactions with both (1) sex (p=0.046) and (2) retinopathy of prematurity (ROP; p=0.004). In 614 males, BSID-III scores were higher by 1.07 points per g/dL (95% CI 1.58 to 4.33; p=0.008), not differing significantly among the three subscales (cognitive, language and motor; p=0.94). In 247 infants with ROP, BSID-III scores were higher by 2.95 points per g/dL (95% CI 0.28 to 1.87; p<0.0001), uniformly across subscales (p=0.73). These associations were non-significant in 676 females (p=0.96) and 1043 infants without ROP (p=0.81).

CONCLUSIONS

This study demonstrates sex-specific associations between mean pretransfusion haemoglobin (a marker of the severity of anaemia throughout the neonatal intensive care unit [NICU] hospitalisation) and early neurodevelopmental function at 22-26 months' corrected age.

Non-High-Density Lipoprotein Cholesterol Levels From Childhood to Adulthood and Cardiovascular Disease Events

Importance

Elevated non-high-density lipoprotein cholesterol (non-HDL-C; a recommended measure of lipid-related cardiovascular risk) is common in children and increases risk of adult cardiovascular disease (CVD). Whether resolution of elevated childhood non-HDL-C levels by adulthood is associated with reduced risk of clinical CVD events is unknown.

Objective

To examine the associations of non-HDL-C status between childhood and adulthood with incident CVD events.

Design, Setting, and Participants

Individual participant data from 6 prospective cohorts of children (mean age at baseline, 10.7 years) in the US and Finland. Recruitment took place between 1970 and 1996, with a final follow-up in 2019.

Exposures

Child (age 3-19 years) and adult (age 20-40 years) non-HDL-C age- and sex-specific z scores and categories according to clinical guideline-recommended cutoffs for dyslipidemia.

Main Outcomes and Measures

Incident fatal and nonfatal CVD events adjudicated by medical records.

Results

Over a mean length of follow-up of 8.9 years after age 40 years, 147 CVD events occurred among 5121 participants (60% women; 15% Black). Both childhood and adult non-HDL-C levels were associated with increased risk of CVD events (hazard ratio [HR], 1.42 [95% CI, 1.18-1.70] and HR, 1.50 [95% CI, 1.26-1.78] for a 1-unit increase in z score, respectively), but the association for childhood non-HDL-C was reduced when adjusted for adult levels (HR, 1.12 [95% CI, 0.89-1.41]). A complementary analysis showed that both childhood non-HDL-C levels and the change between childhood and adulthood were independently associated with the outcome, suggesting that from a preventive perspective, both childhood non-HDL-C levels and the change into adulthood are informative. Compared with those whose non-HDL-C levels remained within the guideline-recommended range in childhood and adulthood, participants who had incident non-HDL-C dyslipidemia from childhood to adulthood and those with persistent dyslipidemia had increased risks of CVD events (HR, 2.17 [95% CI, 1.00-4.69] and HR, 5.17 [95% CI, 2.80-9.56], respectively). Individuals who had dyslipidemic non-HDL-C in childhood but whose non-HDL-C levels were within the guideline-recommended range in adulthood did not have a significantly increased risk (HR, 1.13 [95% CI, 0.50-2.56]).

Conclusions and Relevance

Individuals with persistent non-HDL-C dyslipidemia from childhood to adulthood had an increased risk of CVD events, but those in whom dyslipidemic non-HDL-C levels resolve by adulthood have similar risk to individuals who were never dyslipidemic. These findings suggest that interventions to prevent and reduce elevated childhood non-HDL-C levels may help prevent premature CVD.

Adjust for non-ignorable panel attrition in the analysis of leaving the parental home

Leaving the parental home is an important life event that has received significant attention in the literature. Research on this topic relies heavily on panel data; however, panel data faces the issue of serious non-ignorable panel attrition associated with leaving the parental home. This paper addresses this issue using the UK Household Longitudinal Study (UKHLS) as a case study. It introduces an adjustment procedure that combines panel gap imputation via the next observation carried backward and inverse probability weighting based on the retrieved information about leaving the parental home. The results show that this adjustment method yields more precise model estimates for leaving the parental home, and after the adjustment, the positive marginal effects of age and living with non-biological parents, as well as the negative marginal effects of Asian ethnicity and regional house prices, become more pronounced. This adjustment method has the potential to be applied to address non-ignorable panel attrition associated with other events in different panel data.

Factors associated with PrEP-era HIV seroconversion in a 4-year U.S. national cohort of n = 6059 sexual and gender minority individuals who have sex with men, 2017-2022

INTRODUCTION

Community-based cohort studies of HIV seroconversion can identify important avenues for enhancing HIV prevention efforts in the era of pre-exposure prophylaxis (PrEP). Within individuals, one can assess exposure and outcome variables repeatedly and with increased certainty regarding temporal ordering. This cohort study examined the association of several risk factors with subsequent HIV seroconversion.

METHODS

We report data from a 4-year study (2017-2022) of 6059 HIV seronegative sexual and gender minority individuals who have sex with men who had indications for-, but were not using-, PrEP at enrolment. Participants completed repeat exposure assessments and self-collection of biospecimens for HIV testing. We examined the roles of race and ethnicity, socio-economic status, methamphetamine use and PrEP uptake over the course of follow-up in relation to HIV seroconversion.

RESULTS

Over 4 years, 303 of the participants seroconverted across 18,421 person-years (incidence rate = 1.64 [95% CI: 1.59-1.70] per 100 person-years). In multivariable discrete-time survival analysis, factors independently associated with elevated HIV seroconversion risk included being Black/African American (adjusted risk ratio [aRR]: 2.44, 1.79-3.28), Hispanic/Latinx (1.53, 1.19-1.96), housing instability (1.58, 1.22-2.05) and past year methamphetamine use (3.82, 2.74-5.33). Conversely, time since study enrolment (24 vs. 12 months, 0.67, 0.51-0.87; 36 months, 0.60, 0.45-0.80; 48 months, 0.48, 0.35-0.66) and higher education (master's degree or higher vs. less than or equal to high school, 0.36, 0.17-0.66) were associated with reduced seroconversion risk. Compared to non-PrEP users in the past 2 years without a current clinical indication, those who started PrEP but then discontinued had higher seroconversion risk, irrespective of clinical indication (3.23, 1.74-6.46) or lack thereof (4.30, 1.85-9.88). However, those who initiated PrEP in the past year (0.14, 0.04-0.39) or persistently used PrEP in the past 2 years (0.33, 0.14-0.74) had a lower risk of seroconversion. Of all HIV seroconversions observed during follow-up assessments (12, 24, 36 and 48 months), methamphetamine was reported in the 12 months prior 128 (42.2%) times (overall).

CONCLUSIONS

Interventions that acknowledge race and ethnicity, economic variables such as education and housing instability, and methamphetamine use are critically needed. Not only are interventions to engage individuals in PrEP care needed, but those that retain them, and re-engage those who may fall out of care are essential, given the exceptionally high risk of seroconversion in these groups.

Dietary Inflammatory Index and Magnetic Resonance Imaging-Detected Knee Structural Change and Pain: A 10.7-Year Follow-up Study

OBJECTIVE

To determine whether the dietary inflammatory index (DII) scores were associated with knee structural changes and pain over a 10.7-year follow-up.

METHODS

This study used data from a prospective population-based cohort study (mean age 63 years, 51% female) in which 1,099, 875, 768, and 566 participants completed assessments at baseline, 2.6, 5.1, and 10.7 years, respectively. T1-weighted and T2-weighted magnetic resonance imaging was performed to measure cartilage volume (CV) and bone marrow lesions (BMLs) at baseline and 10.7 years. The Western Ontario and McMaster Universities Osteoarthritis Index pain questionnaire was used to measure knee pain at each visit. Pain trajectories ("minimal pain," "mild pain," and "moderate pain") were previously identified. Baseline energy-adjusted DII (E-DII) scores were calculated. Linear, log-binomial regression, linear mixed-effects modeling, and multi-nominal logistic regression were used for analyses.

RESULTS

The mean ± SD E-DII score at baseline was -0.48 ± 1.39. In multivariable analyses, higher E-DII scores were not associated with tibial CV loss or BML size increase except for medial tibial BML size increase. Higher E-DII scores were associated with a higher pain score (β = 0.21; 95% confidence interval [CI] 0.004-0.43) and an increased risk of belonging to the "moderate pain" compared to the "minimal pain" trajectory group (relative risk ratio 1.19; 95% CI 1.02-1.39).

CONCLUSION

A proinflammatory diet, as indicated by a higher DII score, may be associated with a greater pain score and higher risk of more severe pain trajectory over 10 years. However, inconsistent findings related to structural changes suggest a discordance between the potential impact of diet on structural damage and pain in knee OA.

Ophthalmic quality of life in the adult Danish population: an epidemiological study

BACKGROUND

Ophthalmic quality of life (OQoL) has been investigated in selected parts of general populations and in patients with ocular disease, but OQoL in unselected general populations has not been studied in detail. The present study reports OQoL obtained from a representative sample of the adult Danish population 2020-2022.

METHODS

The FORSYN study invited 10 350 citizen representatives for the adult Danish population for a non-mydriatic eye examination and answer the National Eye Institute Visual Function Questionnaire with 39 items in the validated Danish translation. The results from the 3384 (32.7%) persons who participated in the study were weighted on the basis of relevant socio-economic factors, and data were projected to represent the total population. Binocular visual acuity was below 0.1 corresponding to legal blindness in 0.22% of this population.

RESULTS

OQoL was positively correlated with binocular visual acuity up to better than 93 ETDRS letters, negatively correlated with age for persons younger than 60 years of age and again positively correlated with age for persons older than 60 years. OQoL was negatively correlated with increasing ametropia and refractive error above 1 dioptre and encompassed more OQoL parameters for hyperopic than for myopic persons.

CONCLUSIONS

The study underlines the benefits of improving visual acuity even within the normal range and of adjusting uncorrected refraction errors in the general population. OQoL is positively correlated with age in older persons independently of visual acuity, sex, refractive power and previous cataract surgery.

A Principled Approach to Characterize and Analyze Partially Observed Confounder Data from Electronic Health Records

Objective

Partially observed confounder data pose challenges to the statistical analysis of electronic health records (EHR) and systematic assessments of potentially underlying missingness mechanisms are lacking. We aimed to provide a principled approach to empirically characterize missing data processes and investigate performance of analytic methods.

Methods

Three empirical sub-cohorts of diabetic SGLT2 or DPP4-inhibitor initiators with complete information on HbA1c, BMI and smoking as confounders of interest (COI) formed the basis of data simulation under a plasmode framework. A true null treatment effect, including the COI in the outcome generation model, and four missingness mechanisms for the COI were simulated: completely at random (MCAR), at random (MAR), and two not at random (MNAR) mechanisms, where missingness was dependent on an unmeasured confounder and on the value of the COI itself. We evaluated the ability of three groups of diagnostics to differentiate between mechanisms: 1)-differences in characteristics between patients with or without the observed COI (using averaged standardized mean differences [ASMD]), 2)-predictive ability of the missingness indicator based on observed covariates, and 3)-association of the missingness indicator with the outcome. We then compared analytic methods including "complete case", inverse probability weighting, single and multiple imputation in their ability to recover true treatment effects.

Results

The diagnostics successfully identified characteristic patterns of simulated missingness mechanisms. For MAR, but not MCAR, the patient characteristics showed substantial differences (median ASMD 0.20 vs 0.05) and consequently, discrimination of the prediction models for missingness was also higher (0.59 vs 0.50). For MNAR, but not MAR or MCAR, missingness was significantly associated with the outcome even in models adjusting for other observed covariates. Comparing analytic methods, multiple imputation using a random forest algorithm resulted in the lowest root-mean-squared-error.

Conclusion

Principled diagnostics provided reliable insights into missingness mechanisms. When assumptions allow, multiple imputation with nonparametric models could help reduce bias.

A new and unified method for regression analysis of interval-censored failure time data under semiparametric transformation models with missing covariates

This paper discusses regression analysis of interval-censored failure time data arising from semiparametric transformation models in the presence of missing covariates. Although some methods have been developed for the problem, they either apply only to limited situations or may have some computational issues. Corresponding to these, we propose a new and unified two-step inference procedure that can be easily implemented using the existing or standard software. The proposed method makes use of a set of working models to extract partial information from incomplete observations and yields a consistent estimator of regression parameters assuming missing at random. An extensive simulation study is conducted and indicates that it performs well in practical situations. Finally, we apply the proposed approach to an Alzheimer's Disease study that motivated this study.

Health and Wellbeing of Regional and Rural Australian Healthcare Workers during the COVID-19 Pandemic: Baseline Cross-Sectional Findings from the Loddon Mallee Healthcare Worker COVID-19 Study-A Prospective Cohort Study

BACKGROUND

Coronavirus 19 (COVID-19) has created complex pressures and challenges for healthcare systems worldwide; however, little is known about the impacts COVID-19 has had on regional/rural healthcare workers. The Loddon Mallee Healthcare Worker COVID-19 Study (LMHCWCS) cohort was established to explore and describe the immediate and long-term impacts of the COVID-19 pandemic on regional and rural healthcare workers.

METHODS

Eligible healthcare workers employed within 23 different healthcare organisations located in the Loddon Mallee region of Victoria, Australia, were included. In this cohort study, a total of 1313 participants were recruited from November 2020-May 2021. Symptoms of depression, anxiety, post-traumatic stress, and burnout were measured using the Patient Health Questionnaire-9 (PHQ-9), Generalised Anxiety Disorder-7 (GAD-7), Impact of Events Scale-6 (IES-6), and Copenhagen Burnout Inventory (CBI), respectively. Resilience and optimism were measured using the Brief Resilience Scale and Life Orientation Test-Revised (LOT-R), respectively. Subjective fear of COVID-19 was measured using the Fear of COVID-19 Scale.

RESULTS

These cross-sectional baseline findings demonstrate that regional/rural healthcare workers were experiencing moderate/severe depressive symptoms (n = 211, 16.1%), moderate to severe anxiety symptoms (n = 193, 14.7%), and high personal or patient/client burnout with median total scores of 46.4 (IQR = 28.6) and 25.0 (IQR = 29.2), respectively. There was a moderate degree of COVID-19-related fear. However, most participants demonstrated a normal/high degree of resilience (n = 854, 65.0%). Based on self-reporting, 15.4% had a BMI from 18.5 to 24.9 kgm2 and 37.0% have a BMI of 25 kgm2 or over. Overall, 7.3% of participants reported they were current smokers and 20.6% reported alcohol consumption that is considered moderate/high-risk drinking. Only 21.2% of the sample reported consuming four or more serves of vegetables daily and 37.8% reported consuming two or more serves of fruit daily. There were 48.0% the sample who reported having poor sleep quality measured using the Pittsburgh Sleep Quality Index (PSQI).

CONCLUSION

Regional/rural healthcare workers in Victoria, Australia, were experiencing a moderate to high degree of psychological distress during the early stages of the pandemic. However, most participants demonstrated a normal/high degree of resilience. Findings will be used to inform policy options to support healthcare workers in responding to future pandemics.

Invited Commentary: Mixing multiple imputation and bootstrapping for variance estimation

In 2023, Martinez et al. examined trends in the inclusion, conceptualization, operationalization and analysis of race and ethnicity among studies published in US epidemiology journals. Based on a random sample of papers (N=1,050) published from 1995-2018, the authors describe the treatment of race, ethnicity, and ethnorace in the analytic sample (N=414, 39% of baseline sample) over time. Between 32% and 19% of studies in each time stratum lacked race data; 61% to 34% lacked ethnicity data. The review supplies stark evidence of the routine omission and variability of measures of race and ethnicity in epidemiologic research. Informed by public health critical race praxis (PHCRP), this commentary discusses the implications of four problems the findings suggest pervade epidemiology: 1) a general lack of clarity about what race and ethnicity are; 2) the limited use of critical race or other theory; 3) an ironic lack of rigor in measuring race and ethnicity; and, 4) the ordinariness of racism and white supremacy in epidemiology. The identified practices reflect neither current publication guidelines nor the state of the knowledge on race, ethnicity and racism; therefore, we conclude by offering recommendations to move epidemiology toward more rigorous research in an increasingly diverse society.

Health-related quality of life and impact of socioeconomic status among primary and secondary school students after the third COVID-19 wave in Berlin, Germany

In the earlier phases of the COVID-19 pandemic, studies in Germany and elsewhere found an overall reduction in health-related quality of life (HRQoL) among students. However, there is little evidence on later pandemic stages as well as socioeconomic influencing factors. We aimed to (1) describe HRQoL in a Berlin student cohort at two time points in mid-2021, and to (2) analyze the effects of household income and education. We assessed HRQoL of students from 24 randomly selected primary and secondary schools in Berlin, Germany, with the KIDSCREEN-10 index in June and September 2021. To adjust for non-response bias, inverse probability weighting was applied. The potential effects of both household income and education (lower vs. higher) were estimated in generalized linear mixed models, based on prior assumptions presented in directed acyclic graphs. Our cohort comprised 660 students aged 7-19 years. In June 2021, 11.3% [95% CI = 9.0% - 14.0%] reported low HRQoL, whereas in September 2021, this increased to 13.7% [95% CI = 11.1% - 16.5%], with adolescent girls more frequently reporting low HRQoL at both time points (20% [95% CI = 17.1% - 23.3%] and 29% [95% CI = 25.5% - 32.5%]) compared to boys and younger children. While there was no statistically significant total effect of lower household income on HRQoL, a negative effect of lower household education was statistically significant (β = -2.15, SE 0.95, 95% CI = -4.01 to -0.29, p = 0.024). In summary, students' HRQoL in mid-2021 was better than that documented in other studies conducted at pandemic onset using KIDSCREEN-10. Female adolescents reported low HRQoL more often, and lower household education significantly reduced children's HRQoL. Support strategies for psychosocial wellbeing should consider socioeconomically disadvantaged children as important target groups.

Natural history of depression up to 18 years after stroke: a population-based South London Stroke Register study

Background

Current evidence on the long-term natural history of post-stroke depression (PSD) is limited. We aim to determine the prevalence, incidence, duration and recurrence rates of depression to 18-years after stroke and assess differences by onset-time and depression severity.

Methods

Data were from the South London Stroke Register (1995-2019, N = 6641 at registration). Depression was defined using the Hospital Anxiety and Depression scale (scores > 7 = depression) at 3-months, then annually to 18-years after stroke. We compared early- (3-months post-stroke) vs late-onset depression (1-year) and initial mild (HADS scores > 7) vs severe depression (scores > 10).

Findings

3864 patients were assessed for depression at any time-points during the follow-up (male:55.4% (2141), median age: 68.0 (20.4)), with the number ranging from 2293 at 1-year to 145 at 18-years after stroke. Prevalence of PSD ranged from 31.3% (28.9-33.8) to 41.5% (33.6-49.3). The cumulative incidence of depression was 59.4% (95% CI 57.8-60.9), of which 87.9% (86.5-89.2) occurred within 5-years after stroke. Of patients with incident PSD at 3-months after stroke, 46.6% (42.1-51.2) recovered after 1 year. Among those recovered, 66.7% (58.0-74.5) experienced recurrent depression and 94.4% (87.5-98.2) of recurrences occurred within 5-years since recovery. Similar estimates were observed in patients with PSD at 1-year. 34.3% (27.9-41.1) of patients with severe depression had recovered at the next time-point, compared to 56.7% (50.5-62.8) with mild depression. Recurrence rate at 1-year after recovery was higher in patients with severe depression (52.9% (35.1-70.2)) compared to mild depression (23.5% (14.1-35.4)) (difference: 29.4% (7.6-51.2), p = 0.003).

Interpretation

Long-term depressive status may be established by 5-years post-onset. Early- and late-onset depression presented similar natural history, while severe depression had a longer duration and quicker recurrence than mild depression. These estimates were limited to alive patients completing the depression assessment, who tended to have less severe stroke than excluded patients, so may be underestimated and not generalizable to all stroke survivors.

Funding

National Institute for Health and Care Research (NIHR202339).

Association of Healthy Lifestyle and Incident Polypharmacy

BACKGROUND

Polypharmacy, commonly defined as taking ≥5 medications, is an undesirable state associated with lower quality of life. Strategies to prevent polypharmacy may be an important priority for patients. We sought to examine the association of healthy lifestyle, a modifiable risk factor, with incident polypharmacy.

METHODS

We performed a secondary analysis of the REasons for Geographic and Racial Differences in Stroke (REGARDS) cohort study, including 15,478 adults aged ≥45 years without polypharmacy at baseline. The primary exposure was healthy lifestyle at baseline as measured by the Healthy Behavior Score (HBS), a cumulative assessment of diet, exercise frequency, tobacco smoking, and sedentary time. HBS ranges from 0-8, whereby 0-2 indicates low HBS, 3-5 indicates moderate HBS, and 6-8 indicates high HBS. We used multinomial logistic regression to examine the association between HBS and incident polypharmacy, survival without polypharmacy, and death.

RESULTS

Higher HBS (i.e., healthier lifestyle) was inversely associated with incident polypharmacy after adjusting for sociodemographic and baseline health variables. Compared with participants with low HBS, those with moderate HBS had lower odds of incident polypharmacy (odds ratio [OR] 0.85; 95% confidence interval [CI], 0.73-0.98) and lower odds of dying (OR 0.74; 95% CI, 0.65-0.83). Participants with high HBS had even lower odds of both incident polypharmacy (OR 0.75; 95% CI, 0.64-0.88) and death (OR 0.62; 95% CI, 0.54-0.70). There was an interaction for age, where the association between HBS and incident polypharmacy was most pronounced for participants aged ≤65 years.

CONCLUSIONS

Healthier lifestyle was associated with lower risk for incident polypharmacy.

Ultra-processed food consumption and mortality among patients with stages I-III colorectal cancer: a prospective cohort study

Background

Ultra-processed foods (UPFs) are emerging as a risk factor for colorectal cancer (CRC), yet how post-diagnostic UPF intake may impact CRC prognosis remains unexplored.

Methods

Data collected from food frequency questionnaires were used to estimate intakes of total UPFs and UPF subgroups (serving/d) at least 6 months but less than 4 years post-diagnosis among 2498 patients diagnosed with stages I-III CRC within the Nurses' Health Study and Health Professionals Follow-up Study during 1980-2016. Hazard ratios (HR) and 95% confidence intervals (CIs) of all-cause, CRC- and cardiovascular disease (CVD)-specific mortality in association with UPF consumption were estimated using an inverse probability weighted multivariable Cox proportional hazards regression model, adjusted for confounders.

Findings

The mean (SD) age of patients at diagnosis was 68.5 (9.4) years. A total of 1661 deaths were documented, including 321 from CRC and 335 from CVD. Compared to those in the lowest quintile (median = 3.6 servings/d), patients in the highest quintile (median = 10 servings/d) of post-diagnostic UPF intake had higher CVD mortality (HR = 1.65, 95% CI = 1.13-2.40) but not CRC or all-cause mortality. Among UPF subgroups, higher consumption of fats/condiments/sauces was associated with a higher risk of CVD-specific mortality (highest vs. lowest quintile of intake, HR = 1.96, 95% CI = 1.41-2.73), and higher intake of ice cream/sherbet was associated with an increased risk of CRC-specific mortality (highest vs. lowest quintile, HR = 1.86, 95% CI: 1.33-2.61). No statistically significant association was found between UPF subgroups and overall mortality.

Interpretation

Higher post-diagnostic intake of total UPFs and fats/condiments/sauces in CRC survivors is associated with higher CVD mortality, and higher ice cream/sherbet intake is linked to higher CRC mortality.

Funding

US National Institutes of Health and the American Cancer Society.

Acetaminophen Use During Pregnancy and Children's Risk of Autism, ADHD, and Intellectual Disability

Importance

Several studies suggest that acetaminophen (paracetamol) use during pregnancy may increase risk of neurodevelopmental disorders in children. If true, this would have substantial implications for management of pain and fever during pregnancy.

Objective

To examine the associations of acetaminophen use during pregnancy with children's risk of autism, attention-deficit/hyperactivity disorder (ADHD), and intellectual disability.

Design, Setting, and Participants

This nationwide cohort study with sibling control analysis included a population-based sample of 2 480 797 children born in 1995 to 2019 in Sweden, with follow-up through December 31, 2021.

Exposure

Use of acetaminophen during pregnancy prospectively recorded from antenatal and prescription records.

Main Outcomes and Measures

Autism, ADHD, and intellectual disability based on International Classification of Diseases, Ninth Revision and International Classification of Diseases, Tenth Revision codes in health registers.

Results

In total, 185 909 children (7.49%) were exposed to acetaminophen during pregnancy. Crude absolute risks at 10 years of age for those not exposed vs those exposed to acetaminophen were 1.33% vs 1.53% for autism, 2.46% vs 2.87% for ADHD, and 0.70% vs 0.82% for intellectual disability. In models without sibling control, ever-use vs no use of acetaminophen during pregnancy was associated with marginally increased risk of autism (hazard ratio [HR], 1.05 [95% CI, 1.02-1.08]; risk difference [RD] at 10 years of age, 0.09% [95% CI, -0.01% to 0.20%]), ADHD (HR, 1.07 [95% CI, 1.05-1.10]; RD, 0.21% [95% CI, 0.08%-0.34%]), and intellectual disability (HR, 1.05 [95% CI, 1.00-1.10]; RD, 0.04% [95% CI, -0.04% to 0.12%]). To address unobserved confounding, matched full sibling pairs were also analyzed. Sibling control analyses found no evidence that acetaminophen use during pregnancy was associated with autism (HR, 0.98 [95% CI, 0.93-1.04]; RD, 0.02% [95% CI, -0.14% to 0.18%]), ADHD (HR, 0.98 [95% CI, 0.94-1.02]; RD, -0.02% [95% CI, -0.21% to 0.15%]), or intellectual disability (HR, 1.01 [95% CI, 0.92-1.10]; RD, 0% [95% CI, -0.10% to 0.13%]). Similarly, there was no evidence of a dose-response pattern in sibling control analyses. For example, for autism, compared with no use of acetaminophen, persons with low (<25th percentile), medium (25th-75th percentile), and high (>75th percentile) mean daily acetaminophen use had HRs of 0.85, 0.96, and 0.88, respectively.

Conclusions and Relevance

Acetaminophen use during pregnancy was not associated with children's risk of autism, ADHD, or intellectual disability in sibling control analysis. This suggests that associations observed in other models may have been attributable to familial confounding.

Incidence of long COVID and associated psychosocial characteristics in a large U.S. city

PURPOSE

Persistent residual effects from Coronavirus Disease-2019 (COVID-19) have been observed with varying definitions of "Long COVID" and little comprehensive examination. This study examined the incidence and psychosocial correlates of Long COVID using different definitions.

METHODS

Data were analyzed from a citywide sample of 3595 adults with lab-confirmed cases of COVID-19 that were surveyed over 3 months. Rates of Long COVID were examined in terms of Post-Acute COVID (PAC), defined as at least one symptom lasting for 4 weeks, and three levels of Post-COVID Syndrome (PCS) that included experiencing at least one symptom for 3 months (PCS-1), experiencing three or more symptoms for 3 months (PCS-2), or experiencing at least one of the same symptoms for 3 months (PCS-3).

RESULTS

Among the 686 participants who completed baseline, 1-month, and 3-month follow-up assessments, 75.7% had PAC, 55.0% had PSC-1, 26.5% had PSC-2, and 19.0% had PSC-3. Comparing participants with PAC and PSC-3 in the total sample with inverse probability weighting, multivariable analyses revealed being female, Asian or Native American, greater reported longlines, and less social support were predictive of PCS-3.

CONCLUSION

Residual effects of COVID-19 are very common and nearly one-fifth of our sample met the most restrictive definition of Long COVID warranting concern as a public health issue. Some demographic and social factors may predispose some adults to Long COVID, which should be considered for prevention and population health.