Intracluster Correlation Research Articles

Assessing treatment effect heterogeneity in the presence of missing effect modifier data in cluster-randomized trials.

Understanding whether and how treatment effects vary across subgroups is crucial to inform clinical practice and recommendations. Accordingly, the assessment of heterogeneous treatment effects based on pre-specified potential effect modifiers has become a common goal in modern randomized trials. However, when one or more potential effect modifiers are missing, complete-case analysis may lead to bias and under-coverage. While statistical methods for handling missing data have been proposed and compared for individually randomized trials with missing effect modifier data, few guidelines exist for the cluster-randomized setting, where intracluster correlations in the effect modifiers, outcomes, or even missingness mechanisms may introduce further threats to accurate assessment of heterogeneous treatment effect. In this article, the performance of several missing data methods are compared through a simulation study of cluster-randomized trials with continuous outcome and missing binary effect modifier data, and further illustrated using real data from the Work, Family, and Health Study. Our results suggest that multilevel multiple imputation and Bayesian multilevel multiple imputation have better performance than other available methods, and that Bayesian multilevel multiple imputation has lower bias and closer to nominal coverage than standard multilevel multiple imputation when there are model specification or compatibility issues.

Sample size estimation for stratified cluster randomization trial with survivalendpoint.

Cluster randomization trials with survival endpoint are predominantly used in drug development and clinical care research when drug treatments or interventions are delivered at a group level. Unlike conventional cluster randomization design, stratified cluster randomization design is generally considered more effective in reducing the impacts of imbalanced baseline prognostic factors and varying cluster sizes between groups when these stratification factors are adopted in the design. Failure to account for stratification and cluster size variability may lead to underpowered analysis and inaccurate sample size estimation. Apart from the sample size estimation in unstratified cluster randomization trials, there are no development of an explicit sample size formula for survival endpoint when a stratified cluster randomization design is employed. In this article, we present a closed-form sample size formula based on the stratified cluster log-rank statistics for stratified cluster randomization trials with survival endpoint. It provides an integrated solution for sample size estimation that account for cluster size variation, baseline hazard heterogeneity, and the estimated intracluster correlation coefficient based on the preliminary data. Simulation studies show that the proposed formula provides the appropriate sample size for achieving the desired statistical power under various parameter configurations. A real example of a stratified cluster randomization trial in the population with stable coronary heart disease is presented to illustrate our method.

Abstract 4846: Geographic and physician-level variation in the use of hypofractionated radiotherapy for breast cancer in the U.S.: A cross-classified multilevel analysis

Abstract Hypofractionated whole-breast irradiation (HF-WBI), a radiotherapy with a larger dose per fraction and a shorter schedule for early-stage breast cancer patients post breast-conserving surgery, shows equivalent safety and efficacy compared to conventionally fractionated whole-breast irradiation (CF-WBI) regarding local recurrence and disease-free survival. While several countries have set HF-WBI as the preferred standard of care for these patients, its uptake in the U.S. has lagged behind expectations. We used the Health Care Cost Institute (HCCI) database 2008-2017, covering a third of the U.S. employer-sponsored insurance population, to identify early-stage breast cancer patients. Patients were clustered at different geographic levels (Census region, state, core-based statistical area (CBSA), hospital referral region of Dartmouth Atlas, and zip code) and the radiation oncologist level. These two levels are cross-nested. We used a Bayesian cross-classified multilevel logistic model to simultaneously model the geographic units variability and physician-level variability in HF-WBI use across the U.S. Variation metrics included intracluster correlation coefficient (ICC) and median odds ratios (MOR). ICC with a value of 0 is equivalent to MOR of 1, indicating identical outcomes across all levels. After accounting for these variations in the cross-classified multilevel model framework, we evaluated the association between HF-WBI use and clinical and demographic factors.Our study included 79,747 women (74.0%) who underwent CF-WBI and 27,999 (26.0%) who received HF-WBI. HF-WBI adoption increased over time (2008-2017). Variability across radiation oncologists (ICC = 0.272, MOR = 2.88) was notably larger than geographic areas variability. Variability across CBSA (ICC = 0.097, MOR = 1.76) was the strongest among all the geographic levels. The variation in HF-WBI use was mainly attributed to physician-level variability (ICC=0.222, MOR=2.59) based on the cross-classified multilevel model. No substantial differences were found between young (aged 64 or younger with commercial health plans) and older patients (aged 65 or older with Medicare Advantage plans) in terms of both geographic and physician-level variations. After accounting for variability in radiation oncologists and CBSAs, older age, non-reception of chemotherapy, and several community-level factors, including longer distance from home to facility and higher community education level, were associated with a higher likelihood of HF-WBI use. These results may lend insights into potential facilitators/barriers to HF-WBI uptake. Future studies can further investigate the impact of the 2018 American Society for Radiation Oncology (ASTRO) guideline and the Covid-19 pandemic on HF-WBI use, which would illuminate potential improvements of healthcare practices. Citation Format: Yijia Sun, Loren Saulsberry, Chuanhong Liao, Donald Hedeker, Dezheng Huo. Geographic and physician-level variation in the use of hypofractionated radiotherapy for breast cancer in the U.S.: A cross-classified multilevel analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4846.

The Validity and Reliability of the Persian Version of the Xerostomia Inventory Questionnaire.

Reduced saliva production or changes in the quality of saliva are major causes of xerostomia as a perceptual disease. This study aimed to measure validity and reliability of the Persian version of the xerostomia inventory (XI) questionnaire. This cross-sectional study was performed in Kerman in 2021. The sample size for this study was 100 people. To test the discriminant validity of XI, 50 healthy people (control group) and 50 people undergoing or recovering from laryngeal radiation (xerostomia group) were chosen. Cronbach's alpha coefficient was used to determine internal consistency, and the intra-cluster correlation (ICC) coefficient was used to determine test-retest reliability after two weeks. To evaluate the concurrent validity, the relationship between the total score of the questionnaire and the golden question, which was defined as "How often do you have dry mouth?" was measured. Cronbach's alpha and ICC coefficient for the total XI score were 0.84 and 0.95, respectively. The mean ages of patients in the radiotherapy group and the healthy participants were 59±7.5 and 41.1±6.6 years, respectively. Participants who were having or had had laryngeal radiotherapy had a significantly higher mean total XI score than healthy individuals (p< 0.001). The Persian version of the Xerostomia Inventory is a valid and reliable tool for assessing xerostomia.

Association of maternal, obstetric, fetal, and neonatal mortality outcomes with Lady Health Worker coverage from a cross-sectional survey of >10,000 households in Gilgit-Baltistan, Pakistan.

Pakistan has among the highest rates of maternal, perinatal, and neonatal mortality globally. Many of these deaths are potentially preventable with low-cost, scalable interventions delivered through community-based health worker programs to the most remote communities. We conducted a cross-sectional survey of 10,264 households during the baseline phase of a cluster randomized controlled trial (cRCT) in Gilgit-Baltistan, Pakistan from June-August 2021. The survey was conducted through a stratified, two-stage sampling design with the objective of estimating the neonatal mortality rate (NMR) within the study catchment area, and informing implementation of the cRCT. Study outcomes were self-reported and included neonatal death, stillbirth, health facility delivery, maternal death, postpartum hemorrhage (PPH), and Lady Health Worker (LHW) coverage. Summary statistics (proportions and rates) were weighted according to the sampling design, and mixed-effects Poisson regression was conducted to explore the relationship between LHW coverage and maternal/newborn outcomes. We identified 7,600 women who gave birth in the past five years, among whom 13% reported experiencing PPH. The maternal mortality ratio was 225 maternal deaths per 100,000 live births (95% confidence interval [CI] 137-369). Among 12,376 total births, the stillbirth rate was 41.4 per 1,000 births (95% CI 36.8-46.7) and the perinatal mortality rate was 53.0 per 1,000 births (95% CI 47.6-59.0). Among 11,863 live births, NMR was 16.2 per 1,000 live births (95% CI 13.6-19.3) and 65% were delivered at a health facility. LHW home visits were associated with declines in PPH (risk ratio [RR] 0.89 per each additional visit, 95% CI 0.83-0.96) and late neonatal mortality (RR 0.80, 95% CI 0.67-0.97). Intracluster correlation coefficients were also estimated to inform the planning of future trials. The high rates of maternal, perinatal, and neonatal death in Gilgit-Baltistan continue to fall behind targets of the 2030 Sustainable Development Goals.

An intervention to improve antibiotic prescription behavior in veterinary students: A protocol based on the multi-theory model to tackle antimicrobial resistance.

Antimicrobial resistance (AMR) is a global health threat. Moreover, incorrect and inappropriate drug prescription behavior is considered a fundamental risk factor. Thus, the present study aims to develop, implement, and evaluate the effectiveness of an educational program based on the multi-theoretical model (MTM) in improving antibiotic prescription behavior in veterinary students of Iran. The present study will include four phases including a qualitative phase, an instrument design and psychometric test phase, and a cross-sectional, and an interventional phase. In the first phase, the sampling will be purposive with a maximum variety. The interviews will be conducted with a sample of veterinarians. The data will be analyzed in MAXQDA 10. In the second phase, the face and content validity will be tested by a panel of experts as field specialists. A confirmatory factor analysis will be used to test construct validity, and Cronbach's alpha coefficient and intracluster correlation coefficient will be used to determine the internal consistency of the instrument. Then, at this stage, a number of veterinary students will be selected through a multi-stage sampling method. In the cross-sectional phase, another sample of veterinary students will complete a researcher-made questionnaire. Then, Spearman's correlation coefficient test will be used to test the relationship between the two stages of behavior initiation and behavior continuation. The data will be analyzed in SPSS 22. In the third phase, some veterinary students will be selected through a census and will be randomly divided into a control and an intervention group. To collect data in the final phase, the researcher-made questionnaire that was designed in the second phase of the study based on a multi-theory model will be used to extract data. To compare demographic characteristics, compare the correlation between the constructs of the multi-theory model with antibiotic prescribing behavior in the cross-sectional phase and compare the scores of the constructs of the MTM in two intervention and control groups paired-samples T test and independent-samples T test will be used. The present study will aim to improve antibiotic prescription behavior in veterinary students based on a MTM. The findings can be used as a model for training students in clinical fields such as veterinary medicine and general medicine at university at a national level. After verification and approval by experts and university professors, we can expect a change in the educational curriculum to include instructions on how to write out prescriptions for students. There are hopes that the present study if conducted accurately and widely to help prevent AMR in livestock, humans, and society.

First impressions matter: The influence of initial assessments on psychological treatment initiation and subsequent dropout

Objective This study investigated if patients’ experience of an initial assessment may be associated with outcome expectations, and with subsequent treatment attendance. Method The sample comprised n = 6051 patients with depression/anxiety disorders, nested within k = 148 assessing therapists. Multilevel modelling (MLM) was used to examine therapist effects on treatment initiation and subsequent dropout, adjusting for patient-level characteristics. We tested associations between early outcome expectancy measured at an initial assessment with attendance at a first therapy session, and with dropout after initiation. Variability in mean expectancy ratings in the caseloads of assessing therapists was examined using the intracluster correlation coefficient (ICC). Results Therapist effects partly explained the variance in treatment initiation and dropout. Pre-treatment outcome expectations significantly predicted treatment initiation but not dropout for the subgroup of patients who started treatment. Approximately 16% of variability in mean expectancy ratings was explained by therapist effects (ICC = 0.159) after controlling for patient-level covariates. Conclusions Patients assessed by some therapists are more likely to have higher outcome expectations, which influences their decision to initiate treatment thereafter. Once patients start therapy, early expectancy measured at assessment no longer influences their attendance, but the “first impression” from an initial assessment does influence their subsequent likelihood of dropout.

Modern contraceptive utilization and its associated factors in East Africa: Findings from multi-country demographic and health surveys.

The use of modern contraceptives has been low in most Sub-Saharan African countries despite high population growth and a sluggish economy. This study aimed to identify the prevalence and determinants of modern contraceptive use among married reproductive-age women in East Africa. For this study, the Demographic and health survey (DHS) data from nine countries in East Africa were analyzed, yielding a weighted sample of 32,925 married women. A multilevel mixed-effect logistic regression model was used to identify characteristics associated with the utilization of modern contraceptives at a p-value less than 0.05. For model comparison, we used the Akaike and Bayesian Information Criteria (AIC and BIC). For assessing variation (random effects), we used community-level variance with standard deviation and intra-cluster correlation coefficient (ICC). The overall prevalence of modern contraceptive use was 45.68%, 95% CI (45.15, 46.21). Women's age, maternal education level, husband education level, media exposure, wealth status, occupation, religion, the total number of children ever born, distance to health facilities, history of termination of pregnancy, couple's desire for children, women's participation in decision making, living country and place of residence were significantly associated with modern contraceptive use in Eastern Africa. Conferring to this study, utilization of modern contraceptives is low in East Africa. Interventions to improve the use of modern contraceptives should encompass disseminating awareness through mass media, enrolment of males in family planning, giving maternal education, building health facilities in remote areas, and encouraging family planning programs in rural areas.

Analysing cluster randomised controlled trials using GLMM, GEE1, GEE2, and QIF: results from four case studies

BackgroundUsing four case studies, we aim to provide practical guidance and recommendations for the analysis of cluster randomised controlled trials.MethodsFour modelling approaches (Generalized Linear Mixed Models with parameters estimated by maximum likelihood/restricted maximum likelihood; Generalized Linear Models with parameters estimated by Generalized Estimating Equations (1st order or second order) and Quadratic Inference Function, for analysing correlated individual participant level outcomes in cluster randomised controlled trials were identified after we reviewed the literature. We systematically searched the online bibliography databases of MEDLINE, EMBASE, PsycINFO (via OVID), CINAHL (via EBSCO), and SCOPUS. We identified the above-mentioned four statistical analytical approaches and applied them to four case studies of cluster randomised controlled trials with the number of clusters ranging from 10 to 100, and individual participants ranging from 748 to 9,207. Results were obtained for both continuous and binary outcomes using R and SAS statistical packages.ResultsThe intracluster correlation coefficient (ICC) estimates for the case studies were less than 0.05 and are consistent with the observed ICC values commonly reported in primary care and community-based cluster randomised controlled trials. In most cases, the four methods produced similar results. However, in a few analyses, quadratic inference function produced different results compared to the generalized linear mixed model, first-order generalized estimating equations, and second-order generalized estimating equations, especially in trials with small to moderate numbers of clusters.ConclusionThis paper demonstrates the analysis of cluster randomised controlled trials with four modelling approaches. The results obtained were similar in most cases, however, for trials with few clusters we do recommend that the quadratic inference function should be used with caution, and where possible a small sample correction should be used. The generalisability of our results is limited to studies with similar features to our case studies, for example, studies with a similar-sized ICC. It is important to conduct simulation studies to comprehensively evaluate the performance of the four modelling approaches.

Practice patterns, time trends and quality of care of uterine cancer in Belgium: An analysis of the EFFECT database

ObjectivesTo investigate the practice patterns and quality of care for uterine cancer on a national level in Belgium, including trends in practice over the period 2012–2016. MethodsQuality indicators were measured using the EFFectiveness of Endometrial Cancer Treatment (EFFECT) database. Multivariable logistic mixed regression was used to test for associations between the quality indicators and year of diagnosis, adjusted for potential confounders and intra-cluster correlations. ResultsThe EFFECT database includes 4178 patients diagnosed with uterine cancer in the period 2012–2016. Minimally invasive surgery (laparoscopic or robotic-assisted) was applied in 61.6% of patients who had surgery for clinical stage I endometrial carcinoma (EC), increasing from 52.9% in 2012 to 66.4% in 2016. At least pelvic lymph node staging was performed in 69.0% of patients with clinical stage I, high-grade EC; and in 63.9% of patients with clinical stage I-II serous carcinoma, clear cell carcinoma or carcinosarcoma. The latter increased from 48.8% in 2012 to 77.2% in 2016. Adjuvant radiotherapy (external beam and/or brachytherapy) was offered to 33.5% of patients who had surgery without lymph node staging for pathological stage I EC at high-intermediate or high risk of recurrence. Adjuvant chemotherapy was administered to 64.4% of patients with pathological stage III-IVA EC. ConclusionsStudy results indicate an overall good quality of care for patients with uterine cancer in Belgium. Treatment areas with potential room for improvement include the use of minimally invasive surgery, comprehensive surgical staging and adjuvant therapy, which confirms the remaining controversies in uterine cancer treatment and the need for further research.

Evaluating the Efficiency of Restricted Pseudo Likelihood Estimation in Balanced and Unbalanced Clustered Binary Data Models

Clustered binary data analysis is a common task in various fields, such as social sciences and epidemiology. Restricted Pseudo Likelihood Estimation (PLE) is a widely used approach for analyzing clustered binary data, providing flexibility in handling complex dependencies within clusters. This study aims to evaluate the efficiency of Restricted Pseudo Likelihood Estimation in balanced and unbalanced clustered binary data models. Using simulated data, we compare the performance of PLE in balanced and unbalanced clustered binary data scenarios. We consider various factors such as the number of clusters, cluster sizes, and intra-cluster correlation. The preferred class of models for clustered binary data is the Hierarchical Generalized Linear Model (HGLM). This article compares the performance of a restricted pseudo-likelihood estimation method of the Hierarchical Generalized Linear Model (HGLM) with equal and unequal cluster sizes. Through comprehensive simulation experiments, we assess the accuracy and precision of PLE estimates in terms of parameter estimation, standard errors, and hypothesis testing. Our findings provide insights into the efficiency of Restricted Pseudo Likelihood Estimation (RPLE) in balanced and unbalanced clustered binary data models. The results highlight the advantages and limitations of PLE in different scenarios, aiding researchers in selecting appropriate modeling approaches for their specific data characteristics.  The results can guide researchers in making informed decisions regarding the selection and application of PLE in their own studies, ultimately enhancing the validity and reliability of statistical analyses in the presence of clustered binary data.

Higher adherence to the Mediterranean Diet is associated with lower micronutrient inadequacy in children: the SENDO project.

To assess whether the Mediterranean Diet (MedDiet) is associated with lower micronutrients inadequacy in a sample of Spanish preschoolers. We conducted a cross-sectional study with 4-5-year-old children participating in the SENDO project. Information was gathered through an online questionnaire completed by parents. Dietary information was collected with a previously validated semi-quantitative FFQ. The estimated average requirements or adequate intake levels as proposed by the Institute of Medicine were used as cut-off point to define inadequate intake. Crude and multivariable adjusted estimates were calculated with generalised estimated equations to account for intra-cluster correlation between siblings. We used baseline information of 1153 participants enrolled in the SENDO project between January 2015 and June 2022. OR and 95 % CI of presenting an inadequate intake of ≥ 3 micronutrients associated with the MedDiet. The adjusted proportion of children with inadequate intake of ≥ 3 micronutrients was 27·2 %, 13·5 % and 8·1 % in the categories of low, medium and high adherence to the MedDiet, respectively. After adjusting for all potential confounders, children who had a low adherence to the MedDiet showed a significant lower odds of inadequate intake of ≥ 3 micronutrients compared to those with a high adherence (OR 9·85; 95 % CI 3·33, 29·09). Lower adherence to the MedDiet is associated with higher odds of nutritional inadequacy.

Multilevel and subnational analysis of the predictors of maternity continuum of care completion in Nigeria: a cross-sectional survey

Understanding population discrepancy in maternity continuum of care (CoC) completion, particularly in sub-Saharan Africa is significant for interventional plan to achieve optimal pregnancy outcome and child survival. This study thus investigated the magnitudes, distribution, and drivers of maternity CoC completion in Nigeria. A secondary analysis of 19,474 reproductive age (15–49 years) women with at least a birth (level 1) in 1400 communities (level 2) across 37 states covered in the 2018 cross-sectional survey. Stepwise regression initially identified important variables at 10% cutoff point. Multilevel analysis was performed to determine the likelihood and significance of individual and community factors. Intra-cluster correlation assessed the degree of clustering and deviance statistics identified the optimal model. Only 6.5% of the women completed the CoC. Completion rate is significantly different between communities “4.3% in urban and 2.2% in rural” (χ2 = 392.42, p < 0.001) and was higher in southern subnational than the north. Education (AOR = 1.61, 95% CI 1.20–2.16), wealth (AOR = 1.73, 95% CI 1.35–2.46), media exposure (AOR = 1.22, 95% CI 1.06–1.40), women deciding own health (AOR = 1.37, 95% CI 1.13–1.66), taking iron drug (AOR = 1.84, 95% CI 1.43–2.35) and at least 2 dose of tetanus-toxoid vaccine during pregnancy (AOR = 1.35, 95% CI 1.02–1.78) are associated individual factors. Rural residency (AOR = 1.84, 95% CI = 1.43–2.35), region (AOR = 1.84, 95% CI 1.43–2.35) and rural population proportion (AOR = 1.84, 95% CI 1.43–2.35) are community predictors of the CoC completion. About 63.2% of the total variation in CoC completion was explained by the community predictors. Magnitude of maternity CoC completion is generally low and below the recommended level in Nigeria. Completion rate in urban is twice rural and more likely in the southern than northern subnational. Women residence and region are harmful and beneficial community drivers respectively. Strengthening women health autonomy, sensitization, and education programs particularly in the rural north are essential to curtail the community disparity and optimize maternity CoC practice.

Adherence to key recommendations for design and analysis of stepped-wedge cluster randomized trials: A review of trials published 2016-2022.

The stepped-wedge cluster randomized trial (SW-CRT), in which clusters are randomized to a time at which they will transition to the intervention condition - rather than a trial arm - is a relatively new design. SW-CRTs have additional design and analytical considerations compared to conventional parallel arm trials. To inform future methodological development, including guidance for trialists and the selection of parameters for statistical simulation studies, we conducted a review of recently published SW-CRTs. Specific objectives were to describe (1) the types of designs used in practice, (2) adherence to key requirements for statistical analysis, and (3) practices around covariate adjustment. We also examined changes in adherence over time and by journal impact factor. We used electronic searches to identify primary reports of SW-CRTs published 2016-2022. Two reviewers extracted information from each trial report and its protocol, if available, and resolved disagreements through discussion. We identified 160 eligible trials, randomizing a median (Q1-Q3) of 11 (8-18) clusters to 5 (4-7) sequences. The majority (122, 76%) were cross-sectional (almost all with continuous recruitment), 23 (14%) were closed cohorts and 15 (9%) open cohorts. Many trials had complex design features such as multiple or multivariate primary outcomes (50, 31%) or time-dependent repeated measures (27, 22%). The most common type of primary outcome was binary (51%); continuous outcomes were less common (26%). The most frequently used method of analysis was a generalized linear mixed model (112, 70%); generalized estimating equations were used less frequently (12, 8%). Among 142 trials with fewer than 40 clusters, only 9 (6%) reported using methods appropriate for a small number of clusters. Statistical analyses clearly adjusted for time effects in 119 (74%), for within-cluster correlations in 132 (83%), and for distinct between-period correlations in 13 (8%). Covariates were included in the primary analysis of the primary outcome in 82 (51%) and were most often individual-level covariates; however, clear and complete pre-specification of covariates was uncommon. Adherence to some key methodological requirements (adjusting for time effects, accounting for within-period correlation) was higher among trials published in higher versus lower impact factor journals. Substantial improvements over time were not observed although a slight improvement was observed in the proportion accounting for a distinct between-period correlation. Future methods development should prioritize methods for SW-CRTs with binary or time-to-event outcomes, small numbers of clusters, continuous recruitment designs, multivariate outcomes, or time-dependent repeated measures. Trialists, journal editors, and peer reviewers should be aware that SW-CRTs have additional methodological requirements over parallel arm designs including the need to account for period effects as well as complex intracluster correlations.

A mixed model approach to estimate the survivor average causal effect in cluster-randomized trials.

In many medical studies, the outcome measure (such as quality of life, QOL) for some study participants becomes informatively truncated (censored, missing, or unobserved) due to death or other forms of dropout, creating a nonignorable missing data problem. In such cases, the use of a composite outcome or imputation methods that fill in unmeasurable QOL values for those who died rely on strong and untestable assumptions and may be conceptually unappealing to certain stakeholders when estimating a treatment effect. The survivor average causal effect (SACE) is an alternative causal estimand that surmounts some of these issues. While principal stratification has been applied to estimate the SACE in individually randomized trials, methods for estimating the SACE in cluster-randomized trials are currently limited. To address this gap, we develop a mixed model approach along with an expectation-maximization algorithm to estimate the SACE in cluster-randomized trials. We model the continuous outcome measure with a random intercept to account for intracluster correlations due to cluster-level randomization, and model the principal strata membership both with and without a random intercept. In simulations, we compare the performance of our approaches with an existing fixed-effects approach to illustrate the importance of accounting for clustering in cluster-randomized trials. The methodology is then illustrated using a cluster-randomized trial of telecare and assistive technology on health-related QOL in the elderly.

Estimating intracluster correlation for ordinal data

In this paper, we consider the estimation of intracluster correlation for ordinal data. We focus on pure-tone audiometry hearing threshold data, where thresholds are measured in 5 decibel increments. We estimate the intracluster correlation for tests from iPhone-based hearing assessment applications as a measure of test/retest reliability. We present a method to estimate the intracluster correlation using mixed effects cumulative logistic and probit models, which assume the outcome data are ordinal. This contrasts with using a mixed effects linear model which assumes that the outcome data are continuous. In simulation studies, we show that using a mixed effects linear model to estimate the intracluster correlation for ordinal data results in a negative finite sample bias, while using mixed effects cumulative logistic or probit models reduces this bias. The estimated intracluster correlation for the iPhone-based hearing assessment application is higher when using the mixed effects cumulative logistic and probit models compared to using a mixed effects linear model. When data are ordinal, using mixed effects cumulative logistic or probit models reduces the bias of intracluster correlation estimates relative to using a mixed effects linear model.

Longer breastfeeding duration is associated with lower consumption of ultra-processed foods in a sample of Spanish preschoolers: The SENDO project.

Breastfeeding has been linked to a higher consumption of fruit and vegetables at 4 to 5 years of age. More recently, it has been suggested that it may also be associated with lower ultra-processed food (UPF) consumption in childhood. The aim of this study was to assess whether breastfeeding duration was associated with consumption of UPF in a sample of Mediterranean preschoolers. This study involved a cross-sectional analysis of baseline information of children in the SENDO cohort. Children were enrolled at 4 to 5 years of age and information was gathered through an online questionnaire completed by parents. Dietary information was collected with a previously validated semi-quantitative food frequency questionnaire (FFQ) and foods were classified based on the degree of processing according to the NOVA classification. This study used baseline information for 806 participants enrolled in the SENDO cohort between January 2015 and June 2021 in Spain. Main study outcome measures were 1) Difference in g/day and in the percentage of total energy intake (TEI) from UPF consumption related to breastfeeding duration, and 2) Odds ratio (OR) that UPF represents a high percentage of TEI. Crude and multivariable adjusted estimates were calculated with generalized estimating equations to account for intra-cluster correlation between siblings. The prevalence of breastfeeding in the sample was 84%. After adjusting for potential confounders, children who were breastfed for some time reported significantly lower consumption of UPF than children who were not breastfed at all. The mean differences were -19.2 g. (95% CI: -44.2, 10.8) for children who were breastfed for <6 months, -42.5 g. (95% CI: -77.2, -7.80) for those who were breastfed for 6-12 months, and -43.6 g. (95% CI: -79.8, -7.48) for those who were breastfed for 12 months or more (p for trend=0.01). After adjusting for potential confounders, compared with children who were not breastfed, those who were breastfed for ≥12 months had consistently lower odds of UPF representing more than 25%, 30%, 35% and 40% of TEI. Breastfeeding is associated with lower consumption of UPF in Spanish preschoolers.

Development and psychometric properties of Iranian Women’s Quality of Life Instrument (IWQOLI): mixed exploratory study

BackgroundTo determine the health needs and promote women's health, their quality of life should be investigated. For this purpose, a valid tool is needed, that has credible validity and reliability, and its concepts are clearly defined and culturally appropriate. This study aimed to develop and assess the psychometric properties of “Iranian Women’s Quality of Life Instrument (IWQOLI)”.MethodsThe items of “IWQOLI” were generated from themes extracted (150 items) from a content analysis approach with the participation of 40 women. Face validity of the questionnaire with the participation of 10 women and content validity by 10 experts was approved. To determine the domains of the questionnaire, exploratory factor analysis (principal component extraction method) was performed. Internal consistency and test—retest reliability methods with 14-day intervals (30 women) were used to assess the reliability of WQOLI.ResultsAfter performing the face and content validity, 32 items were deleted. S-CVI/Ave was obtained for the instrument (0.93). The factor structure of the inventory was identified by undertaking a principal component analysis in a sample of 590 women. Five factors were extracted with a total variance account of 56.24% and 28 items dropped at this point. The IWQOLI score was significantly correlated with the SF-36 (r = 0.717, p < 0.001). Reliability was demonstrated with Cronbach’s alpha coefficient of 0.919 for the entire scale (90-item). Consistency of the instrument was established with test–retest reliability with an interval of 2 weeks (intra-cluster correlation = 0.889, P < 0.001).ConclusionsThe Iranian women’s Quality of life Instrument “IWQOLI”, consisting of 90 items representing 5 domains (sense of peace in life, sense of security, health responsibility, pleasant communication, received comprehensive support), demonstrated excellent psychometric properties, so it may be used for measuring women’s QOL in practical research.

Association between the Carbohydrate Quality Index (CQI) and Nutritional Adequacy in a Pediatric Cohort: The SENDO Project.

Suboptimal micronutrient intake in children remains a public health concern around the world. This study examined the relationship between a previously defined dietary carbohydrate quality index (CQI) and the risk of micronutrient intake inadequacy in a pediatric cohort of Spanish preschoolers. Children aged 4-5 years old were recruited at their medical center or at school, and information on sociodemographic, dietary, and lifestyle variables were collected through a self-administered online questionnaire. Dietary information was obtained from a validated 147-item semi-quantitative food frequency questionnaire. We calculated the CQI and categorized participants into quartiles according to their scores. We assessed the intakes of 20 micronutrients and evaluated the probability of intake inadequacy by using the estimated average requirement cut-off point. Generalized estimating equations were used to adjust for potential confounders and account for the intra-cluster correlations between siblings. The adjusted proportions of children with an inadequate intake of ≥three micronutrients were 23%, 12%, 11%, and 9% in the first, second, third, and fourth quartiles of the CQI, respectively. Children in the highest quartile of the CQI had 0.22-fold lower odds (95% CI 0.10-0.48) of having ≥three inadequate micronutrient intakes than their peers in the lowest quartile. These findings reinforce the relevance of carbohydrate quality in children's diets.

Trends of Exclusive Breastfeeding Practices and Its Determinants in Tanzania from 1999 to 2016.

Introduction: The benefits of exclusive breastfeeding (EBF) are widely reported. However, it is crucial to examine potential disparities in EBF practices across different regions of a country. Our study uses Tanzania demographic and health survey data to report on the trends of EBF across regions from 1999 to 2016, the patterns of the practice based on geographical location and socioeconomic status, and explores its determinants across the years. Methods: Descriptive statistics were used to establish the trends of EBF by geographical location and wealth quintile. A generalized linear mixed model was developed to incorporate both infant and maternal attributes as fixed covariates while considering enumeration areas and regions as clusters. The fitted model facilitated the estimation of EBF proportions at a regional level and identified key determinants influencing EBF practices across the survey periods. Moreover, we designed breastfeeding maps, visually depicting the performance of different regions throughout the surveys. Results: Across the various survey rounds, a notable regional variation in EBF practices was observed, with coastal regions generally exhibiting lower adherence to the practice. There was a linear trend between EBF and geographical residence (p < 0.05) and socioeconomic standing (p < 0.05) across the survey periods. Rural-dwelling women and those from the least affluent backgrounds consistently showcased a higher proportion of EBF. The prevalence of EBF declined as infants aged (p < 0.001), a trend consistent across all survey waves. The associations between maternal attributes and EBF practices displayed temporal variations. Furthermore, a correlation between exclusive breastfeeding and attributes linked to both regional disparities and enumeration areas was observed. The intra-cluster correlation ranged from 18% to 41.5% at the regional level and from 40% to 58.5% at the enumeration area level. Conclusions: While Tanzania's progress in EBF practices is laudable, regional disparities persist, demanding targeted interventions. Sustaining achievements while addressing wealth-based disparities and the decline in EBF with infant age is vital. The study highlights the need for broad national strategies and localized investigations to understand and enhance EBF practices across different regions and socioeconomic contexts.