Tracking of Maternal Physical Activity and Sport Participation over 11 Years: Findings from the Czech ELSPAC Study

Purpose: Tracking of physical activity (PA) and sport participation (SP) during motherhood is poorly understood. The purpose of the study was to analyze the extent of tracking of maternal PA and SP. Methods: In this investigation, data were collected from the Czech ELSPAC study subsample of 4811 and 2609 women measured postnatally (1991–1992) and after 11 years of follow-up (2002–2003), respectively. The structured questionnaire was used to assess the participation and average weekly time spent in PA, and the frequency of engaging in different sports (running, cycling, strength training, racket sports, swimming, and team sports). Tracking was calculated using generalized estimating equations (GEE) with beta coefficients (β), odds ratios (ORs), and 95% confidence intervals (95% CI). Results: Moderately high tracking coefficients were observed for cycling (β = 0.69, 95% 0.67–0.72), strength training (β = 0.59, 95% 0.56–0.63), and weekly time spent in PA (β = 0.53, 95% 0.38–0.66); meanwhile, moderate tracking coefficients were generated for swimming (β = 0.48, 95% 0.44–0.52), team sports (β = 0.44, 95% 0.39–0.48), racket sports (β = 0.44, 95% 0.39–0.48), and running (β = 0.35, 95% 0.30–0.40). Mothers who did not participate in PA at baseline were 81% more likely not to participate in it at follow-up (OR = 1.81, 95% CI 1.53–2.13). Conclusion: Cycling- and strength-related activities and weekly PA were tracked moderately-to-moderately high during motherhood. Moreover, the strong tracking of physical inactivity indicates that the detection of this risk factor before pregnancy should be advocated.

Mean corrected generalized estimating equations for longitudinal binary outcomes with report bias

Cocaine addiction is an important public health problem worldwide. Cognitive-behavioral therapy is a counseling intervention for supporting cocaine-dependent individuals through recovery and relapse prevention. It may reduce patients’ cocaine uses by improving their motivations and enabling them to recognize risky situations. To study the effect of cognitive behavioral therapy on cocaine dependence, the self-reported cocaine use with urine test data were collected at the Primary Care Center of Yale-New Haven Hospital. Its outcomes are binary, including both the daily self-reported drug uses and weekly urine test results. To date, the generalized estimating equations are widely used to analyze binary data with repeated measures. However, due to the existence of significant self-report bias in the self-reported cocaine use with urine test data, a direct application of the generalized estimating equations approach may not be valid. In this paper, we proposed a novel mean corrected generalized estimating equations approach for analyzing longitudinal binary outcomes subject to reporting bias. The mean corrected generalized estimating equations can provide consistently and asymptotically normally distributed estimators under true contamination probabilities. In the self-reported cocaine use with urine test study, accurate weekly urine test results are used to detect contamination. The superior performances of the proposed method are illustrated by both simulation studies and real data analysis.

Concordance correlation coefficients estimated by modified variance components and generalized estimating equations for longitudinal overdispersed Poisson data

For longitudinal overdispersed Poisson data sets, estimators of the intra-, inter-, and total concordance correlation coefficient through variance components have been proposed. However, biased estimators of quadratic forms are used in concordance correlation coefficient estimation. In addition, the generalized estimating equations approach has been used in estimating agreement for longitudinal normal data and not for longitudinal overdispersed Poisson data. Therefore, this paper proposes a modified variance component approach to develop the unbiased estimators of the concordance correlation coefficient for longitudinal overdispersed Poisson data. Further, the indices of intra-, inter-, and total agreement through generalized estimating equations are also developed considering the correlation structure of longitudinal count repeated measurements. Simulation studies are conducted to compare the performance of the modified variance component and generalized estimating equation approaches for longitudinal Poisson and overdispersed Poisson data sets. An application of corticospinal diffusion tensor tractography study is used for illustration. In conclusion, the modified variance component approach performs outstandingly well with small mean square errors and nominal 95% coverage rates. The generalized estimating equation approach provides in model assumption flexibility of correlation structures for repeated measurements to produce satisfactory concordance correlation coefficient estimation results.

Lipids Changes After Switch from TDF to TAF in the OPERA Cohort: LDL-cholesterol and Triglycerides

Background Increases in lipids have been observed in people living with HIV (PLWH) switching from tenofovir disoproxil fumarate (TDF) to tenofovir alafenamide (TAF). We assessed changes in low-density lipoprotein cholesterol (LDL-C) and triglycerides (TG) following a switch from TDF to TAF. Methods Adults with ≥1 lipid measure before and after switch from TDF-to-TAF were identified in the OPERA® cohort. Multivariable linear regression using generalized estimating equations were used to estimate predicted changes in lipids over time on TAF, modeled flexibly with linear splines. Results 6,451 PLWH switched from TDF-to-TAF, of whom 4,328 maintained all other agents. LDL-C increased significantly by 1.40 mg/dL/month over the first 3 months on TAF, by 0.33 mg/dL/month between 3-9 months and plateaued beyond 9 months. TG increased significantly by 3.52 mg/dL/month over the first 3 months of TAF, by 0.91 mg/mL/month between 3-9 months, and by 0.72 mg/mL/month between 9-16 months but decreased thereafter. Similar patterns were observed in analyses restricted to PLWH who switched from TDF-to-TAF but maintained all other agents. Discussion TDF-to-TAF switch was associated with LDL-C and TG increases over the first 9 to 16 months on TAF. The dynamic patterns observed cannot be attributed to changes in other agents.

Power and sample size calculations for cluster randomized trials with binary outcomes when intracluster correlation coefficients vary by treatment arm

Background/Aims Generalized estimating equations are commonly used to fit logistic regression models to clustered binary data from cluster randomized trials. A commonly used correlation structure assumes that the intracluster correlation coefficient does not vary by treatment arm or other covariates, but the consequences of this assumption are understudied. We aim to evaluate the effect of allowing variation of the intracluster correlation coefficient by treatment or other covariates on the efficiency of analysis and show how to account for such variation in sample size calculations. Methods We develop formulae for the asymptotic variance of the estimated difference in outcome between treatment arms obtained when the true exchangeable correlation structure depends on the treatment arm and the working correlation structure used in the generalized estimating equations analysis is: (i) correctly specified, (ii) independent, or (iii) exchangeable with no dependence on treatment arm. These formulae require a known distribution of cluster sizes; we also develop simplifications for the case when cluster sizes do not vary and approximations that can be used when the first two moments of the cluster size distribution are known. We then extend the results to settings with adjustment for a second binary cluster-level covariate. We provide formulae to calculate the required sample size for cluster randomized trials using these variances. Results We show that the asymptotic variance of the estimated difference in outcome between treatment arms using these three working correlation structures is the same if all clusters have the same size, and this asymptotic variance is approximately the same when intracluster correlation coefficient values are small. We illustrate these results using data from a recent cluster randomized trial for infectious disease prevention in which the clusters are groups of households and modest in size (mean 9.6 individuals), with intracluster correlation coefficient values of 0.078 in the control arm and 0.057 in an intervention arm. In this application, we found a negligible difference between the variances calculated using structures (i) and (iii) and only a small increase (typically [Formula: see text]) for the independent correlation structure (ii), and hence minimal effect on power or sample size requirements. The impact may be larger in other applications if there is greater variation in the ICC between treatment arms or with an additional covariate. Conclusion The common approach of fitting generalized estimating equations with an exchangeable working correlation structure with a common intracluster correlation coefficient across arms likely does not substantially reduce the power or efficiency of the analysis in the setting of a large number of small or modest-sized clusters, even if the intracluster correlation coefficient varies by treatment arm. Our formulae, however, allow formal evaluation of this and may identify situations in which variation in intracluster correlation coefficient by treatment arm or another binary covariate may have a more substantial impact on power and hence sample size requirements.

A comparative study of R functions for clustered data analysis

Background Clustered or correlated outcome data is common in medical research studies, such as the analysis of national or international disease registries, or cluster-randomized trials, where groups of trial participants, instead of each trial participant, are randomized to interventions. Within-group correlation in studies with clustered data requires the use of specific statistical methods, such as generalized estimating equations and mixed-effects models, to account for this correlation and support unbiased statistical inference. Methods We compare different approaches to estimating generalized estimating equations and mixed effects models for a continuous outcome in R through a simulation study and a data example. The methods are implemented through four popular functions of the statistical software R, “geese”, “gls”, “lme”, and “lmer”. In the simulation study, we compare the mean squared error of estimating all the model parameters and compare the coverage proportion of the 95% confidence intervals. In the data analysis, we compare estimation of the intervention effect and the intra-class correlation. Results In the simulation study, the function “lme” takes the least computation time. There is no difference in the mean squared error of the four functions. The “lmer” function provides better coverage of the fixed effects when the number of clusters is small as 10. The function “gls” produces close to nominal scale confidence intervals of the intra-class correlation. In the data analysis and the “gls” function yields a positive estimate of the intra-class correlation while the “geese” function gives a negative estimate. Neither of the confidence intervals contains the value zero. Conclusions The “gls” function efficiently produces an estimate of the intra-class correlation with a confidence interval. When the within-group correlation is as high as 0.5, the confidence interval is not always obtainable.

Worse Self-Reported Hearing Ability Is Associated With Greater Perceived Physical and Mental Fatigability

Older adults with hearing loss often report higher fatigue due to effortful listening. We evaluated whether self-reported hearing ability is associated with perceived physical and mental fatigability (a more sensitive measure than fatigue) using the Pittsburgh Fatigability Scale (PFS). Older adults (N=2,558) from the Long Life Family Study Visit 2 (71.5±11.4 years; 54.8% women) completed PFS and self-reported hearing ability (worse=[fair,poor,very poor,deaf] or better=[good, excellent]). Age-adjusted PFS Physical and Mental scores were 2.3 and 2.5 lower, respectively, for worse vs. better hearing (p<.0001). Generalized estimating equations adjusted for family-relatedness, site, age, sex, cognitive function (Mini-Mental State Examination), education, and self-reported health. Compared to individuals with better hearing, those with worse hearing had a 42% and 44% greater odds of physical (≥15) (CI:1.12-1.80,p=0.0042) and mental(≥13) (CI:1.13-1.84,p=0.0034) fatigability, respectively. These observed associations may potentially be explained via complex psychosocial and cognitive aging pathways (e.g. effortful listening) to be examined in future work.

Greater Adolescent Cognitive Ability Linked to Lower Risk of Earlier Mortality

There have been few investigations of the role that adolescent cognitive ability might play in predicting physical resilience across the life course, including decreased risk of early mortality. Our limited knowledge of how multiple cognitive ability domains shape trajectories of longevity is due, in part, to a lack of aging cohorts with early life cognitive assessments, and family data that allow for examination of shared family and genetic characteristics that may play a role in cognitive ability-health links. We capitalized on data from the 1960 Project Talent high school cohort (n>360,000, born 1942-1946) and mortality data (n=22,584; 5,497 deceased) collected as part of two recent follow-ups, the Project Talent Twin & Sibling Study and the Project Talent Aging Study, to examine these potential associations. In 1960, ability was assessed in multiple cognitive domains (e.g., general aptitude, quantitative, reasoning). Mortality status was ascertained through 2016. Binary logistic generalized estimating equations with race, age, sex, and adolescent family SES covariates, indicated that each 1 standard deviation higher ability in multiple cognitive domains in adolescence predicted lower odds of earlier mortality (ORs of 0.79 - 0.87). Co-sibling control models indicated a similar pattern, suggesting that benefits associated with higher cognitive performance do not simply reflect shared environmental and genetic background, but may represent a direct protective effect. These findings indicate that better performance in multiple cognitive domains in adolescence, above and beyond the influence of genetic and family environmental factors, may be or point to modifiable protective factors against risk of early mortality.

Examining the Associations Between Online Interactions and Momentary Affect in Depressed Adolescents

Concern has been raised over parallel increases in youth depression and online interactive media use over the past two decades. The aim of this study was to determine whether online interactions are associated with users’ affective states. Using ecological momentary assessment, we measured depressed adolescents’ momentary affect during and residual feelings following online interactions with offline friends and family, online friends, and acquaintances/strangers. We found that depressed adolescents use texting services and social networking sites to interact online, most frequently with offline friends and family, followed by online friends. Results of generalized estimating equations showed associations between negative affect and digital interactions with offline friends and family. Participants were less likely to report feeling better after interacting with online friends than after interacting with any other relationship type. Our findings highlight the heterogeneity of depressed adolescents’ online interactions and suggest that their affective experience varies depending on the nature of the relationships they have with those with whom they interact.