Generalizability Of Trials Research Articles

Abstract 4006: Effect of Prior Cancer on Trial Eligibility and Treatment Outcomes in Nasopharyngeal Carcinoma: Implications for Clinical Trial Accrual

Abstract Background: In cancer trials, prior cancer is a common exclusion criterion. We evaluated the characteristics of prior cancer exclusion criteria in nasopharyngeal carcinoma (NPC) trials and determined its prognostic effect on patients with NPC. Methods: We reviewed NPC trials for prior cancer exclusion criteria. Then we estimated the effect of prior cancer among NPC patients using the Surveillance, Epidemiology, and End Results database. Propensity score-matching was used to compensate for differences in baseline characteristics between patients with and without prior cancer. Results: There were 109 clinical trials involving 10,437 patients; 49 trials (45%) excluded patients with prior cancer. Prior cancer exclusion was more common in recent or phase III trials. We identified 10,195 NPC patients; 6.2% had prior cancer. More than 70% of these cancers were in situ/localized/regional and diagnosed relatively close to the NPC diagnosis (median 3.3 years). Patients with certain prior cancer type (prostate, breast, gynecological, hematological), time of diagnosis (>5 years ago), or stage (in situ/localized) did not have inferior survival compared with patients with no prior cancer. We tested one form of prior cancer exclusion criteria in an NPC cohort resembling a modern trial population: it did not adversely affect overall and NPC-specific survival. Conclusions: Many NPC trials excluded patients with prior cancer, which impacts trial accrual and generalizability. Our findings suggest that broader inclusion in trials of patients with NPC with prior cancer might not affect trial outcomes. More research is needed to understand the appropriateness of this exclusion policy across cancer types and trials. Citation Format: Ya-Qin Wang, Jia-Wei Lv, Ling-Long Tang, Xiao-Jing Du, Lei Chen, Wen-Fei Li, Xu Liu, Ying Guo, Ai-Hua Lin, Yan-Ping Mao, Ying Sun, Yu-Pei Chen, Jun Ma. Effect of Prior Cancer on Trial Eligibility and Treatment Outcomes in Nasopharyngeal Carcinoma: Implications for Clinical Trial Accrual [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4006.

Do clinical trials truly mirror their target population? An external validity analysis of national register versus trial data from the Swedish prospective SENOMIC trial on sentinel node micrometastases in breast cancer

PurposeIncreasing evidence suggests that completion axillary lymph node dissection (ALND) may be omitted in breast cancer patients with limited axillary nodal metastases. However, the representativeness of trial participants for the original clinical practice population, and thus, the generalizability of published trials have been questioned. We propose the use of background data from national registers as a means to assess whether trial participants mirror their target population and to strengthen the generalizability and implementation of trial outcomes.MethodsThe Swedish prospective SENOMIC trial, omitting a completion ALND in breast cancer patients with sentinel lymph node micrometastases, reached full target accrual in 2017. To assess the generalizability of trial results for the target population, a comparative analysis of trial participants versus cases reported to the Swedish National Breast Cancer Register (NKBC) was performed.ResultsComparing 548 trial participants and 1070 NKBC cases, there were no significant differences in age, tumor characteristics, breast surgery, or adjuvant treatment. Only the mean number of sentinel lymph nodes with micrometastasis per individual was lower in trial participants than in register cases (1.06 vs. 1.09, p = 0.037).ConclusionsPatients included in the SENOMIC trial are acceptably representative of the Swedish breast cancer target population. There were some minor divergences between trial participants and the NKBC population, but taking these into consideration, upcoming trial outcomes should be generalizable to breast cancer patients with micrometastases in their sentinel lymph node biopsy.

Sex-Based Differences in Heart Failure Across the Ejection Fraction Spectrum: Phenotyping, and Prognostic and Therapeutic Implications

ObjectivesThis study assessed sex-related differences in a large cohort of unselected patients with heart failure (HF) across the ejection fraction (EF) spectrum. BackgroundFemales are under-represented in randomized clinical trials. Potential sex-related differences in HF may question the generalizability of trials. MethodsIn the Swedish Heart Failure Registry population multivariate Cox and logistic regression models were fitted to investigate differences in prognosis, prognostic predictors, and treatments across males and females. ResultsOf 42,987 patients, 37% were females (55% with HF with preserved EF [HFpEF], 39% with HF with mid-range EF [HFmrEF], and 29% with HF with reduced EF [HFrEF]). Females were older and more symptomatic and more likely to have hypertension and kidney disease but less likely to have diabetes and ischemic heart disease. After adjustments, females were more likely to use beta-blockers and digoxin but less likely to receive HF device therapy. Crude mortality/HF hospitalization rates for HFpEF (hazard ratio [HR]: 1.16) and HFmrEF (HR: 1.14) were significantly higher in females but lower in females with HFrEF (HR: 0.95). After adjustments, the risk was significantly lower in females regardless of EF (HR: 0.80 in HFrEF, HR: 0.91 in HFmrEF, and HR: 0.93 in HFpEF). The main sex-related differences in prognostic predictors concerned diabetes in HFrEF and anemia in HFmrEF. ConclusionsMales and females with HF showed different characteristics across the EF spectrum. Males reported a lower crude risk of mortality/morbidity in HFpEF and HFmrEF but higher risk of HFrEF, although after adjustments, prognosis was better in females regardless of EF. The observed sex-related differences highlight the need for an adequate representation of females in HF randomized controlled trials to improve generalizability.

High risk patients with diffuse large B cell lymphoma are not enrolled on clinical trials.

e19058 Background: Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma. In this study, we explore reasons for non-enrollment in clinical trials for DLBCL and implications on trial design and interpretation. Methods: This is a retrospective analysis of patients (pts) with a pathological diagnosis of DLBCL or high grade B-cell lymphoma (HGBL) at University of Rochester (4/14-6/16) and New York-Presbyterian Hospital/Weill Cornell Medicine (NYP/WCM) (4/14-4/17).Ten clinical trials were opened during this time. Participants were divided into 3 groups: those treated in trial, those not enrolled in trial because of need for urgent treatment, and those not enrolled in trial for any other reason. We used a center-stratified Cox proportional hazards model to estimate association of trial enrollment with progression-free survival (PFS; time from start of treatment until progression/death or the last date the pt was known to be progression free) and overall survival (OS). Results: We identified 263 pts; 17% (n = 45) enrolled in a trial. Reasons for non-enrollment included not meeting eligibility criteria (n = 98), physician choice (n = 50), and pt choice (n = 38). For 32 pts, reasons were unclear. Of the 50 pts who were not enrolled because of physician choice, the primary reason for non-enrollment was the need for urgent treatment (n = 46). Pts who needed urgent treatment had higher risk clinical features compared with pts in trial (Table). Compared with those treated in trial and those not enrolled in trial for any other reason, those not enrolled in trial due to need for urgent treatment had an inferior PFS (HR 2.61, 95% CI 1.23–5.16) and OS (HR 2.27, 95% CI 1.21–4.06). Conclusions: At 2 academic institutions, 52% of patients with DLBCL or HGBL required urgent chemotherapy and failed to enroll on trials. Exclusion of such patients limits the applicability and generalizability of clinical trials in DLBCL. This barrier must be overcome so clinical trials may better reflect true DLBCL demographics. [Table: see text]

FDA analysis: Demographic trends in bladder and renal cancer trials.

e18144 Background: Many publications report under representation of minorities in certain subgroups, which may limit the generalizability of clinical trial (CT) results. This analysis, investigates and reports enrollment trends in CTs submitted between 2006-2017 in support of marketing applications for drugs indicated for the treatment of urothelial (UC) and renal cancer carcinoma (RCC), and compares them to incidence rates of these diseases by Surveillance, Epidemiology, and End Results (SEER) registry and the US census bureau. Methods: We identified all marketing applications for the treatment of UC and RCC that provided the primary evidence of safety and efficacy and aggregated the demographic data across trials and disease. Using these two pooled datasets, we compared the patient proportions enrolled in each of the race, sex and age categories to the corresponding rates in US cancer population estimated based on the corresponding incidence rates reported by SEER and the US census bureau using a Chi-squared test. Results: The pooled seven UC and 14 RCC CTs provided 2035 and 6757 patients respectively. The results are summarized below for the 939 (46%) UC and 1489 (22%) RCC patients enrolled in the US. Conclusions: Our findings indicate that majority of the patients were enrolled outside of the US. There were lower proportion of Black patients (4% vs 8%), older patients, age ≥ 75 years (30% vs 48%) and males (74% vs 80%) enrolled in UC population in the US. Higher proportions were observed in both White (89% vs 85%) and Asian (4% vs 2%) patients in UC and in White (90% vs 79%) patients in RCC.[Table: see text]

External Generalizability of Clinical Trials to Developing Countries

External Generalizability of Clinical Trials to Developing Countries

Pragmaticism of Randomized Controlled Trials of Biologic Treatment With Methotrexate in Rheumatoid Arthritis: ASystematic Review.

Randomized controlled trials (RCTs) exist along a spectrum, from explanatory, designed to evaluate interventions under ideal conditions, to pragmatic, designed to reflect usual care. This study assessed the pragmatism of RCTs of advanced therapeutics in rheumatoid arthritis (RA). A systematic review was conducted to identify RA RCTs comparing biologic or targeted synthetic therapy in combination with methotrexate, to placebo or any other disease-modifying antirheumatic drugs (DMARDs). Trials were rated using the Pragmatic Explanatory Continuum Indicator Summary-2 (PRECIS-2) tool in 9 domains, each rated from 1 (very explanatory) to 5 (very pragmatic). Latent class and regression analyses examined the heterogeneity in PRECIS-2 scores and the relationship to trial characteristics. In total, 96 trials were included. Eligibility, follow-up, and flexibility of delivery of the intervention were rated as explanatory, with mean ± SD PRECIS-2 scores of 2.0 ± 0.7, 2.0 ± 1.1, and 2.1 ± 0.7, respectively, reflecting strict inclusion criteria, intensive follow-up, and rigid protocols. Studies were rated as pragmatic in setting (3.6 ± 1.5) because many were international, multicenter trials, and in primary analysis (4.1 ± 1.3), because most used intent-to-treat analyses. In latent class analyses, 2 groups were identified; the majority (74%) were rated as explanatory for most domains assessed. These trials had larger sample sizes, were more likely to be industry-funded, and enrolled patients with higher Disease Activity Score in 28 joints and Health Assessment Questionnaire disability index scores, but were less likely to be at high risk of bias. RCTs of biologic DMARD treatment in combination with methotrexate for RA were rated as predominantly explanatory, which may affect the generalizability of trial results to clinical practice.

Geographic Trends in Clinical Trials in Lymphoma, 1997-2017

BackgroundIn order to fulfil their medical and academic mission, clinical trials have to be available to the very populations they are intended to serve. A broad and homogenous geographic distribution is thus paramount to ensure patient fairness as well as to allow optimal generalizability of trial results to real-world populations. We aimed to investigate trends in the geographic distribution of clinical trials in lymphoma performed over the past 2 decades.MethodsWe systematically searched the Clinicaltrials.gov registry for phase 3 and 4 clinical trials in lymphoma conducted between 01/01/1997 and 12/31/2017. We divided the time period into two intervals and analysed the geographic distribution of trials during these time periods. The Gini coefficient, a statistical measure of inequality ranging from 0 (equal distribution) to 1 (most unequal distribution), was calculated for each period.ResultsWe found 633 clinical trials in lymphoma conducted between 1997 and 2017 globally. 284 and 349 trials were performed in the 1997-2007 and 2008-2017 time periods, respectively. 127 (50%) and 138 (39.5%) of the trials were done in the US. Hodgkin Lymphoma accounted for 55 (19.3%) and 42 (12.0%) studies in the two time periods. CA, TX, IL and NY were consistently in the upper one-fifth of trial-abundant states (each accounting for 3.5-4.5% of all trials). AK, WY, MT and RI were consistently in the lower one-fifth of trial-abundant states (each accounting for 0.13-0.84% of trials). The trial participation of NY and AK mildly increased during the time period while that of TX and RI decreased. In each of the time periods, the 12 most trial-intense states accounted for 41.5% of all investigations. The 12 least trial-intense states were responsible for 8.8% of all the trials. The extent of geographic inequality increased over the years with the Gini coefficient rising from 0.246 to 0.3 between the first and second study period. The geographical pattern of state-specific trial intensity was strongly mitigated after adjusting for each state's population size.ConclusionsThe geographic distribution of clinical trials in lymphoma in the U.S is not uniform. Our data suggests that the trend of geographic mal-distribution is worsening, a phenomenon that may lead to deleterious public health outcomes. As patients' enrolment continues to be linked to geographic factors, patients' ability to enter clinical trials may be limited and the representation of real-world participants may be skewed (towards particular zip codes, socio-economical status and education levels, etc), consequently affecting the yield and generalizability of clinical trials. Innovative solutions must be devised to minimise any barriers to trial participation for suitable patients. DisclosuresNo relevant conflicts of interest to declare.

A data-zone scoring system to assess the generalizability of clinical trial results to individual patients.

Evaluating the applicability of a clinical trial to a specific patient is difficult. A novel framework, the Trial Score, was created to quantify the generalizability of a trial's result based on participants' baseline characteristics and not on the trial's inclusion and exclusion criteria. For each Systolic Blood Pressure Intervention Trial (SPRINT) participant, the Euclidean distance in six-dimensional space from the theoretical "average" participant was calculated to produce an individual Trial Score that incorporates multiple distinct continuous-variable baseline characteristics. We prospectively defined the "data-rich," "data-limited," and "data-free" zones as Trial Scores < 90th percentile, the 90th-97.5th percentile, and >97.5th percentile, respectively. Trial Scores were then calculated for National Health and Nutrition Examination Survey participants to map data zones of the general population. Individual participant data from the Action to Control Cardiovascular Risk in Diabetes blood pressure trial (ACCORD-BP) was used to test if participants further from the average SPRINT participant behave differently than the overall SPRINT results. The National Health and Nutrition Examination Survey cohort and the ACCORD-BP trial demonstrate large percentages of participants in SPRINT's data-free and data-limited zones. Time-to-event rates seen with intensive and standard blood pressure control in SPRINT were the same as ACCORD-BP participants within SPRINT's data-rich zone (hazard ratio 0.97, p = 0.84 and hazard ratio 0.95, p = 0.70). However, these rates were significantly different than those of ACCORD-BP participants outside SPRINT's data-rich zone (hazard ratio 0.64, p < 0.01 and hazard ratio 0.77, p < 0.01). ACCORD-BP participants with SPRINT Trial Scores in the 90th percentile or below have similar event rates to SPRINT participants in both the intensive and standard blood pressure groups. Quantifying the difference between an individual patient and the average clinical trial participant holds promise as a tool to more precisely determine applicability of a specific trial to individual patients.

Challenges of conducting clinical trials in Asia

&lt;p class="abstract"&gt;Unlike North America and European Union (EU), Asian continent appears to be an ideal destination for conducting cost-effective clinical trials utilizing the pool of treatment naïve subjects. With a population approaching 4.5 billion, recruitment of subjects can be done without a fear of limited patient pool across Asia. The burden of infectious and chronic diseases is also higher in Asian countries. The emerging clinical trial markets particularly China, South Korea, and Taiwan offers genetic diversity in population group, thus promoting the quality and generalizability of clinical trial’s data. Nonetheless, several challenges also exist for the Western sponsors in majority of the Asian countries; regulatory, operational and infrastructural challenges are at the forefront. Challenges under the heads regulatory, operational, infrastructural, language and cultural, ethics, and future challenges have been discussed. SWOT analysis of Asian clinical trial’s market exhibits enormous opportunities for study sponsors with manageable threats.&lt;/p&gt;

Comparison of Methods to Generalize Randomized Clinical Trial Results Without Individual-Level Data for the Target Population.

Our study explored the application of methods to generalize randomized controlled trial results to a target population without individual-level data. We compared 4 methods using aggregate data for the target population to generalize results from the international trial, Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER), to a target population of trial-eligible patients in the UK Clinical Practice Research Datalink (CPRD). The gold-standard method used individual data from both the trial and CPRD to predict probabilities of being sampled in the trial and to reweight trial participants to reflect CPRD patient characteristics. Methods 1 and 2 used weighting methods based on simulated individual data or the method of moments, respectively. Method 3 weighted the trial's subgroup-specific treatment effects to match the distribution of an effect modifier in CPRD. Method 4 calculated the expected absolute benefits in CPRD assuming homogeneous relative treatment effect. Methods based on aggregate data for the target population generally yielded results between the trial and gold-standard estimates. Methods 1 and 2 yielded estimates closest to the gold-standard estimates when continuous effect modifiers were represented as categorical variables. Although individual data or data on joint distributions remains the best approach to generalize trial results, these methods using aggregate data might be useful tools for timely assessment of randomized trial generalizability.

Pilot and feasibility studies in exercise, physical activity, or rehabilitation research

BackgroundClinical trials of physical activity and rehabilitation interventions can be challenging. Pilot or feasibility studies can be conducted prior to a definitive randomized controlled trial (RCT), to improve the chances of conducting a high-quality RCT of a physical activity intervention.Main bodyPhysical activity interventions or trials present unique challenges at the population, intervention, comparator and outcome levels. At each level, we present guidance for researchers on the design considerations for pilot or feasibility studies of physical activity interventions. When it comes to defining study population, physical activity trials often exclude participants with certain health conditions or other characteristics (e.g., age, gender) because of uncertainty of the safety of the exercise intervention or presumed differences in responsiveness, at the expense of trial generalizability. A pilot trial could help investigators determine refined inclusion and exclusion criteria to balance safety, adequate recruitment, and generalizability. At the intervention level, because exercise can be a complex intervention, pilot trials allow investigators to evaluate participant adherence and instructor fidelity to the intervention and participant experience. At the comparator level, control group dissatisfaction and post-randomization drop-out can occur, because of the desire to be randomized to the exercise group, and the difficulty with blinding to group allocation; an active control or deception could be used. Finally, at the outcome level, there should be an emphasis on the pilot or feasibility outcomes such as recruitment rate, adherence to exercise, and retention or fidelity, than the efficacy of the exercise intervention.ConclusionPhysical activity and rehabilitation researchers can use pilot and feasibility studies to enhance the rigor of future trials, while also publishing the results of their pilot work to move the field forward. Researchers in this field are encouraged to use published reporting guidelines for pilot and feasibility studies and to consider the challenges discussed in this paper.

Generalizability of clinical trials of advanced melanoma in the real-world, population-based setting

Results from novel therapeutics trials are not always generalizable to real-world patients. We aimed to determine the pattern in which trial findings are applied in a population-based setting of melanoma patients and consequent treatment outcomes. Patients with unresectable disease during 2011-2014 and referred to cancer centers in a large Canadian province were retrospectively reviewed. Based on eligibility criteria as described in registration trials of vemurafenib (Vem) and ipilimumab (Ipi), we classified patients into trial-eligible and ineligible and those treated and untreated with these agents. We identified 290 patients with known BRAF status for the Vem analysis and 212 patients previously treated with first-line agents for the Ipi analysis. For the Vem cohort, a total of 49 patients were considered trial-eligible, of whom 36 (73%) received treatment. For the Ipi cohort, there were 119 trial-eligible cases of whom 43 (36%) received therapy. Factors other than eligibility criteria most frequently associated with non-treatment in these cohorts included concerns regarding treatment harm and patient preferences. In multivariable analysis, overall survival was improved in Vem cohort patients considered trial-eligible and treated compared to those who were ineligible. Within the Ipi cohort, survival was improved in trial-eligible patients regardless of whether they received Ipi compared to ineligible patients. Real-world uptake of new melanoma treatments was suboptimal, and non-use in trial-eligible patients was frequent. Future clinical trials that are more pragmatically designed to include participants who better reflect the real-world population may facilitate increased uptake of novel therapeutics into routine clinical practice.

An evaluation of race, ethnicity, age, and sex-based representation in phase I to II renal cell carcinoma clinical trials in the United States

IntroductionEnrollment of a representative study population permits generalizable and reliable results for clinical trials. We sought to evaluate whether patients enrolled in trials for advanced renal cell carcinoma (RCC) are representative of the overall population of advanced RCC patients in the United States. Materials and methodsThe clinicaltrials.gov results database was queried for interventional clinical trials directed at clinically advanced (stage III/IV) RCC that enrolled patients from the US only. We identified 375 patients from 18 phase I to II trials that met eligibility criteria. The American College of Surgeons′ National Cancer Database (NCDB) which includes data on approximately 70% of all US cancer diagnoses was queried and we identified 75,308 patients with advanced (stage III/IV) RCC. Demographic characteristics were summarized and compared between the 2 populations. ResultsCompared to the US population of advanced RCC (NCDB), significant under-representation in clinical trials was observed for patients aged 65+ (26.3% vs. 50.4%; P<0.001) and among those with Hispanic ethnicity (2.7% vs. 7.2%; P = 0.005). A trend toward under-representation was observed for black patients (7.0% vs. 9.8%, P = 0.076) but not for white patients (89.9% vs. 87.0%, P = 0.107) or other racial groups (P>0.05 for all). Female patients made up 30.3% of trial enrollees and 33.3% of the US advanced RCC population (P = 0.221). ConclusionSignificant under-representation was observed for elderly and Hispanic patients with a trend toward under-representation for black and female patients in phase I to II RCC clinical trials. Greater efforts to include underrepresented populations are necessary to improve the effectiveness and generalizability of clinical trials in kidney cancer.

DO COMPASS TRIAL PARTICIPANTS REPRESENT PATIENTS IN EVERYDAY PRACTICE?

In the COMPASS trial, the addition of rivaroxaban to aspirin reduced cardiovascular events but increased major bleeding in patients with coronary artery disease (CAD) or peripheral arterial disease (PAD). We aimed to assess the trial generalizability by investigating the prevalence and risk profile

GENERALIZABILITY OF THE FOURIER TRIAL TO ROUTINE CLINICAL CARE: DO TRIAL PARTICIPANTS REPRESENT PATIENTS IN EVERYDAY PRACTICE?

In the FOURIER trial, evolocumab, a proprotein convertase subtilisin-kexin type 9 inhibitor (PCSK9i), reduced cardiovascular events when added to statins in patients with atherosclerotic cardiovascular disease (ASCVD). We aimed to assess the trial generalizability by investigating the prevalence and

Application and impact of run-in studies.

A run-in phase is often employed prior to randomization in a clinical trial to exclude non-adherent patients, placebo responders, active drug non-responders, or patients who do not tolerate the active drug. This may impact the generalizability of trial results. To determine if clinical outcomes differed between randomized controlled trials with run-in phases compared with randomized controlled trials of the same medication without run-in phases. From 2006 to 2014, the Food and Drug Administration approved 258 new medications. Sitaglitpin, saxagliptin, linagliptin, and alogliptin were among the only drugs with a common mechanism of action that each had multiple clinical trials, some of which had run-in phases and some of which did not. We identified all published randomized controlled trials for these four medications from MEDLINE and EMBASE as well as prior systematic reviews. We extracted key measures of medication efficacy (reduction in hemoglobin A1C) and safety (serious adverse events) from qualifying trials. Study results were pooled for each medication using random effects meta-analysis. We identified 106 qualifying trials for DPP4 inhibitors, of which 88 had run-in phases and 18 did not. The average run-in phase duration was 4.0weeks (range 1-21), and 73% of run-in phases administered placebo rather than active drug. The reduction in hemoglobin A1C compared to baseline was similar for trials with and without run-in phases (0.70%, 95% confidence interval [CI] 0.65-0.75 vs 0.76%, 95% CI 0.69-0.84, p = 0.27). The proportion of patients with serious adverse events was also similar for trials with and without run-in phases (4%, 95% CI: 3-5% vs 3%, 95% CI: 1-4%, p = 0.35). Trials with run-in phases provided similar estimates for medication efficacy and safety compared to trials without run-in phases. Because run-in phases are costly and time-consuming, these results call their utility into question for clinical trials of short duration.

Generalizability of Clinical Trial Results for Adolescent Major Depressive Disorder.

Although there have been a number of clinical trials evaluating treatments for adolescents with major depressive disorder (MDD), the generalizability of those trials to samples of depressed adolescents who present for routine clinical care is unknown. Examining the generalizability of clinical trials of pharmacological and psychotherapy interventions for adolescent depression can help administrators and frontline practitioners determine the relevance of these studies for their patients and may also guide eligibility criteria for future clinical trials in this clinical population. Data on nationally representative adolescents were derived from the National Comorbidity Survey: Adolescent Supplement. To assess the generalizability of adolescent clinical trials for MDD, we applied a standard set of eligibility criteria representative of clinical trials to all adolescents in the National Comorbidity Survey: Adolescent Supplement with a Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnosis of MDD (N = 592). From the overall MDD sample, 61.9% would have been excluded from a typical pharmacological trial, whereas 42.2% would have been excluded from a psychotherapy trial. Among those who sought treatment (n = 412), the corresponding exclusion rates were 72.7% for a pharmacological trial and 52.2% for a psychotherapy trial. The criterion leading to the largest number of exclusions was "significant risk of suicide" in both pharmacological and psychotherapy trials. Pharmacological and, to a lesser extent, psychotherapy clinical trials likely exclude most adolescents with MDD. Careful consideration should be given to balancing eligibility criteria and internal validity with applicability in routine clinical care while ensuring patient safety.

Estimation of Population Average Treatment Effects in the FIRST Trial: Application of a Propensity Score-Based Stratification Approach.

To estimate and compare sample average treatment effects (SATE) and population average treatment effects (PATE) of a resident duty hour policy change on patient and resident outcomes using data from the Flexibility in Duty Hour Requirements for Surgical Trainees Trial ("FIRST Trial"). Secondary data from the National Surgical Quality Improvement Program and the FIRST Trial (2014-2015). The FIRST Trial was a cluster-randomized pragmatic noninferiority trial designed to evaluate the effects of a resident work hour policy change to permit greater flexibility in scheduling on patient and resident outcomes. We estimated hierarchical logistic regression models to estimate the SATE of a policy change on outcomes within an intent-to-treat framework. Propensity score-based poststratification was used to estimate PATE. This study was a secondary analysis of previously collected data. Although SATE estimates suggested noninferiority of outcomes under flexible duty hour policy versus standard policy, the noninferiority of a policy change was inconclusively noninferior based on PATE estimates due to imprecision. Propensity score-based poststratification can be valuable tools to address trial generalizability but may yield imprecise estimates of PATE when sparse strata exist.

Generalizing Randomized Clinical Trial Results: Implementation and Challenges Related to Missing Data in the Target Population

Statins are indicated in patients with elevated levels of high-sensitivity C-reactive protein and normal low-density lipoprotein cholesterol based on results of the multicountry trial, Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) (2003-2008), but the benefit in real-world populations remains unknown. We sought to generalize JUPITER results to trial-eligible population using data from the UK Clinical Practice Research Datalink (CPRD), 2001-2014. We multiply imputed missing baseline characteristics for the CPRD population and selected the trial-eligible participants as the target population based on observed and imputed values. Trial participants were weighted to be representative of the CPRD population (n = 383,418) based on individual predicted probability of selection into the trial. Trial participants were also standardized to the CPRD population without missing values (n = 2,677). In JUPITER, rosuvastatin reduced cardiovascular risk with a 3-year risk difference of -2.0% (95% confidence interval (CI): -2.9, -1.1). The rosuvastatin effect was muted in the first 2 years but remained strong at 3 years after standardizing to the imputed CPRD population (3-year risk difference = -2.7%; 95% CI: -5.8, 0.4) and the CPRD population without missing data (3-year risk difference = -1.7%; 95% CI: -3.5, 0.1). The study serves as an illustration of possible approaches to understanding generalizability of trials using real-world databases given limitations due to missing data on inclusion/exclusion criteria.