scholarly journals Beyond total treatment effects in randomised controlled trials: Baseline measurement of intermediate outcomes needed to reduce confounding in mediation investigations

2018 ◽  
Vol 15 (3) ◽  
pp. 247-256 ◽  
Author(s):  
Sabine Landau ◽  
Richard Emsley ◽  
Graham Dunn
Keyword(s):  
Clinical Outcomes ◽  
Treatment Effect ◽  
Average Treatment Effect ◽  
Post Treatment ◽  
Analysis Of Covariance ◽  
Change Scores ◽  
Intermediate Variable ◽  
Common Causes ◽  
Total Treatment ◽  
Baseline Values

Background: Random allocation avoids confounding bias when estimating the average treatment effect. For continuous outcomes measured at post-treatment as well as prior to randomisation (baseline), analyses based on (A) post-treatment outcome alone, (B) change scores over the treatment phase or (C) conditioning on baseline values (analysis of covariance) provide unbiased estimators of the average treatment effect. The decision to include baseline values of the clinical outcome in the analysis is based on precision arguments, with analysis of covariance known to be most precise. Investigators increasingly carry out explanatory analyses to decompose total treatment effects into components that are mediated by an intermediate continuous outcome and a non-mediated part. Traditional mediation analysis might be performed based on (A) post-treatment values of the intermediate and clinical outcomes alone, (B) respective change scores or (C) conditioning on baseline measures of both intermediate and clinical outcomes. Methods: Using causal diagrams and Monte Carlo simulation, we investigated the performance of the three competing mediation approaches. We considered a data generating model that included three possible confounding processes involving baseline variables: The first two processes modelled baseline measures of the clinical variable or the intermediate variable as common causes of post-treatment measures of these two variables. The third process allowed the two baseline variables themselves to be correlated due to past common causes. We compared the analysis models implied by the competing mediation approaches with this data generating model to hypothesise likely biases in estimators, and tested these in a simulation study. We applied the methods to a randomised trial of pragmatic rehabilitation in patients with chronic fatigue syndrome, which examined the role of limiting activities as a mediator. Results: Estimates of causal mediation effects derived by approach (A) will be biased if one of the three processes involving baseline measures of intermediate or clinical outcomes is operating. Necessary assumptions for the change score approach (B) to provide unbiased estimates under either process include the independence of baseline measures and change scores of the intermediate variable. Finally, estimates provided by the analysis of covariance approach (C) were found to be unbiased under all the three processes considered here. When applied to the example, there was evidence of mediation under all methods but the estimate of the indirect effect depended on the approach used with the proportion mediated varying from 57% to 86%. Conclusion: Trialists planning mediation analyses should measure baseline values of putative mediators as well as of continuous clinical outcomes. An analysis of covariance approach is recommended to avoid potential biases due to confounding processes involving baseline measures of intermediate or clinical outcomes, and not simply for increased precision.

scholarly journals Statistical analysis of two arm randomized pre-post designs with one post-treatment measurement

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Fei Wan
Keyword(s):  
Treatment Effect ◽  
Repeated Measures ◽  
Informed Choice ◽  
Clinical Effectiveness ◽  
Change Score ◽  
Post Treatment ◽  
Analysis Of Covariance ◽  
Alternative Methods ◽  
Main Effect ◽  
Comparable Performance

Abstract Background Randomized pre-post designs, with outcomes measured at baseline and after treatment, have been commonly used to compare the clinical effectiveness of two competing treatments. There are vast, but often conflicting, amount of information in current literature about the best analytic methods for pre-post designs. It is challenging for applied researchers to make an informed choice. Methods We discuss six methods commonly used in literature: one way analysis of variance (“ANOVA”), analysis of covariance main effect and interaction models on the post-treatment score (“ANCOVAI” and “ANCOVAII”), ANOVA on the change score between the baseline and post-treatment scores (“ANOVA-Change”), repeated measures (“RM”) and constrained repeated measures (“cRM”) models on the baseline and post-treatment scores as joint outcomes. We review a number of study endpoints in randomized pre-post designs and identify the mean difference in the post-treatment score as the common treatment effect that all six methods target. We delineate the underlying differences and connections between these competing methods in homogeneous and heterogeneous study populations. Results ANCOVA and cRM outperform other alternative methods because their treatment effect estimators have the smallest variances. cRM has comparable performance to ANCOVAI in the homogeneous scenario and to ANCOVAII in the heterogeneous scenario. In spite of that, ANCOVA has several advantages over cRM: i) the baseline score is adjusted as covariate because it is not an outcome by definition; ii) it is very convenient to incorporate other baseline variables and easy to handle complex heteroscedasticity patterns in a linear regression framework. Conclusions ANCOVA is a simple and the most efficient approach for analyzing pre-post randomized designs.

scholarly journals Assumption Trade-Offs When Choosing Identification Strategies for Pre-Post Treatment Effect Estimation: An Illustration of a Community-Based Intervention in Madagascar

2015 ◽  
Vol 3 (1) ◽  
pp. 109-130 ◽  
Author(s):  
Ann M. Weber ◽  
Mark J. van der Laan ◽  
Maya L. Petersen
Keyword(s):  
Treatment Outcome ◽  
Nutritional Status ◽  
Treatment Effect ◽  
Large Scale ◽  
Simulated Data ◽  
Directed Acyclic Graphs ◽  
Average Treatment Effect ◽  
Post Treatment ◽  
Nutrition Program ◽  
Trade Offs

AbstractFailure (or success) in finding a statistically significant effect of a large-scale intervention may be due to choices made in the evaluation. To highlight the potential limitations and pitfalls of some common identification strategies used for estimating causal effects of community-level interventions, we apply a roadmap for causal inference to a pre-post evaluation of a national nutrition program in Madagascar. Selection into the program was non-random and strongly associated with the pre-treatment (lagged) outcome. Using structural causal models (SCM), directed acyclic graphs (DAGs) and simulated data, we illustrate that an estimand with the outcome defined as the post-treatment outcome controls for confounding by the lagged outcome but not by possible unmeasured confounders. Two separate differencing estimands (of the pre- and post-treatment outcome) have the potential to adjust for a certain type of unmeasured confounding, but introduce bias if the additional identification assumptions they rely on are not met. In order to illustrate the practical impact of choice between three common identification strategies and their corresponding estimands, we used observational data from the community nutrition program in Madagascar to estimate each of these three estimands. Specifically, we estimated the average treatment effect of the program on the community mean nutritional status of children 5 years and under and found that the estimate based on the post-treatment estimand was about a quarter of the magnitude of either of the differencing estimands (0.066 SD vs. 0.26–0.27 SD increase in mean weight-for-age z-score). Choice of estimand clearly has important implications for the interpretation of the success of the program to improve nutritional status of young children. A careful appraisal of the assumptions underlying the causal model is imperative before committing to a statistical model and progressing to estimation. However, knowledge about the data-generating process must be sufficient in order to choose the identification strategy that gets us closest to the truth.

Effectiveness of transdiagnostic seminars to support patients with common mental disorders: a multi-service practice research network study

2021 ◽  
pp. 1-13
Author(s):  
Chris Gaskell ◽  
Ryan Askey-Jones ◽  
Martin Groom ◽  
Jaime Delgadillo
Keyword(s):  
Treatment Effect ◽  
High Intensity ◽  
Post Treatment ◽  
Waiting List ◽  
Effect Sizes ◽  
Research Network ◽  
Randomised Control Trial ◽  
Practice Research ◽  
Matched Sample ◽  
Post Treatment Effect

Abstract Background: This was a multi-site evaluation of psycho-educational transdiagnostic seminars (TDS) as a pre-treatment intervention to enhance the effectiveness and utilisation of high-intensity cognitive behavioural therapy (CBT). Aims: To evaluate the effectiveness of TDS combined with high-intensity CBT (TDS+CBT) versus a matched sample receiving CBT only. Second, to determine the consistency of results across participating services which employed CBT+TDS. Finally, to determine the acceptability of TDS across patients with different psychological disorders. Method: 106 patients across three services voluntarily attended TDS while on a waiting list for CBT (TDS+CBT). Individual and pooled service pre–post treatment effect sizes were calculated using measures of depression, anxiety and functional impairment. Effectiveness and completion rates for TDS+CBT were compared with a propensity score matched sample from an archival dataset of cases who received high-intensity CBT only. Results: Pre–post treatment effect sizes for TDS+CBT were comparable to the matched sample. Recovery rates were greater for the group receiving TDS; however, this was not statistically significant. Greater improvements were observed during the waiting-list period for patients who had received TDS for depression (d = 0.49 compared with d = 0.07) and anxiety (d = 0.36 compared with d = 0.04). Conclusions: Overall, this new evidence found a trend for TDS improving symptoms while awaiting CBT across three separate IAPT services. The effectiveness of TDS now warrants further exploration through an appropriately sized randomised control trial.

scholarly journals Evaluación de impacto académico y social en egresados universitarios

2019 ◽  
Vol 30 (3) ◽  
pp. 695-712
Author(s):  
Gabriel González ◽  
Luisa Díez-Echavarría ◽  
Elkin Zapa ◽  
Danilo Eusse
Keyword(s):  
Treatment Effect ◽  
Average Treatment Effect ◽  
Educación Superior ◽  
Educacion Superior ◽  
Average Treatment

Las instituciones de educación superior deben formar a sus estudiantes según requerimientos del contexto en que se desenvuelven, ya que, sobre la base de su desempeño, es donde se medirá si las políticas de desarrollo socioeconómico son efectivas. Para lograrlo, es necesario identificar el impacto de esa educación en sus egresados, y hacer los ajustes necesarios que generen mejora continua. El objetivo de este artículo es estimar el impacto académico y social de egresados del Instituto Tecnológico Metropolitano – Medellín, a través de un análisis multivariado y la estimación del modelo Average Treatment Effect (ATE). Se encontró que la educación ofrecida a esta población ha generado un impacto académico, asociado a los estudios de actualización, y dos impactos sociales, asociados a la situación laboral y al nivel de ingresos percibidos por los egresados. Se recomienda usar esta metodología en otras instituciones, ya que suele arrojar resultados más informativos y precisos que los estudios tradicionales de caracterización, y se puede medir el efecto de cualquier estrategia.

scholarly journals Low Back Pain in Adolescents: A Comparison of Clinical Outcomes in Sports Participants and Nonparticipants

2010 ◽  
Vol 45 (1) ◽  
pp. 61-66 ◽  
Author(s):  
Julie M. Fritz ◽  
Shannon N. Clifford
Keyword(s):  
Low Back Pain ◽  
Back Pain ◽  
Physical Therapy ◽  
Treatment Outcomes ◽  
Clinical Outcomes ◽  
Sports Participation ◽  
Mean Difference ◽  
Analysis Of Covariance ◽  
Low Back ◽  
Baseline Characteristics

Abstract Context: Back pain is common in adolescents. Participation in sports has been identified as a risk factor for the development of back pain in adolescents, but the influence of sports participation on treatment outcomes in adolescents has not been adequately examined. Objective: To examine the clinical outcomes of rehabilitation for adolescents with low back pain (LBP) and to evaluate the influence of sports participation on outcomes. Design: Observational study. Setting: Outpatient physical therapy clinics. Patients or Other Participants: Fifty-eight adolescents (age  =  15.40 ± 1.44 years; 56.90% female) with LBP referred for treatment. Twenty-three patients (39.66%) had developed back pain from sports participation. Intervention(s): Patients completed the Modified Oswestry Disability Questionnaire and numeric pain rating before and after treatment. Treatment duration and content were at the clinician's discretion. Adolescents were categorized as sports participants if the onset of back pain was linked to organized sports. Additional data collected included diagnostic imaging before referral, clinical characteristics, and medical diagnosis. Main Outcome Measure(s): Baseline characteristics were compared based on sports participation. The influence of sports participation on outcomes was examined using a repeated-measures analysis of covariance with the Oswestry and pain scores as dependent variables. The number of sessions and duration of care were compared using t tests. Results: Many adolescents with LBP receiving outpatient physical therapy treatment were involved in sports and cited sports participation as a causative factor for their LBP. Some differences in baseline characteristics and clinical treatment outcomes were noted between sports participants and nonparticipants. Sports participants were more likely to undergo magnetic resonance imaging before referral (P  =  .013), attended more sessions (mean difference  =  1.40, 95% confidence interval [CI]  =  0.21, 2.59, P  =  .022) over a longer duration (mean difference  =  12.44 days, 95% CI  =  1.28, 23.10, P  =  .024), and experienced less improvement in disability (mean Oswestry difference  =  6.66, 95% CI  =  0.53, 12.78, P  =  .048) than nonparticipants. Overall, the pattern of clinical outcomes in this sample of adolescents with LBP was similar to that of adults with LBP. Conclusions: Adolescents with LBP due to sports participation received more treatment but experienced less improvement in disability than nonparticipants. This may indicate a worse prognosis for sports participants. Further research is required.

Sign in / Sign up

Export Citation Format

Share Document