The Complier Average Causal Effect Parameter for Multiarmed RCTs

2020 ◽  
pp. 0193841X2097920
Author(s):  
Peter Z. Schochet
Keyword(s):  
Causal Effect ◽  
Control Group ◽  
Multiple Treatment ◽  
Intention To Treat ◽  
Treatment Services ◽  
Average Causal Effect ◽  
Key Issues ◽  
Effect Parameter ◽  
Complier Average Causal Effect ◽  
And Control

In randomized controlled trials, the complier average causal effect (CACE) parameter is often of policy interest because it pertains to intervention effects for study units that comply with their research assignments and receive a meaningful dose of treatment services. Causal inference methods for identifying and estimating the CACE parameter using an instrumental variables (IV) framework are well established for designs with a single treatment and control group. This article uses a parallel IV framework to discuss and build on the much smaller literature on estimation of CACE parameters for designs with multiple treatment groups. The key finding is that the conditions to identify and estimate CACE parameters are much more complex for multiarmed designs and may not be tractable in some cases. Practical steps are provided on how to proceed, and a case study demonstrates key issues. The results suggest that ensuring compliance is particularly important in multiarmed trials so that intention-to-treat estimates on the offer of intervention services (which can be identified) can provide meaningful information on the CACE parameters.

Role of good oral hygiene on clinical evolution of rheumatoid arthritis: a randomized study nested in the ESPOIR cohort

Rheumatology ◽  
2019 ◽  
Vol 59 (5) ◽  
pp. 988-996 ◽  
Author(s):  
Xavier Mariette ◽  
Elodie Perrodeau ◽  
Christian Verner ◽  
Xavier Struillou ◽  
Nicolas Picard ◽  
...  
Keyword(s):  
Periodontal Disease ◽  
Oral Hygiene ◽  
Causal Effect ◽  
Randomized Study ◽  
Intervention Group ◽  
Control Group ◽  
Periodontal Pathogens ◽  
Average Causal Effect ◽  
Complier Average Causal Effect ◽  
And Control

Abstract Objective There is a relationship between RA and periodontal disease. We aimed to investigate if a good oral hygiene could improve activity of RA. Methods The patients with RA according to ACR/EULAR 2010 criteria and included in the French early arthritis ESPOIR cohort were included in a randomized nested study into: (i) intervention group: general recommendations of good oral hygiene including teeth brushing, daily antiseptic mouthwash and twice a year scaling; and (ii) control group: no intervention. The primary end point was the delta DAS28-ESR. Results Four hundred and seventy-two patients were randomized (238 in intervention and 234 in control). 92/238 from the intervention group accepted the procedure and 81 had a first visit to the dentist. 56% of patients had periodontal disease at baseline. Duration of RA was 9.0±0.7 years. Baseline DAS28-ESR was 2.7±1.3. After a median duration of 24 months, delta DAS28-ESR was −0.17±1.29 and −0.09±1.28 in intervention and control groups, respectively (mean difference (complier average causal effect): −0.37 (95% CI −1.12, 0.37), P = 0.33). In the intervention group, there was a significant decrease of the bacteria involved in the red complex: Porphyromonas gingivalis (P = 0.002), Tannerella forsythia (P = 0.002) and Treponema denticola (P = 0.019). The patients with baseline periodontal disease and those who became negative for one red complex bacterium had a slightly more important decrease of DAS28-ESR. Conclusion Oral hygiene instruction together with regular scaling and polishing of the teeth significantly decreased the load of periodontal pathogens but did not decrease RA activity. This intervention should be tested in patients with earlier RA and more active disease. Trial registration ClinicalTrials.gov, http://clinicaltrials.gov, NCT01831648.

Azithromycin distribution and childhood mortality in compliance-related subgroups in Niger: complier average causal effect and spillovers in a cluster-randomized, placebo-controlled trial

2021 ◽  
Author(s):  
Kieran S O’Brien ◽  
Ahmed M Arzika ◽  
Ramatou Maliki ◽  
Abdou Amza ◽  
Farouk Manzo ◽  
...  
Keyword(s):  
Incidence Rate ◽  
Rate Ratio ◽  
Incidence Rate Ratio ◽  
Causal Effect ◽  
Controlled Trial ◽  
Effect Analysis ◽  
Intention To Treat ◽  
Average Causal Effect ◽  
Cluster Randomized ◽  
Complier Average Causal Effect

Abstract Background Biannual azithromycin distribution to children 1–59 months old reduced all-cause mortality by 18% [incidence rate ratio (IRR) 0.82, 95% confidence interval (CI): 0.74, 0.90] in an intention-to-treat analysis of a randomized controlled trial in Niger. Estimation of the effect in compliance-related subgroups can support decision making around implementation of this intervention in programmatic settings. Methods The cluster-randomized, placebo-controlled design of the original trial enabled unbiased estimation of the effect of azithromycin on mortality rates in two subgroups: (i) treated children (complier average causal effect analysis); and (ii) untreated children (spillover effect analysis), using negative binomial regression. Results In Niger, 594 eligible communities were randomized to biannual azithromycin or placebo distribution and were followed from December 2014 to August 2017, with a mean treatment coverage of 90% [standard deviation (SD) 10%] in both arms. Subgroup analyses included 2581 deaths among treated children and 245 deaths among untreated children. Among treated children, the incidence rate ratio comparing mortality in azithromycin communities to placebo communities was 0.80 (95% CI: 0.72, 0.88), with mortality rates (deaths per 1000 person-years at risk) of 16.6 in azithromycin communities and 20.9 in placebo communities. Among untreated children, the incidence rate ratio was 0.91 (95% CI: 0.69, 1.21), with rates of 33.6 in azithromycin communities and 34.4 in placebo communities. Conclusions As expected, this analysis suggested similar efficacy among treated children compared with the intention-to-treat analysis. Though the results were consistent with a small spillover benefit to untreated children, this trial was underpowered to detect spillovers.

School-based obesity interventions in the metropolitan area of Rio De Janeiro, Brazil: pooled analysis from five randomised studies

2021 ◽  
pp. 1-7
Author(s):  
Renata da R. M. Rodrigues ◽  
Bruna K. Hassan ◽  
Michele R. Sgambato ◽  
Bárbara da S. N. Souza ◽  
Diana B. Cunha ◽  
...  
Keyword(s):  
Eating Habits ◽  
Intervention Group ◽  
Pooled Analysis ◽  
Control Group ◽  
Intention To Treat ◽  
School Year ◽  
School Based ◽  
Obesity Interventions ◽  
The One ◽  
And Control

Abstract School-based studies, despite the large number of studies conducted, have reported inconclusive results on obesity prevention. The sample size is a major constraint in such studies by requiring large samples. This pooled analysis overcomes this problem by analysing 5926 students (mean age 11·5 years) from five randomised school-based interventions. These studies focused on encouraging students to change their drinking and eating habits, and physical activities over the one school year, with monthly 1-h sessions in the classroom; culinary class aimed at developing cooking skills to increase healthy eating and attempts to family engagement. Pooled intention-to-treat analysis using linear mixed models accounted for school clusters. Control and intervention groups were balanced at baseline. The overall result was a non-significant change in BMI after one school year of positive changes in behaviours associated with obesity. Estimated mean BMI changed from 19·02 to 19·22 kg/m2 in the control group and from 19·08 to 19·32 kg/m2 in the intervention group (P value of change over time = 0·09). Subgroup analyses among those overweight or with obesity at baseline also did not show differences between intervention and control groups. The percentage of fat measured by bioimpedance indicated a small reduction in the control compared with intervention (P = 0·05). This large pooled analysis showed no effect on obesity measures, although promising results were observed about modifying behaviours associated with obesity.

Paracetamol bei Rückenschmerzen nicht empfohlen

2020 ◽  
Vol 24 (02) ◽  
pp. 54-55
Author(s):  
Arne Vielitz
Keyword(s):  
Randomized Controlled Trial ◽  
Low Back Pain ◽  
Causal Effect ◽  
Controlled Trial ◽  
Low Back ◽  
Effect Analysis ◽  
Acute Low Back Pain ◽  
Randomized Controlled ◽  
Average Causal Effect ◽  
Complier Average Causal Effect

Schreijenberg M, Lin CC, McLachlan AJ et al. Paracetamol is Ineffective for Acute Low Back Pain even for Patients Who Comply with Treatment: Complier Average Causal Effect Analysis of a Randomized Controlled Trial. Pain 2019; 160: 2848–2854. doi: 10.1097/j.pain.0000000000001685

scholarly journals The Effectiveness of a Community-Based Mentoring Program for Children Aged 5–11 Years: Results from a Randomized Controlled Trial

2020 ◽  
Vol 22 (1) ◽  
pp. 100-112 ◽  
Author(s):  
Nick Axford ◽  
Gretchen Bjornstad ◽  
Justin Matthews ◽  
Laura Whybra ◽  
Vashti Berry ◽  
...  
Keyword(s):  
Primary Outcome ◽  
Causal Effect ◽  
Controlled Trial ◽  
Well Being ◽  
Mentoring Program ◽  
Post Intervention ◽  
Intention To Treat ◽  
Randomized Controlled ◽  
Average Causal Effect ◽  
Superiority Trial

AbstractThe study, a two-arm, randomized controlled, parallel group, superiority trial, aimed to evaluate the implementation and effectiveness of a 12-month one-to-one volunteer mentoring program designed to improve behavioral and emotional outcomes in children aged 5 to 11 years who have teacher- and parent/carer-reported behavioral difficulties. Participants were 246 children (123 intervention, 123 control; mean age 8.4 years; 87% boys) in five sites in London, UK, scoring in the “abnormal” range on the teacher-rated Strengths and Difficulties Questionnaire (SDQ) Total Difficulties measure and in the “borderline” or abnormal range on the parent-rated SDQ Total Difficulties measure. Randomization on a 1:1 ratio took place using a computer-generated sequence and stratifying by site. Data collectors and statisticians were blind to participant allocation status. Outcome measures focused on parent- and teacher-rated child behavior and emotions, and child-rated self-perception and hope. Intention-to-treat analysis on all 246 randomized participants (using imputed data where necessary) showed that at post-intervention (16 months after randomization), there were no statistically significant effects on the primary outcome—parent-rated SDQ Total Difficulties (adjusted standardized mean difference = − 0.12; 95% CI: −0.38 to 0.13; p = 0.33)—or any secondary outcomes. Results from complier average causal effect (CACE) analysis using the primary outcome indicated the intervention was not effective for children who received the recommended duration of mentoring. Exploratory analyses found no sub-group effects on the primary outcome. The article concludes that the mentoring program had no effect on children’s behavior or emotional well-being, and that program content needs revising to satisfactorily address key risk and protective factors.

Sign in / Sign up

Export Citation Format

Share Document