Debate: Subgroup analyses in clinical trials: fun to look at - but don't believe them!

Reporting of Demographics & Subgroup Analyses in Premarketing Studies of FDA Approved High-Risk Cardiovascular Devices, 2014-2022

Background

Representation of diverse study populations in pivotal clinical trials for medical devices and subgroup analyses for demographic groups to explore differences in safety and effectiveness are essential to understanding the benefits and risks in diverse populations. The US Food and Drug Administration (FDA) has taken many steps to improve transparency and subgroup analyses over the past decade, but there has not been a recent evaluation of demographic reporting and subgroup analyses.

Methods

We reviewed all FDA Premarket Approvals for high-risk cardiovascular devices from 2014 to 2022, focusing on pivotal studies supporting device approval. We abstracted detailed demographic data about the age, sex, race, ethnicity, and socioeconomic position of study participants. We also assessed the presence and results of subgroup analyses to understand the safety and effectiveness of devices across trial populations.

Results

Analysis of 92 pivotal studies revealed that age and sex were reported in 96.7% of the studies, while race and ethnicity were reported in 71.7% and 58.7%, respectively. However, only 7.9% of studies explicitly detailed the participation of older adults (≥65 years) and no studies reported patients' socioeconomic position. Subgroup analyses by sex were conducted in 70.7% of studies, with 12.3% reporting significant differences. In contrast, analyses by race and ethnicity were performed in only 12.0% of the studies, with 9.1% reporting significant differences.

Conclusion

Approximately one-third of pivotal studies for high-risk cardiovascular devices approved by the FDA from 2014 to 2022 did not report the race of study participants, nearly 40% did not report ethnicity, and more than 90% did not report the participation of older adults (≥65 years). Subgroup analyses were infrequently conducted by age or race and ethnicity. There is a need for better trial demographic reporting and conduct of subgroup analyses in premarketing studies to ensure the safety and effectiveness of medical devices for all patients.

The credibility of subgroup analyses reported in stroke trials is low: A systematic review

BACKGROUND

Subgroup analyses are widely used to evaluate the heterogeneity of treatment effects in randomized clinical trials. However, there is a limited investigation of the quality of prespecified and reported subgroup analyses in stroke trials. This study evaluated the credibility of subgroup analyses in stroke trials.

METHODS AND ANALYSIS

We searched Medline/PubMed, Embase, the Cochrane Central Register of Controlled Trials, and the Web of Science from inception to 24 March 2021. Three reviewers screened, extracted, and analyzed the data from the publications. Primary publications of stroke trials that reported at least one subgroup effect and had published corresponding study protocols were included. The Instrument for Assessing the Credibility of Effect Modification Analyses (ICEMAN) was used to examine the quality of the subgroup effects reported, with each subgroup effect assigned a credibility rating ranging from very low to high. Subgroup effects with two or more "definitely no" responses received a low credibility rating. The risk of bias was assessed using the Cochrane Risk-of-Bias tool for randomized trials version 2.

RESULTS

Seventy-four articles met the inclusion criteria and reported a combined total of 647 subgroup effects. The median sample size was 1264 (interquartile range (IQR): 380-3876), and the median number of subgroups prespecified in the protocol was 6 (IQR: 2-10). Sixty-one (82%) studies used the univariate test of interaction. Of the total 647 subgroup effects reported in these studies, 319 (49%) were reported in acute stroke trials, while 423 (65%) had low credibility.

CONCLUSION

The quality of subgroup analysis reporting in stroke trials remains poor. More effort is needed to train trialists on the best methods for designing and performing subgroup analyses, and how to report the results.

TRIAL REGISTRATION NUMBER

We prospectively registered the review with International Prospective Register for Systematic Reviews (registration number: CRD42020223133).

Dyslipidemia in Patients with Chronic Kidney Disease: An Updated Overview

Dyslipidemia is a potentially modifiable cardiovascular risk factor. Whereas the recommendations for the treatment target of dyslipidemia in the general population are being more and more rigorous, the 2013 Kidney Disease: Improving Global Outcomes clinical practice guideline for lipid management in chronic kidney disease (CKD) presented a relatively conservative approach with respect to the indication of lipid lowering therapy and therapeutic monitoring among the patients with CKD. This may be largely attributed to the lack of high-quality evidence derived from CKD population, among whom the overall feature of dyslipidemia is considerably distinctive to that of general population. In this review article, we cover the characteristic features of dyslipidemia and impact of dyslipidemia on cardiovascular outcomes in patients with CKD. We also review the current evidence on lipid lowering therapy to modify the risk of cardiovascular events in this population. We finally discuss the association between dyslipidemia and CKD progression and the potential strategy to delay the progression of CKD in relation to lipid lowering therapy.

A shrinkage estimator for subgroup analysis without the exchangeability assumption

Shrinkage estimators for exploratory subgroup analyses are intuitively appealing and can greatly improve estimation over standard analysis approaches; however, adoption of these estimators has been limited by reliance on the exchangeability assumption. This paper describes a new shrinkage estimator that does not rely on this assumption. Rather than assuming that treatment effect sizes within subgroups are randomly distributed around an overall mean, this new estimator assumes that the difference between the effect sizes in any given pair of subgroups is randomly distributed around zero. The estimator is illustrated using data from a clinical trial in which the treatment effect size in one region was substantially different from the sizes in other regions. Simulation results show that the estimator has properties that are comparable to or superior to a standard shrinkage estimator when exchangeability is assumed, while allowing the flexibility to handle situations where exchangeability cannot be assumed.

Efficacy, Safety, and Immunogenicity of Insulin Aspart Biosimilar SAR341402 Compared with Originator Insulin Aspart in Adults with Diabetes (GEMELLI 1): A Subgroup Analysis by Prior Type of Mealtime Insulin

INTRODUCTION

The biosimilar SAR341402 insulin aspart (SAR-Asp) was compared to its originator NovoLog®/NovoRapid® insulin aspart (NN-Asp) in terms of efficacy, safety, and immunogenicity, in adults with type 1 or type 2 diabetes switching from different rapid-acting insulin analogs.

METHODS

This phase 3, randomized, open-label, multinational, 52-week study (GEMELLI 1) enrolled participants with type 1 or type 2 diabetes (n = 597). At randomization, participants transitioned from NovoLog/NovoRapid (n = 380) or Humalog®/Liprolog® (n = 217) to equivalent (1:1) doses (or a dose at the discretion of the investigator) of either SAR-Asp or NN-Asp (1:1 randomization). Participants were treated with multiple daily injections in combination with insulin glargine 100 U/mL (Lantus®). In this subgroup analysis, efficacy measures (change in hemoglobin A1c [HbA1c], insulin dose [total, basal and mealtime]), and safety outcomes (hypoglycemia incidence, adverse events, anti-insulin aspart antibodies) of SAR-Asp were compared with those of NN-Asp separately according to the participants' prestudy mealtime insulin.

RESULTS

At week 26 (primary efficacy endpoint), change in HbA1c was similar between SAR-Asp and NN-Asp in those participants pre-treated with NovoLog/NovoRapid (least squares [LS] mean difference - 0.04%, 95% confidence interval [CI] - 0.182 to 0.106%) or Humalog/Liprolog (LS mean difference - 0.15%, 95% CI - 0.336 to 0.043%) (P value for treatment by subgroup interaction = 0.36). This HbA1c response persisted over the 52 weeks of the study similarly for both treatments within each subgroup. In both subgroups, changes in insulin doses were similar between treatments over 26 weeks and 52 weeks, as were the incidences of severe or any hypoglycemia, adverse events (including hypersensitivity and injection site reactions), and anti-insulin aspart antibodies.

CONCLUSIONS

Efficacy and safety (including immunogenicity) profiles of SAR-Asp are similar to those of NN-Asp over 52 weeks in adults with diabetes irrespective of prior type of mealtime insulin.

TRIAL REGISTRATION

ClinicalTrials.gov identifier: NCT03211858.

A systematic review of subgroup analyses in randomised clinical trials in cardiovascular disease

Background:

Subgroup analyses are frequently used to assess heterogeneity of treatment effects in randomised clinical trials. Inconsistent, improper and incomplete implementation, reporting and interpretation have been identified as ongoing challenges. Further, subgroup analyses were frequently criticised because of unreliable or potentially misleading results. More recently, recommendations and guidelines have been provided to improve the reporting of data in this regard.

Methods:

This systematic review was based on a literature search within the digital archives of three selected medical journals, The New England Journal of Medicine, The Lancet and Circulation. We reviewed articles of randomised clinical trials in the domain of cardiovascular disease which were published in 2015 and 2016. We screened and evaluated the selected articles for the mode of implementation and reporting of subgroup analyses.

Results:

We were able to identify a total of 130 eligible publications of randomised clinical trials. In 89/130 (68%) articles, results of at least one subgroup analysis were presented. This was dependent on the considered journal (p < 0.001), the number of included patients (p < 0.001) and the lack of statistical significance of a trial’s primary analysis (p < 0.001). The number of reported subgroup analyses ranged from 1 to 101 (median = 13). We were able to comprehend the specification time of reported subgroup analyses for 71/89 (80%) articles, with 55/89 (62%) articles presenting exclusively pre-specified analyses. This information was not always traceable on the basis of provided trial protocols and often did not include the pre-definition of cut-off values for the categorization of subgroups. The use of interaction tests was reported in 84/89 (94%) articles, with 36/89 (40%) articles reporting heterogeneity of the treatment effect for at least one primary or secondary trial outcome. Subgroup analyses were reported more frequently for larger randomised clinical trials, and if primary analyses did not reach statistical significance. Information about the implementation of subgroup analyses was reported most consistently for articles from The New England Journal of Medicine, since it was also traceable on the basis of provided trial protocols. We were able to comprehend whether subgroup analyses were pre-specified in a majority of the reviewed publications. Even though results of multiple subgroup analyses were reported for most published trials, a corresponding adjustment for multiple testing was rarely considered.

Conclusion:

Compared to previous reviews in this context, we observed improvements in the reporting of subgroup analyses of cardiovascular randomised clinical trials. Nonetheless, critical shortcomings, such as inconsistent reporting of the implementation and insufficient pre-specification, persist.

Systematic Incorporation of Sex-Specific Information Into Clinical Practice Guidelines for the Management of ST -Segment-Elevation Myocardial Infarction: Feasibility and Outcomes

Background Clinical practice guideline ( CPG ) developers have yet to endorse a consistent and systematic approach for considering sex-specific cardiovascular information in CPG s. This article describes an initiative led by the Canadian Cardiovascular Society to determine the feasibility and outcomes of a structured process for considering sex in a CPG for the management of ST-segment-elevation myocardial infarction. Methods and Results A sex and gender champion was appointed to the guideline development committee. The feasibility of tailoring the CPG to sex was ascertained by recording (1) the male-female distribution of the study population, (2) the adequacy of sex-specific representation in each study using the participation/prevalence ratio, and (3) whether data were disaggregated by sex. The outcome was to determine whether recommendations for CPG s based on an assessment of the evidence should differ by sex. In total, 175 studies were included. The mean percentage of female participants reported in the studies was 24.5% ( SD : 6.6%; minimum: 0%; maximum: 51%). The mean participation/prevalence ratio was 0.62 ( SD : 0.16; minimum: 0.00; maximum: 1.19). Eighteen (10.2%) studies disaggregated the data by sex. Based on the participation/prevalence ratio and the sex-specific analyses presented, only 1 study provided adequate evidence to confidently inform the applicability of the CPG recommendations to male and female patients. Conclusions Implementing a systematic process for critically appraising sex-specific evidence for CPG s was straightforward and feasible. Inadequate enrollment and reporting by sex hindered comprehensive sex-specific assessment of the quality of evidence and strength of recommendations for a CPG on the management of ST-segment-elevation myocardial infarction.

Lipid management in patients with chronic kidney disease

An increased risk of cardiovascular disease, independent of conventional risk factors, is present even at minor levels of renal impairment and is highest in patients with end-stage renal disease (ESRD) requiring dialysis. Renal dysfunction changes the level, composition and quality of blood lipids in favour of a more atherogenic profile. Patients with advanced chronic kidney disease (CKD) or ESRD have a characteristic lipid pattern of hypertriglyceridaemia and low HDL cholesterol levels but normal LDL cholesterol levels. In the general population, a clear relationship exists between LDL cholesterol and major atherosclerotic events. However, in patients with ESRD, LDL cholesterol shows a negative association with these outcomes at below average LDL cholesterol levels and a flat or weakly positive association with mortality at higher LDL cholesterol levels. Overall, the available data suggest that lowering of LDL cholesterol is beneficial for prevention of major atherosclerotic events in patients with CKD and in kidney transplant recipients but is not beneficial in patients requiring dialysis. The 2013 Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guideline for Lipid Management in CKD provides simple recommendations for the management of dyslipidaemia in patients with CKD and ESRD. However, emerging data and novel lipid-lowering therapies warrant some reappraisal of these recommendations.

Assessment of effect modification of statins on new-onset diabetes based on various medical backgrounds: a retrospective cohort study

BACKGROUND

The aim of this study was to investigate the association between statin use and new-onset diabetes in clinical settings and to assess its effect modification (heterogeneity) among patients with various medical histories and current medications.

METHODS

In a total of 12,177 Japanese patients without diabetes, from December 2004 to November 2012, we identified 500 statin users and 500 matched non-users using propensity-score matching. Patients were followed until December 2017. We estimated the hazard ratios of new-onset diabetes associated with statin use. We also tested the heterogeneity of the treatment effect by evaluating subgroup interactions in subgroups according to sex, age, medical history, and current medication.

RESULTS

New-onset diabetes had occurred in 71 patients (13.6%) with statin use and 43 patients (8.3%) with non-use at 5 years (hazard ratio, 1.66; 95% confidence interval [CI], 1.11 to 2.48; P = 0.0143), and in 78 patients (15.6%) with statin use and 48 patients (9.6%) with non-use at 10 years (hazard ratio, 1.61; 95% CI, 1.10 to 2.37; P = 0.0141). There were no significant treatment-by-subgroup interactions in all subgroups defined according to sex, age, medical history, and current medication.

CONCLUSIONS

In patients with various clinical backgrounds, those who received statin therapy had a higher risk of new-onset diabetes at 5 and 10 years than those who did not receive it. Effect modification of statins on new-onset diabetes was not found in patient populations defined according to various comorbid diseases or concomitant drugs.

Studying treatment-effect heterogeneity in precision medicine through induced subgroups

Precision medicine, in the sense of tailoring the choice of medical treatment to patients' pretreatment characteristics, is nowadays gaining a lot of attention. Preferably, this tailoring should be realized in an evidence-based way, with key evidence in this regard pertaining to subgroups of patients that respond differentially to treatment (i.e., to subgroups involved in treatment-subgroup interactions). Often a-priori hypotheses on subgroups involved in treatment-subgroup interactions are lacking or are incomplete at best. Therefore, methods are needed that can induce such subgroups from empirical data on treatment effectiveness in a post hoc manner. Recently, quite a few such methods have been developed. So far, however, there is little empirical experience in their usage. This may be problematic for medical statisticians and statistically minded medical researchers, as many (nontrivial) choices have to be made during the data-analytic process. The main purpose of this paper is to discuss the major concepts and considerations when using these methods. This discussion will be based on a systematic, conceptual, and technical analysis of the type of research questions at play, and of the type of data that the methods can handle along with the available software, and a review of available empirical evidence. We will illustrate all this with the analysis of a dataset comparing several anti-depressant treatments.

Implications for paediatric shock management in resource-limited settings: a perspective from the FEAST trial

BACKGROUND

Although the African "Fluid Expansion as Supportive therapy" (FEAST) trial showed fluid resuscitation was harmful in children with severe febrile illness managed in resource-limited hospitals, the most recent evidence reviewed World Health Organization (WHO) guidelines continue to recommend fluid boluses in children with shock according to WHO criteria "WHO shock", arguing that the numbers included in the FEAST trial were too small to provide reasonable certainty.

METHODS

We re-analysed the FEAST trial results for all international definitions for paediatric shock including hypotensive (or decompensated shock) and the WHO criteria. In addition, we examined the clinical relevance of the WHO criteria to published and unpublished observational studies reporting shock in resource-limited settings.

RESULTS

We established that hypotension was rare in children with severe febrile illness complicating only 29/3170 trial participants (0.9%). We confirmed that fluid boluses were harmful irrespective of the definitions of shock including the very small number with WHO shock (n = 65). In this subgroup 48% of bolus recipients died at 48 h compared to 20% of the non-bolus control group, an increased absolute risk of 28%, but translating to an increased relative risk of 240% (p = 0.07 (two-sided Fisher's exact test)). Examining studies describing the prevalence of the stringent WHO shock criteria in children presenting to hospital we found this was rare (~ 0.1%) and in these children mortality was very high (41.5-100%).

CONCLUSIONS

The updated WHO guidelines continue to recommend boluses for a very limited number of children presenting at hospital with the strict definition of WHO shock. Nevertheless, the 3% increased mortality from boluses seen across FEAST trial participants would also include this subgroup of children receiving boluses. Recommendations aiming to differentiate WHO shock from other definitions will invariably lead to "slippage" at the bedside, with the potential of exposing a wider group of children to the harm of fluid-bolus therapy.

Social value and individual choice: The value of a choice-based decision-making process in a collectively funded health system

Evidence about cost-effectiveness is increasingly being used to inform decisions about the funding of new technologies that are usually implemented as guidelines from centralized decision-making bodies. However, there is also an increasing recognition for the role of patients in determining their preferred treatment option. This paper presents a method to estimate the value of implementing a choice-based decision process using the cost-effectiveness analysis toolbox. This value is estimated for 3 alternative scenarios. First, it compares centralized decisions, based on population average cost-effectiveness, against a decision process based on patient choice. Second, it compares centralized decision based on patients' subgroups versus an individual choice-based decision process. Third, it compares a centralized process based on average cost-effectiveness against a choice-based process where patients choose according to a different measure of outcome to that used by the centralized decision maker. The methods are applied to a case study for the management of acute coronary syndrome. It is concluded that implementing a choice-based process of treatment allocation may be an option in collectively funded health systems. However, its value will depend on the specific health problem and the social values considered relevant to the health system.

Stress Echocardiography Positivity Predicts Cancer Death

BACKGROUND

Stress echocardiography (SE) predicts cardiac death, but an increasing share of cardiac patients eventually die of cancer. The aim of the study was to assess whether SE positivity predicts cancer death.

METHODS AND RESULTS

In a retrospective analysis of prospectively acquired single-center, observational data, we evaluated 4673 consecutive patients who underwent SE from 1983 to 2009. All patients were cancer-free at index SE and were followed up for a median of 131 months (interquartile range 134). We separately analyzed predetermined end points: cardiovascular, cancer, and noncardiovascular, noncancer death, with and without competing risk. SE was positive in 1757 and negative in 2916 patients; 869 cardiovascular, 418 cancer, and 625 noncardiovascular, noncancer deaths were registered. The 25-year mortality was higher in SE-positive than in SE-negative patients, considering cardiovascular (40% versus 31%; P<0.001) and cancer mortality (26% versus 17%; P<0.01). SE positivity was a strong predictor of cancer (cause-specific hazard ratio 1.19; 95% confidence interval, 1.16-1.73; P=0.05) and cardiovascular mortality (1.18; 95% confidence interval, 1.03-1.35; P=0.02). Fine-Gray analysis to account for competing risk gave similar results. Cancer risk diverged after 15 years, whereas differences were already significant at 5 years for cardiovascular risk.

CONCLUSIONS

SE results predict cardiovascular and cancer mortality. SE may act as a proxy of the shared risk factor milieu for cancer or cardiovascular death.

Subgroup analyses with special reference to the effect of antiplatelet agents in acute coronary syndromes

Controlled trials estimate treatment effects averaged over the reference population of subjects. However, physicians are interested in whether the treatment effect varies across subgroups (effect heterogeneity) in order to target specific subgroups to maximise the benefit of treatment and minimise harm. Therefore, large clinical trials of antiplatelet agents include subgroup analyses that examine whether treatment effects differ between subgroups of subjects identified by baseline characteristics. Reporting subgroup is pervasive and often accompanied by claims of difference of treatment effects between subgroups with potential important implications for clinical practice. However, subgroup-specific analyses of clinical trial data have inherent limitations that reduce their reliability. These include reduced statistical power, failure to specify the subgroups of interest a priori, failure to account for examining large numbers of subgroups, lack of strong rationale for biological response modification, and performing analyses based on variables measured post randomisation or in trials showing no overall difference between treatments. Rules for interpretation of subgroup findings in subgroups have been suggested but are frequently not applied. In this article we draw attention to the pitfalls of subgroup analyses in the context of recent trials of antiplatelet agents.

Evaluating the impact of treating the optimal subgroup

Suppose we have a binary treatment used to influence an outcome. Given data from an observational or controlled study, we wish to determine whether or not there exists some subset of observed covariates in which the treatment is more effective than the standard practice of no treatment. Furthermore, we wish to quantify the improvement in population mean outcome that will be seen if this subgroup receives treatment and the rest of the population remains untreated. We show that this problem is surprisingly challenging given how often it is an (at least implicit) study objective. Blindly applying standard techniques fails to yield any apparent asymptotic results, while using existing techniques to confront the non-regularity does not necessarily help at distributions where there is no treatment effect. Here, we describe an approach to estimate the impact of treating the subgroup which benefits from treatment that is valid in a nonparametric model and is able to deal with the case where there is no treatment effect. The approach is a slight modification of an approach that recently appeared in the individualized medicine literature.

Post Hoc Analyses of ApoE Genotype-Defined Subgroups in Clinical Trials

BACKGROUND

Many post hoc analyses of clinical trials in Alzheimer's disease (AD) and mild cognitive impairment (MCI) are in small Phase 2 trials. Subject heterogeneity may lead to statistically significant post hoc results that cannot be replicated in larger follow-up studies.

OBJECTIVE

We investigated the extent of this problem using simulation studies mimicking current trial methods with post hoc analyses based on ApoE4 carrier status.

METHODS

We used a meta-database of 24 studies, including 3,574 subjects with mild AD and 1,171 subjects with MCI/prodromal AD, to simulate clinical trial scenarios. Post hoc analyses examined if rates of progression on the Alzheimer's Disease Assessment Scale-cognitive (ADAS-cog) differed between ApoE4 carriers and non-carriers.

RESULTS

Across studies, ApoE4 carriers were younger and had lower baseline scores, greater rates of progression, and greater variability on the ADAS-cog. Up to 18% of post hoc analyses for 18-month trials in AD showed greater rates of progression for ApoE4 non-carriers that were statistically significant but unlikely to be confirmed in follow-up studies. The frequency of erroneous conclusions dropped below 3% with trials of 100 subjects per arm. In MCI, rates of statistically significant differences with greater progression in ApoE4 non-carriers remained below 3% unless sample sizes were below 25 subjects per arm.

CONCLUSIONS

Statistically significant differences for ApoE4 in post hoc analyses often reflect heterogeneity among small samples rather than true differential effect among ApoE4 subtypes. Such analyses must be viewed cautiously. ApoE genotype should be incorporated into the design stage to minimize erroneous conclusions.

Tutorial on statistical considerations on subgroup analysis in confirmatory clinical trials

Clinical trials target patients who are expected to benefit from a new treatment under investigation. However, the magnitude of the treatment benefit, if it exists, often depends on the patient baseline characteristics. It is therefore important to investigate the consistency of the treatment effect across subgroups to ensure a proper interpretation of positive study findings in the overall population. Such assessments can provide guidance on how the treatment should be used. However, great care has to be taken when interpreting consistency results. An observed heterogeneity in treatment effect across subgroups can arise because of chance alone, whereas true heterogeneity may be difficult to detect by standard statistical tests because of their low power. This tutorial considers issues related to subgroup analyses and their impact on the interpretation of findings of completed trials that met their main objectives. In addition, we provide guidance on the design and analysis of clinical trials that account for the expected heterogeneity of treatment effects across subgroups by establishing treatment benefit in a pre-defined targeted subgroup and/or the overall population. Copyright © 2016 John Wiley & Sons, Ltd.

Quantifying App Store Dynamics: Longitudinal Tracking of Mental Health Apps

Background

For many mental health conditions, mobile health apps offer the ability to deliver information, support, and intervention outside the clinical setting. However, there are difficulties with the use of a commercial app store to distribute health care resources, including turnover of apps, irrelevance of apps, and discordance with evidence-based practice.

Objective

The primary aim of this study was to quantify the longevity and rate of turnover of mental health apps within the official Android and iOS app stores. The secondary aim was to quantify the proportion of apps that were clinically relevant and assess whether the longevity of these apps differed from clinically nonrelevant apps. The tertiary aim was to establish the proportion of clinically relevant apps that included claims of clinical effectiveness. We performed additional subgroup analyses using additional data from the app stores, including search result ranking, user ratings, and number of downloads.

Methods

We searched iTunes (iOS) and the Google Play (Android) app stores each day over a 9-month period for apps related to depression, bipolar disorder, and suicide. We performed additional app-specific searches if an app no longer appeared within the main search

Results

On the Android platform, 50% of the search results changed after 130 days (depression), 195 days (bipolar disorder), and 115 days (suicide). Search results were more stable on the iOS platform, with 50% of the search results remaining at the end of the study period. Approximately 75% of Android and 90% of iOS apps were still available to download at the end of the study. We identified only 35.3% (347/982) of apps as being clinically relevant for depression, of which 9 (2.6%) claimed clinical effectiveness. Only 3 included a full citation to a published study.

Conclusions

The mental health app environment is volatile, with a clinically relevant app for depression becoming unavailable to download every 2.9 days. This poses challenges for consumers and clinicians seeking relevant and long-term apps, as well as for researchers seeking to evaluate the evidence base for publicly available apps.

Subgroup analyses of randomized clinical trials in heart failure: facts and numbers

Subgroup analyses of major randomized clinical trials in heart failure are published frequently, but their impact on medical knowledge and practice guidelines has not been previously reported. In a novel analysis, we determined number of citations, impact factors, number of authors, and citations in guidelines of both parent trials and sub‐studies; we also qualitatively assessed whether the analyses were described as post‐hoc and non‐pre‐specified. A total of 229 sub‐studies evaluating outcomes in patient subgroups were published (median 6, range 0–36 per trial). The number of subjects in the parent trials positively correlated with number of sub‐studies (rho = 0.51, P = 0.009). The subgroups are frequently not pre‐specified. The impact factors of sub‐studies were lower in comparison to the parent trials as were the number of citations two years after the publication date; in addition, parent trials were cited more frequently in European and American professional guidelines compared with the sub‐studies. We maintain that the sub‐studies derived from major heart failure trials are frequently published, but their contribution to clinical guidelines and medical knowledge are highly debatable.

Perils and pitfalls of reporting sex differences

The idea of sex differences in the brain both fascinates and inflames the public. As a result, the communication and public discussion of new findings is particularly vulnerable to logical leaps and pseudoscience. A new US National Institutes of Health policy to consider both sexes in almost all preclinical research will increase the number of reported sex differences and thus the risk that research in this important area will be misinterpreted and misrepresented. In this article, I consider ways in which we might reduce that risk, for example, by (i) employing statistical tests that reveal the extent to which sex explains variation, rather than whether or not the sexes 'differ', (ii) properly characterizing the frequency distributions of scores or dependent measures, which nearly always overlap, and (iii) avoiding speculative functional or evolutionary explanations for sex-based variation, which usually invoke logical fallacies and perpetuate sex stereotypes. Ultimately, the factor of sex should be viewed as an imperfect, temporary proxy for yet-unknown factors, such as hormones or sex-linked genes, that explain variation better than sex. As scientists, we should be interested in discovering and understanding the true sources of variation, which will be more informative in the development of clinical treatments.

Risks and Benefits of Nalmefene in the Treatment of Adult Alcohol Dependence: A Systematic Literature Review and Meta-Analysis of Published and Unpublished Double-Blind Randomized Controlled Trials

BACKGROUND

Nalmefene is a recent option in alcohol dependence treatment. Its approval was controversial. We conducted a systematic review and meta-analysis of the aggregated data (registered as PROSPERO 2014:CRD42014014853) to compare the harm/benefit of nalmefene versus placebo or active comparator in this indication.

METHODS AND FINDINGS

Three reviewers searched for published and unpublished studies in Medline, the Cochrane Library, Embase, ClinicalTrials.gov, Current Controlled Trials, and bibliographies and by mailing pharmaceutical companies, the European Medicines Agency (EMA), and the US Food and Drug Administration. Double-blind randomized clinical trials evaluating nalmefene to treat adult alcohol dependence, irrespective of the comparator, were included if they reported (1) health outcomes (mortality, accidents/injuries, quality of life, somatic complications), (2) alcohol consumption outcomes, (3) biological outcomes, or (4) treatment safety outcomes, at 6 mo and/or 1 y. Three authors independently screened the titles and abstracts of the trials identified. Relevant trials were evaluated in full text. The reviewers independently assessed the included trials for methodological quality using the Cochrane Collaboration tool for assessing risk of bias. On the basis of the I2 index or the Cochrane's Q test, fixed or random effect models were used to estimate risk ratios (RRs), mean differences (MDs), or standardized mean differences (SMDs) with 95% CIs. In sensitivity analyses, outcomes for participants who were lost to follow-up were included using baseline observation carried forward (BOCF); for binary measures, patients lost to follow-up were considered equal to failures (i.e., non-assessed patients were recorded as not having responded in both groups). Five randomized controlled trials (RCTs) versus placebo, with a total of 2,567 randomized participants, were included in the main analysis. None of these studies was performed in the specific population defined by the EMA approval of nalmefene, i.e., adults with alcohol dependence who consume more than 60 g of alcohol per day (for men) or more than 40 g per day (for women). No RCT compared nalmefene with another medication. Mortality at 6 mo (RR = 0.39, 95% CI [0.08; 2.01]) and 1 y (RR = 0.98, 95% CI [0.04; 23.95]) and quality of life at 6 mo (SF-36 physical component summary score: MD = 0.85, 95% CI [-0.32; 2.01]; SF-36 mental component summary score: MD = 1.01, 95% CI [-1.33; 3.34]) were not different across groups. Other health outcomes were not reported. Differences were encountered for alcohol consumption outcomes such as monthly number of heavy drinking days at 6 mo (MD = -1.65, 95% CI [-2.41; -0.89]) and at 1 y (MD = -1.60, 95% CI [-2.85; -0.35]) and total alcohol consumption at 6 mo (SMD = -0.20, 95% CI [-0.30; -0.10]). An attrition bias could not be excluded, with more withdrawals for nalmefene than for placebo, including more withdrawals for safety reasons at both 6 mo (RR = 3.65, 95% CI [2.02; 6.63]) and 1 y (RR = 7.01, 95% CI [1.72; 28.63]). Sensitivity analyses showed no differences for alcohol consumption outcomes between nalmefene and placebo, but the weight of these results should not be overestimated, as the BOCF approach to managing withdrawals was used.

CONCLUSIONS

The value of nalmefene for treatment of alcohol addiction is not established. At best, nalmefene has limited efficacy in reducing alcohol consumption.

Challenges and solutions to pre- and post-randomization subgroup analyses

Subgroup analyses are commonly performed in the clinical trial setting with the purpose of illustrating that the treatment effect was consistent across different patient characteristics or identifying characteristics that should be targeted for treatment. There are statistical issues involved in performing subgroup analyses, however. These have been given considerable attention in the literature for analyses where subgroups are defined by a pre-randomization feature. Although subgroup analyses are often performed with subgroups defined by a post-randomization feature—including analyses that estimate the treatment effect among compliers—discussion of these analyses has been neglected in the clinical literature. Such analyses pose a high risk of presenting biased descriptions of treatment effects. We summarize the challenges of doing all types of subgroup analyses described in the literature. In particular, we emphasize issues with post-randomization subgroup analyses. Finally, we provide guidelines on how to proceed across the spectrum of subgroup analyses.

A systematic review and meta-analysis of 130,000 individuals shows smoking does not modify the association of APOE genotype on risk of coronary heart disease

Background

Conflicting evidence exists on whether smoking acts as an effect modifier of the association between APOE genotype and risk of coronary heart disease (CHD).

Methods and results

We searched PubMed and EMBASE to June 11, 2013 for published studies reporting APOE genotype, smoking status and CHD events and added unpublished data from population cohorts. We tested for presence of effect modification by smoking status in the relationship between APOE genotype and risk of CHD using likelihood ratio test.

In total 13 studies (including unpublished data from eight cohorts) with 10,134 CHD events in 130,004 individuals of European descent were identified. The odds ratio (OR) for CHD risk from APOE genotype (ε4 carriers versus non-carriers) was 1.06 (95% confidence interval (CI): 1.01, 1.12) and for smoking (present vs. past/never smokers) was OR 2.05 (95%CI: 1.95, 2.14). When the association between APOE genotype and CHD was stratified by smoking status, compared to non-ε4 carriers, ε4 carriers had an OR of 1.11 (95%CI: 1.02, 1.21) in 28,789 present smokers and an OR of 1.04 (95%CI 0.98, 1.10) in 101,215 previous/never smokers, with no evidence of effect modification (P-value for heterogeneity = 0.19). Analysis of pack years in individual participant data of >60,000 with adjustment for cardiovascular traits also failed to identify evidence of effect modification.

Conclusions

In the largest analysis to date, we identified no evidence for effect modification by smoking status in the association between APOE genotype and risk of CHD.

•

We examined evidence for an interaction between APOE genotype, smoking and risk of coronary heart disease.

•

This was conducted in the largest meta-analysis of published and unpublished data sets to date (>130,000 individuals).

•

Our analysis did not identify evidence of interaction.

•

These findings bring into question presence of a clinically meaningful interaction between APOE genotype and smoking.

Comparison of adjunctive use of aripiprazole with bupropion or selective serotonin reuptake inhibitors/serotonin-norepinephrine reuptake inhibitors: analysis of patients beginning adjunctive treatment in a 52-week, open-label study

Background

This post hoc analysis assessed the safety, tolerability and effectiveness of long-term treatment with aripiprazole adjunctive to either bupropion or selective serotonin reuptake inhibitors (SSRIs)/serotonin–norepinephrine reuptake inhibitors (SNRIs) in patients with major depressive disorder (MDD).

Methods

Data from de novo patients (did not participate in 2 previous studies) in a 52-week, open-label safety study of adjunctive aripiprazole after documented inadequate response to 1–4 antidepressant treatments (ADTs; SSRI, SNRI, or bupropion) were analyzed post hoc. Assessments included safety and tolerability, sexual functioning (Massachusetts General Hospital Sexual Functioning Inventory [MGH-SFI]) and Clinical Global Impressions–Severity (CGI-S).

Results

Forty-seven patients received bupropion plus aripiprazole and 245 received an SSRI/SNRI plus aripiprazole; 19 (40.4%) and 78 (31.8%), respectively, completed 52 weeks of treatment, and 46 and 242, respectively, received ≥1 dose of study medication (safety sample). Median time to discontinuation (any reason) was 184.0 days. Overall, 97.8% of patients in the bupropion group and 93.8% in the SSRI/SNRI group experienced ≥1 adverse event. The most common treatment-emergent adverse events were fatigue (26.1%) and somnolence (21.7%) with bupropion and fatigue (23.6%) and akathisia (23.6%) with an SSRI/SNRI. Mean change in body weight at week 52 (observed cases) was +3.1 kg for bupropion and +2.4 kg for an SSRI/SNRI. Treatment-emergent, potentially clinically relevant abnormalities in fasting glucose occurred in 8.3% of patients with bupropion and 17.4% with an SSRI/SNRI; for abnormalities in fasting total cholesterol, the incidence was 25.0% and 34.7%, respectively. Mean (SE) change from baseline in fasting glucose was 1.4 (1.9) mg/dL with bupropion and 2.7 (1.5) mg/dL with an SSRI/SNRI. Baseline MGH-SFI item scores indicated less severe impairment with bupropion versus an SSRI/SNRI; in both groups most MGH-SFI items exhibited improvement at week 52. Mean CGI-S improvement at week 52 (last observation carried forward) was -1.4 with bupropion and -1.5 with an SSRI/SNRI (efficacy sample).

Conclusions

There were no unexpected AEs with long-term adjunctive aripiprazole therapy when added to either bupropion or SSRIs/SNRIs, and symptom improvement was similar between ADT groups. Sexual functioning in patients with MDD on antidepressants was also modestly improved after adding aripiprazole.

Trial registration

ClinicalTrials.gov: NCT00095745 (November 9, 2004).

Subgroup analyses in cost-effectiveness analyses to support health technology assessments

'Success' in drug development is bringing to patients a new medicine that has an acceptable benefit-risk profile and that is also cost-effective. Cost-effectiveness means that the incremental clinical benefit is deemed worth paying for by a healthcare system, and it has an important role in enabling manufacturers to obtain new medicines to patients as soon as possible following regulatory approval. Subgroup analyses are increasingly being utilised by decision-makers in the determination of the cost-effectiveness of new medicines when making recommendations. This paper highlights the statistical considerations when using subgroup analyses to support cost-effectiveness for a health technology assessment. The key principles recommended for subgroup analyses supporting clinical effectiveness published by Paget et al. are evaluated with respect to subgroup analyses supporting cost-effectiveness. A health technology assessment case study is included to highlight the importance of subgroup analyses when incorporated into cost-effectiveness analyses. In summary, we recommend planning subgroup analyses for cost-effectiveness analyses early in the drug development process and adhering to good statistical principles when using subgroup analyses in this context. In particular, we consider it important to provide transparency in how subgroups are defined, be able to demonstrate the robustness of the subgroup results and be able to quantify the uncertainty in the subgroup analyses of cost-effectiveness.

Effect of symptom severity on efficacy and safety of aripiprazole adjunctive to antidepressant monotherapy in major depressive disorder: a pooled analysis

BACKGROUND

There is a paucity of evidence for outcome predictors in patients with major depressive disorder (MDD) not responding to initial antidepressant therapy (ADT). This post-hoc analysis evaluated whether MDD severity affects response to adjunctive aripiprazole.

METHODS

Data from 3 randomized, double-blind, placebo-controlled trials of adjunctive aripiprazole in adults with MDD and inadequate response to 1 to 3 ADT trials were pooled and stratified based on Montgomery-Åsberg Depression Rating Scale (MADRS) total score (mild, ≤24; moderate, 25-30; severe, ≥31). Treatment differences in change in MADRS total score and rates of response (≥50% MADRS improvement) and remission (response with MADRS total score ≤10) were analyzed at endpoint. Adverse events were assessed within each subgroup.

RESULTS

Aripiprazole produced greater improvement than placebo in the MADRS total score regardless of MDD severity at baseline (between-treatment difference [95% CI]: mild, -2.5 [-4.0 to -1.1]; moderate, -3.2 [-4.9 to -1.6]; severe, -4.5 [-6.8 to -2.2]). Compared with placebo, adjunctive aripiprazole increased the likelihood of response in all subgroups (risk ratio [95% CI]: mild, 1.50 [1.15, 1.95]; moderate, 1.51 [1.09, 2.11]; severe, 1.95 [1.23, 3.10]). Common treatment-emergent adverse events included akathisia and restlessness.

LIMITATIONS

The original studies were not designed to assess the efficacy of adjunctive aripiprazole by baseline severity, and this post-hoc analysis was not powered to evaluate differences in severity subgroups.

CONCLUSIONS

In patients who failed to respond to initial ADT, adjunctive aripiprazole was more effective than placebo in mild, moderate, and severe MDD strata.

CLINICAL TRIAL REGISTRATION

ClinicalTrial.gov: NCT00095823, NCT00105196, and NCT00095758.

Ticagrelor: positive, negative and misunderstood properties as a new antiplatelet agent

Dual antiplatelet therapy is essential for the management of acute coronary syndrome. In particular, combination therapy using aspirin with a platelet ADP (i.e. P2Y12 ) receptor inhibitor, such as clopidogrel, prasugrel or, more recently, ticagrelor, has been recommended for patients with acute coronary syndrome. Pharmacological agents that reversibly inhibit platelet aggregation without metabolic activation in the liver are believed to reduce cardiovascular mortality compared with the current drug of choice for antiplatelet therapy, namely clopidogrel. These findings are based on a multicentre, double-blind, double-dummy, randomized controlled trial. Numerous factors are postulated to contribute to the improved survival of patients who take ticagrelor compared with those taking clopidogrel, including the risk of myocardial infarction, heart failure, arrhythmia and bleeding. In addition, clopidogrel may lead to a much higher incidence of infection. Although ticagrelor has recently been approved for use in the US and exhibits superiority over other antiplatelet agents, certain concerns remain regarding its use, including lung injury and dyspnoea, thus raising the issue of its true superiority over clopidogrel or prasugrel. Recent studies into ticagrelor report conflicting data, with certain aspects of its mechanisms of action still not fully understood. Ticagrelor has beneficial effects following its clinical application, such as achieving overall higher reductions in mortality compared with the use of clopidogrel and prasugrel. Harmful effects associated with the use of ticagrelor include a higher incidence of dyspnoea and major bleeding compared with clopidogrel.