Level of evidence for promising subgroup findings: The case of trends and multiple subgroups.

Subgroup analyses are an essential part of fully understanding the complete results from confirmatory clinical trials. However, they come with substantial methodological challenges. In case no statistically significant overall treatment effect is found in a clinical trial, this does not necessarily indicate that no patients will benefit from treatment. Subgroup analyses could be conducted to investigate whether a treatment might still be beneficial for particular subgroups of patients. Assessment of the level of evidence associated with such subgroup findings is primordial as it may form the basis for performing a new clinical trial or even drawing the conclusion that a specific patient group could benefit from a new therapy. Previous research addressed the overall type I error and the power associated with a single subgroup finding for continuous outcomes and suitable replication strategies. The current study aims at investigating two scenarios as part of a nonconfirmatory strategy in a trial with dichotomous outcomes: (a) when a covariate of interest is represented by ordered subgroups, eg, in case of biomarkers, and thus, a trend can be studied that may reflect an underlying mechanism, and (b) when multiple covariates, and thus multiple subgroups, are investigated at the same time. Based on simulation studies, this paper assesses the credibility of subgroup findings in overall nonsignificant trials and provides practical recommendations for evaluating the strength of evidence of subgroup findings in these settings.

Estimates of subgroup treatment effects in overall nonsignificant trials: To what extent should we believe in them?

In drug development, it sometimes occurs that a new drug does not demonstrate effectiveness for the full study population but appears to be beneficial in a relevant subgroup. In case the subgroup of interest was not part of a confirmatory testing strategy, the inflation of the overall type I error is substantial and therefore such a subgroup analysis finding can only be seen as exploratory at best. To support such exploratory findings, an appropriate replication of the subgroup finding should be undertaken in a new trial. We should, however, be reasonably confident in the observed treatment effect size to be able to use this estimate in a replication trial in the subpopulation of interest. We were therefore interested in evaluating the bias of the estimate of the subgroup treatment effect, after selection based on significance for the subgroup in an overall "failed" trial. Different scenarios, involving continuous as well as dichotomous outcomes, were investigated via simulation studies. It is shown that the bias associated with subgroup findings in overall nonsignificant clinical trials is on average large and varies substantially across plausible scenarios. This renders the subgroup treatment estimate from the original trial of limited value to design the replication trial. An empirical Bayesian shrinkage method is suggested to minimize this overestimation. The proposed estimator appears to offer either a good or a conservative correction to the observed subgroup treatment effect hence provides a more reliable subgroup treatment effect estimate for adequate planning of future studies.

Ovulatory status and menstrual cycle duration assessed by self-collection of urine on pH strips in a population-based sample of French women not using hormonal contraception.

BACKGROUND: Assessing menstrual cycle function in the general population using a non-invasive method is challenging, both in non-industrialized and industrialized countries. SUBJECTS AND METHODS: The Observatory of Fecundity in France (Obseff) recruited on a nationwide basis a random sample of 943 women aged 18-44 years with unprotected intercourse. A sub-study was set up to assess the characteristics of a menstrual cycle by using a non-invasive method adapted to the general population. Voluntary women were sent a collection kit by the post and requested to collect urine samples on pH strips, together with daily recording of reproductive-related information during a full menstrual cycle. A total of 48 women collected urine every day, whereas 160 women collected urine every other day. Immunoassays were used to measure pregnanediol-3-α-glucuronide, estrone-3-glucuronide and creatinine. Ovulation occurrence and follicular phase duration were estimated using ovulation detection algorithms, compared to a gold standard consisting of three external experts in reproductive medicine. RESULTS: Every other day urine collection gave consistent results in terms of ovulation detection with every day collection (intraclass coefficient of correlation, 0.84, 95% confidence interval, 0.76-0.98). The proportion of anovulatory menstrual cycles was 8%. The characteristics of the ovulatory cycles were length 28 (26-34), follicular phase 16 (12-23), luteal phase 13 (10-16) days median (10th-90th percentiles). DISCUSSION-CONCLUSION: Assessing menstrual cycle characteristics based on urine sample spot only collected every other day in population-based studies through a non-invasive, well accepted and cost-limited procedure not requiring any direct contact with the survey team appears feasible and accurate.

Subgroup analyses in confirmatory clinical trials: time to be specific about their purposes.

BACKGROUND: It is well recognized that treatment effects may not be homogeneous across the study population. Subgroup analyses constitute a fundamental step in the assessment of evidence from confirmatory (Phase III) clinical trials, where conclusions for the overall study population might not hold. Subgroup analyses can have different and distinct purposes, requiring specific design and analysis solutions. It is relevant to evaluate methodological developments in subgroup analyses against these purposes to guide health care professionals and regulators as well as to identify gaps in current methodology. METHODS: We defined four purposes for subgroup analyses: (1) Investigate the consistency of treatment effects across subgroups of clinical importance, (2) Explore the treatment effect across different subgroups within an overall non-significant trial, (3) Evaluate safety profiles limited to one or a few subgroup(s), (4) Establish efficacy in the targeted subgroup when included in a confirmatory testing strategy of a single trial. We reviewed the methodology in line with this "purpose-based" framework. The review covered papers published between January 2005 and April 2015 and aimed to classify them in none, one or more of the aforementioned purposes. RESULTS: In total 1857 potentially eligible papers were identified. Forty-eight papers were selected and 20 additional relevant papers were identified from their references, leading to 68 papers in total. Nineteen were dedicated to purpose 1, 16 to purpose 4, one to purpose 2 and none to purpose 3. Seven papers were dedicated to more than one purpose, the 25 remaining could not be classified unambiguously. Purposes of the methods were often not specifically indicated, methods for subgroup analysis for safety purposes were almost absent and a multitude of diverse methods were developed for purpose (1). CONCLUSIONS: It is important that researchers developing methodology for subgroup analysis explicitly clarify the objectives of their methods in terms that can be understood from a patient's, health care provider's and/or regulator's perspective. A clear operational definition for consistency of treatment effects across subgroups is lacking, but is needed to improve the usability of subgroup analyses in this setting. Finally, methods to particularly explore benefit-risk systematically across subgroups need more research.

Impact of anthropometric data and physical activity level on the closed kinetic chain upper extremity stability test (CKCUEST) score: A cross-sectional study.

OBJECTIVE: The aim of this study was to evaluate the impact of anthropometric data and the level of physical activity on the CKCUEST score on uninjured subjects. METHOD: Subjects were eligible for the study if they were aged between 18 and 25 years old. Anthropometric data were collected, and the physical activity level was assessed using the Global Physical Activity Questionnaire (GPAQ). The primary endpoint was the CKCUEST score. The CKCUEST was performed three times and the score was the mean of the 3 tries. The compared groups were posteriori dichotomized according to the median. A multivariate model has been built by step-by-step selection with a threshold set at 0.05. RESULTS: 82 subjects were included. The difference between groups determined for each variable was statistically significant for CKCUEST score (p < 0.0001) and for physical activity level (p = 0.0245). The multivariate analysis showed that arm span, sex and weight are, in order, three main variables influencing the CKCUEST score. A predictive equation was established based on these 3 factors (R2 = 0.51). CONCLUSION: The CKCUEST score seems to be impacted by sex, weight and arm span. A predictive equation for the CKCUEST score was proposed: (0.271 × Upper Limb Span)-(0.103 × Weight)-(3.219 × Sex)-17.719.

European regulatory use and impact of subgroup evaluation in marketing authorisation applications.

Marketing authorisation application dossiers relating to medicinal products containing new active substances and evaluated by the European Medicines Agency (EMA) over the period 2012-2015 were examined. Major objections and other concerns relating to efficacy and safety of the day 80 assessment reports were reviewed. Overall, approved products have more subgroup concerns than nonapproved products, which seems to be a consistent pattern. Subgroup analyses are mainly assessed to have the insurance that subgroups of patients that might lack a positive benefit: risk ratio will not be wrongly included in the approved treatment indication.

Level of evidence for promising subgroup findings in an overall non-significant trial.

In drug development and drug licensing, it sometimes occurs that a new drug does not demonstrate effectiveness for the full study population, but there appears to be benefit in a relevant, pre-defined subgroup. This raises the question, how strong the evidence from such a subgroup is, and which confirmatory testing strategies are the most appropriate ones. Hence, we considered the type I error and the power of a subgroup result in a trial with non-significant overall results and of suitable replication strategies. In the case of a single trial, the inflation of the overall type I error is substantial and can be up to twice as large, especially in relatively small subgroups. This also increases to the risk of starting a replication trial that should not be done, if such a second trial is not already available. The overall type I error is almost controlled by using an appropriate replication strategy. This confirms the required cautious interpretation of promising subgroups, even in the case that overall trial results were perceived to be close to significance.

A model for the distribution of daily number of births in obstetric clinics based on a descriptive retrospective study.

OBJECTIVE: To test whether the relatively unpredictable nature of labour onset can be described by the Poisson distribution. DESIGN: A descriptive retrospective study. SETTING: From the Danish Birth Registry, we identified births from all seven obstetric clinics in the capital region of Denmark (n=211 290) between 2000 and the end of 2009. On each date, the number of births at each department was registered. Births are categorised based on whether an elective caesarean section or induction of labour has been performed, and among the remaining 'non-elective births', acute caesareans were registered. METHODS: After the exclusion of elective caesarean sections and births after induction of labour, only 'non-elective' births (n=171 009) were included for the main statistical analysis. Simple descriptive plots and one-way analysis of variance were used to analyse the distribution of 'non-elective' births for each day of the week. MAIN OUTCOME MEASURES: The daily number of 'non-elective' births. RESULTS: The number of 'non-elective' births varies considerably over the days of the week and over the year for each obstetric clinic regardless of clinic size. However, for each fixed day of the week, the variation over the year is well described by a Poisson distribution, allowing simple prediction of the variability. For births at each fixed day of the week, the Poisson distribution is indistinguishable from a normal distribution. CONCLUSIONS: The number of 'non-elective' births for each day of the week is well described by a Poisson distribution. Consequently, the Poisson model is suitable for estimating the variation in the daily number of 'non-elective' births and could be used for planning of staffing in obstetric clinics. The model can be used in smaller as well as larger clinics.