Two implementation strategies to support the integration of depression screening and treatment into hypertension and diabetes care in Malawi (SHARP): parallel, cluster-randomised, controlled, implementation trial

BACKGROUND

Although evidence-based treatments for depression in low-resource settings are established, implementation strategies to scale up these treatments remain poorly defined. We aimed to compare two implementation strategies in achieving high-quality integration of depression care into chronic medical care and improving mental health outcomes in patients with hypertension and diabetes.

METHODS

We conducted a parallel, cluster-randomised, controlled, implementation trial in ten health facilities across Malawi. Facilities were randomised (1:1) by covariate-constrained randomisation to either an internal champion alone (ie, basic strategy group) or an internal champion plus external supervision with audit and feedback (ie, enhanced strategy group). Champions integrated a three-element, evidence-based intervention into clinical care: universal depression screening; peer-delivered psychosocial counselling; and algorithm-guided, non-specialist antidepressant management. External supervision involved structured facility visits by Ministry officials and clinical experts to assess quality of care and provide supportive feedback approximately every 4 months. Eligible participants were adults (aged 18-65 years) seeking hypertension and diabetes care with signs of depression (Patient Health Questionnaire-9 score ≥5). Primary implementation outcomes were depression screening fidelity, treatment initiation fidelity, and follow-up treatment fidelity over the first 3 months of treatment, analysed by intention to treat. This trial is registered with ClinicalTrials.gov, NCT03711786, and is complete.

FINDINGS

Five (50%) facilities were randomised to the basic strategy and five (50%) to the enhanced strategy. Between Oct 1, 2019, and Nov 30, 2021, in the basic group, 587 patients were assessed for eligibility, of whom 301 were enrolled; in the enhanced group, 539 patients were assessed, of whom 288 were enrolled. All clinics integrated the evidence-based intervention and were included in the analyses. Of 60 774 screening-eligible visits, screening fidelity was moderate (58% in the enhanced group vs 53% in the basic group; probability difference 5% [95% CI -38% to 47%]; p=0·84) and treatment initiation fidelity was high (99% vs 98%; 0% [-3% to 3%]; p=0·89) in both groups. However, treatment follow-up fidelity was substantially higher in the enhanced group than in the basic group (82% vs 20%; 62% [36% to 89%]; p=0·0020). Depression remission was higher in the enhanced group than in the basic group (55% vs 36%; 19% [3% to 34%]; p=0·045). Serious adverse events were nine deaths (five in the basic group and four in the enhanced group) and 26 hospitalisations (20 in the basic group and six in the enhanced group); none were treatment-related.

INTERPRETATION

The enhanced implementation strategy led to an increase in fidelity in providers' follow-up treatment actions and in rates of depression remission, consistent with the literature that follow-up decisions are crucial to improving depression outcomes in integrated care models. These findings suggest that external supervision combined with an internal champion could offer an important advance in integrating depression treatment into general medical care in low-resource settings.

FUNDING

The National Institute of Mental Health.

Multi-ancestry Genome-Wide Association Study of Early Childhood Caries

Early childhood caries (ECC) is the most common non-communicable childhood disease. It is an important health problem with known environmental and social/behavioral influences that lacks evidence for specific associated genetic risk loci. To address this knowledge gap, we conducted a genome-wide association study of ECC in a multi-ancestry population of U.S. preschool-age children (n=6,103) participating in a community-based epidemiologic study of early childhood oral health. Calibrated examiners used ICDAS criteria to measure ECC with the primary trait using the dmfs index with decay classified as macroscopic enamel loss (ICDAS ≥3). We estimated heritability, concordance rates, and conducted genome-wide association analyses to estimate overall genetic effects; the effects stratified by sex, household water fluoride, and dietary sugar; and leveraged the combined gene/gene-environment effects using the 2-degree-of-freedom (2df) joint test. The common genetic variants explained 24% of the phenotypic variance (heritability) of the primary ECC trait and the concordance rate was higher with a higher degree of relatedness. We identified 21 novel non-overlapping genome-wide significant loci for ECC. Two loci, namely RP11-856F16 . 2 (rs74606067) and SLC41A3 (rs71327750) showed evidence of association with dental caries in external cohorts, namely the GLIDE consortium adult cohort (n=∼487,000) and the GLIDE pediatric cohort (n=19,000), respectively. The gene-based tests identified TAAR6 as a genome-wide significant gene. Implicated genes have relevant biological functions including roles in tooth development and taste. These novel associations expand the genomics knowledge base for this common childhood disease and underscore the importance of accounting for sex and pertinent environmental exposures in genetic investigations of oral health.

Life-Space of Older Adults after Discharge from Skilled Nursing Facilities

OBJECTIVES

Describe (1) patient or caregiver perceptions of physical function in 30 days after skilled nursing facility (SNF) discharge indicated by Life-Space Assessment (LSA) scores, and (2) patient and caregiver factors associated with LSA scores.

DESIGN

Secondary analysis of baseline and outcomes data from the cluster randomized trial of the Connect-Home transitional care intervention.

SETTING AND PARTICIPANTS

Six SNFs in North Carolina. Patient and caregiver dyads with LSA scores (N = 245).

METHODS

SNF patients or their caregivers serving as proxy reported the life-space of the SNF patient using the LSA tool, a measure of environmental and social factors that influence physical mobility. Simple scores for highest life-space attained depending on equipment and/or caregiver support range from 0 to 5, with higher scores indicating greater mobility. Multiple linear regression models for simple LSA scores and Composite Life-Space (0-120), adjusted for treatment, time via a COVID pandemic indicator, and treatment × COVID effect as fixed effects, were used to estimate the association of patient and caregiver variables and life-space.

RESULTS

Patients had a mean age of 76.3 years, 62.6% were female, and 74.7% were white. Caregivers were commonly female (73.9%) and adult children of the patient (46.5%). The mean Composite Life-Space score was 22.6 (16.09). The mean Assisted Life-Space score (range: 0-5) was 1.6 (1.47), and 76.3% of patients could not move beyond their bedroom, house, and yard without assistance of another person. Higher Composite Life-Space scores were associated with lower levels of cognitive impairment and shorter SNF length of stay.

CONCLUSIONS AND IMPLICATIONS

SNF patients and their caregivers reported very low LSA scores in 30 days after SNF care. Findings indicate the need for care redesign to promote recovery of physical function of older adults after SNF discharge, such as optimizing SNF rehabilitative therapy and adding postdischarge rehabilitative supports at home.

Patterns of site-level periodontal disease and within-mouth correlation among older adults in the Hispanic Community Health Study/Study of Latinos

OBJECTIVES

Clinical measures of periodontal disease such as attachment loss (CAL) and probing depth (PD) vary considerably between and within individuals with periodontitis and are known to be influenced by person-level factors (e.g. age and race/ethnicity) as well as intraoral characteristics (e.g. tooth type and location). This study sought to characterize site-level disease patterns and correlations using both person-level and intraoral factors through a model-based approach.

METHODS

This study used full-mouth, six sites per tooth, periodontal examination data collected from 2301 Hispanic/Latino adults aged 60-74 years in the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). The presence of site-level CAL ≥3 mm and PD ≥4 mm was estimated using generalized estimating equations (GEE), explicitly modelling pairwise periodontal site correlations, while adjusting for number of teeth, sex and Hispanic/Latino background. Subsequently tooth- and tooth-site patterns of intraoral CAL ≥3 mm and PD ≥4 mm were estimated and visualized in the HCHS/SOL population.

RESULTS

The findings showed that posterior sites had the highest odds of CAL ≥3 mm and PD ≥4 mm. Sites located in the interproximal space had higher odds of PD ≥4 mm but lower odds of CAL ≥3 mm than non-interproximal sites. Mexicans had the lowest odds of CAL ≥3 mm among all Hispanic/Latino backgrounds. While Mexicans had lower odds of PD ≥4 mm than Central Americans and Cubans, they had higher odds than Dominicans and Puerto Ricans. Site-level proportions and pairwise correlations of PD ≥4 mm were generally smaller than those of CAL ≥3 mm.

CONCLUSIONS

The patterns of site-level probabilities of clinical measures of periodontal disease can be defined based on tooth, site and individual-level characteristics. Intraoral correlation patterns, while complex, are quantifiable. The risk factors for site-level CAL ≥3 mm may differ from those of PD ≥4 mm. Likewise, participant risk factors for site-level clinical measures of periodontal disease are distinct from those that affect individual-level periodontitis prevalence. Future epidemiological investigations should consider model-based approaches when examining site-level disease probabilities to identify intra-oral patterns of periodontal disease and make inferences about the larger population.

Generalizing the information content for stepped wedge designs: A marginal modeling approach

Stepped wedge trials are increasingly adopted because practical constraints necessitate staggered roll-out. While a complete design requires clusters to collect data in all periods, resource and patient-centered considerations may call for an incomplete stepped wedge design to minimize data collection burden. To study incomplete designs, we expand the metric of information content to discrete outcomes. We operate under a marginal model with general link and variance functions, and derive information content expressions when data elements (cells, sequences, periods) are omitted. We show that the centrosymmetric patterns of information content can hold for discrete outcomes with the variance-stabilizing link function. We perform numerical studies under the canonical link function, and find that while the patterns of information content for cells are approximately centrosymmetric for all examined underlying secular trends, the patterns of information content for sequences or periods are more sensitive to the secular trend, and may be far from centrosymmetric.

Ten-year cross-sectional and longitudinal assessment and factors associated with unfavourable self-rated oral health in older adults in the United States

OBJECTIVE

To investigate the association of demographic and socio-economic characteristics with self-reported oral health (SROH) among older adults who participated in the Health and Retirement Study (HRS) in 2008, 2018, or both, and to describe temporal changes.

METHODS

Data were from the University of Michigan's Health and Retirement Study (HRS), a nationally representative longitudinal survey of Americans aged 51 and older. Responses from participants who completed the Core HRS survey and Dental Module (DM) in 2008 (n = 1310), 2018 (n = 1330), and the "common group" at both timepoints (n = 559) were analysed. Using the common group, the outcome measure was 2018 self-rated oral health (Favourable vs Unfavourable). Potential explanatory variables included 2008 self-rated oral health (SROH), sociodemographic, and dental utilisation-related factors. Survey logistic regression analysis was used to identify factors that were associated with unfavourable 2018 SROH in 2018.

RESULTS

Unfavourable SROH prevalence was 28.5% and 31.6% in 2008 and 2018, respectively. Among the common, longitudinal group, the unfavourable prevalence remained the same, 26.1% at both timepoints. A positive association was seen between 2018 unfavourable SROH and baseline variables of 2008 unfavourable SROH, male gender, less education, and lower levels of wealth.

CONCLUSIONS

Over a quarter of participants reported unfavourable SROH. There was little change in SROH during this period. Sociodemographic factors influence the SROH of the older population. Policies and programs to promote and protect the oral health of older adults should be designed and implemented to reduce social inequalities and improve the SROH of disadvantaged older adults.

Non-esterified erythrocyte linoleic acid, arachidonic acid, and subjective sleep outcomes

OBJECTIVE

This study investigated whether non-esterified erythrocyte omega-6 PUFAs were associated with subjective assessment of sleep quality and duration, and risk for obstructive sleep apnea.

METHODS

In this secondary analysis of the cross-sectional OPPERA-II study, 538 adults completed the Pittsburgh Sleep Quality Index (PSQI), reported their usual hours of sleep, and answered STOP screening questions for obstructive sleep apnea. Circulating non-esterified erythrocyte concentrations of omega-6 PUFA linoleic acid and arachidonic acid were quantified by liquid chromatography tandem mass spectroscopy. Sleep outcomes were dichotomized as poor (PSQI ≤5) vs good (PSQI ≥6) sleep quality, insufficient or excessive (≤6 or >9 h) vs good (7-9 h) sleep duration, and high (≥2 affirmative responses) vs low (<2 affirmative responses) risk for obstructive sleep apnea. Non-esterified omega-6 PUFAs and the continuous covariates of body mass index, Short Form (SF) 12 Health Survey Physical and Mental Component scores and resting measures of systolic and diastolic blood pressure were standardized for multivariable analysis. Categorical covariates were study site, age, sex, and race/ethnicity. Multivariable-adjusted logistic regression first estimated odds ratios (OR) and 95% confidence limits (CL) for sleep outcomes using linoleic acid as the main exposure. Analysis was then repeated using arachidonic acid as the main exposure.

RESULTS

In the multivariable-adjusted model, each standard deviation increase in non-esterified erythrocyte linoleic acid was associated with higher odds of poor sleep quality (OR=1.2, 95% CL: 1.1, 1.5), insufficient or excessive sleep (OR= 1.3, 95% CL: 1.1, 1.6) and high-risk for obstructive sleep apnea (OR=1.3, 95% CL: 1.1, 1.6). Likewise, for each standard deviation increase in non-esterified erythrocyte arachidonic acid, odds increased of poor sleep quality (OR=1.2, 95% CL: 1.1, 1.5), and insufficient or excessive sleep (OR=1.2, 95% CL: 1.1, 1.5). Odds of being high risk for obstructive sleep apnea increased with greater circulating arachidonic acid, but the association did not reach statistical significance (OR=1.1, 95% CL: 0.9, 1.4).

CONCLUSION

Non-esterified erythrocyte linoleic acid and arachidonic acid were associated with poor sleep quality and insufficient or excessive sleep duration. Linoleic acid, but not arachidonic acid, was also associated with high risk for obstructive sleep apnea.

Patient-level information underlying overdiagnosis of urinary tract infections in nursing homes: A discrete choice experiment

The overdiagnosis of urinary tract infections (UTIs) in nursing home residents is a significant public health threat. Using a discrete choice experiment and a diagnostic guideline, we examined which patient-level information was associated with the overdiagnosis of UTIs and found that urinalysis results and lower urinary tract status were most associated.

A bivariate zero-inflated negative binomial model and its applications to biomedical settings

The zero-inflated negative binomial distribution has been widely used for count data analyses in various biomedical settings due to its capacity of modeling excess zeros and overdispersion. When there are correlated count variables, a bivariate model is essential for understanding their full distributional features. Examples include measuring correlation of two genes in sparse single-cell RNA sequencing data and modeling dental caries count indices on two different tooth surface types. For these purposes, we develop a richly parametrized bivariate zero-inflated negative binomial model that has a simple latent variable framework and eight free parameters with intuitive interpretations. In the scRNA-seq data example, the correlation is estimated after adjusting for the effects of dropout events represented by excess zeros. In the dental caries data, we analyze how the treatment with Xylitol lozenges affects the marginal mean and other patterns of response manifested in the two dental caries traits. An R package "bzinb" is available on Comprehensive R Archive Network.

Edentulism Predicts Cognitive Decline in the US Health and Retirement Cohort Study

This longitudinal cohort study examines if 1) cognitive decline varies by birth cohort, adjusting for covariates, and 2) edentulism and nonuse of dental care predict 10-y cognitive decline (2008-2018). The Health and Retirement Study (HRS) features a representative sample of US adults over age 50. Eligibility criteria included having cognitive interview data available and responding to the question, "Have you lost all of your upper and lower natural permanent teeth?" at 2+ time points between 2006 and 2018. Use of dental care in the past 2 y was assessed. Linear mixed models for repeated measures estimated the trajectories of mean cognition over time for the birth cohorts, adjusted for baseline cognition, dentition status, dental care use, and covariates (demographic characteristics, health behaviors, and medical conditions). Cohort-by-time interaction terms were included to assess if cognitive decline varied by birth cohort. Ten-year change in cognition status (measured by HRS Cogtot27)-categorized as dementia (<7); cognitive impairment, not demented (7-11) 7≤Cogtot27<12; and normal (≥12)-was also investigated according to birth cohort, dentition status, and dental care use. Mean (SD) baseline age was 63.4 (10.1) y (n = 22,728). Older birth cohorts had greater cognitive decline than younger cohorts. Linear mixed-model estimates and 95% confidence intervals for protective factors for cognitive decline included higher baseline cognition (HRS Cogtot27) (0.49; 0.48-0.50), use of dental care in the past 2 y (0.17; 0.10-0.23), and covariates such as greater household wealth and being married. Risk increased with being edentulous (-0.42; -0.56 to -0.28), history of stroke or diabetes, less education, Medicaid recipient, current smoker, loneliness, and poor/fair self-rated health. Edentulism and irregular dental care are among important predictors of cognitive decline. Tooth retention and regular dental care throughout life appear to be important for maintaining oral and cognitive health.

ORTH.Ord: An R package for analyzing correlated ordinal outcomes using alternating logistic regressions with orthogonalized residuals

BACKGROUND AND OBJECTIVES

Marginal models with generalized estimating equations (GEE) are usually recommended for analyzing correlated ordinal outcomes which are commonly seen in a longitudinal study or clustered randomized trial (CRT). Within-cluster association is often of interest in longitudinal studies or CRTs, and can be estimated with paired estimating equations. However, the estimators for within-cluster association parameters and variances may be subject to finite-sample biases when the number of clusters is small. The objective of this article is to introduce a newly developed R package ORTH.Ord for analyzing correlated ordinal outcomes using GEE models with finite-sample bias corrections.

METHODS

The R package ORTH.Ord implements a modified version of alternating logistic regressions with estimation based on orthogonalized residuals (ORTH), which use paired estimating equations to jointly estimate parameters in marginal mean and association models. The within-cluster association between ordinal responses is modeled by global pairwise odds ratios (POR). The R package also provides a finite-sample bias correction to POR parameter estimates based on matrix multiplicative adjusted orthogonalized residuals (MMORTH) for correcting estimating equations, and bias-corrected sandwich estimators with different options for covariance estimation.

RESULTS

A simulation study shows that MMORTH provides less biased global POR estimates and coverage of their 95% confidence intervals closer to the nominal level than uncorrected ORTH. An analysis of patient-reported outcomes from an orthognathic surgery clinical trial illustrates features of ORTH.Ord.

CONCLUSIONS

This article provides an overview of the ORTH method with bias-correction on both estimating equations and sandwich estimators for analyzing correlated ordinal data, describes the features of the ORTH.Ord R package, evaluates the performance of the package using a simulation study, and finally illustrates its application in an analysis of a clinical trial.

Connect-Home transitional care from skilled nursing facilities to home: A stepped wedge, cluster randomized trial

BACKGROUND

Skilled nursing facility (SNF) patients and their caregivers who transition to home experience complications and frequently return to acute care. We tested the efficacy of the Connect-Home transitional care intervention on patient and caregiver preparedness for care at home, and other patient and caregiver-reported outcomes.

METHODS

We used a stepped wedge, cluster-randomized trial design to test the intervention against standard discharge planning (control). The setting was six SNFs and six home health offices in one agency. Participants were 327 dyads of patients discharged from SNF to home and their caregivers; 11.1% of dyads in the control condition and 81.2% in the intervention condition were enrolled after onset of COVID-19. Patients were 63.9% female and mean age was 76.5 years. Caregivers were 73.7% female and mean age was 59.5 years. The Connect-Home intervention includes tools, training, and technical assistance to deliver transitional care in SNFs and patients' homes. Primary outcomes measured at 7 days included patient and caregiver measures of preparedness for care at home, the Care Transitions Measure-15 (patient) and the Preparedness for Caregiving Scale (caregiver). Secondary outcomes measured at 30 and 60 days included the McGill Quality of Life Questionnaire, Life Space Assessment, Zarit Caregiver Burden Scale, Distress Thermometer, and self-reported number of patient days in the ED or hospital in 30 and 60 days following SNF discharge.

RESULTS

The intervention was not associated with improvement in patient or caregiver outcomes in the planned analyses. Post-hoc analyses that distinguished between pre- and post-pandemic effects suggest the intervention may be associated with increased patient preparedness for discharge and decreased number of acute care days.

CONCLUSIONS

Connect-Home transitional care did not improve outcomes in the planned statistical analysis. Post-hoc findings accounting for COVID-19 impact suggest SNF transitional care has potential to increase patient preparedness and decrease return to acute care.

Circulating Polyunsaturated Fatty Acids and Pain Intensity in Five Chronic Pain Conditions

Pain intensity is well-known to be influenced by a wide range of biobehavioral variables. Nutritional factors, however, have not been generally considered for their potential importance. This cross-sectional study examined associations between erythrocyte omega-6 (n-6) and omega-3 (n-3) polyunsaturated fatty acids (PUFAs) and pain intensity in 605 adults. Pain intensity was computed on a 0 to 100 numeric rating scale from questions about 5 chronic pain conditions: orofacial pain, headache, low back pain, irritable bowel syndrome, and bodily pain. For each pain condition, multiple linear regression tested the hypothesis that a higher ratio of n-6 arachidonic acid to the sum of n-3 eicosapentaenoic acid and docosahexaenoic acid (AA/(EPA+DHA) was associated with greater pain intensity. In covariate-adjusted analysis, orofacial pain intensity increased 5.7 points (95% CI: 1.4, 9.9) per unit increase in n-6/n-3 PUFA ratio. Likewise, a 1 unit increase in n-6/n-3 PUFA ratio was associated with significant increases in pain intensity (range 5-8 points) of headache pain, low back pain, and bodily pain, but not abdominal pain. Separate multiple linear regression models investigated the independent strength of association of individual PUFAs to the intensity of each pain condition. Overall, n-3 docosahexaenoic acid was most strongly, and inversely, associated with pain intensity. PERSPECTIVE: A higher ratio of n-6/n-3 long-chain polyunsaturated fatty acids was associated greater pain intensity for orofacial pain, headache, low back pain, and bodily pain, but not abdominal pain. The n-6/n-3 PUFA ratio was more consistently associated with pain intensity than any individual constituent of the long-chain PUFA ratio.

GEEMAEE: A SAS macro for the analysis of correlated outcomes based on GEE and finite-sample adjustments with application to cluster randomized trials

BACKGROUND AND OBJECTIVES

Generalized estimating equations (GEE) are used to analyze correlated outcomes in marginal regression models with population-averaged interpretations of exposure effects. Limitations of popular software for GEE include: (i) user choice is restricted to a small set of within-cluster pairwise correlation (intra-class correlation; ICC) structures; and (ii) inference on ICC parameters is usually not possible because the precision of their estimates is not quantified. This is important because ICC values inform the design of cluster randomized trials. Beyond the standard GEE implementation, use of paired estimating equations (Prentice 1988) provides: (i) flexible specification of models for pairwise correlations and (ii) standard errors for ICC estimates. However, most GEEs give biased estimates of standard errors and correlations when the number of clusters is small (roughly, ≤40). Consequently, there is a need for software to provide GEE analysis with finite-sample bias-corrections.

METHODS

The SAS macro GEEMAEE implements paired estimating equations to simultaneously estimate parameters in marginal mean and ICC models. It provides bias-corrected standard errors and uses matrix-adjusted estimating equations (MAEE) for bias-corrected estimation of correlations. Several built-in correlation matrix options, rarely found in software, are offered for multi-period, cluster randomized trials and similarly structured longitudinal observational data structures. Additional options include user-specified correlation structures and deletion diagnostics, namely Cooks' Distance and DBETA statistics that estimate the influence of observations, cluster-periods (when applicable) and clusters.

RESULTS

GEEMAEE is illustrated for a binary and a count outcome in two stepped wedge cluster randomized trials and a binary outcome in a longitudinal study of disease surveillance. Use of MAEE resulted in larger values of correlation estimates compared to uncorrected estimating equations. Use of bias-corrected variance estimators resulted in (appropriately) larger values of standard errors compared to the usual sandwich estimators. Deletion diagnostics identified the clusters and cluster-periods having the most influence.

CONCLUSIONS

The SAS macro GEEMAEE provides regression analysis for clustered or longitudinal responses, and is particularly useful when the number of clusters is small. Flexible specification and bias-corrected estimation of pairwise correlation parameters and standard errors are key features of the software to provide valid inference in real-world settings.

A multi-site pilot study of a parent-centered tool to promote shared decision-making in hypospadias care

BACKGROUND

Using a user-centered design approach, we conducted a two-site pilot study to evaluate a decision aid (DA) website, the Hypospadias Hub, for parents of hypospadias patients.

OBJECTIVES

The objectives were to assess the Hub's acceptability, remote usability, and feasibility of study procedures, and to evaluate its preliminary efficacy.

METHODS

From June 2021-February 2022, we recruited English-speaking parents (≥18 years old) of hypospadias patients (≤5 years) and delivered the Hub electronically ≤2 months before their hypospadias consultation. We collected website analytic data using an ad tracker plug-in. We inquired about treatment preference, hypospadias knowledge, and decisional conflict (Decisional Conflict Scale) at baseline, after viewing the Hub (pre-consultation), and post-consultation. We administered the Decision Aid Acceptability Questionnaire (DAAQ) and the Preparation for Decision-Making Scale (PrepDM) which assessed how well the Hub prepared parents for decision-making with the urologist. Post-consultation, we assessed participants' perception of involvement in decision-making with the Shared Decision-making Questionnaire (SDM-Q-9) and the Decision Regret Scale (DRS). A bivariate analysis compared participants' baseline and pre/post-consultation hypospadias knowledge, decisional conflict, and treatment preference. Using a thematic analysis, we analyzed our semi-structured interviews to uncover how the Hub impacted the consultation and what influenced participants' decisions.

RESULTS

Of 148 parents contacted, 134 were eligible and 65/134 (48.5%) enrolled: mean age 29.2, 96.9% female, 76.6% White (Extended Summary Figure). Pre/post-viewing the Hub, there was a statistically significant increase in hypospadias knowledge (54.3 vs. 75.6, p < 0.001) and decrease in decisional conflict (36.0 vs. 21.9, p < 0.001). Most participants (83.3%) thought Hub's length and amount of information (70.4%) was "about right", and 93.0% found most or everything was clear. Pre/post-consultation, there was a statistically significant decrease in decisional conflict (21.9 vs. 8.8, p < 0.001). PrepDM's mean score was 82.6/100 (SD = 14.1); SDM-Q-9's mean score was 82.5/100 (SD = 16.7). DCS's mean score was 25.0/100 (SD = 47.03). Each participant spent an average of 25.75 min reviewing the Hub. Based on thematic analysis, the Hub helped participants feel prepared for the consultation.

DISCUSSION

Participants engaged extensively with the Hub and demonstrated improved hypospadias knowledge and decision quality. They felt prepared for the consultation and perceived a high degree of involvement in decision-making.

CONCLUSION

As the first pilot test of a pediatric urology DA, the Hub was acceptable and study procedures were feasible. We plan to conduct a randomized controlled trial of the Hub versus usual care to test its efficacy to improve the quality of shared decision-making and reduce long-term decisional regret.

A general method for calculating power for GEE analysis of complete and incomplete stepped wedge cluster randomized trials

Stepped wedge designs have uni-directional crossovers at randomly assigned time points (steps) where clusters switch from control to intervention condition. Incomplete stepped wedge designs are increasingly used in cluster randomized trials of health care interventions and have periods without data collection due to logistical, resource and patient-centered considerations. The development of sample size formulae for stepped wedge trials has primarily focused on complete designs and continuous responses. Addressing this gap, a general, fast, non-simulation based power procedure is proposed for generalized estimating equations analysis of complete and incomplete stepped wedge designs and its predicted power is compared to simulated power for binary and continuous responses. An extensive set of simulations for six and twelve clusters is based upon the Connect-Home trial with an incomplete stepped wedge design. Results show that empirical test size is well controlled using a t-test with bias-corrected sandwich variance estimator for as few as six clusters. Analytical power agrees well with a simulated power in scenarios with twelve clusters. For six clusters, analytical power is similar to simulated power with estimation using the correctly specified model-based variance estimator. To explore the impact of study design choice on power, the proposed fast GEE power method is applied to the Connect-Home trial design, four alternative incomplete stepped wedge designs and one complete design.

power swgee: GEE-based power calculations in stepped wedge cluster randomized trials

Stepped wedge cluster randomized trials are increasingly being used to evaluate interventions in medical, public health, educational, and social science contexts. With the longitudinal and crossover nature of a SW-CRT, complex analysis techniques are often needed which makes appropriately powering SW-CRTs challenging. In this paper, we introduce a newly-developed SW-CRT power calculator, embedded within the power command in Stata. The power calculator assumes a marginal model (i.e., generalized estimating equations [GEE]) for the primary analysis of SW-CRTs, for which other currently available SW-CRT power calculators may not be suitable. The program accommodates complete cross-sectional and closed-cohort designs, and includes multilevel correlation structures appropriate for such designs. We discuss the methods and formulae underlying our SW-CRT calculator, and provide illustrative examples of the use of power swgee. We provide suggestions about the choice of parameters in power swgee, and conclude by discussing areas of future research which may improve the program.

Decline in Lung Function From Mid-to Late-Life With Central Arterial Stiffness: The Atherosclerosis Risk in Communities Study

We investigated the association of lung function at mid-life, later in life, and its 20-year decline, with arterial stiffness later in life. We examined 5720 Atherosclerosis Risk in Communities Study participants who attended Visits 1 (1987-1989) and 5 (2011-2013). Lung function measures were forced expiratory volume in one second (FEV1) and forced vital capacity (FVC), obtained at Visits 1, 2 (1990-1992), and 5. Central artery stiffness (carotid-femoral pulse wave velocity [cfPWV]) was measured at Visit 5. We evaluated associations of lung function with later-life central artery stiffness and cfPWV >75th percentile by multivariable linear and logistic regressions. Lung function at Visit 1 (FEV1 β: -26, 95% Confidence Interval [CI]: -48, -5; FVC β: -14, 95% CI: -32, 5) and Visit 5 (FEV1 β: -22, 95% CI: -46, 2; FVC β: -18, 95% CI: -38, 2) were inversely associated with cfPWV at Visit 5, and with odds of high cfPWV in fully adjusted models. Twenty-year decline in lung function was not associated with continuous or dichotomous measures of arterial stiffness (FEV1 β: 11, 95% CI: -46, 68; FVC β: -4, 95% CI: -52, 43). Lung function at mid-life and late-life was inversely associated with arterial stiffness in later life.

Ratio of Omega-6/Omega-3 Polyunsaturated Fatty Acids Associated With Somatic and Depressive Symptoms in People With Painful Temporomandibular Disorder and Irritable Bowel Syndrome

Somatic symptom disturbance is among the strongest predictors of painful temporomandibular disorder (TMD). Related psychological constructs, such as anxiety and depression, respond therapeutically to omega-3 polyunsaturated fatty acids (PUFAs) in clinical trials. This cross-sectional study investigated associations between the omega-6/omega-3 PUFA ratio and somatic symptom disturbance and depressive symptoms in a community-based sample of 501 adults and determined whether these associations differed between adults with and without TMD or irritable bowel syndrome (IBS). Liquid chromatography tandem mass spectrometry quantified PUFAs in circulating erythrocytes. Somatic symptoms and depression were quantified using Symptom Checklist-90-Revised subscales. Presence or absence of TMD and IBS, respectively, were determined by clinical examination and Rome III screening questions. The standardized beta coefficient for the omega-6/omega-3 long-chain PUFA ratio was 0.26 (95% confidence limits (CL): 0.08, 0.43) in a multivariable linear regression model in which somatic symptom disturbance was the dependent variable. When modelling depressive symptoms as the dependent variable, the standardized beta coefficient was 0.17 (95% CL:0.01, 0.34). Both associations were stronger among TMD cases and IBS cases than among non-cases. Future randomized control trials that lower the omega-6/omega-3 PUFA ratio could consider somatic or depressive symptoms as a therapeutic target for TMD or IBS pain. PERSPECTIVE: In people with TMD or IBS, a high n-6/n-3 PUFA ratio was positively associated with somatic symptom disturbance and depressive symptoms. Both measures of psychological distress were elevated in people with painful TMD and IBS. Future randomized clinical trials will determine whether lowering the n-6/n-3 ratio is therapeutic for pain.

Circulating Omega-6 and Omega-3 Polyunsaturated Fatty Acids in Painful Temporomandibular Disorder and Low Back Pain

Preclinical studies demonstrate opposing effects of long-chain polyunsaturated fatty acid (PUFA) metabolites on inflammation and nociception. Omega-6 (n-6) PUFAs amplify both processes while omega-3 (n-3) PUFAs inhibit them. This cross-sectional study examined relationships between PUFAs in circulating erythrocytes and 2 chronic idiopathic pain conditions: temporomandibular disorder (TMD) and low back pain in a community-based sample of 503 U.S. adults. Presence or absence of TMD and low back pain, respectively, were determined by clinical examination and by responses to established screening questions. Liquid chromatography-tandem mass spectrometry quantified PUFAs. In multivariable logistic regression models, a higher ratio of n-6/n-3 long-chain PUFAs was associated with greater odds of TMD (odds ratio ((OR) = 1.75, 95% confidence limits (CL): 1.16, 2.64) and low back pain (OR = 1.63, 95% CL: 1.07, 2.49). Higher levels of the pronociceptive n-6 long-chain arachidonic acid (AA) were associated with a greater probability of both pain conditions for women, but not men. Higher levels of the antinociceptive long-chain n-3 PUFAs eicosapentaenoic and docosahexaenoic acids were associated with a lower probability of both pain conditions for men, but not women. As systemic inflammation is not a hallmark of these conditions, PUFAs may influence idiopathic pain through other mechanisms. PERSPECTIVE: This cross-sectional clinical study found that a higher ratio of circulating n-6/n-3 long-chain PUFAs was associated with greater odds of 2 common chronic overlapping pain conditions. This suggests that the pro and antinociceptive properties of n-6 and n-3 PUFAs, respectively, influence pain independently of their well-established inflammatory pathways.

Circulating polyunsaturated fatty acids, pressure pain thresholds, and nociplastic pain conditions

OBJECTIVE

Polyunsaturated fatty acids (PUFAs) play a role in pain regulation. This study sought to determine whether free PUFAs found in red blood cells also play a role in nociceptive processing. We examined associations between circulating PUFAs and nociceptive thresholds to noxious mechanical stimuli. We also determined whether nociceptive thresholds were associated with nociplastic pain conditions.

METHODS

This cross-sectional study used stored red bloods cells and data from 605 adult participants in the OPPERA-2 study of chronic overlapping pain conditions. In OPPERA-2 adults completed quantitative sensory testing in which pressure algometry measured deep muscular tissue sensitivity at six anatomical sites. Standardized protocols classified adults for presence or absence of five nociplastic pain conditions: temporomandibular disorder, headache, low back pain, irritable bowel syndrome and fibromyalgia. Liquid chromatography tandem mass spectroscopy quantified erythrocyte PUFAs. We conducted three sets of analyses. First, a multivariable linear regression model assessed the association between n-6/n-3 PUFA ratio and the number of overlapping nociplastic pain conditions. Second, a series of 36 multivariable linear regression models assessed covariate-adjusted associations between PUFAs and nociceptive thresholds at each of six anatomical sites. Third, a series of 30 multivariable linear regression models assessed covariate-adjusted associations between nociceptive thresholds at six anatomical sites and each of five pain conditions.

RESULTS

In multiple linear regression, each unit increase in n-6/n-3 PUFA ratio was associated with more pain conditions (β = 0.30, 95% confidence limits: 0.07, 0.53, p = 0.012). Omega-6 linoleic acid and arachidonic acid were negatively associated with lower nociceptive thresholds at three and at five, respectively, anatomical sites. In contrast, omega-3 alpha-linolenic acid, eicosapentaenoic acid, docosahexaenoic acid and the n-6/n-3 PUFA ratio were not associated with nociceptive thresholds at any site. Pain cases had significantly lower nociceptive thresholds than non-case controls at all anatomical sites.

CONCLUSION

A higher n-6/n-3 PUFA ratio was associated with more pain conditions. Omega-6 PUFAs may promote a generalized upregulation of nociceptive processing.

Adolescent Experiences of Violence Victimizations Among Minors Who Exchange Sex/Experience Minor Sex Trafficking

This work investigates the associations between experiences of domestic minor sex trafficking and adolescent interpersonal violence victimizations, including intimate partner violence (IPV) and community violence. Abuse and violence in childhood are commonly proposed as risk factors for domestic minor sex trafficking. However, less is known about how interpersonal violence victimizations in adolescence connect to domestic minor sex trafficking experiences. The poly-victimization framework provides a means to understand domestic minor sex trafficking as a type of violence amid a web of additional interconnected violence victimizations. Efforts to better understand the interpersonal violence experienced by survivors of domestic minor sex trafficking are valuable in contextualizing trafficking experiences for adolescents. Data from The National Longitudinal Study of Adolescent to Adult Health, a population-based sample of adolescents in the United States (n = 12,605) were used to examine experiences of domestic minor sex trafficking for minor respondents, as measured through questions about exchanging sex for money or drugs. A multivariable logistic regression model was used to estimate the associations between domestic minor sex trafficking and IPV or community violence, while controlling for demographic variables and adolescent risk behaviors. Minors who experience community violence had significantly greater odds of having exchanged sex (aOR: 1.86; 95% CI: 1.32 -2.64). However, IPV was not significantly associated with minors' experiences of sex exchange (aOR: 1.14; 95% CI: 0.85 -1.54). Alcohol or drug use (aOR: 1.87; 95% CI: 1.32 -2.65) and having run away (aOR: 2.04; 95% CI: 1.53 -2.72) were also significantly associated with minor sex exchange. As experiences of domestic minor sex trafficking were significantly associated with community violence victimizations, prevention and intervention efforts targeting youth at high risk for or survivors of domestic minor sex trafficking should consider how community violence victimizations impact these adolescent populations, and programming/messaging should be adjusted to account for these additional violence victimizations.

Marginal modeling of cluster-period means and intraclass correlations in stepped wedge designs with binary outcomes

Stepped wedge cluster randomized trials (SW-CRTs) with binary outcomes are increasingly used in prevention and implementation studies. Marginal models represent a flexible tool for analyzing SW-CRTs with population-averaged interpretations, but the joint estimation of the mean and intraclass correlation coefficients (ICCs) can be computationally intensive due to large cluster-period sizes. Motivated by the need for marginal inference in SW-CRTs, we propose a simple and efficient estimating equations approach to analyze cluster-period means. We show that the quasi-score for the marginal mean defined from individual-level observations can be reformulated as the quasi-score for the same marginal mean defined from the cluster-period means. An additional mapping of the individual-level ICCs into correlations for the cluster-period means further provides a rigorous justification for the cluster-period approach. The proposed approach addresses a long-recognized computational burden associated with estimating equations defined based on individual-level observations, and enables fast point and interval estimation of the intervention effect and correlations. We further propose matrix-adjusted estimating equations to improve the finite-sample inference for ICCs. By providing a valid approach to estimate ICCs within the class of generalized linear models for correlated binary outcomes, this article operationalizes key recommendations from the CONSORT extension to SW-CRTs, including the reporting of ICCs.

Sample size calculators for planning stepped-wedge cluster randomized trials: a review and comparison

Recent years have seen a surge of interest in stepped-wedge cluster randomized trials (SW-CRTs). SW-CRTs include several design variations and methodology is rapidly developing. Accordingly, a variety of power and sample size calculation software for SW-CRTs has been developed. However, each calculator may support only a selected set of design features and may not be appropriate for all scenarios. Currently, there is no resource to assist researchers in selecting the most appropriate calculator for planning their trials. In this paper, we review and classify 18 existing calculators that can be implemented in major platforms, such as R, SAS, Stata, Microsoft Excel, PASS and nQuery. After reviewing the main sample size considerations for SW-CRTs, we summarize the features supported by the available calculators, including the types of designs, outcomes, correlation structures and treatment effects; whether incomplete designs, cluster-size variation or secular trends are accommodated; and the analytical approach used. We then discuss in more detail four main calculators and identify their strengths and limitations. We illustrate how to use these four calculators to compute power for two real SW-CRTs with a continuous and binary outcome and compare the results. We show that the choice of calculator can make a substantial difference in the calculated power and explain these differences. Finally, we make recommendations for implementing sample size or power calculations using the available calculators. An R Shiny app is available for users to select the calculator that meets their requirements (https://douyang.shinyapps.io/swcrtcalculator/).

Dementia and COVID-19 Infection Control in Assisted Living in Seven States

Background

Assisted living (AL) is the largest residential long‐term care provider in the United States, including for persons with Alzheimer's disease and related dementias. Despite recognizing the challenge of infection control for persons with dementia, this study of 119 AL communities is the first to describe dementia‐relevant COVID‐19 infection control across different types of AL communities, and to discuss implications for the future.

Methods

From a parent study sampling frame of 244 AL communities across seven states, 119 administrators provided data about COVID‐19 infection control practices and resident behaviors. Data were collected from July 2020 through September 2021. Communities were differentiated based on the presence of beds/units dedicated for persons living with dementia, as being either dementia‐specific, mixed, or integrated. Data obtained from administrators related to feasibility of implementing seven infection control practices, and the extent to which residents themselves practiced infection prevention. Analyses compared practices across the three community types.

Results

Less than half of administrators found it feasible to close indoor common areas, all community types reported a challenge organizing group activities for safe distancing, and more than half of residents with dementia did not wear a face covering or maintain physical distance from other residents when indicated. Dementia‐specific AL communities were generally the most challenged with infection control during COVID‐19.

Conclusion

All AL community types experienced infection control challenges, more so in dementia‐specific communities (which generally provide care to persons with more advanced dementia and have fewer private beds). Results indicate a need to bolster infection prevention capacity when caring for this especially vulnerable population, and have implications for care in nursing homes as well.

Lexicogrammatical skills in 2-year-old children with and without repaired cleft palate

The purpose of the current research was to compare the lexical-grammatical skills of two-year-old children with and without repaired cleft palate (CP), accounting for the effect of variables such as vocabulary size at 18 months of age, maternal education level, and gender. Participants included 52 children with CP and 25 typically developing (TD) children. The CDI-WS was employed to measure vocabulary and grammatical skills. Significant differences were observed between the CP and TD groups with respect to the number of words, word forms (irregular nouns and verbs), word endings (overuse of plural (-s) and past tense (-ed) markers), the mean number of morphemes in their three longest utterances (M3L), and sentence complexity. In addition, compared to TD children, significantly smaller proportions of children with CP were observed to use words to talk about past and future events or use words to talk about an absent object. The difference between the CP and TD groups in terms of the size of vocabulary at 24 months of age remained statistically significant in the multivariable model. Among all predictors, the size of vocabulary at 18 months of age was identified as the most robust precursor of lexical and grammatical skills at 24 months of age. Gender was identified as a predictor of the M3L measure as an index for syntactic ability.

PARTITIONING AROUND MEDOIDS CLUSTERING AND RANDOM FOREST CLASSIFICATION FOR GIS-INFORMED IMPUTATION OF FLUORIDE CONCENTRATION DATA

Community water fluoridation is an important component of oral health promotion, as fluoride exposure is a well-documented dental caries-preventive agent. Direct measurements of domestic water fluoride content provide valuable information regarding individuals' fluoride exposure and thus caries risk; however, they are logistically challenging to carry out at a large scale in oral health research. This article describes the development and evaluation of a novel method for the imputation of missing domestic water fluoride concentration data informed by spatial autocorrelation. The context is a state-wide epidemiologic study of pediatric oral health in North Carolina, where domestic water fluoride concentration information was missing for approximately 75% of study participants with clinical data on dental caries. A new machine-learning-based imputation method that combines partitioning around medoids clustering and random forest classification (PAMRF) is developed and implemented. Imputed values are filtered according to allowable error rates or target sample size, depending on the requirements of each application. In leave-one-out cross-validation and simulation studies, PAMRF outperforms four existing imputation approaches-two conventional spatial interpolation methods (i.e., inverse-distance weighting, IDW and universal kriging, UK) and two supervised learning methods (k-nearest neighbors, KNN and classification and regression trees, CART). The inclusion of multiply imputed values in the estimation of the association between fluoride concentration and dental caries prevalence resulted in essentially no change in PAMRF estimates but substantial gains in precision due to larger effective sample size. PAMRF is a powerful new method for the imputation of missing fluoride values where geographical information exists.

Overdiagnosis of urinary tract infections by nursing home clinicians versus a clinical guideline

PURPOSE

To inform overprescribing and antibiotic stewardship in nursing homes (NHs), we examined the concordance between clinicians' (NH primary care providers and registered nurses) diagnosis of suspected UTI with a clinical guideline treated as the gold standard, and whether clinician characteristics were associated with diagnostic classification.

METHODS

We conducted a cross-sectional web-based survey of a U.S. national convenience sample of NH clinicians. The survey included a discrete choice experiment with 19 randomly selected clinical scenarios of NH residents with possible UTIs. For each scenario, participants were asked if they thought a UTI was likely. Responses were compared to the guideline to determine the sensitivity and specificity of clinician judgment and performance indicators. Multivariable logistic mixed effects regression analysis of demographic, work, personality, and UTI knowledge/attitudes characteristics was conducted.

RESULTS

One thousand seven hundred forty-eight NH clinicians responded to 33,212 discrete choice scenarios; 867 (50%) were NH primary care providers and 881 (50%) were NH registered nurses, 39% were male, and the mean age was 45 years. Participants were uncertain about diagnosis in 30% of scenarios. Correct classification occurred for 66% of all scenarios (providers: 70%; nurses: 62%). Respondent judgment had a sensitivity of 78% (providers: 81%; nurses: 74%) and specificity of 54% (providers: 59%; nurses: 49%) compared to the clinical guideline. Adjusting for covariates in multivariable models, being a nurse and having higher closemindedness were associated higher odds of false positive UTI (odds ratio [OR] 1.61, p < 0.001; and OR 1.09, p = 0.039, respectively), although higher UTI knowledge and conscientiousness were associated with lower odds of false positive UTI ratings (OR 0.80, p < 0.001; OR 0.90, p = 0.005, respectively).

CONCLUSIONS

Clinicians tend to over-diagnose urinary tract infections, necessitating systems-based interventions to augment clinical decision-making. Clinician type, UTI knowledge, and personality traits may also influence behavior and deserve further study.

As-Needed Prescribing and Administration of Psychotropic Medications in Assisted Living: A 7-State Study

OBJECTIVES

Psychotropic medications administered on an as-needed basis, often designated as pro re nata (or PRN, hereafter as-needed), can alleviate acute symptoms and facilitate deprescribing, although as-needed use is associated with negative outcomes such as polypharmacy and drug interactions. The aim of this study was to examine the extent to which PRN psychotropic medications are prescribed and administered to assisted living (AL) residents, overall and in relation to resident- and community-level characteristics.

DESIGN

Cross-sectional study.

SETTING AND PARTICIPANTS

All AL residents and select staff in 250 AL communities in 7 states.

METHODS

Data regarding prescribing and administration of 5 types of PRN psychotropic medications in the prior 7 days (antipsychotics, mood stabilizers and antiepileptics, antidepressants, anxiolytics/hypnotics, and cognitive enhancers) were abstracted and compared across resident and community characteristics.

RESULTS

Prescribing of PRN psychotropics in AL is low (10.3%). Of prescribed PRN medications, 2.5% of residents were administered a PRN and 8% had PRN that was not administered in the previous 7 days. Anxiolytics were administered PRN more commonly compared with antipsychotics (2.0% vs 0.2%). Of all PRN psychotropic prescriptions (n = 1039), 70.5% had a written indication describing the reason for administration. Among PRN medications administered (n = 242), the proportion with an indication was lower (62.0%). PRN psychotropic medication prescribing was higher among residents with dementia and a psychiatric diagnosis, and in larger AL communities and those with a higher proportion of dementia care beds.

CONCLUSIONS AND IMPLICATIONS

The prescribing and administration of PRN psychotropic medications in AL is relatively rare, although more common among residents with dementia. Emerging psychotropic medication policies should be expanded to address mental health care, anxiolytic/hypnotic use for residents living with dementia, PRN prescribing in chart review, and the use of detailed indications for PRN use, especially when medications are administered by unlicensed care staff.

Power considerations for generalized estimating equations analyses of four-level cluster randomized trials

In this article, we develop methods for sample size and power calculations in four-level intervention studies when intervention assignment is carried out at any level, with a particular focus on cluster randomized trials (CRTs). CRTs involving four levels are becoming popular in healthcare research, where the effects are measured, for example, from evaluations (level 1) within participants (level 2) in divisions (level 3) that are nested in clusters (level 4). In such multilevel CRTs, we consider three types of intraclass correlations between different evaluations to account for such clustering: that of the same participant, that of different participants from the same division, and that of different participants from different divisions in the same cluster. Assuming arbitrary link and variance functions, with the proposed correlation structure as the true correlation structure, closed-form sample size formulas for randomization carried out at any level (including individually randomized trials within a four-level clustered structure) are derived based on the generalized estimating equations approach using the model-based variance and using the sandwich variance with an independence working correlation matrix. We demonstrate that empirical power corresponds well with that predicted by the proposed method for as few as eight clusters, when data are analyzed using the matrix-adjusted estimating equations for the correlation parameters with a bias-corrected sandwich variance estimator, under both balanced and unbalanced designs.

Impact of unequal cluster sizes for GEE analyses of stepped wedge cluster randomized trials with binary outcomes

The stepped wedge (SW) design is a type of unidirectional crossover design where cluster units switch from control to intervention condition at different prespecified time points. While a convention in study planning is to assume the cluster‐period sizes are identical, SW cluster randomized trials (SW‐CRTs) involving repeated cross‐sectional designs frequently have unequal cluster‐period sizes, which can impact the efficiency of the treatment effect estimator. In this paper, we provide a comprehensive investigation of the efficiency impact of unequal cluster sizes for generalized estimating equation analyses of SW‐CRTs, with a focus on binary outcomes as in the Washington State Expedited Partner Therapy trial. Several major distinctions between our work and existing work include the following: (i) we consider multilevel correlation structures in marginal models with binary outcomes; (ii) we study the implications of both the between‐cluster and within‐cluster imbalances in sizes; and (iii) we provide a comparison between the independence working correlation versus the true working correlation and detail the consequences of ignoring correlation estimation in SW‐CRTs with unequal cluster sizes. We conclude that the working independence assumption can lead to substantial efficiency loss and a large sample size regardless of cluster‐period size variability in SW‐CRTs, and recommend accounting for correlations in the analysis. To improve study planning, we additionally provide a computationally efficient search algorithm to estimate the sample size in SW‐CRTs accounting for unequal cluster‐period sizes, and conclude by illustrating the proposed approach in the context of the Washington State study.

GEECORR: A SAS macro for regression models of correlated binary responses and within-cluster correlation using generalized estimating equations

BACKGROUND AND OBJECTIVES

Generalized estimating equations (GEE) provide population-averaged model inference for longitudinal and clustered outcomes via a generalized linear model for the effect of explanatory variables on the marginal mean, while intra-cluster correlations are ordinarily treated as nuisance parameters. Software to richly parameterize and conduct inference for complex correlation structures in the marginal modeling framework is scarce.

METHODS

A SAS macro, GEECORR, has been developed for the analysis of clustered binary data based on GEE to include additional estimating equations for modeling pairwise correlation between binary variates as a function of covariates.

RESULTS

We illustrate the macro in a surveillance study with repeated measures, a longitudinal study, and a study with biological clustering.

CONCLUSIONS

This article provides an overview of the GEE method consisting of a pair of estimating equations, describes the features and capabilities of the GEECORR macro including regression diagnostics and finite-sample bias-corrected covariance estimators, and demonstrates the macro usage for three studies.

Children's oral health-related behaviours and early childhood caries: A latent class analysis

OBJECTIVE

In this cross-sectional study in a large community-based sample of preschool-age children, we sought to identify distinct clusters of modifiable early childhood oral health-related behaviours (OHBs) and quantify their association with clinical and parent-reported measures of early childhood oral health.

METHODS

We relied upon a questionnaire (n = 8033; 11% in Spanish) and clinical oral health data (n = 6404; early childhood caries [ECC] prevalence = 54%] collected in the context of an epidemiologic study of early childhood oral health among 3- to 5-year-old children in North Carolina. Latent class analysis was used to identify clusters of modifiable OHBs based on parents' responses to 6 questionnaire items pertaining to their children's oral hygiene, diet and dental home. The optimal number of clusters was determined based on measures of model fit and interpretability. We examined associations of OHB clusters with clinical and parent-reported child oral health status (ie, ECC prevalence, severity and proportion with untreated disease) using bivariate association tests and multivariable regression modelling with marginal effects estimation accounting for clustered data. We used Mplus v.8.6 (Muthén & Muthén, Los Angeles, CA, USA) and Stata v.16.1 (StataCorp, College Station, TX, USA) for data analyses.

RESULTS

We identified 2 OHB clusters, a favourable (74%) and an unfavourable (26%) one. Children in the favourable OHB cluster had better oral hygiene practices (ie, tooth brushing frequency and fluoridated toothpaste use), lower consumption frequency of sugar-containing snacks and beverages, less frequent reports of night-time bottle-feeding history and a higher likelihood of a dental home. Children in the unfavourable cluster had significantly higher ECC prevalence (57% vs 53%), caries burden (mean dmfs = 9.3 vs 7.6), untreated disease (43% vs 33%) and worse parent-reported oral health status than the favourable cluster.

CONCLUSIONS

Our findings demonstrate the importance and utility of clustering common, modifiable ECC risk factors in population studies - health promotion efforts may centre on groups of people rather than individual behavioural risk factors.

Nontraditional Small House Nursing Homes Have Fewer COVID-19 Cases and Deaths

OBJECTIVES

Green House and other small nursing home (NH) models are considered "nontraditional" due to their size (10-12 beds), universal caregivers, and other home-like features. They have garnered great interest regarding their potential benefit to limit Coronavirus Disease 2019 (COVID-19) infections due to fewer people living, working, visiting, and being admitted to Green House/small NHs, and private rooms and bathrooms, but this assumption has not been tested. If they prove advantageous compared with other NHs, they may constitute an especially promising model as policy makers and providers reinvent NHs post-COVID.

DESIGN

This cohort study compared rates of COVID-19 infections, COVID-19 admissions/readmissions, and COVID-19 mortality, among Green House/small NHs with rates in other NHs between January 20, 2020 and July 31, 2020.

SETTING AND PARTICIPANTS

All Green House homes that held a skilled nursing license and received Medicaid or Medicare payment were invited to participate; other small NHs that replicate Green House physical design and operational practices were eligible if they had the same licensure and payer sources. Of 57 organizations, 43 (75%) provided complete data, which included 219 NHs. Comparison NHs (referred to as "traditional NHs") were up to 5 most geographically proximate NHs within 100 miles that had <50 beds and ≥50 beds for which data were available from the Centers for Medicare and Medicaid Services (CMS). Because Department of Veterans Affairs organizations are not required to report to CMS, they were not included.

METHODS

Rates per 1000 resident days were derived for COVID-19 cases and admissions, and per 100 COVID-19 positive cases for mortality. A log-rank test compared rates between Green House/small NHs and traditional NHs with <50 beds and ≥50 beds.

RESULTS

Rates of all outcomes were significantly lower in Green House/small NHs than in traditional NHs that had <50 beds and ≥50 beds (log-rank test P < .025 for all comparisons). The median (middle value) rates of COVID-19 cases per 1000 resident days were 0 in both Green House/small NHs and NHs <50 beds, while they were 0.06 in NHs ≥50 beds; in terms of COVID-19 mortality, the median rates per 100 positive residents were 0 (Green House/small NHs), 10 (<50 beds), and 12.5 (≥50 beds). Differences were most marked in the highest quartile: 25% of Green House/small NHs had COVID-19 case rates per 1000 resident days higher than 0.08, with the corresponding figures for other NHs being 0.15 (<50 beds) and 0.74 (≥50 beds).

CONCLUSIONS AND IMPLICATIONS

COVID-19 incidence and mortality rates are less in Green House/small NHs than rates in traditional NHs with <50 and ≥50 beds, especially among the higher and extreme values. Green House/small NHs are a promising model of care as NHs are reinvented post-COVID.

Transitional care from skilled nursing facilities to home: study protocol for a stepped wedge cluster randomized trial

Background

Skilled nursing facility (SNF) patients are medically complex with multiple, advanced chronic conditions. They are dependent on caregivers and have experienced recent acute illnesses. Among SNF patients, the rate of mortality or acute care use is over 50% within 90 days of discharge, yet these patients and their caregivers often do not receive the quality of transitional care that prepares them to manage serious illnesses at home.

Methods

The study will test the efficacy of Connect-Home, a successfully piloted transitional care intervention targeting seriously ill SNF patients discharged to home and their caregivers. The study setting will be SNFs in North Carolina, USA, and, following discharge, in patients’ home. Using a stepped wedge cluster randomized trial design, six SNFs will transition at randomly assigned intervals from standard discharge planning to the Connect-Home intervention. The SNFs will contribute data for patients (N = 360) and their caregivers (N = 360), during both the standard discharge planning and Connect-Home time periods. Connect-Home is a two-step intervention: (a) SNF staff create an individualized Transition Plan of Care to manage the patient’s illness at home; and (b) a Connect-Home Activation RN visits the patient’s home to implement the written Transition Plan of Care. A key feature of the trial includes training of the SNF and Home Care Agency staff to complete the transition plan rather than using study interventionists. The primary outcomes will be patient preparedness for discharge and caregiver preparedness for caregiving role. With the proposed sample and using a two-sided test at the 5% significance level, we have 80% power to detect a 18% increase in the patient’s preparedness for discharge score. We will employ linear mixed models to compare observations between intervention and usual care periods to assess primary outcomes. Secondary outcomes include (a) patients’ quality of life, functional status, and days of acute care use and (b) caregivers’ burden and distress.

Discussion

Study results will determine the efficacy of an intervention using existing clinical staff to (a) improve transitional care for seriously ill SNF patients and their caregivers, (b) prevent avoidable days of acute care use in a population with persistent risks from chronic conditions, and (c) advance the science of transitional care within end-of-life and palliative care trajectories of SNF patients and their caregivers. While this study protocol was being implemented, the COVID-19 pandemic occurred and this protocol was revised to mitigate COVID-related risks of patients, their caregivers, SNF staff, and the study team. Thus, this paper includes additional material describing these modifications.

Trial registration

ClinicalTrials.gov NCT03810534. Registered on January 18, 2019.

Emergence of Prevocalic Stop Consonants in Children With Repaired Cleft Palate

Purpose This study determined the time course of the emergence of prevocalic stop consonants in young children with cleft palate following surgical repair. Method A total of 120 children in four cohorts from three institutions were followed from 12 to 24 months of age: (a) 24 with repaired cleft lip and palate (CLP), (b) 36 with repaired cleft palate only (CP), (c) 33 without clefts but with histories of frequent otitis media and ventilation tubes (OM), and (d) 27 typically developing (TD) children without clefts or OM. Emergence of prevocalic stops and symbolic language skills were determined during administration of the Communication and Symbolic Behavioral Scales Developmental Profile. Parametric survival models were fitted with and without covariates-recruitment site, gender, maternal education level, middle ear status, language ability, and age at surgery for children with clefts-to describe the time course of the emergence of prevocalic stops. Results The estimated age at which 80% of children demonstrated prevocalic stop emergence was 15.0, 15.3, 18.9, and 21.8 months for TD, OM, CP, and CLP groups, respectively (p < .001, unadjusted model). Both CP and CLP cohorts had a significantly longer time to stop emergence than either the TD or OM cohorts, even after adjusting for covariates. Abnormal middle ear status, lower symbolic language ability, and older age at palatal surgery were significantly associated with delayed stop emergence. Conclusions Survival model estimates show that four out of five children with repaired cleft palate will achieve emergence of prevocalic stop consonants by 19-22 months of age, corresponding to 9-12 months following palate repair. Clinical implications are discussed.

Cohort Profile: ZOE 2.0-A Community-Based Genetic Epidemiologic Study of Early Childhood Oral Health

Early childhood caries (ECC) is an aggressive form of dental caries occurring in the first five years of life. Despite its prevalence and consequences, little progress has been made in its prevention and even less is known about individuals' susceptibility or genomic risk factors. The genome-wide association study (GWAS) of ECC ("ZOE 2.0") is a community-based, multi-ethnic, cross-sectional, genetic epidemiologic study seeking to address this knowledge gap. This paper describes the study's design, the cohort's demographic profile, data domains, and key oral health outcomes. Between 2016 and 2019, the study enrolled 8059 3-5-year-old children attending public preschools in North Carolina, United States. Participants resided in 86 of the state's 100 counties and racial/ethnic minorities predominated-for example, 48% (n = 3872) were African American, 22% white, and 20% (n = 1611) were Hispanic/Latino. Seventy-nine percent (n = 6404) of participants underwent clinical dental examinations yielding ECC outcome measures-ECC (defined at the established caries lesion threshold) prevalence was 54% and the mean number of decayed, missing, filled surfaces due to caries was eight. Nearly all (98%) examined children provided sufficient DNA from saliva for genotyping. The cohort's community-based nature and rich data offer excellent opportunities for addressing important clinical, epidemiologic, and biological questions in early childhood.

Is Fluorescence Technology a Promising Tool for Detecting Infected Dentin in Deep Carious Lesions?

The purpose of this study was to determine if the degree of fluorescence detected by fluorescence-aided caries excavation (FACE) correlates with dentin bacterial microbiome diversity, as assessed by 16S rRNA gene amplicon sequencing, and with traditional tactile dentin caries assessment. Unidentified human teeth were obtained from a dental facility. The included teeth had a carious lesion two-thirds into the dentin, verified by radiography, and were red-fluorescing (RF) using FACE technology (SIROInspect; Sirona, Bensheim, Germany). Two independent examiners performed visual/tactile assessment of the lesions. RF sites were sampled with a sterile spoon excavator and dentin characteristics were evaluated. Once RF dentin was removed, a second sample of pink-fluorescing (PF) dentin was obtained. After excavation with a sterile round bur to nonfluorescing (NF) dentin, a third sample was collected with a slow-speed round bur. The samples were processed at the UNC (University of North Carolina at Chapel Hill) Microbiome Core Facility. Out of 134 extracted teeth collected, 21 fit the inclusion criteria, yielding 61 dentin samples. RF samples had the highest number of observed operational taxonomic units (n = 154), followed by PF (n = 109) and NF (n = 100). RF carious dentin was primarily "soft," and NF dentin was assessed as "hard" 100% of the time by both examiners (rank correlation χ2: p < 0.001). However, approximately one-third of the tactile assessments of hard dentin still displayed some fluorescence, either pink or red. We concluded that the sampled fluorescing (RF and PF) and NF carious dentin layers displayed diverse bacterial taxa, and tactile assessments of soft, leathery, and hard corresponded with RF, PF, and NF.

Effectiveness of a Mouth Care Program Provided by Nursing Home Staff vs Standard Care on Reducing Pneumonia Incidence: A Cluster Randomized Trial

Importance

Pneumonia affects more than 250 000 nursing home (NH) residents annually. A strategy to reduce pneumonia is to provide daily mouth care, especially to residents with dementia.

Objective

To evaluate the effectiveness of Mouth Care Without a Battle, a program that increases staff knowledge and attitudes regarding oral hygiene, changes mouth care, and improves oral hygiene, in reducing the incidence of pneumonia among NH residents.

Design, Setting, and Participants

This pragmatic cluster randomized trial observing 2152 NH residents for up to 2 years was conducted from September 2014 to May 2017. Data collectors were masked to study group. The study included 14 NHs from regions of North Carolina that evidenced proportionately high rehospitalization rates for pneumonia and long-term care residents. Nursing homes were pair matched and randomly assigned to intervention or control groups.

Intervention

Mouth Care Without a Battle is a standardized program that teaches that mouth care is health care, provides instruction on individualized techniques and products for mouth care, and trains caregivers to provide care to residents who are resistant and in special situations. The control condition was standard mouth care.

Main Outcomes and Measures

Pneumonia incidence (primary) and hospitalization and mortality (secondary), obtained from medical records.

Results

Overall, the study enrolled 2152 residents (mean [SD] age, 79.4 [12.4] years; 1281 [66.2%] women; 1180 [62.2%] white residents). Participants included 1219 residents (56.6%) in 7 intervention NHs and 933 residents (43.4%) in 7 control NHs. During the 2-year study period, the incidence rate of pneumonia per 1000 resident-days was 0.67 and 0.72 in the intervention and control NHs, respectively. Neither the primary (unadjusted) nor secondary (covariate-adjusted) analyses found a significant reduction in pneumonia due to Mouth Care Without a Battle during 2 years (unadjusted incidence rate ratio, 0.90; upper bound of 1-sided 95% CI, 1.24; P = .27; adjusted incidence rate ratio, 0.92; upper bound of 1-sided 95% CI, 1.27; P = .30). In the second year, the rate of pneumonia was nonsignificantly higher in intervention NHs. Adjusted post hoc analyses limited to the first year found a significant reduction in pneumonia incidence in intervention NHs (IRR, 0.69; upper bound of 1-sided 95% CI, 0.94; P = .03).

Conclusions and Relevance

This matched-pairs cluster randomized trial of a mouth care program compared with standard care was not effective in reducing pneumonia incidence at 2 years, although reduction was found during the first year. The lack of significant results in the second year may be associated with sustainability. Improving mouth care in US NHs may require the presence and support of dedicated oral care aides.

Trial Registration

ClinicalTrials.gov Identifier: NCT03817450.

Flossing Is Associated with Improved Oral Health in Older Adults

The effect of preventive oral habits is largely unexplored in older individuals. The purpose of this study was to evaluate the associations between home use of flossing and prevalence of periodontal disease and caries in older adults. Five-year incident tooth loss was also evaluated. Data on 686 individuals ≥65 y-old from the Piedmont 65+ Dental Study were examined including: 1) interproximal clinical attachment level (iCAL), 2) interproximal probing depth (iPD), 3) numbers of caries, and 4) missing teeth. Flossing behavior was evaluated according to the Periodontal Profile Class (PPC) system. Five-year follow-up data (n = 375) was evaluated for incident tooth loss. Dichotomous and categorical variables were analyzed using Pearson chi-square tests as well as covariate-adjusted Cochran-Mantel-Haenszel tests. Multiple linear regression compared clinical parameters based on flossing behavior. Elderly flossers had lower (mean, SE) %iCAL≥3 mm (38.2, 2.38 vs. 48.8, 1.56) and %iPD≥4 mm (8.70, 1.41 vs. 14.4, 0.93) compared to nonflossers (P ≤ 0.005). Flossers showed less coronal caries compared to nonflossers (P = 0.02). Baseline number of missing teeth (mean, SE) was 11.5 (0.35) in nonflossers compared to 8.6 (0.53) in flossers (P < 0.0001). Regular dental visitors had lower oral disease levels compared to episodic dental users. The majority of flossers classified into PPC-Stage I (health) whereas nonflossers classified as PPC-Stages V, VI, and VII (disease). At the 5-y follow-up visit, the average tooth loss for flossers was ~1 tooth compared to ~4 teeth lost for nonflossers (P < 0.0001). Among all teeth, molars showed the highest benefit (>40%) for flossing behavior (P = 0.0005). In conclusion, the extent of oral disease for older individuals was significantly less in flossers than in nonflossers. Flossers showed less periodontal disease, fewer dental caries, and loss of fewer teeth over a 5-y period. These findings further support flossing as an important oral hygiene behavior to prevent oral disease progression in older adults.

Vocabulary Growth From 18 to 24 Months of Age in Children With and Without Repaired Cleft Palate

Purpose This study investigated vocabulary growth from 18 to 24 months of age in young children with repaired cleft palate (CP), children with otitis media, and typically developing (TD) children. In addition, the contributions of factors such as hearing level, middle ear status, size of consonant inventory, maternal education level, and gender to the development of expressive vocabulary were explored. Method Vocabulary size of 40 children with repaired CP, 29 children with otitis media, and 25 TD children was measured using the parent report on MacArthur-Bates Communicative Development Inventories: Words and Sentences (Fenson et al., 2007) at 18 and 24 months of age. All participants underwent sound field audiometry at 12 months of age and tympanometry at 18 months of age. A multiple linear regression with and without covariates was used to model vocabulary growth from 18 to 24 months of age across the 3 groups. Results Children with CP produced a significantly smaller number of words at 24 months of age and showed significantly slower rate of vocabulary growth from 18 to 24 months of age when compared to TD children (p < .05). Although middle ear status was found to predict vocabulary growth from 18 to 24 months of age across the 3 groups (p < .05), the confidence interval was large, suggesting the effect should be interpreted with caution. Conclusions Children with CP showed slower expressive vocabulary growth relative to their age-matched TD peers. Middle ear status may be associated with development of vocabulary skills for some children.

The Effect of Physician Oral Health Services on Dental Use and Expenditures under General Anesthesia

BACKGROUND

Despite early evidence touting the effectiveness of physician-provided oral health services (POHS), recent evidence suggests these services might have little impact on caries-related outcomes in children. General anesthesia (GA) is often used to treat early childhood caries and may be considered the most extreme utilization outcome. We sought to assess the impact of POHS utilization on dental GA utilization and expenditures.

METHODS

We used the Medicaid claims of a birth cohort of children born in 2008 in North Carolina (N = 32,558) to determine the impact of POHS on dental utilization and expenditures under GA for individual children. Children were followed until their eighth birthday. We analyzed the association of the number of prior POHS visits with visit-specific outcomes of dental treatment under GA using population-averaged models fit with generalized estimating equations with exchangeable working correlation structure.

RESULTS

Children with 2 or more previous POHS visits had reduced odds of GA (odds ratio [OR] = 0.93; confidence interval [CI], 0.87-0.99; P = 0.029) and expenditures ($114; CI,-$152.61 to -$75.19; P < 0.001) compared to those without physician-provided oral health visits, adjusting for age, sex, race/ethnicity, and geographic residence. Dental expenditures did not differ between POHS and non-POHS subjects at non-GA visits.

CONCLUSIONS

POHS decreased the odds of having dental GA treatment and dental expenditures at GA visits. The role of physicians in oral health care can reduce the impact on the most severe outcome-treatment under general anesthesia.

KNOWLEDGE TRANSFER STATEMENT

The results of this study have important financial implications for state Medicaid programs and disease management programs trying to mitigate the costs of treating early childhood caries under general anesthesia. Children who receive physician oral health care are less likely to use and more likely to save money on general anesthesia to complete dental treatment.

A 2-Year Pragmatic Trial of Antibiotic Stewardship in 27 Community Nursing Homes

OBJECTIVES

To determine if antibiotic prescribing in community nursing homes (NHs) can be reduced by a multicomponent antibiotic stewardship intervention implemented by medical providers and nursing staff and whether implementation is more effective if performed by a NH chain or a medical provider group.

DESIGN

Two-year quality improvement pragmatic implementation trial with two arms (NH chain and medical provider group).

SETTING

A total of 27 community NHs in North Carolina that are typical of NHs statewide, conducted before announcement of the US Centers for Medicare and Medicaid Services antibiotic stewardship mandate.

PARTICIPANTS

Nursing staff and medical care providers in the participating NHs.

INTERVENTION

Standardized antibiotic stewardship quality improvement program, including training modules for nurses and medical providers, posters, algorithms, communication guidelines, quarterly information briefs, an annual quality improvement report, an informational brochure for residents and families, and free continuing education credit.

MEASUREMENTS

Antibiotic prescribing rates per 1000 resident days overall and by infection type; rate of urine test ordering; and incidence of Clostridium difficile and methicillin-resistant Staphylococcus aureus (MRSA) infections.

RESULTS

Systemic antibiotic prescription rates decreased from baseline by 18% at 12 months (incident rate ratio [IRR] = 0.82; 95% confidence interval [CI] = 0.69-0.98) and 23% at 24 months (IRR = 0.77; 95% CI = 0.65-0.90). A 10% increase in the proportion of residents with the medical director as primary physician was associated with a 4% reduction in prescribing (IRR = 0.96; 95% CI = 0.92-0.99). Incidence of C. difficile and MRSA infections, hospitalizations, and hospital readmissions did not change significantly. No adverse events from antibiotic nonprescription were reported. Estimated 2-year implementation costs per NH, exclusive of medical provider time, ranged from $354 to $3653.

CONCLUSIONS

Antibiotic stewardship programs can be successfully disseminated in community NHs through either NH administration or medical provider groups and can achieve significant reductions in antibiotic use for at least 2 years. Medical director involvement is an important element of program success. J Am Geriatr Soc 68:46-54, 2019.

Impact of an Individually Tailored Light Mask on Sleep Parameters in Older Adults With Advanced Phase Sleep Disorder

Objective: This study investigated whether light delivered through the eyelids of sleeping persons might create phase delay in older adults who are adversely affected by advanced sleep phase disorder. Participants: Thirty-two cognitively intact, community-dwelling participants aged ≥ 50 years (20 females, 12 males) with Pittsburgh Sleep Quality Index scores ≥ 5 (poor sleep) completed the study. Methods: This within-subjects, randomized, two-treatment crossover design study exposed participants to an active "blue" (λ_max  =  480 nm) lighting intervention or a placebo "red" (λ_max = 640 nm) control through closed eyelids during sleep for 8 weeks. Conditions were administered 1 hr after bedtime using custom-built light masks delivering a train of 2-s duration light pulses presented every 30 s for ≤ 2 hr (approximately 240 pulses/night). Dependent variables were subjective measures of sleep and depression (questionnaires) and objective measures of sleep (wrist actigraphy), analyzed using linear mixed models with treatment, period, and carryover as fixed effects. Results: The actigraphy analysis found no effect of the intervention or the control condition on sleep start time, total sleep time, number of sleep bouts, or sleep efficiency, either compared to baseline or to one another. Subjective responses of study participants, however, indicated statistically significant (p < 0.05) improvement in seven of eight reported measures of sleep quality with both the intervention and the control condition, but no difference between the two conditions. Conclusions: The participants reported improvement in sleep quality, but the intervention did not confer additional advantages after adjusting for period and carryover effects.

Improving Nursing Home Residents' Oral Hygiene: Results of a Cluster Randomized Intervention Trial

OBJECTIVE

A 2-year cluster randomized trial of Mouth Care Without a Battle (MCWB) was conducted in nursing homes (NHs) to determine if recommended mouth care practices provided by NH staff could improve residents' oral hygiene and denture outcomes.

DESIGN

Cluster randomized trial of NHs.

SETTING AND PARTICIPANTS

Seven MCWB NHs and 6 control NHs. A total of 219 NH residents completed baseline and 24-month oral examinations and, if applicable, denture assessments (control = 98, intervention = 121).

INTERVENTION

The intervention consisted of training NH staff in the MCWB protocol, and providing support in its use for 2 years.

MEASURES

Descriptive data from the Minimum Data Set and clinical oral health assessments: the Plaque Index for Long-Term Care (range 0‒3), the Gingival Index for Long-Term Care (range 0‒4), and the Denture Plaque Index (range 0‒4), with lower scores indicating better oral health.

RESULTS

There were no significant demographic or health differences between groups at baseline. Residents' mean age (standard deviation) was 77.8 years (13.5), 71% were female, and 49% had cognitive impairment. At 24 months, there were significant improvements in oral and denture hygiene in the intervention group compared with control (all P < .05) with mean changes in indices that were 0.44 (Plaque Index for Long-Term Care), 0.55 (Gingival Index for Long-Term Care), and 0.67 (Denture Plaque Index) points lower in intervention NHs than control NHs.

CONCLUSIONS AND IMPLICATIONS FOR PRACTICE

Training NH staff to attend to residents' oral hygiene and denture care had a sustained, favorable impact on residents' oral and denture hygiene after 24 months compared with usual care. The protocol, MCWB, can be used by direct caregivers to improve the oral hygiene and denture care of NH residents.

Oral health trajectories in community-dwelling older adults in the last 3 years of life

BACKGROUND

Systemic health and physical, cognitive, and social function gradually decline at the end of life. How oral health change at the end of life in community-dwelling older adults remains unknown, increasing the difficulty in treatment planning for these individuals.

METHODS

A total of 250 community-dwelling decedents of a longitudinal study were included in the analysis. Decedents' clinical and interview data collected at baseline, 18, 36, 60, and 84 months were linked to their death dates and plotted on an inverted time scale to study their oral health changes in the 3 years prior to death. Mixed effect models with random intercepts and slopes of months from death and sociodemographic variables were used to model the subjective and objective oral health trajectories in the last 3 years of life. The association of the change in the outcomes with a linear trend of months from death were assessed using the F-test.

RESULTS

Number of missing teeth was significantly associated with a linear trend of months from death. With every month in proximity to death, the expected number of missing teeth increased by 0.03 (SE = 0.004; p < 0.001), resulting in an average of 1.08 missing teeth in 3 years. Among subjective outcomes, the odds of good versus poor or "neither" self-rated systemic health (p = 0.009) and mouth appearance (p = 0.008) significantly decreased as death approached.

CONCLUSION

Although other oral health measures remained relatively stable, tooth loss and dissatisfaction with mouth appearance steadily increased in community-dwelling older adults as death approached. These changes were gradual and parallel the end-of-life general health decline.

Power and sample size requirements for GEE analyses of cluster randomized crossover trials

The cluster randomized crossover design has been proposed to improve efficiency over the traditional parallel cluster randomized design, which often involves a limited number of clusters. In recent years, the cluster randomized crossover design has been increasingly used to evaluate the effectiveness of health care policy or programs, and the interest often lies in quantifying the population-averaged intervention effect. In this paper, we consider the two-treatment two-period crossover design, and develop sample size procedures for continuous and binary outcomes corresponding to a population-averaged model estimated by generalized estimating equations, accounting for both within-period and interperiod correlations. In particular, we show that the required sample size depends on the correlation parameters through an eigenvalue of the within-cluster correlation matrix for continuous outcomes and through two distinct eigenvalues of the correlation matrix for binary outcomes. We demonstrate that the empirical power corresponds well with the predicted power by the proposed formulae for as few as eight clusters, when outcomes are analyzed using the matrix-adjusted estimating equations for the correlation parameters concurrently with a suitable bias-corrected sandwich variance estimator.