Assessing efficacy in non-inferiority trials with non-adherence to interventions: Are intention-to-treat and per-protocol analyses fit for purpose?

BACKGROUND: Non-inferiority trials comparing different active drugs are often subject to treatment non-adherence. Intention-to-treat (ITT) and per-protocol (PP) analyses have been advocated in such studies but are not guaranteed to be unbiased in the presence of differential non-adherence. METHODS: The REMoxTB trial evaluated two 4-month experimental regimens compared with a 6-month control regimen for newly diagnosed drug-susceptible TB. The primary endpoint was a composite unfavorable outcome of treatment failure or recurrence within 18 months post-randomization. We conducted a simulation study based on REMoxTB to assess the performance of statistical methods for handling non-adherence in non-inferiority trials, including: ITT and PP analyses, adjustment for observed adherence, multiple imputation (MI) of outcomes, inverse-probability-of-treatment weighting (IPTW), and a doubly-robust (DR) estimator. RESULTS: When non-adherence differed between trial arms, ITT, and PP analyses often resulted in non-trivial bias in the estimated treatment effect, which consequently under- or over-inflated the type I error rate. Adjustment for observed adherence led to similar issues, whereas the MI, IPTW and DR approaches were able to correct bias under most non-adherence scenarios; they could not always eliminate bias entirely in the presence of unobserved confounding. The IPTW and DR methods were generally unbiased and maintained desired type I error rates and statistical power. CONCLUSIONS: When non-adherence differs between trial arms, ITT and PP analyses can produce biased estimates of efficacy, potentially leading to the acceptance of inferior treatments or efficacious regimens being missed. IPTW and the DR estimator are relatively straightforward methods to supplement ITT and PP approaches.

Why estimands are needed to define treatment effects in clinical trials.

BACKGROUND: The estimand for a clinical trial is a precise definition of the treatment effect to be estimated. Traditionally, estimates of treatment effects are based on either an ITT analysis or a per-protocol analysis. However, there are important clinical questions which are not addressed by either of these analyses. For example, consider a trial where patients take a rescue medication. The ITT analysis includes data after use of rescue, while the per-protocol analysis excludes these patients altogether. Neither of these analyses addresses the important question of what the treatment effect would have been if patients did not take rescue medication. MAIN TEXT: Trial estimands provide a broader perspective compared to the limitations of ITT and per-protocol analysis. Trial treatment effects depend on how events occurring after treatment initiation such as use of alternative medication or discontinuation of the intervention are included in the definition. These events can be accounted for in different ways, depending on the clinical question of interest. CONCLUSION: The estimand framework is an important step forward in improving the clarity and transparency of clinical trials. The centrality of estimands to clinical trials is currently not reflected in methods recommended by the Cochrane group or the CONSORT statement, the current standard for reporting clinical trials in medical journals. We encourage revisions to these guidelines.

Adherence, per-protocol effects, and the estimands framework.

In the statistical literature, treatment effects in clinical trials are frequently described as either ITT or per-protocol effects. The estimand given for the per-protocol effect is the effect in adherers, where adherers are typically defined as adhering to the intervention as specified in the trial protocol. This dichotomy of treatment effects is unhelpful when there are in reality multiple treatment effects that can be of clinical interest and relevance. The terms "per-protocol" and "adherence" are confusing to non-statisticians. Protocols always allow for discontinuation of randomized treatment so participants discontinuing have actually followed the protocol. When rescue or additional medication is available, the effect in adherers could mean the effect regardless of use of these medications or the effect in a counterfactual world where the participant did not take the medication. Adherence can mean continuing to be prescribed a treatment or some arbitrary level of compliance with a medication that has been prescribed. The ICH E9 (R1) estimands framework provides an improved alternative for the description of treatment effects in clinical trials. Identification of important intercurrent events and the strategy used to handle these events is key to determining the treatment effect. When designing a trial, estimands should be properly defined according to this framework. It is time the statistical literature abandoned describing treatment effects as the effect in adherers or the per-protocol effect.

Invited Commentary: Conducting and Emulating Trials to Study Effects of Social Interventions.

All else being equal, if we had 1 causal effect we wished to estimate, we would conduct a randomized trial with a protocol that mapped onto that causal question, or we would attempt to emulate that target trial with observational data. However, studying the social determinants of health often means there are not just 1 but several causal contrasts of simultaneous interest and importance, and each of these related but distinct causal questions may have varying degrees of feasibility in conducting trials. With this in mind, we discuss challenges and opportunities that arise when conducting and emulating such trials. We describe designing trials with the simultaneous goals of estimating the intention-to-treat effect, the per-protocol effect, effects of alternative protocols or joint interventions, effects within subgroups, and effects under interference, and we describe ways to make the most of all feasible randomized trials and emulated trials using observational data. Our comments are grounded in the study results of Courtin et al. (Am J Epidemiol. 2022;191(8):1444-1452).

Rethinking intercurrent events in defining estimands for tuberculosis trials.

BACKGROUND/AIMS: Tuberculosis remains one of the leading causes of death from an infectious disease globally. Both choices of outcome definitions and approaches to handling events happening post-randomisation can change the treatment effect being estimated, but these are often inconsistently described, thus inhibiting clear interpretation and comparison across trials. METHODS: Starting from the ICH E9(R1) addendum's definition of an estimand, we use our experience of conducting large Phase III tuberculosis treatment trials and our understanding of the estimand framework to identify the key decisions regarding how different event types are handled in the primary outcome definition, and the important points that should be considered in making such decisions. A key issue is the handling of intercurrent (i.e. post-randomisation) events (ICEs) which affect interpretation of or preclude measurement of the intended final outcome. We consider common ICEs including treatment changes and treatment extension, poor adherence to randomised treatment, re-infection with a new strain of tuberculosis which is different from the original infection, and death. We use two completed tuberculosis trials (REMoxTB and STREAM Stage 1) as illustrative examples. These trials tested non-inferiority of new tuberculosis treatment regimens versus a control regimen. The primary outcome was a binary composite endpoint, 'favourable' or 'unfavourable', which was constructed from several components. RESULTS: We propose the following improvements in handling the above-mentioned ICEs and loss to follow-up (a post-randomisation event that is not in itself an ICE). First, changes to allocated regimens should not necessarily be viewed as an unfavourable outcome; from the patient perspective, the potential harms associated with a change in the regimen should instead be directly quantified. Second, handling poor adherence to randomised treatment using a per-protocol analysis does not necessarily target a clear estimand; instead, it would be desirable to develop ways to estimate the treatment effects more relevant to programmatic settings. Third, re-infection with a new strain of tuberculosis could be handled with different strategies, depending on whether the outcome of interest is the ability to attain culture negativity from infection with any strain of tuberculosis, or specifically the presenting strain of tuberculosis. Fourth, where possible, death could be separated into tuberculosis-related and non-tuberculosis-related and handled using appropriate strategies. Finally, although some losses to follow-up would result in early treatment discontinuation, patients lost to follow-up before the end of the trial should not always be classified as having an unfavourable outcome. Instead, loss to follow-up should be separated from not completing the treatment, which is an ICE and may be considered as an unfavourable outcome. CONCLUSION: The estimand framework clarifies many issues in tuberculosis trials but also challenges trialists to justify and improve their outcome definitions. Future trialists should consider all the above points in defining their outcomes.

The ring vaccination trial design for the estimation of vaccine efficacy and effectiveness during infectious disease outbreaks.

The ring vaccination trial is a recently developed approach for evaluating the efficacy and effectiveness of vaccines, modeled after the surveillance and containment strategy of ring vaccination. Contacts and contacts of contacts of a newly identified disease case form a ring, and these rings are randomized as part of a cluster-randomized trial or with individual randomization within rings. Key advantages of the design include its flexibility to follow the epidemic as it progresses and the targeting of high-risk participants to increase power. We describe the application of the design to estimate the efficacy and effectiveness of an Ebola vaccine during the 2014-2016 West African Ebola epidemic. The design has several notable statistical features. Because vaccination occurs around the time of exposure, the design is particularly sensitive to the choice of per protocol analysis period. If incidence wanes before the per protocol analysis period begins (due to a slow-acting vaccine or a fast-moving pathogen), power can be substantially reduced. Mathematical modeling is valuable for exploring the suitability of the approach in different disease settings. Another statistical feature is zero inflation, which can occur if the chain of transmission does not take off within a ring. In the application to Ebola, the majority of rings had zero subsequent cases. The ring vaccination trial can be extended in several ways, including the definition of rings (e.g. contact-based, spatial, and occupational). The design will be valuable in settings where the spatio-temporal spread of the pathogen is highly focused and unpredictable.

Adjuvant chemotherapy is superior to chemoradiation after D2 surgery for gastric cancer in the per-protocol analysis of the randomized CRITICS trial.

BACKGROUND: The Intergroup 0116 and the MAGIC trials changed clinical practice for resectable gastric cancer in the Western world. In these trials, overall survival improved with post-operative chemoradiotherapy (CRT) and perioperative chemotherapy (CT). Intention-to-treat analysis in the CRITICS trial of post-operative CT or post-operative CRT did not show a survival difference. The current study reports on the per-protocol (PP) analysis of the CRITICS trial. PATIENTS AND METHODS: The CRITICS trial was a randomized, controlled trial in which 788 patients with stage Ib-Iva resectable gastric or esophagogastric adenocarcinoma were included. Before start of preoperative CT, patients from the Netherlands, Sweden and Denmark were randomly assigned to receive post-operative CT or CRT. For the current analysis, only patients who started their allocated post-operative treatment were included. Since it is uncertain that the two treatment arms are balanced in such PP analysis, adjusted proportional hazards regression analysis and inverse probability weighted analysis were used to minimize the risk of selection bias and to estimate and compare overall and event-free survival. RESULTS: Of the 788 patients, 478 started post-operative treatment according to protocol, 233 (59%) patients in the CT group and 245 (62%) patients in the CRT group. Patient and tumor characteristics between the groups before start of the post-operative treatment were not different. After a median follow-up of 6.7 years since the start of post-operative treatment, the 5-year overall survival was 57.9% (95% confidence interval: 51.4% to 64.3%) in the CT group versus 45.5% (95% confidence interval: 39.2% to 51.8%) in the CRT group (adjusted hazard ratio CRT versus CT: 1.62 (1.24-2.12), P = 0.0004). Inverse probability weighted analysis resulted in similar hazard ratios. CONCLUSION: After adjustment for all known confounding factors, the PP analysis of patients who started the allocated post-operative treatment in the CRITICS trial showed that the CT group had a significantly better 5-year overall survival than the CRT group (NCT00407186).

Intention-to-treat analysis may be more conservative than per protocol analysis in antibiotic non-inferiority trials: a systematic review.

BACKGROUND: In non-inferiority trials, there is a concern that intention-to-treat (ITT) analysis, by including participants who did not receive the planned interventions, may bias towards making the treatment and control arms look similar and lead to mistaken claims of non-inferiority. In contrast, per protocol (PP) analysis is viewed as less likely to make this mistake and therefore preferable in non-inferiority trials. In a systematic review of antibiotic non-inferiority trials, we compared ITT and PP analyses to determine which analysis was more conservative. METHODS: In a secondary analysis of a systematic review, we included non-inferiority trials that compared different antibiotic regimens, used absolute risk reduction (ARR) as the main outcome and reported both ITT and PP analyses. All estimates and confidence intervals (CIs) were oriented so that a negative ARR favored the control arm, and a positive ARR favored the treatment arm. We compared ITT to PP analyses results. The more conservative analysis between ITT and PP analyses was defined as the one having a more negative lower CI limit. RESULTS: The analysis included 164 comparisons from 154 studies. In terms of the ARR, ITT analysis yielded the more conservative point estimate and lower CI limit in 83 (50.6%) and 92 (56.1%) comparisons respectively. The lower CI limits in ITT analysis favored the control arm more than in PP analysis (median of - 7.5% vs. -6.9%, p = 0.0402). CIs were slightly wider in ITT analyses than in PP analyses (median of 13.3% vs. 12.4%, p < 0.0001). The median success rate was 89% (interquartile range IQR 82 to 93%) in the PP population and 44% (IQR 23 to 60%) in the patients who were included in the ITT population but excluded from the PP population (p < 0.0001). CONCLUSIONS: Contrary to common belief, ITT analysis was more conservative than PP analysis in the majority of antibiotic non-inferiority trials. The lower treatment success rate in the ITT analysis led to a larger variance and wider CI, resulting in a more conservative lower CI limit. ITT analysis should be mandatory and considered as either the primary or co-primary analysis for non-inferiority trials. TRIAL REGISTRATION: PROSPERO registration number CRD42020165040 .

Adjusting for adherence in randomized trials when adherence is measured as a continuous variable: An application to the Lipid Research Clinics Coronary Primary Prevention Trial.

BACKGROUND: Clinicians and patients may be more interested in per-protocol effect estimates than intention-to-treat effect estimates from randomized trials. However, per-protocol effect estimates may be biased due to insufficient adjustment for prognostic factors that predict adherence. Adjustment for this bias is possible when appropriate methods, such as inverse probability weighting, are used. But, when adherence is measured as a continuous variable, constructing these weights can be challenging. METHODS: In the placebo arm of the Lipid Research Clinics Coronary Primary Prevention Trial, we estimated the 7-year cumulative incidence of coronary heart disease under 100% adherence and 0% adherence to placebo. We used dose-response discrete-hazards models with inverse probability weighting to adjust for pre- and post-randomization covariates. We considered several continuous distributions for constructing the inverse probability weights. RESULTS: The risk difference estimate for 100% adherence compared with 0% adherence ranged from -7.7 to -6.1 percentage points without adjustment for baseline and post-baseline covariates, and ranged from -1.8 to 2.2 percentage points with adjustment using inverse probability weights, depending on the dose-response model and inverse probability weight distribution used. CONCLUSIONS: Methods which appropriately adjust for time-varying post-randomization variables can explain away much of the bias in the "effect" of adherence to placebo. When considering continuous adherence, investigators should consider several models as estimates may be sensitive to the model chosen.

Intention to treat and per protocol analysis in clinical trials.

In clinical epidemiology, experimental studies usually take the form of randomized controlled clinical trials (RCTs). The data analysis of an RCT can be performed by using two complementary strategies, that is according to the intention to treat (ITT) principle and the per protocol (PP) analysis. By using the ITT approach, investigators aim to assess the effect of assigning a drug whereas by adopting the PP analysis, researchers investigate the effect of receiving the assigned treatment, as specified in the protocol. Both ITT and PP analyses are essentially valid but they have different scopes and interpretations dependent on the context.

Acide hyaluronique intra articulaire dans la gonarthrose: une étude long-terme, ouverte, de Phase IV, avec analyse détaillée par stade de Kellgren-Lawrence.

INTRODUCTION: Les injections intra articulaires (IA) d'acide hyaluronique (HA) désignées sous le nom de viscosupplémentation (VS), sont fréquemment utilisées dans le traitement symptomatique de la gonarthrose (OA), une affection ostéo-articulaire chronique douloureuse et handicapante, qui touche une fraction importante de la population âgée. La sévérité de la gonarthrose est en général décrite par la classification en stades radiologiques de Kellgren-Lawrence (KL). La VS a été largement étudiée à travers de nombreux essais cliniques; cependant, les résultats sont rarement analysés en détail, en fonction du stade KL. MÉTHODE: Une étude ouverte importante, portant sur 1 177 patients souffrant de gonarthrose, fut réalisée de 2004 à 2007. Chaque patient a reçu un traitement de VS consistant en 3 injections d'ARTHRUM H 2% (LCA Pharmaceutical, Chartres, France). A l'inclusion, les patients ont été décrits par leur profil démographique, leur indice de masse corporelle (IMC), leur stade KL et leur état clinique selon les sous-scores douleur et fonction de l'indice Western Ontario and McMaster Universities (WOMAC). Les visites de suivi étaient à M3, M6 et M9 (mois) après la VS. Cette large base de données a été entièrement retraitée en 2019, de manière à fournir une analyse séparée par stade KL, et fut complétée par l'évaluation des taux de patients répondeurs (%) au traitement, selon l'Outcome Measures in Rheumatoid Arthritis Clinical Trials & Osteoarthritis Research Society International (OMERACT-OARSI). L'analyse fut menée à la fois sur les populations en intention de traiter (ITT) et per protocole (PP) ayant terminé l'étude. RÉSULTATS: En analyse ITT du critère principal, les variations du sous-score WOMAC A (douleur) depuis l'inclusion jusqu'à la fin de l'étude, ont été respectivement de 19,8 ; 19,8 ; 17,8 et 14,2, sur une échelle de 0-100, pour les patients des stades KL I à KL IV. En analyse PP dans les mêmes conditions, ces variations ont été de 20,6 ; 19,9 ; 17,1 et 11,7. Tous ces résultats étaient significatifs par rapport aux valeurs à l'inclusion (p<0.001) et cliniquement pertinents à chaque stade KL. Des améliorations significatives ont été également observées pour le sous-score WOMAC C (fonction), et pour les autres critères secondaires. Le taux de répondeurs OMERACT-OARSI variait de 72 à 82% pour les patients KL I à III à M6 et M9. Pour les patients KL IV, le maximum atteint a été 47.7% à M6. Les autres paramètres tels que le sexe, l'IMC ou l'âge, ne furent pas identifiés comme des facteurs de pronostic pour la réponse à la VS. CONCLUSIONS: L'analyse détaillée par stade KL d'une large cohorte de patients suivis en ouvert, suggère le traitement de VS avec ARTHRUM H 2% est applicable à une grande variété de patients gonarthrosiques.

Effect Estimates in Randomized Trials and Observational Studies: Comparing Apples With Apples.

Effect estimates from randomized trials and observational studies might not be directly comparable because of differences in study design, other than randomization, and in data analysis. We propose a 3-step procedure to facilitate meaningful comparisons of effect estimates from randomized trials and observational studies: 1) harmonization of the study protocols (eligibility criteria, treatment strategies, outcome, start and end of follow-up, causal contrast) so that the studies target the same causal effect, 2) harmonization of the data analysis to estimate the causal effect, and 3) sensitivity analyses to investigate the impact of discrepancies that could not be accounted for in the harmonization process. To illustrate our approach, we compared estimates of the effect of immediate with deferred initiation of antiretroviral therapy in individuals positive for the human immunodeficiency virus from the Strategic Timing of Antiretroviral Therapy (START) randomized trial and the observational HIV-CAUSAL Collaboration.

Diagnosing Covariate Balance Across Levels of Right-Censoring Before and After Application of Inverse-Probability-of-Censoring Weights.

Covariate balance is a central concept in the potential outcomes literature. With selected populations or missing data, balance across treatment groups can be insufficient for estimating marginal treatment effects. Recently, a framework for using covariate balance to describe measured confounding and selection bias for time-varying and other multivariate exposures in the presence of right-censoring has been proposed. Here, we revisit this framework to consider balance across levels of right-censoring over time in more depth. Specifically, we develop measures of covariate balance that can describe what is known as "dependent censoring" in the literature, along with its associated selection bias, under multiple mechanisms for right censoring. Such measures are interesting because they substantively describe the evolution of dependent censoring mechanisms. Furthermore, we provide weighted versions that can depict how well such dependent censoring has been eliminated when inverse-probability-of-censoring weights are applied. These results provide a conceptually grounded way to inspect covariate balance across levels of right-censoring as a validity check. As a motivating example, we applied these measures to a study of hypothetical "static" and "dynamic" treatment protocols in a sequential multiple-assignment randomized trial of antipsychotics with high dropout rates.

Generalizing the per-protocol treatment effect: The case of ACTG A5095.

BACKGROUND: Intention-to-treat comparisons of randomized trials provide asymptotically consistent estimators of the effect of treatment assignment, without regard to compliance. However, decision makers often wish to know the effect of a per-protocol comparison. Moreover, decision makers may also wish to know the effect of treatment assignment or treatment protocol in a user-specified target population other than the sample in which the trial was fielded. Here, we aimed to generalize results from the ACTG A5095 trial to the US recently HIV-diagnosed target population. METHODS: We first replicated the published conventional intention-to-treat estimate (2-year risk difference and hazard ratio) comparing a four-drug antiretroviral regimen to a three-drug regimen in the A5095 trial. We then estimated the intention-to-treat effect that accounted for informative dropout and the per-protocol effect that additionally accounted for protocol deviations by constructing inverse probability weights. Furthermore, we employed inverse odds of sampling weights to generalize both intention-to-treat and per-protocol effects to a target population comprising US individuals with HIV diagnosed during 2008-2014. RESULTS: Of 761 subjects in the analysis, 82 dropouts (36 in the three-drug arm and 46 in the four-drug arm) and 59 protocol deviations (25 in the three-drug arm and 34 in the four-drug arm) occurred during the first 2 years of follow-up. A total of 169 subjects incurred virologic failure or death. The 2-year risks were similar both in the trial and in the US HIV-diagnosed target population for estimates from the conventional intention-to-treat, dropout-weighted intention-to-treat, and per-protocol analyses. In the US target population, the 2-year conventional intention-to-treat risk difference (unit: %) for virologic failure or death comparing the four-drug arm to the three-drug arm was -0.4 (95% confidence interval: -6.2, 5.1), while the hazard ratio was 0.97 (95% confidence interval: 0.70, 1.34); the 2-year risk difference was -0.9 (95% confidence interval: -6.9, 5.3) for the dropout-weighted intention-to-treat comparison (hazard ratio = 0.95, 95% confidence interval: 0.68, 1.32) and -0.7 (95% confidence interval: -6.7, 5.5) for the per-protocol comparison (hazard ratio = 0.96, 95% confidence interval: 0.69, 1.34). CONCLUSION: No benefit of four-drug antiretroviral regimen over three-drug regimen was found from the conventional intention-to-treat, dropout-weighted intention-to-treat or per-protocol estimates in the trial sample or target population.

Intention-to-treat and transparency of related practices in randomized, controlled trials of anti-infectives.

BACKGROUND: Intention-to-treat (ITT) analysis is commonly recommended for use, due to its benefits on external validity, in randomized, controlled trials (RCTs). No published reports describe how ITT analysis, as well as alternative approaches, are used in anti-infective RCTs. The purpose of this study is to describe the extent to which ITT analysis and alternative data approaches are used, the practices used to handle missing subject data, and whether non-inferiority trials present both ITT and per protocol (PP) analyses. Results of this analysis will help guide end users of infectious diseases primary drug literature. METHODS: A cross-sectional study of RCTs of anti-infectives published from January 1, 2013 through December 31, 2014 was conducted. A PubMed search identified relevant articles published in five specialty infectious diseases journals and four general medical journals. Each article was reviewed by two independent investigators with discrepancies resolved by consensus. Descriptive statistics were used to quantify results. RESULTS: One hundred four articles met study criteria. The most common medication classes represented in the RCTs were hepatitis C antivirals (26 %), antibacterials (25 %), and antiretrovirals (21 %). Thirty studies (29 %) were non-inferiority trials. Most studies (77 %) described use of ITT or modified ITT (mITT) in their methods. Of the ITT and mITT studies, most (73 %) did not describe practices used to handle missing data. Most (97 %) non-inferiority trials described use of ITT, mITT, or both; however, only 15 (50 %) also described use of PP. CONCLUSIONS: RCTs of anti-infectives commonly employ ITT and mITT. Most do not describe how missing data were addressed. Non-inferiority trials of anti-infectives do not consistently employ both ITT and PP populations.

Adherence adjustment in the Coronary Drug Project: A call for better per-protocol effect estimates in randomized trials.

BACKGROUND: In many randomized controlled trials, patients and doctors are more interested in the per-protocol effect than in the intention-to-treat effect. However, valid estimation of the per-protocol effect generally requires adjustment for prognostic factors associated with adherence. These adherence adjustments have been strongly questioned in the clinical trials community, especially after 1980 when the Coronary Drug Project team found that adherers to placebo had lower 5-year mortality than non-adherers to placebo. METHODS: We replicated the original Coronary Drug Project findings from 1980 and re-analyzed the Coronary Drug Project data using technical and conceptual developments that have become established since 1980. Specifically, we used logistic models for binary outcomes, decoupled the definition of adherence from loss to follow-up, and adjusted for pre-randomization covariates via standardization and for post-randomization covariates via inverse probability weighting. RESULTS: The original Coronary Drug Project analysis reported a difference in 5-year mortality between adherers and non-adherers in the placebo arm of 9.4 percentage points. Using modern approaches, we found that this difference was reduced to 2.5 (95% confidence interval: -2.1 to 7.0). CONCLUSION: Valid estimation of per-protocol effects may be possible in randomized clinical trials when analysts use appropriate methods to adjust for post-randomization variables.

Theory-Driven Process Evaluation of the SHINE Trial Using a Program Impact Pathway Approach.

Two reasons for the lack of success of programs or interventions are poor alignment of interventions with the causes of the problem targeted by the intervention, leading to poor efficacy (theory failure), and failure to implement interventions as designed (program failure). These failures are important for both public health programs and randomized trials. In the Sanitation Hygiene and Infant Nutrition Efficacy (SHINE) Trial, we utilize the program impact pathway (PIP) approach to track intervention implementation and behavior uptake. In this article, we present the SHINE PIP including definitions and measurements of key mediating domains, and discuss the implications of this approach for randomized trials. Operationally, the PIP can be used for monitoring and strengthening intervention delivery, facilitating course-correction at various stages of implementation. Analytically, the PIP can facilitate a richer understanding of the mediating and modifying determinants of intervention impact than would be possible from an intention-to-treat analysis alone.

Simple Estimation of Patient-Oriented Effects From Randomized Trials: An Open and Shut CACE.

In randomized controlled trials, the intention-to-treat estimator provides an unbiased estimate of the causal effect of treatment assignment on the outcome. However, patients often want to know what the effect would be if they were to take the treatment as prescribed (the patient-oriented effect), and several researchers have suggested that the more relevant causal effect for this question is the complier average causal effect (CACE), also referred to as the local average treatment effect. Sophisticated approaches to estimating the CACE include Bayesian and frequentist methods for principal stratification, inverse-probability-of-treatment-weighted estimators, and instrumental-variable (IV) analysis. All of these approaches exploit information about adherence to assigned treatment to improve upon the intention-to-treat estimator, but they are rarely used in practice, probably because of their complexity. The IV principal stratification estimator is simple to implement but has had limited use in practice, possibly due to lack of familiarity. Here, we show that the IV principal stratification estimator is a modified per-protocol estimator that should be obtainable from any randomized controlled trial, and we provide a closed form for its robust variance (and its uncertainty). Finally, we illustrate sensitivity analyses we conducted to assess inference in light of potential violations of the exclusion restriction assumption.

Correcting for non-compliance in randomized non-inferiority trials with active and placebo control using structural models.

The three-arm clinical trial design, which includes a test treatment, an active reference, and placebo control, is the gold standard for the assessment of non-inferiority. In the presence of non-compliance, one common concern is that an intent-to-treat (ITT) analysis (which is the standard approach to non-inferiority trials), tends to increase the chances of erroneously concluding non-inferiority, suggesting that the per-protocol (PP) analysis may be preferable for non-inferiority trials despite its inherent bias. The objective of this paper was to develop statistical methodology for dealing with non-compliance in three-arm non-inferiority trials for censored, time-to-event data. Changes in treatment were here considered the only form of non-compliance. An approach using a three-arm rank preserving structural failure time model and G-estimation analysis is here presented. Using simulations, the impact of non-compliance on non-inferiority trials was investigated in detail using ITT, PP analyses, and the present proposed method. Results indicate that the proposed method shows good characteristics, and that neither ITT nor PP analyses can always guarantee the validity of the non-inferiority conclusion. A Statistical Analysis System program for the implementation of the proposed test procedure is available from the authors upon request.

Effect of the probiotic strain Bifidobacterium animalis subsp. lactis, BB-12®, on defecation frequency in healthy subjects with low defecation frequency and abdominal discomfort: a randomised, double-blind, placebo-controlled, parallel-group trial.

The aim of the present study was to investigate the effect of Bifidobacterium animalis subsp. lactis, BB-12®, on two primary end points - defecation frequency and gastrointestinal (GI) well-being - in healthy adults with low defecation frequency and abdominal discomfort. A total of 1248 subjects were included in a randomised, double-blind, placebo-controlled trial. After a 2-week run-in period, subjects were randomised to 1 or 10 billion colony-forming units/d of the probiotic strain BB-12® or a matching placebo capsule once daily for 4 weeks. Subjects completed a diary on bowel habits, relief of abdominal discomfort and symptoms. GI well-being, defined as global relief of abdominal discomfort, did not show significant differences. The OR for having a defecation frequency above baseline for ≥50% of the time was 1·31 (95% CI 0·98, 1·75), P=0·071, for probiotic treatment overall. Tightening the criteria for being a responder to an increase of ≥1 d/week for ≥50 % of the time resulted in an OR of 1·55 (95% CI 1·22, 1·96), P=0·0003, for treatment overall. A treatment effect on average defecation frequency was found (P=0·0065), with the frequency being significantly higher compared with placebo at all weeks for probiotic treatment overall (all P<0·05). Effects on defecation frequency were similar for the two doses tested, suggesting that a ceiling effect was reached with the one billion dose. Overall, 4 weeks' supplementation with the probiotic strain BB-12® resulted in a clinically relevant benefit on defecation frequency. The results suggest that consumption of BB-12® improves the GI health of individuals whose symptoms are not sufficiently severe to consult a doctor (ISRCTN18128385).