Designing three-level cluster randomized trials to assess treatment effect heterogeneity.

Cluster randomized trials often exhibit a three-level structure with participants nested in subclusters such as health care providers, and subclusters nested in clusters such as clinics. While the average treatment effect has been the primary focus in planning three-level randomized trials, interest is growing in understanding whether the treatment effect varies among prespecified patient subpopulations, such as those defined by demographics or baseline clinical characteristics. In this article, we derive novel analytical design formulas based on the asymptotic covariance matrix for powering confirmatory analyses of treatment effect heterogeneity in three-level trials, that are broadly applicable to the evaluation of cluster-level, subcluster-level, and participant-level effect modifiers and to designs where randomization can be carried out at any level. We characterize a nested exchangeable correlation structure for both the effect modifier and the outcome conditional on the effect modifier, and generate new insights from a study design perspective for conducting analyses of treatment effect heterogeneity based on a linear mixed analysis of covariance model. A simulation study is conducted to validate our new methods and two real-world trial examples are used for illustrations.

Association Between Rental Assistance Programs and Undiagnosed Diabetes Among U.S.

BACKGROUND: Rental assistance programs have been linked to better housing quality, stability, healthcare access, and reduced likelihood of uncontrolled diabetes. However, its direct association with diabetes screening is uncertain. OBJECTIVE: To determine whether federal rental assistance programs are associated with lower odds of undiagnosed diabetes. DESIGN: We used a quasi-experimental approach, comparing outcomes among adults receiving rental assistance to those who entered assisted housing within 2 years after their health data were collected. We test the a priori hypothesis that rental assistance will be associated with decreased odds of undiagnosed diabetes. PARTICIPANTS: Participants in the National Health and Nutrition Examination Survey 1999-2018 who received rental assistance and who had diabetes. INTERVENTION: Current rental assistance participation, including specific housing programs. MAIN MEASURES: Undiagnosed diabetes based on having hemoglobin A1c ≥ 6.5% but answering no to the survey question of being diagnosed with diabetes. KEY RESULTS: Among 435 eligible adults (median age 54.5 years, female 68.5%, non-Hispanic white 32.5%), 80.7% were receiving rental assistance programs at the time of the interview, and 19.3% went on to receive rental assistance within 2 years. The rates of undiagnosed diabetes were 15.0% and 25.3% among those receiving rental assistance programs vs. those in the future assistance group (p-value = 0.07). In an adjusted logistic regression model, adults receiving rental assistance had lower odds of undiagnosed diabetes (OR 0.52, 95% CI 0.28-0.94) than those in future assistance groups. Sex, race and ethnic group, educational level, and poverty ratio were not significantly associated with having undiagnosed diabetes, but individuals aged 45-64 years had significantly lower odds of undiagnosed diabetes (OR 0.21, 95% CI 0.08-0.53) compared with those aged 18-44. CONCLUSIONS: Rental assistance was linked to lower odds of undiagnosed diabetes, suggesting that affordable housing programs can aid in early recognition and diagnosis, which may improve long-term outcomes.

Group sequential two-stage preference designs.

The two-stage preference design (TSPD) enables inference for treatment efficacy while allowing for incorporation of patient preference to treatment. It can provide unbiased estimates for selection and preference effects, where a selection effect occurs when patients who prefer one treatment respond differently than those who prefer another, and a preference effect is the difference in response caused by an interaction between the patient's preference and the actual treatment they receive. One potential barrier to adopting TSPD in practice, however, is the relatively large sample size required to estimate selection and preference effects with sufficient power. To address this concern, we propose a group sequential two-stage preference design (GS-TSPD), which combines TSPD with sequential monitoring for early stopping. In the GS-TSPD, pre-planned sequential monitoring allows investigators to conduct repeated hypothesis tests on accumulated data prior to full enrollment to assess study eligibility for early trial termination without inflating type I error rates. Thus, the procedure allows investigators to terminate the study when there is sufficient evidence of treatment, selection, or preference effects during an interim analysis, thereby reducing the design resource in expectation. To formalize such a procedure, we verify the independent increments assumption for testing the selection and preference effects and apply group sequential stopping boundaries from the approximate sequential density functions. Simulations are then conducted to investigate the operating characteristics of our proposed GS-TSPD compared to the traditional TSPD. We demonstrate the applicability of the design using a study of Hepatitis C treatment modality.

A Bayesian platform trial design with hybrid control based on multisource exchangeability modelling.

Enrolling patients to the standard of care (SOC) arm in randomized clinical trials, especially for rare diseases, can be very challenging due to the lack of resources, restricted patient population availability, and ethical considerations. As the therapeutic effect for the SOC is often well documented in historical trials, we propose a Bayesian platform trial design with hybrid control based on the multisource exchangeability modelling (MEM) framework to harness historical control data. The MEM approach provides a computationally efficient method to formally evaluate the exchangeability of study outcomes between different data sources and allows us to make better informed data borrowing decisions based on the exchangeability between historical and concurrent data. We conduct extensive simulation studies to evaluate the proposed hybrid design. We demonstrate the proposed design leads to significant sample size reduction for the internal control arm and borrows more information compared to competing Bayesian approaches when historical and internal data are compatible.

Hierarchical Bayesian modeling of heterogeneous outcome variance in cluster randomized trials.

BACKGROUND: Heterogeneous outcome correlations across treatment arms and clusters have been increasingly acknowledged in cluster randomized trials with binary endpoints, where analytical methods have been developed to study such heterogeneity. However, cluster-specific outcome variances and correlations have yet to be studied for cluster randomized trials with continuous outcomes. METHODS: This article proposes models fitted in the Bayesian setting with hierarchical variance structure to quantify heterogeneous variances across clusters and explain it with cluster-level covariates when the outcome is continuous. The models can also be extended to analyzing heterogeneous variances in individually randomized group treatment trials, with arm-specific cluster-level covariates, or in partially nested designs. Simulation studies are carried out to validate the performance of the newly introduced models across different settings. RESULTS: Simulations showed that overall the newly introduced models have good performance, reporting low bias and approximately 95% coverage for the intraclass correlation coefficients and regression parameters in the variance model. When variances are heterogeneous, our proposed models had improved model fit over models with homogeneous variances. When used to analyze data from the Kerala Diabetes Prevention Program study, our models identified heterogeneous variances and intraclass correlation coefficients across clusters and examined cluster-level characteristics associated with such heterogeneity. CONCLUSION: We proposed new hierarchical Bayesian variance models to accommodate cluster-specific variances in cluster randomized trials. The newly developed methods inform the understanding of how an intervention strategy is implemented and disseminated differently across clusters and can help improve future trial design.

A compressed large language model embedding dataset of ICD 10 CM descriptions.

This paper presents novel datasets providing numerical representations of ICD-10-CM codes by generating description embeddings using a large language model followed by a dimension reduction via autoencoder. The embeddings serve as informative input features for machine learning models by capturing relationships among categories and preserving inherent context information. The model generating the data was validated in two ways. First, the dimension reduction was validated using an autoencoder, and secondly, a supervised model was created to estimate the ICD-10-CM hierarchical categories. Results show that the dimension of the data can be reduced to as few as 10 dimensions while maintaining the ability to reproduce the original embeddings, with the fidelity decreasing as the reduced-dimension representation decreases. Multiple compression levels are provided, allowing users to choose as per their requirements, download and use without any other setup. The readily available datasets of ICD-10-CM codes are anticipated to be highly valuable for researchers in biomedical informatics, enabling more advanced analyses in the field. This approach has the potential to significantly improve the utility of ICD-10-CM codes in the biomedical domain.

A Randomized Trial of a Multifactorial Strategy to Prevent Serious Fall Injuries.

BACKGROUND: Injuries from falls are major contributors to complications and death in older adults. Despite evidence from efficacy trials that many falls can be prevented, rates of falls resulting in injury have not declined. METHODS: We conducted a pragmatic, cluster-randomized trial to evaluate the effectiveness of a multifactorial intervention that included risk assessment and individualized plans, administered by specially trained nurses, to prevent fall injuries. A total of 86 primary care practices across 10 health care systems were randomly assigned to the intervention or to enhanced usual care (the control) (43 practices each). The participants were community-dwelling adults, 70 years of age or older, who were at increased risk for fall injuries. The primary outcome, assessed in a time-to-event analysis, was the first serious fall injury, adjudicated with the use of participant report, electronic health records, and claims data. We hypothesized that the event rate would be lower by 20% in the intervention group than in the control group. RESULTS: The demographic and baseline characteristics of the participants were similar in the intervention group (2802 participants) and the control group (2649 participants); the mean age was 80 years, and 62.0% of the participants were women. The rate of a first adjudicated serious fall injury did not differ significantly between the groups, as assessed in a time-to-first-event analysis (events per 100 person-years of follow-up, 4.9 in the intervention group and 5.3 in the control group; hazard ratio, 0.92; 95% confidence interval [CI], 0.80 to 1.06; P = 0.25). The rate of a first participant-reported fall injury was 25.6 events per 100 person-years of follow-up in the intervention group and 28.6 events per 100 person-years of follow-up in the control group (hazard ratio, 0.90; 95% CI, 0.83 to 0.99; P = 0.004). The rates of hospitalization or death were similar in the two groups. CONCLUSIONS: A multifactorial intervention, administered by nurses, did not result in a significantly lower rate of a first adjudicated serious fall injury than enhanced usual care. (Funded by the Patient-Centered Outcomes Research Institute and others; STRIDE ClinicalTrials.gov number, NCT02475850.).

Maximin optimal cluster randomized designs for assessing treatment effect heterogeneity.

Cluster randomized trials (CRTs) are studies where treatment is randomized at the cluster level but outcomes are typically collected at the individual level. When CRTs are employed in pragmatic settings, baseline population characteristics may moderate treatment effects, leading to what is known as heterogeneous treatment effects (HTEs). Pre-specified, hypothesis-driven HTE analyses in CRTs can enable an understanding of how interventions may impact subpopulation outcomes. While closed-form sample size formulas have recently been proposed, assuming known intracluster correlation coefficients (ICCs) for both the covariate and outcome, guidance on optimal cluster randomized designs to ensure maximum power with pre-specified HTE analyses has not yet been developed. We derive new design formulas to determine the cluster size and number of clusters to achieve the locally optimal design (LOD) that minimizes variance for estimating the HTE parameter given a budget constraint. Given the LODs are based on covariate and outcome-ICC values that are usually unknown, we further develop the maximin design for assessing HTE, identifying the combination of design resources that maximize the relative efficiency of the HTE analysis in the worst case scenario. In addition, given the analysis of the average treatment effect is often of primary interest, we also establish optimal designs to accommodate multiple objectives by combining considerations for studying both the average and heterogeneous treatment effects. We illustrate our methods using the context of the Kerala Diabetes Prevention Program CRT, and provide an R Shiny app to facilitate calculation of optimal designs under a wide range of design parameters.

Rental assistance improves food security and nutrition: An analysis of National Survey Data.

The U.S. is experiencing a severe housing affordability crisis, resulting in households having to make difficult trade-offs between paying for a place to live and basic health necessities such as food. Rental assistance may mitigate these strains, improving food security and nutrition. However, only one in five eligible individuals receive assistance, with an average wait time of two years. Existing waitlists create a comparable control group, allowing us to examine the causal impact of improved housing access on health and well-being. This national quasi-experimental study utilizes linked NHANES-HUD data (1999-2016) to investigate the impacts of rental assistance on food security and nutrition using cross-sectional regression. Tenants with project-based assistance were less likely to experience food insecurity (B = -0.18, p = 0.02) and rent-assisted individuals consumed 0.23 more cups of daily fruits and vegetables compared the pseudo-waitlist group. These findings suggest that the current unmet need for rental assistance and resulting long waitlists have adverse health implications, including decreased food security and fruit and vegetable consumption.

A qualitative study identifying challenges resulting from complex evidence on lifestyle factors and cancer: perspectives from Black and Latina cancer survivors and healthcare providers.

PURPOSE: Most breast cancer survivors have challenges with adopting healthy lifestyle behaviors. This may be due to contextual challenges that result from the complex nature of the evidence. To address this gap, we explored the experiences of breast cancer survivors of color and oncology healthcare providers. METHODS: Content analysis with inductive and deductive approaches was used for semi-structured interviews with 26 female breast cancer survivors and 10 oncology healthcare providers from Greater New Haven, Connecticut. RESULTS: Survivors identified substantial confusion on the evidence regarding lifestyle behaviors and breast cancer, stemming from inadequate healthcare provider counseling and an overreliance on informal sources of information. Providers identified lack of evidence-based knowledge as a barrier to counseling on these topics. There was a mixed perspective regarding the consistency of evidence, stemming from a combination of gaps in the available evidence and accessing evidence-based knowledge from a wide range of professional resources. Some providers perceived the guidelines as consistent; others felt guidelines were constantly changing, impacting how and on what they counseled. Therefore, many healthcare providers in oncology care relied on generic messaging on lifestyle behaviors after a cancer diagnosis. CONCLUSIONS: Inconsistent information sources, the rapidly changing evidence, and gaps in the current evidence contribute to generic messaging about lifestyle behaviors and may inhibit a survivor's ability to engage in behavior change. Consistent and uniform healthy lifestyle guidelines for cancer outcomes may address both provider and patient level barriers to knowledge.

A competing risk analysis of adherence to the American Cancer Society Guidelines on Nutrition and Physical Activity for Cancer Prevention and obesity-related cancer risk in Hispanic/Latino adults in the NIH-AARP Diet and Health Study.

Incidence of obesity-related cancers (ORCs) is rising among US Hispanic/Latino adults, which may be partly due to inadequate engagement in healthy lifestyle behaviors. Prior research on cancer prevention guideline adherence and cancer risk has not considered competing events that may lead to misinterpreting the magnitude of risk between guideline adherence and cancer incidence. Among Hispanic/Latino adults (N = 9204) in the NIH-AARP Diet and Health Study, we examined the association between adherence to the 2012 American Cancer Society (ACS) guidelines (high, moderate, low) on nutrition and physical activity for cancer prevention and risk of any first observed ORC using Fine and Gray methods for competing risk analysis. Over a median of 10.5 years of follow-up, there were 619 first ORCs. The cumulative risk of ORC over a 15-year period was not significantly different across ACS guideline adherence categories (high cumulative incidence function [CIF]: 2.2%-5.8%; moderate CIF: 2.2%-6.6%; low CIF: 2.3%-6.7%, PGray's log rank  = .690). In competing risk analysis, high (compared to low) adherence to the ACS guidelines was associated with reduced probability of ORC (subdistribution hazard [SHR]: 0.76, 95% CI: 0.58-0.996, P = .047), with evidence of a linear trend for increasing adherence (Ptrend  = .039). Our findings were consistent with hypothesized inverse associations between ACS guideline adherence and ORC incidence accounting for competing risks. These findings suggest a need for continued public health efforts focused on promoting engagement in healthy lifestyle behaviors to reduce ORC incidence among US Hispanic/Latino adults.

Adherence to the American Cancer Society Guidelines on nutrition and physical activity for cancer prevention and obesity-related cancer risk and mortality in Black and Latina Women's Health Initiative participants.

BACKGROUND: Although adherence to the American Cancer Society (ACS) Guidelines on Nutrition and Physical Activity for Cancer Prevention associates with lower risk of obesity-related cancer (ORC) incidence and mortality, evidence in Black and Latina women is limited. This association was examined in Black and Latina participants in the Women's Health Initiative (WHI). METHODS: Semi-Markov multistate model examined the association between ACS guideline adherence and ORC incidence and mortality in the presence of competing events, combined and separately, for 9301 Black and 4221 Latina postmenopausal women. Additionally, ACS guideline adherence was examined in a subset of less common ORCs and potential effect modification by neighborhood socioeconomic status and smoking. RESULTS: Over a median of 11.1, 12.5, and 3.7 years of follow-up for incidence, nonconditional mortality, and conditional mortality, respectively, 1191 ORCs (Black/Latina women: 841/269), 1970 all-cause deaths (Black/Latina women: 1576/394), and 341 ORC-related deaths (Black/Latina women: 259/82) were observed. Higher ACS guideline adherence was associated with lower ORC incidence for both Black (cause-specific hazard ratio [CSHR]highvs.low : 0.72; 95% CI, 0.55-0.94) and Latina (CSHRhighvs.low : 0.58, 95% CI, 0.36-0.93) women; but not conditional all-cause mortality (Black hazard ratio [HR]highvs.low : 0.86; 95% CI, 0.53-1.39; Latina HRhighvs.low : 0.81; 95% CI, 0.32-2.06). Higher adherence was associated with lower incidence of less common ORC (Ptrend  = .025), but conditional mortality events were limited. Adherence and ORC-specific deaths were not associated and there was no evidence of effect modification. CONCLUSIONS: Adherence to the ACS guidelines was associated with lower risk of ORCs and less common ORCs but was not for conditional ORC-related mortality. LAY SUMMARY: Evidence on the association between the American Cancer Society Guidelines on Nutrition and Physical Activity for Cancer Prevention and cancer remains scarce for women of color. Adherence to the guidelines and risk of developing one of 13 obesity-related cancers among Black and Latina women in the Women's Health Initiative was examined. Women who followed the lifestyle guidelines had 28% to 42% lower risk of obesity-related cancer. These findings support public health interventions to reduce growing racial/ethnic disparities in obesity-related cancers.

Accounting for unequal cluster sizes in designing cluster randomized trials to detect treatment effect heterogeneity.

Unequal cluster sizes are common in cluster randomized trials (CRTs). While there are a number of previous investigations studying the impact of unequal cluster sizes on the power for testing the average treatment effect in CRTs, little is known about the impact of unequal cluster sizes on the power for testing the heterogeneous treatment effect (HTE) in CRTs. In this work, we expand the sample size procedures for studying HTE in CRTs to accommodate cluster size variation under the linear mixed model framework. Through analytical derivation and graphical exploration, we show that the sample size for the HTE with an individual-level effect modifier is less affected by unequal cluster sizes than with a cluster-level effect modifier. The impact of cluster size variability jointly depends on the mean and coefficient of variation of cluster sizes, covariate intraclass correlation coefficient (ICC) and the conditional outcome ICC. In addition, we demonstrate that the HTE-motivated analysis of covariance framework can be used for analyzing the average treatment effect, and offer a more efficient sample size procedure for studying the average treatment effect adjusting for the effect modifier. We use simulations to confirm the accuracy of the proposed sample size procedures for both the average treatment effect and HTE in CRTs. Extensions to multivariate effect modifiers are provided and our procedure is illustrated in the context of the Strategies to Reduce Injuries and Develop Confidence in Elders trial.

Design and analysis of cluster randomized trials with time-to-event outcomes under the additive hazards mixed model.

A primary focus of current methods for cluster randomized trials (CRTs) has been for continuous, binary, and count outcomes, with relatively less attention given to right-censored, time-to-event outcomes. In this article, we detail considerations for sample size requirement and statistical inference in CRTs with time-to-event outcomes when the intervention effect parameter is specified through the additive hazards mixed model (AHMM), which includes a frailty term to explicitly account for the dependency between the failure times. First, we discuss improved inference for the treatment effect parameter via bias-corrected sandwich variance estimators and randomization-based test under AHMM, addressing potential small-sample biases in CRTs. Next, we derive a new sample size formula for AHMM analysis of CRTs accommodating both equal and unequal cluster sizes. When the cluster sizes vary, our sample size formula depends on the mean and coefficient of variation of cluster sizes, based on which we articulate the impact of cluster size variation in CRTs with time-to-event outcomes. Furthermore, we obtain the insight that the classical variance inflation factor for CRTs with a non-censored outcome can in fact apply to CRTs with a time-to-event outcome, providing that an appropriate definition of the intraclass correlation coefficient is considered under AHMM. Simulation studies are carried out to illustrate key design and analysis considerations in CRTs with a small to moderate number of clusters. The proposed sample size procedure and analytical methods are further illustrated using the context of the STrategies to Reduce Injuries and Develop Confidence in Elders CRT.

Bayesian local exchangeability design for phase II basket trials.

We propose an information borrowing strategy for the design and monitoring of phase II basket trials based on the local multisource exchangeability assumption between baskets (disease types). In our proposed local-MEM framework, information borrowing is only allowed to occur locally, that is, among baskets with similar response rate and the amount of information borrowing is determined by the level of similarity in response rate, whereas baskets not considered similar are not allowed to share information. We construct a two-stage design for phase II basket trials using the proposed strategy. The proposed method is compared to competing Bayesian methods and Simon's two-stage design in a variety of simulation scenarios. We demonstrate the proposed method is able to maintain the family-wise type I error rate at a reasonable level and has desirable basket-wise power compared to Simon's two-stage design. In addition, our method is computationally efficient compared to existing Bayesian methods in that the posterior profiles of interest can be derived explicitly without the need for sampling algorithms. R scripts to implement the proposed method are available at https://github.com/yilinyl/Bayesian-localMEM.

Sample size calculation in hierarchical 2×2 factorial trials with unequal cluster sizes.

Motivated by a suicide prevention trial with hierarchical treatment allocation (cluster-level and individual-level treatments), we address the sample size requirements for testing the treatment effects as well as their interaction. We assume a linear mixed model, within which two types of treatment effect estimands (controlled effect and marginal effect) are defined. For each null hypothesis corresponding to an estimand, we derive sample size formulas based on large-sample z-approximation, and provide finite-sample modifications based on a t-approximation. We relax the equal cluster size assumption and express the sample size formulas as functions of the mean and coefficient of variation of cluster sizes. We show that the sample size requirement for testing the controlled effect of the cluster-level treatment is more sensitive to cluster size variability than that for testing the controlled effect of the individual-level treatment; the same observation holds for testing the marginal effects. In addition, we show that the sample size for testing the interaction effect is proportional to that for testing the controlled or the marginal effect of the individual-level treatment. We conduct extensive simulations to validate the proposed sample size formulas, and find the empirical power agrees well with the predicted power for each test. Furthermore, the t-approximations often provide better control of type I error rate with a small number of clusters. Finally, we illustrate our sample size formulas to design the motivating suicide prevention factorial trial. The proposed methods are implemented in the R package H2x2Factorial.

Impact of complex, partially nested clustering in a three-arm individually randomized group treatment trial: A case study with the wHOPE trial.

BACKGROUND/AIMS: When participants in individually randomized group treatment trials are treated by multiple clinicians or in multiple group treatment sessions throughout the trial, this induces partially nested clusters which can affect the power of a trial. We investigate this issue in the Whole Health Options and Pain Education trial, a three-arm pragmatic, individually randomized clinical trial. We evaluate whether partial clusters due to multiple visits delivered by different clinicians in the Whole Health Team arm and dynamic participant groups due to changing group leaders and/or participants across treatment sessions during treatment delivery in the Primary Care Group Education arm may impact the power of the trial. We also present a Bayesian approach to estimate the intraclass correlation coefficients. METHODS: We present statistical models for each treatment arm of Whole Health Options and Pain Education trial in which power is estimated under different intraclass correlation coefficients and mapping matrices between participants and clinicians or treatment sessions. Power calculations are based on pairwise comparisons. In practice, sample size calculations depend on estimates of the intraclass correlation coefficients at the treatment sessions and clinician levels. To accommodate such complexities, we present a Bayesian framework for the estimation of intraclass correlation coefficients under different participant-to-session and participant-to-clinician mapping scenarios. We simulated continuous outcome data based on various clinical scenarios in Whole Health Options and Pain Education trial using a range of intraclass correlation coefficients and mapping matrices and used Gibbs samplers with conjugate priors to obtain posteriors of the intraclass correlation coefficients under those different scenarios. Posterior means and medians and their biases are calculated for the intraclass correlation coefficients to evaluate the operating characteristics of the Bayesian intraclass correlation coefficient estimators. RESULTS: Power for Whole Health Team versus Primary Care Group Education is sensitive to the intraclass correlation coefficient in the Whole Health Team arm. In these two arms, an increased number of clinicians, more evenly distributed workload of clinicians, or more homogeneous treatment group sizes leads to increased power. Our simulation study for the intraclass correlation coefficient estimation indicates that the posterior mean intraclass correlation coefficient estimator has less bias when the true intraclass correlation coefficients are large (i.e. 0.10), but when the intraclass correlation coefficient is small (i.e. 0.01), the posterior median intraclass correlation coefficient estimator is less biased. CONCLUSION: Knowledge of intraclass correlation coefficients and the structure of clustering are critical to the design of individually randomized group treatment trials with partially nested clusters. We demonstrate that the intraclass correlation coefficient of the Whole Health Team arm can affect power in the Whole Health Options and Pain Education trial. A Bayesian approach provides a flexible procedure for estimating the intraclass correlation coefficients under complex scenarios. More work is needed to educate the research community about the individually randomized group treatment design and encourage publication of intraclass correlation coefficients to help inform future trial 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.

Estimation of ascertainment bias and its effect on power in clinical trials with time-to-event outcomes.

While the gold standard for clinical trials is to blind all parties-participants, researchers, and evaluators-to treatment assignment, this is not always a possibility. When some or all of the above individuals know the treatment assignment, this leaves the study open to the introduction of postrandomization biases. In the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) trial, we were presented with the potential for the unblinded clinicians administering the treatment, as well as the individuals enrolled in the study, to introduce ascertainment bias into some but not all events comprising the primary outcome. In this article, we present ways to estimate the ascertainment bias for a time-to-event outcome, and discuss its impact on the overall power of a trial vs changing of the outcome definition to a more stringent unbiased definition that restricts attention to measurements less subject to potentially differential assessment. We found that for the majority of situations, it is better to revise the definition to a more stringent definition, as was done in STRIDE, even though fewer events may be observed.

A case study of ascertainment bias for the primary outcome in the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) trial.

BACKGROUND/AIM: In clinical trials, there is potential for bias from unblinded observers that may influence ascertainment of outcomes. This issue arose in the Strategies to Reduce Injuries and Develop Confidence in Elders trial, a cluster randomized trial to test a multicomponent intervention versus enhanced usual care (control) to prevent serious fall injuries, originally defined as a fall injury leading to medical attention. An unblinded nurse falls care manager administered the intervention, while the usual care arm did not involve contact with a falls care manager. Thus, there was an opportunity for falls care managers to refer participants reporting falls to seek medical attention. Since this type of observer bias could not occur in the usual care arm, there was potential for additional falls to be reported in the intervention arm, leading to dilution of the intervention effect and a reduction in study power. We describe the clinical basis for ascertainment bias, the statistical approach used to assess it, and its effect on study power. METHODS: The prespecified interim monitoring plan included a decision algorithm for assessing ascertainment bias and adapting (revising) the primary outcome definition, if necessary. The original definition categorized serious fall injuries requiring medical attention into Type 1 (fracture other than thoracic/lumbar vertebral, joint dislocation, cut requiring closure) and Type 2 (head injury, sprain or strain, bruising or swelling, other). The revised definition, proposed by the monitoring plan, excluded Type 2 injuries that did not necessarily require an overnight hospitalization since these would be most subject to bias. These injuries were categorized into those with (Type 2b) and without (Type 2c) medical attention. The remaining Type 2a injuries required medical attention and an overnight hospitalization. We used the ratio of 2b/(2b + 2c) in intervention versus control as a measure of ascertainment bias; ratios > 1 indicated the likelihood of falls care manager bias. We determined the effect of ascertainment bias on study power for the revised (Types 1 and 2a) versus original definition (Types 1, 2a, and 2b). RESULTS: The estimate of ascertainment bias was 1.14 (95% confidence interval: 0.98, 1.30), providing evidence of the likelihood of falls care manager bias. We estimated that this bias diluted the hazard ratio from the hypothesized 0.80 to 0.86 and reduced power to under 80% for the original primary outcome definition. In contrast, adapting the revised definition maintained study power at nearly 90%. CONCLUSION: There was evidence of ascertainment bias in the Strategies to Reduce Injuries and Develop Confidence in Elders trial. The decision to adapt the primary outcome definition reduced the likelihood of this bias while preserving the intervention effect and study power.