The analysis of multivariate longitudinal data: A review

Verbeke, Geert; Fieuws, Steffen; Molenberghs, Geert; Davidian, Marie

doi:10.1177/0962280212445834

Cited by 232 publications

(175 citation statements)

References 97 publications

Supporting

Mentioning

172

Contrasting

Unclassified

Order By: Relevance

“…In order to incorporate both baseline and intervention into the model, we use a shared parameter, multivariate linear mixed model [33,34] given by

(\begin{matrix} {boldy}_{0 i} \\ {boldy}_{1 i} \end{matrix}) = (\begin{matrix} X_{0 i} \\ X_{1 i} \end{matrix}) bold-italicβ + (\begin{matrix} Z_{0 i} \\ Z_{1 i} \end{matrix}) b_{i} + (\begin{matrix} e_{0 i} \\ e_{1 i} \end{matrix}),

where y 0 i is the 24 × 1 vector of observations for baseline BP, y 1 i is the 24 × 1 vector of observations for intervention BP, X 0 i and X 1 i the 24 × q fixed effects design matrices for baseline and intervention periods, and Z 0 i and Z 1 i the 24 × m random effects design matrices for baseline and intervention periods. The shared parameter model is appropriate since the response is the same for baseline and intervention.…”

Section: Resultsmentioning

confidence: 99%

An analysis of 24-h ambulatory blood pressure monitoring data using orthonormal polynomials in the linear mixed model

Edwards

Simpson

2014

Blood Pressure Monitoring

View full text Add to dashboard Cite

Background The use of 24-hour ambulatory blood pressure monitoring (ABPM) in clinical practice and observational epidemiological studies has grown considerably in the past 25 years. ABPM is a very effective technique for assessing biological, environmental, and drug effects on blood pressure. Objectives In order to enhance the effectiveness of ABPM for clinical and observational research studies via analytical and graphical results, developing alternative data analysis approaches using modern statistical techniques are important. Methods The linear mixed model for the analysis of longitudinal data is particularly well-suited for the estimation of, inference about, and interpretation of both population (mean) and subject-specific trajectories for ABPM data. We propose using a linear mixed model with orthonormal polynomials across time in both the fixed and random effects to analyze ABPM data. Results We demonstrate the proposed analysis technique using data from the Dietary Approaches to Stop Hypertension (DASH) study, a multicenter, randomized, parallel arm feeding study that tested the effects of dietary patterns on blood pressure. Conclusions The linear mixed model is relatively easy to implement (given the complexity of the technique) using available software, allows for straight-forward testing of multiple hypotheses, and the results can be presented to research clinicians using both graphical and tabular displays. Using orthonormal polynomials provides the ability to model the nonlinear trajectories of each subject with the same complexity as the mean model (fixed effects).

show abstract

“…In order to incorporate both baseline and intervention into the model, we use a shared parameter, multivariate linear mixed model [33,34] given by

(\begin{matrix} {boldy}_{0 i} \\ {boldy}_{1 i} \end{matrix}) = (\begin{matrix} X_{0 i} \\ X_{1 i} \end{matrix}) bold-italicβ + (\begin{matrix} Z_{0 i} \\ Z_{1 i} \end{matrix}) b_{i} + (\begin{matrix} e_{0 i} \\ e_{1 i} \end{matrix}),

Section: Resultsmentioning

confidence: 99%

An analysis of 24-h ambulatory blood pressure monitoring data using orthonormal polynomials in the linear mixed model

Edwards

Simpson

2014

Blood Pressure Monitoring

View full text Add to dashboard Cite

show abstract

“…We will use repeated measures mixed models 56 to compare biomarker changes between study arms. We will use the F test from a multivariate mixed model 57,58 as an omnibus test to assess group effects across the three biomarkers, namely HOMA-IR, CRP and bioavailable estrogen. Mixed models will include subject-specific intercepts; fixed effects will be included for time (baseline, 6 months), treatment, and treatment*time interactions.…”

Section: Methodsmentioning

confidence: 99%

Recruitment strategies, design, and participant characteristics in a trial of weight-loss and metformin in breast cancer survivors

Patterson

Marinac

Hartman

et al. 2016

Contemporary Clinical Trials

View full text Add to dashboard Cite

Weight loss and metformin are hypothesized to improve breast cancer outcomes; however the joint impacts of these treatments have not been investigated. Reach for Health is a randomized trial using a 2×2 factorial design to investigate the effects of weight loss and metformin on biomarkers associated with breast cancer prognosis among overweight/obese postmenopausal breast cancer survivors. This paper describes the trial recruitment strategies, design, and baseline sample characteristics. Participants were randomized in equal numbers to (1) placebo, (2) metformin, (3) weight loss intervention and placebo, or (4) weight-loss intervention and metformin. The lifestyle intervention was a personalized, telephone-based program targeting a 7% weight-loss in the intervention arm. The metformin dose was 1500 mg/day. The duration of the intervention was 6 months. Main outcomes were biomarkers representing 3 metabolic systems putatively related to breast cancer mortality: glucoregulation, inflammation, and sex hormones. Between August 2011 and May 2015, we randomized 333 breast cancer survivors. Mass mailings from the California Cancer Registry were the most successful recruitment strategy with over 25,000 letters sent at a cost of $191 per randomized participant. At baseline, higher levels of obesity were significantly associated with worse sleep disturbance and impairment scores, lower levels of physical activity and higher levels of sedentary behavior, hypertension, hypercholesterolemia, and lower quality of life (p<0.05 for all). These results illustrate the health burden of obesity. Results of this trial will provide mechanistic data on biological pathways and circulating biomarkers associated with lifestyle and pharmacologic interventions to improve breast cancer prognosis.

show abstract

“…The minimum δ difference in numerical scale pain between dextromethorphan and placebo groups at 4 weeks is estimated at 1.6 and σ standard deviation at 1.5, estimated from published data of the literature [45][46][47], with α = 0.05 two-sided situation and β = 0.10.…”

Section: Sample Size Calculationmentioning

confidence: 99%

Rationale and design of a randomized double-blind clinical trial in breast cancer: Dextromethorphan in chemotherapy-induced peripheral neuropathy

Martin¹,

Morel²,

Joly³

et al. 2015

Contemporary Clinical Trials

View full text Add to dashboard Cite

The analysis of multivariate longitudinal data: A review

Cited by 232 publications

References 97 publications

An analysis of 24-h ambulatory blood pressure monitoring data using orthonormal polynomials in the linear mixed model

An analysis of 24-h ambulatory blood pressure monitoring data using orthonormal polynomials in the linear mixed model

Recruitment strategies, design, and participant characteristics in a trial of weight-loss and metformin in breast cancer survivors

Rationale and design of a randomized double-blind clinical trial in breast cancer: Dextromethorphan in chemotherapy-induced peripheral neuropathy

Contact Info

Product

Resources

About