1
|
Ouyang Y, Taljaard M, Forbes AB, Li F. Maintaining the validity of inference from linear mixed models in stepped-wedge cluster randomized trials under misspecified random-effects structures. Stat Methods Med Res 2024:9622802241248382. [PMID: 38807552 DOI: 10.1177/09622802241248382] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 05/30/2024]
Abstract
Linear mixed models are commonly used in analyzing stepped-wedge cluster randomized trials. A key consideration for analyzing a stepped-wedge cluster randomized trial is accounting for the potentially complex correlation structure, which can be achieved by specifying random-effects. The simplest random effects structure is random intercept but more complex structures such as random cluster-by-period, discrete-time decay, and more recently, the random intervention structure, have been proposed. Specifying appropriate random effects in practice can be challenging: assuming more complex correlation structures may be reasonable but they are vulnerable to computational challenges. To circumvent these challenges, robust variance estimators may be applied to linear mixed models to provide consistent estimators of standard errors of fixed effect parameters in the presence of random-effects misspecification. However, there has been no empirical investigation of robust variance estimators for stepped-wedge cluster randomized trials. In this article, we review six robust variance estimators (both standard and small-sample bias-corrected robust variance estimators) that are available for linear mixed models in R, and then describe a comprehensive simulation study to examine the performance of these robust variance estimators for stepped-wedge cluster randomized trials with a continuous outcome under different data generators. For each data generator, we investigate whether the use of a robust variance estimator with either the random intercept model or the random cluster-by-period model is sufficient to provide valid statistical inference for fixed effect parameters, when these working models are subject to random-effect misspecification. Our results indicate that the random intercept and random cluster-by-period models with robust variance estimators performed adequately. The CR3 robust variance estimator (approximate jackknife) estimator, coupled with the number of clusters minus two degrees of freedom correction, consistently gave the best coverage results, but could be slightly conservative when the number of clusters was below 16. We summarize the implications of our results for the linear mixed model analysis of stepped-wedge cluster randomized trials and offer some practical recommendations on the choice of the analytic model.
Collapse
Affiliation(s)
- Yongdong Ouyang
- Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, ON, Canada
- School of Epidemiology and Public Health, University of Ottawa, Ottawa, ON, Canada
| | - Monica Taljaard
- Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, ON, Canada
- School of Epidemiology and Public Health, University of Ottawa, Ottawa, ON, Canada
| | - Andrew B Forbes
- School of Public Health and Preventive Medicine, Monash University, Melbourne, VIC, Australia
| | - Fan Li
- Department of Biostatistics, Yale School of Public Health, New Haven, CT, USA
- Center for Methods in Implementation and Prevention Science, Yale School of Public Health, New Haven, CT, USA
| |
Collapse
|
2
|
Hamaker ME, Wildiers H, Ardito V, Arsandaux J, Barthod-Malat A, Davies P, Degol L, Ferrara L, Fourrier C, Kenis C, Kret M, Lalet C, Pelissier SM, O'Hanlon S, Rostoft S, Seghers N, Saillour-Glénisson F, Staines A, Schwimmer C, Thevenet V, Wallet C, Soubeyran P. Study protocol for two stepped-wedge interventional trials evaluating the effects of holistic information technology-based patient-oriented management in older multimorbid patients with cancer: The GERONTE trials. J Geriatr Oncol 2024; 15:101761. [PMID: 38581958 DOI: 10.1016/j.jgo.2024.101761] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/08/2024] [Revised: 03/07/2024] [Accepted: 03/26/2024] [Indexed: 04/08/2024]
Abstract
INTRODUCTION Current hospital-based care pathways are generally single-disease centred. As a result, coexisting morbidities are often suboptimally evaluated and managed, a deficiency becoming increasingly apparent among older patients who exhibit heterogeneity in health status, functional abilities, frailty, and other geriatric impairments. To address this issue, our study aims to assess a newly developed patient-centred care pathway for older patients with multimorbidity and cancer. The new care pathway was based on currently available evidence and co-designed by end-users including health care professionals, patients, and informal caregivers. Within this care pathway, all healthcare professionals involved in the care of older patients with multimorbidity and cancer will form a Health Professional Consortium (HPC). The role of the HPC will be to centralise oncologic and non-oncologic treatment recommendations in accordance with the patient's priorities. Moreover, an Advanced Practice Nurse will act as case-manager by being the primary point of contact for the patient, thus improving coordination between specialists, and by organising and leading the consortium. Patient monitoring and the HPC collaboration will be facilitated by digital communication tools designed specifically for this purpose, with the added benefit of being customisable for each patient. MATERIALS AND METHODS The GERONTE study is a prospective international, multicentric study consisting of two stepped-wedge trials performed at 16 clinical sites across three European countries. Each trial will include 720 patients aged 70 years and over with a new or progressive cancer (breast, lung, colorectal, prostate) and at least one moderate or severe multimorbidity. The patients in the intervention group will receive the new care pathway whereas patients in the control group will receive usual oncologic care. DISCUSSION GERONTE will evaluate whether this kind of holistic, patient-oriented healthcare management can improve quality of life (primary outcome) and other valuable endpoints in older patients with multimorbidity and cancer. An ancillary study will assess in depth the socio-economic impact of the intervention and deliver concrete implementation guidelines for the GERONTE intervention care pathway. TRIAL REGISTRATION FRONE: NCT05720910 TWOBE: NCT05423808.
Collapse
Affiliation(s)
- Marije E Hamaker
- Department of Geriatric Medicine, Diakonessenhuis Utrecht, the Netherlands.
| | - Hans Wildiers
- Department of General Medical Oncology, University Hospitals Leuven, Leuven, Belgium
| | - Vittoria Ardito
- Department SDA Bocconi, Government, Health and Not for profit Division, CERGAS, Bocconi University, Milan, Italy
| | - Julie Arsandaux
- Nantes Université, Univ Angers, Laboratoire de psychologie des Pays de la Loire, LPPL, UR 4638, F-44000 Nantes, France; Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Aurore Barthod-Malat
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Paul Davies
- School of Nursing, Psychotherapy and Community Health, Dublin City University, Glasnevin, Dublin 9, Ireland
| | - Lien Degol
- Department of General Medical Oncology, University Hospitals Leuven, Leuven, Belgium
| | - Lucia Ferrara
- Department SDA Bocconi, Government, Health and Not for profit Division, CERGAS, Bocconi University, Milan, Italy
| | - Celia Fourrier
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Cindy Kenis
- Department of General Medical Oncology and Geriatric Medicine, University Hospitals Leuven, Leuven, Belgium; Department of Public Health and Primary Care, Academic Centre for Nursing and Midwifery, KU Leuven, Leuven, Belgium
| | - Marion Kret
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France; CHU de Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Caroline Lalet
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Simone Mathoulin Pelissier
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France; Univ Bordeaux, Inserm BordHEalth eaux Population U1219 Epicene Team, France
| | - Shane O'Hanlon
- Department of Geriatric Medicine, St Vincent's University Hospital, D04 T6F4 Dublin, Ireland; Department of Geriatric Medicine, University College Dublin, D04 V1W8 Dublin, Ireland
| | - Siri Rostoft
- Department of Geriatric Medicine, Oslo University Hospital, 0424 Oslo, Norway; Institute of Clinical Medicine, University of Oslo, 0318 Oslo, Norway
| | - Nelleke Seghers
- Department of Geriatric Medicine, Diakonessenhuis Utrecht, the Netherlands
| | - Florence Saillour-Glénisson
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France; CHU de Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Anthony Staines
- School of Nursing, Psychotherapy and Community Health, Dublin City University, Glasnevin, Dublin 9, Ireland
| | - Christine Schwimmer
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France; CHU de Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Vincent Thevenet
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Cedric Wallet
- Univ. Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France; CHU de Bordeaux, INSERM, Institut Bergonié, CIC 1401, Euclid/F-CRIN clinical trials platform, F-33000 Bordeaux, France
| | - Pierre Soubeyran
- Department of Medical Oncology, Institut Bergonié, Inserm U1312, SIRIC BRIO, Université de Bordeaux, 33076 Bordeaux, France
| |
Collapse
|
3
|
Zhu Y, Li S, Zhang R, Bao L, Zhang J, Xiao X, Jiang D, Chen W, Hu C, Zou C, Zhang J, Zhu Y, Wang J, Liang J, Yang Q. Enhancing doctor-patient relationships in community health care institutions: the Patient Oriented Four Habits Model (POFHM) trial-a stepped wedge cluster randomized trial protocol. BMC Psychiatry 2023; 23:476. [PMID: 37380993 DOI: 10.1186/s12888-023-04948-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/05/2023] [Accepted: 06/10/2023] [Indexed: 06/30/2023] Open
Abstract
BACKGROUND The poor relationship between doctors and patients is a long-standing, global problem. However, current interventions tend to focus on the training of physicians, while patient-targeted interventions still need to be improved. Considering that patients play a significant role in outpatient consultations, we developed a protocol to assess the effectiveness of the Patient Oriented Four Habits Model (POFHM) in improving doctor-patient relationships. METHODS A cross-sectional incomplete stepped-wedge cluster randomized trial design will be conducted in 8 primary healthcare institutions (PHCs). Following phase I of "usual care" as control measures for each PHC, either a patient- or doctor-only intervention will be implemented in phase II. In phase III, both patients and doctors will be involved in the intervention. This study will be conducted simultaneously in Nanling County and West Lake District. The primary outcomes will be evaluated after patients complete their visit: (1) patient literacy, (2) sense of control and (3) quality of doctor-patient communication. Finally, a mixed-effects model and subgroup analysis will be used to evaluate the effectiveness of the interventions. DISCUSSION Fostering good consultation habits for the patient is a potentially effective strategy to improve the quality of doctor-patient communication. This study evaluates the implementation process and develops a rigorous quality control manual using a theoretical domain framework under the collective culture of China. The results of this trial will provide substantial evidence of the effectiveness of patient-oriented interventions. The POFHM can benefit the PHCs and provide a reference for countries and regions where medical resources are scarce and collectivist cultures dominate. TRIAL REGISTRATION AsPredicted #107,282 on Sep 18, 2022; https://aspredicted.org/QST_MHW.
Collapse
Affiliation(s)
- Yunying Zhu
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Sisi Li
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Ruotong Zhang
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Lei Bao
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Jin Zhang
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Xiaohua Xiao
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Dongdong Jiang
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Wenxiao Chen
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China
| | - Chenying Hu
- Community Health Service Center in Jiangcun Street, Hangzhou, 310050, Zhejiang Province, China
| | - Changli Zou
- Community Health Service Center in Sandun Town, Hangzhou, 310030, Zhejiang Province, China
| | - Jingna Zhang
- Community Health Service Center in Liuxia Street, Hangzhou, Zhejiang Province, 310050, China
| | - Yong Zhu
- Xu Zhen Town Center Health Center, Wuhu, 241306, Anhui Province, China
| | - Jianqiu Wang
- Community Health Service Center in Jishan Town, Wuhu, 241307, Anhui Province, China
| | - Jinchun Liang
- Nanling County Traditional Chinese Medicine Hospital, Wuhu, 241307, Anhui Province, China
| | - Qian Yang
- School of Public Health, and Department of Geriatrics of the Fourth Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, Zhejiang Province, 310058, China.
| |
Collapse
|
4
|
Kenny A, Voldal E, Xia F, Heagerty PJ, Hughes JP. Analysis of stepped wedge cluster randomized trials in the presence of a time-varying treatment effect. Stat Med 2022; 41:4311-4339. [PMID: 35774016 PMCID: PMC9481733 DOI: 10.1002/sim.9511] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/09/2021] [Revised: 06/13/2022] [Accepted: 06/20/2022] [Indexed: 11/11/2022]
Abstract
Stepped wedge cluster randomized controlled trials are typically analyzed using models that assume the full effect of the treatment is achieved instantaneously. We provide an analytical framework for scenarios in which the treatment effect varies as a function of exposure time (time since the start of treatment) and define the "effect curve" as the magnitude of the treatment effect on the linear predictor scale as a function of exposure time. The "time-averaged treatment effect" (TATE) and "long-term treatment effect" (LTE) are summaries of this curve. We analytically derive the expectation of the estimatorδ ^ $$ \hat{\delta} $$ resulting from a model that assumes an immediate treatment effect and show that it can be expressed as a weighted sum of the time-specific treatment effects corresponding to the observed exposure times. Surprisingly, although the weights sum to one, some of the weights can be negative. This implies thatδ ^ $$ \hat{\delta} $$ may be severely misleading and can even converge to a value of the opposite sign of the true TATE or LTE. We describe several models, some of which make assumptions about the shape of the effect curve, that can be used to simultaneously estimate the entire effect curve, the TATE, and the LTE. We evaluate these models in a simulation study to examine the operating characteristics of the resulting estimators and apply them to two real datasets.
Collapse
Affiliation(s)
- Avi Kenny
- Department of Biostatistics, University of Washington, Seattle, Washington
| | - Emily Voldal
- Department of Biostatistics, University of Washington, Seattle, Washington
| | - Fan Xia
- Department of Biostatistics, University of Washington, Seattle, Washington
| | | | - James P. Hughes
- Department of Biostatistics, University of Washington, Seattle, Washington
| |
Collapse
|
5
|
Voldal EC, Xia F, Kenny A, Heagerty PJ, Hughes JP. Random effect misspecification in stepped wedge designs. Clin Trials 2022; 19:380-383. [PMID: 35257614 DOI: 10.1177/17407745221084702] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Stepped wedge cluster randomized trials are often analysed using linear mixed effects models that may include random effects for cluster, time and/or treatment. We investigate the impact of misspecification of the random effects structure of the model. Specifically, we considered two cases of misspecification of the random effects in a cross-sectional stepped wedge cluster randomized trials model - fit a linear mixed effects model with random time effects but the true model includes random treatment effects (case 1) or fit a linear mixed effects model with random treatment effect but the true model includes random time effects (case 2) - and derived the variance of the estimated treatment effect under misspecification. We defined two measures of the effect of misspecification: validity and efficiency. Validity is the ratio of the model-based variance of the treatment effect from the mis-specified model divided by the true variance of the treatment effect from the mis-specified model (based on a sandwich estimate of the variance). Efficiency is the ratio of the model-based variance of the treatment effect from the correctly specified model divided by the true variance of the treatment effect from the mis-specified model. We found that validity is less than 1.0 (anti-conservative) in almost all situations investigated with the exception of case 1 with two sequences, when validity could be greater than 1.0. Efficiency is less than 1 in all cases and depends on the intracluster correlation coefficient, the relative magnitude of the variance of the misclassified variance component, and the number of sequences. In general, there is no universal recommendation as to the most robust approach except for the case of a classic stepped wedge cluster randomized trial with only 2 sequences, where fitting a random time model is less likely to lead to anti-conservative inference compared with fitting a random intervention model.
Collapse
Affiliation(s)
- Emily C Voldal
- Department of Biostatistics, University of Washington, Seattle, WA, USA
| | - Fan Xia
- Department of Epidemiology, University of Washington, Seattle, WA, USA
| | - Avi Kenny
- Department of Biostatistics, University of Washington, Seattle, WA, USA
| | | | - James P Hughes
- Department of Biostatistics, University of Washington, Seattle, WA, USA
| |
Collapse
|