Methods for the analysis of multiple endpoints in small populations: A review

Ristl, Robin; Urach, Susanne; Rosenkranz, G; Posch, Martin

doi:10.1080/10543406.2018.1489402

Cited by 39 publications

(45 citation statements)

References 132 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moving beyond the work presented in this paper, our framework would allow combining simulation models with other sources of data describing geographical variation in disease, for example, modelled health systems or modelled prevalence ( 41, 42 ) and to incorporate interactions of interventions with novel interventions for surveillance. Clinical trials for new interventions could thereby be prioritized to geographical settings, where public health impact is likely to be maximized and where appropriate, to inform decisions on achieving non-inferiority or superiority endpoints ( 43, 44 ). A significant extension is incorporating economic considerations which may affect development decisions, including both costs of R&D, as well as implementation and systems costs for final deployment.…”

Section: Discussionmentioning

confidence: 99%

Leveraging mathematical models of disease dynamics and machine learning to improve development of novel malaria interventions

Golumbeanu

Yang

Stuckey

et al. 2021

Preprint

View full text Add to dashboard Cite

The development of novel interventions against a disease entails optimising their specifications to achieve desired health goals such as disease reduction. As testing is limited early in development, it is difficult to predefine these optimal specifications, prioritize or continue investment in candidate interventions. Mathematical models of disease can provide quantitative evidence as they can simulate deployment and predict impact of a new intervention considering deployment, health-system, population and disease characteristics. However, due to large uncertainty early in development, as well as model complexity, testing all possible combinations of interventions and deployments becomes infeasible. As a result, mathematical models have been only marginally used during intervention development to date. Here, we present a new approach where machine learning enables the use of detailed disease models to identify optimal properties of candidate interventions to reach a desired health goal and guide development. We demonstrate the power of our approach by application to five novel malaria interventions under development. For various targeted reductions of malaria prevalence, we quantify and rank intervention characteristics which are key determinants of health impact. Furthermore, we identify minimal requirements and tradeoffs between operational factors, intervention efficacy and duration to achieve different levels of impact and show how these vary across disease transmission settings. When single interventions cannot achieve significant impact, our method allows finding optimal combinations of interventions fulfilling the desired health goals. By enabling efficient use of disease models, our approach supports decision-making and resource investment in the development of new interventions for infectious diseases.Significance StatementDuring development of novel disease interventions (e.g. vaccines), a target product profile (TPP) document defines intervention characteristics required to meet health goals. As clinical trials are limited early in development, mathematical models simulating disease dynamics can help define TPPs. However, testing all combinations of intervention, delivery and environment characteristics is infeasible and so complex mathematical models have not been used until now. We introduce a new approach to define TPPs, combining models of disease with machine learning. We examined several novel malaria interventions, identifying key characteristics, minimum efficacy and duration of effect that ensure significant reductions in malaria prevalence. This approach therefore enabled mathematical models of disease to support intervention development, by identifying intervention requirements that ensure public health impact.One Sentence SummaryDefining quantitative profiles of novel disease interventions by combining machine learning with mathematical models of disease transmission

show abstract

Section: Discussionmentioning

confidence: 99%

Leveraging mathematical models of disease dynamics and machine learning to improve development of novel malaria interventions

Golumbeanu

Yang

Stuckey

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Moreover, a resampling approach for the analysis of correlated multiple endpoints is taken, e.g. as recently described by Ristl et al [ 57 ]. Time-to-event (e.g., survival) distributions are summarized by the Kaplan–Meier method and compared by the (stratified) log-rank test.…”

Section: Discussionmentioning

confidence: 99%

Efficacy of team-based collaborative care for distressed patients in secondary prevention of chronic coronary heart disease (TEACH): study protocol of a multicenter randomized controlled trial

Herrmann‐Lingen

Albus

Zwaan

et al. 2020

BMC Cardiovasc Disord

View full text Add to dashboard Cite

Background Coronary heart disease (CHD) is the leading cause of death and years of life lost worldwide. While effective treatments are available for both acute and chronic disease stages there are unmet needs for effective interventions to support patients in health behaviors required for secondary prevention. Psychosocial distress is a common comorbidity in patients with CHD and associated with substantially reduced health-related quality of life (HRQoL), poor health behavior, and low treatment adherence. Methods In a confirmatory, randomized, controlled, two-arm parallel group, multicenter behavioral intervention trial we will randomize 440 distressed CHD patients with at least one insufficiently controlled cardiac risk factor to either their physicians' usual care (UC) or UC plus 12-months of blended collaborative care (TeamCare = TC). Trained nurse care managers (NCM) will proactively support patients to identify individual sources of distress and risk behaviors, establish a stepwise treatment plan to improve self-help and healthy behavior, and actively monitor adherence and progress. Additional e-health resources are available to patients and their families. Intervention fidelity is ensured by a treatment manual, an electronic patient registry, and a specialist team regularly supervising NCM via videoconferences and recommending protocol and guideline-compliant treatment adjustments as indicated. Recommendations will be shared with patients and their physicians who remain in charge of patients’ care. Since HRQoL is a recommended outcome by both, several guidelines and patient preference we chose a ≥ 50% improvement over baseline on the HeartQoL questionnaire at 12 months as primary outcome. Our primary hypothesis is that significantly more patients receiving TC will meet the primary outcome criterion compared to the UC group. Secondary hypotheses will evaluate improvements in risk factors, psychosocial variables, health care utilization, and durability of intervention effects over 18–30 months of follow-up. Discussion TEACH is the first study of a blended collaborative care intervention simultaneously addressing distress and medical CHD risk factors conducted in cardiac patients in a European health care setting. If proven effective, its results can improve long-term chronic care of this vulnerable patient group and may be adapted for patients with other chronic conditions. Trial registration: German Clinical Trials Register, DRKS00020824, registered on 4 June, 2020; https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00020824

show abstract

“…This approach was motivated by the goal to obtain a more comprehensive and hopefully more robust characterization of behavior than might be achieved by a single endpoint. It was also motivated by the clinical scores constructed from multiple measurements, and the success of multiple-endpoint analyses in clinical studies (e.g., (60)(61)(62), to cite but a few).…”

Section: Statistical Procedures For Dose-response Studiesmentioning

confidence: 99%

Pharmacological prevention of neonatal opioid withdrawal in a pregnant guinea pig model

Safa

Lau²,

Aten

et al. 2020

Preprint

View full text Add to dashboard Cite

Neonatal opioid dependence is an enormous and growing medical challenge. Newborns exposed to prenatal opioids often experience intense postnatal withdrawal after cessation of the opioid, called neonatal opioid withdrawal syndrome (NOWS). They may also show delays in developmental milestones. Thus, a pharmacotherapy that could be delivered in conjunction with a prenatal opioid, and that could prevent fetal opioid dependence without interfering with pain or opioid use management of the mother, would be highly desirable. In a prior study in mice we demonstrated that the peripherally selective neutral opioid antagonist, 6β-naltrexol (6BN), has properties that make it a promising candidate for such an approach. In the current work we extend our studies of 6BN to pregnant guinea pigs, which are a more suitable model than mice and rats, and allows for the detection of robust spontaneous neonatal withdrawal. We find that prenatal methadone (MTD) significantly aggravates two classic maternal separation stress behaviors in newborn guinea pigs: calling (vocalizing) and searching (locomotion) - natural attachment behaviors thought to be controlled by the endogenous opioid system. In addition, prenatal MTD significantly increases the levels of plasma cortisol in newborns exposed to maternal separation stress, showing that cessation of MTD at birth engages the hypothalamic-pituitary-adrenal (HPA) axis. Most importantly, our work shows that 6BN can prevent all these effects of prenatal methadone with high potency (50% inhibitory dose, ID50, of ~0.02 mg/kg), at doses unlikely to induce maternal withdrawal or to interfere with opioid analgesia based on many prior published studies. This potency is surprising in the face of both the long half-life of methadone and our pharmacokinetic (PK) studies showing slow placental transit of 6BN in guinea pigs. Furthermore, we demonstrate an unexpectedly high potency of 6BN in preventing opioid withdrawal in adult guinea pigs (ID50 = 0.01 mg/kg), in spite of the fact that our PK data in both mice and guinea pigs show the relative exclusion of 6BN from the adult CNS, presumably by virtue of the blood brain barrier and associated extrusion pumps. Thus, novel opioid receptor mechanisms may need to be hypothesized to account for the high potency of 6BN for preventing opioid withdrawal. In conclusion, 6BN is an attractive compound for development of a preventive therapy for NOWS, and guinea pigs are an ideal model to further test it.

show abstract

Methods for the analysis of multiple endpoints in small populations: A review

Cited by 39 publications

References 132 publications

Leveraging mathematical models of disease dynamics and machine learning to improve development of novel malaria interventions

Leveraging mathematical models of disease dynamics and machine learning to improve development of novel malaria interventions

Efficacy of team-based collaborative care for distressed patients in secondary prevention of chronic coronary heart disease (TEACH): study protocol of a multicenter randomized controlled trial

Pharmacological prevention of neonatal opioid withdrawal in a pregnant guinea pig model

Contact Info

Product

Resources

About