Mobiele menu

Promoting tailored healthcare: improving methods to investigate subgroup effects in treatment response when having multiple individual participant datasets

Projectomschrijving

Behandeling op maat
Iedere patiënt verwacht de behandeling te krijgen die het best bij hem of haar past. Studies naar behandelingen rapporteren vaak één, gemiddeld effect. De gemiddelde patiënt bestaat niet omdat patiënten onderling verschillen in zowel gunstige reacties als bijwerkingen. Deze verschillen heten subgroep-effecten. Het aantonen van subgroepen is complex. Onderzoek naar subgroep-effecten kan beter wanneer alle studies worden samengenomen en opnieuw worden onderzocht op aanwezigheid van subgroep-effecten. Dit heet een individuele patiënten-data (IPD) meta-analyse. Dit type onderzoek kent nog veel onduidelijkheden. In dit project ontwikkelen wij methoden om subgroepen op te sporen via IPD meta-analyses. Deze kennis zal worden samengevat in een richtlijn voor onderzoekers en beleidsmakers hoe subgroepen te onderzoeken in IPD meta-analyses. Dit leidt tot meer betrouwbare kennis of relevante subgroepen bestaan met voordelen voor zowel individuele patiënten als voor de gehele gezondheidszorg.

Producten

Titel: The use of prognostic scores for causal inference with general treatment regimes
Auteur: Nguyen TL, Debray TPA.
Magazine: Statistics in Medicine
Titel: Multiple imputation for multilevel data with continuous and binary variables.
Auteur: Audigier V, White IR, Jolani S, Debray T P, Quartagno M, Carpenter J, van Buren S, Resche-Rigon, M.
Magazine: Statistical Science
Titel: The development of CHAMP: a checklist for the appraisal of moderators and predictors.
Auteur: van Hoorn R, Tummers M, Booth A, Gerhardus A, Rehfuess E, Hind D, Bossuyt PM, Welch V, Debray TPA, Underwood M, Cuijpers P, Kraemer H, van der Wilt GJ, Kievit W.
Magazine: BMC Medical Research Methodology
Titel: Framework for the synthesis of non-randomised studies and randomised controlled trials: a guidance on conducting a systematic review and meta-analysis for healthcare decision making.
Auteur: Sarri G, Patorno E, Yuan H, Guo JJ, Bennett D, Wen X, Zullo AR, Largent J, Panaccio M, Gokhale M, Moga DC, Ali MS, Debray TPA.
Magazine: BMJ Evidence-Based Medicine
Titel: National Protocol for Model-Based Selection for Proton Therapy in Head and Neck Cancer.
Auteur: Langendijk JA, Hoebers FJP, de Jong MA, Doornaert P, Terhaard CHJ, Steenbakkers RJHM, Hamming-Vrieze O, van de Kamer JB, Verbakel WFAR, Keskin-Cambay F, Reitsma JB, van der Schaaf A, Boersma LJ, Schuit E.
Magazine: International Journal of Particle Therapy
Titel: On the aggregation of published prognostic scores for causal inference in observational studies.
Auteur: Nguyen TL, Collins GS, Pellegrini F, Moons KGM, Debray TPA.
Magazine: Statistics in Medicine
Titel: Calculating the power to examine treatment-covariate interactions when planning an individual participant data meta-analysis of randomized trials with a binary outcome
Auteur: Riley RD, Hattle M, Collins GS, Whittle R, Ensor J.
Magazine: Statistics in Medicine
Titel: Predicting personalised absolute treatment effects in individual participant data meta-analysis: An introduction to splines
Auteur: Belias M, Rovers MM, Hoogland J, Reitsma JB, Debray TPA, IntHout J
Magazine: Research Synthesis Methods
Titel: Handling missing predictor values when validating and applying a prediction model to new patients.
Auteur: Hoogland J, van Barreveld M, Debray TPA, Reitsma JB, Verstraelen TE, Dijkgraaf MGW, Zwinderman AH
Magazine: Statistics in Medicine
Titel: Identifying adults with acute rhinosinusitis in primary care that benefit most from antibiotics: protocol of an individual patient data meta-analysis using multivariable risk prediction modelling.
Auteur: Venekamp RP, Hoogland J, van Smeden M, Rovers MM, De Sutter AI, Merenstein D, van Essen GA, Kaiser L, Liira H, Little P, Bucher HC, Reitsma JB.
Magazine: BMJ Open
Titel: Statistical approaches to identify subgroups in meta-analysis of individual participant data: a simulation study.
Auteur: Belias M, Rovers MM, Reitsma JB, Debray TPA, IntHout J.
Magazine: BMC Medical Research Methodology
Titel: Guidance from key organisations on exploring, confirming and interpreting subgroup effects of medical treatments: a scoping review.
Auteur: Wijn SRW, Rovers MM, Le LH, Belias M, Hoogland J, IntHout J, Debray T, Reitsma JB.
Magazine: BMJ Open
Titel: Personalizing treatment in end-stage kidney disease: deciding between haemodiafiltration and haemodialysis based on individualized treatment effect prediction.
Auteur: van Kruijsdijk RCM, Vernooij RWM, Bots ML, Peters SAE, Dorresteijn JAN, Visseren FLJ, Blankestijn PJ, Debray TPA; HDF Pooling Project investigators.
Magazine: Clinical Kidney Journal
Titel: Hearing Preservation in Cochlear Implant Surgery: A Meta-Analysis.
Auteur: Snels C, IntHout J, Mylanus E, Huinck W, Dhooge I.
Magazine: Otology Neurotology
Titel: Individual participant data meta-analysis to examine interactions between treatment effect and participant-level covariates: Statistical recommendations for conduct and planning.
Auteur: Riley RD, Debray TPA, Fisher D, Hattle M, Marlin N, Hoogland J, Gueyffier F, Staessen JA, Wang J, Moons KGM, Reitsma JB, Ensor J.
Magazine: Statistics in Medicine
Titel: A tutorial on individualized treatment effect prediction from randomized trials with a binary endpoint.
Auteur: Hoogland J, IntHout J, Belias M, Rovers MM, Riley RD, E Harrell F Jr, Moons KGM, Debray TPA, Reitsma JB.
Magazine: Statistics in Medicine
Titel: The Impact of the Extent of Surgery on the Long-Term Outcomes of Patients with Low-Risk Differentiated Non-Medullary Thyroid Cancer: A Systematic Meta-Analysis
Auteur: Bojoga A, Koot A, Bonenkamp J, de Wilt J, IntHout J, Stalmeier P, Hermens R, Smit J, Ottevanger P, Netea-Maier R.
Magazine: Journal of Clinical Medicine
Titel: Maturation of GFR in Term-Born Neonates: An Individual Participant Data Meta-Analysis
Auteur: Smeets NJL, IntHout J, van der Burgh MJP, Schwartz GJ, Schreuder MF, de Wildt SN.
Magazine: Journal of the American Society of Nephrology
Titel: Risk of Peritoneal Carcinomatosis After Risk-Reducing Salpingo-Oophorectomy: A Systematic Review and Individual Patient Data Meta-Analysis.
Auteur: Steenbeek MP, van Bommel MHD, Bulten J, Hulsmann JA, Bogaerts J, Garcia C, Cun HT, Lu KH, van Beekhuizen HJ, Minig L, Gaarenstroom KN, Nobbenhuis M, Krajc M, Rudaitis V, Norquist BM, Swisher EM, Mourits MJE, Massuger LFAG, Hoogerbrugge N, Hermens RPMG, IntHout J, de Hullu JA
Magazine: Journal of Clinical Oncology
Titel: Combining individual patient data from randomized and non-randomized studies to predict real-world effectiveness of interventions
Auteur: Seo M, Debray TP, Ruffieux Y, Gsteiger S, Bujkiewicz S, Finckh A, Egger M, Efthimiou O.
Magazine: Statistical Methods in Medical Research
Titel: Antiplatelet Therapy After Noncardioembolic Stroke.
Auteur: Greving JP, Diener HC, Reitsma JB, Bath PM, Csiba L, Hacke W, Kappelle LJ, Koudstaal PJ, Leys D, Mas JL, Sacco RL, Algra A.
Magazine: Stroke
Titel: Molecular Processes in Stress Urinary Incontinence: A Systematic Review of Human and Animal Studies.
Auteur: Post WM, Widomska J, Grens H, Coenen MJH, Martens FMJ, Janssen DAW, IntHout J, Poelmans G, Oosterwijk E, Kluivers KB.
Magazine: International Journal of Molecular Sciences
Titel: Real-time imputation of missing predictor values improved the application of prediction models in daily practice.
Auteur: Nijman SWJ, Groenhof TKJ, Hoogland J, Bots ML, Brandjes M, Jacobs JJL, Asselbergs FW, Moons KGM, Debray TPA.
Magazine: Journal of Clinical Epidemiology
Titel: R package metamisc - https://cran.r-project.org/package=metamisc
Auteur: Thomas Debray, Valentijn de Jong
Titel: JASP module - https://jasp-stats.org/2022/05/24/meta-analysis-of-prediction-model-performance/
Auteur: František Bartoš, Thomas Debray

Verslagen


Eindverslag

Gerandomiseerde klinische studies (RCTs) vormen de basis voor uitspraken over welke therapie de voorkeur verdient: op groepsniveau is behandeling A beter dan B. In de praktijk kunnen er subgroepen van patiënten zijn waarbij de grootte van het behandeleffect (verschil tussen A en B) groter of kleiner is. Hierdoor kan de optimale keuze van een behandeling voor een individuele patiënt anders uitvallen.

In de praktijk bestaan er grote verschillen in strategieën om subgroepen te onderzoeken in interventie-onderzoek en zijn er verhitte discussies onder experts wat de beste aanpak is. Het ultieme doel van dit TOP-project was om de huidige verschillen terug te dringen en lacunes in kennis te dichten.

Wij hebben daarom een groot aantal statistische strategieën om subgroepen te onderzoeken onderling vergeleken, vooral in situaties waar men de beschikking heeft over de individuele patiëntendata van meerdere studies (technische naam: IPD meta-analyse).

Onze aanbevelingen leiden tot een betere aanpak om de absolute grootte van behandeleffecten te schatten. Dit zal de besluitvorming omtrent de beste keuze van een behandeling bij een individuele patiënt verbeteren. Onze aanpak is generiek van aard en kan worden ingezet bij alle typen aandoeningen en voor alle soorten behandelingen.

Samenvatting van de aanvraag

Individuals differ in their response to treatment. Personalized or patient-centered healthcare involves tailoring therapeutic decisions to individuals based on patient and disease characteristics. The current widespread initiatives for sharing individual participant data (IPD) will create unique opportunities to investigate whether subgroups differ in their response to treatments. Benefits of using IPD rather than traditional meta-analysis arise from standardization of subgroups and outcomes across studies, the higher validity of subgroup findings through proper adjustment for differences in patient and study characteristics, and the increased possibilities to search for subgroups based on multiple characteristics. In our recent papers in PlosMed, JAMA and BMJ we however identified several remaining challenges in IPD meta-analyses leading to diversity and uncertainty on how to perform and interpret subgroup analyses based on IPD. These gaps include: (i) lack of insight and evidence why and when various distinct statistical subgroup approaches lead to conflicting results; (ii) uncertainty how to combine IPD and aggregate data; (iii) unclear how to incorporate differences in design and population between studies; (iv) uncertainty how to best calculate the probability that a new trial will find the subgroup of interest. In view of the expected rise in open access data, guidance on appropriate methodology is urgently needed to improve the use and uptake of IPD meta-analyses and their findings. Our overall aim is to evaluate and improve prevailing approaches, and to develop novel methods where needed, for investigating and interpreting subgroup effects in treatment response when multiple IPD sets are available. It should lead to more credible evidence about whether a relevant subgroup effect exists. To achieve this, we have the following key objectives: A. To compare the performance of various proposed methodological approaches for investigating clinically relevant subgroup effects in response to treatment when having multiple IPD-sets. B. To improve and develop strategies for dealing with challenges and opportunities specific to IPD meta-analysis. C. To generate guidance on how to apply these new and improved methodologies, and to develop a multidimensional instrument to assist health care professionals how to grade the credibility of subgroup effects reported by IPD meta-analyses. In our previous work we have identified the relevant variations in methodological approaches for investigating subgroups. The performance of these prevailing approaches will be compared in our IPD-sets and in extensive simulations. Both institutions together possess at least 12 large international empirical IPD-sets on various medical conditions and treatments (number of studies varying between 3 and 24, and on average 3,639 participants). The big advantage of using simulated data is that we will know the “truth” with regard to the subgroup effects, whereas we can control and deliberately vary other key factors. Key factors that we will examine come from (i) differences in the type of subgroup question (e.g. subgroup defined by single or multiple characteristics, measurement level of characteristics, number of subgroups examined) and (ii) variation in the underlying data (e.g. number and sample size of included studies, magnitude and consistency of subgroup effect). Differences in performance of the prevailing approaches will be analyzed in these scenarios. Performance measures of interest are: the bias and precision in the estimated subgroup effect, the ability to detect true effects from noise signals, the frequency of false positive findings, and the consistency in performance over various scenarios. Particular challenges specific to IPD meta-analyses that will be investigated include handling missing data (including studies not providing IPD), integrating randomized and observational data, and investigating inconsistencies in subgroup effects across studies. Each IPD meta-analysis will face similar challenges, but the degree in which they will be present and their potential impact will vary. We will therefore develop a multidimensional instrument that will help health care professionals to judge the credibility of subgroup effects reported in an IPD meta-analysis. Our TOP proposal will provide an integrated framework with corresponding guidance on how to perform, report and judge IPD meta-analyses investigating subgroup effects. Our novel framework will be comprehensive and context-specific in order to address subgroup effects across the whole range of increasing complexity of subgroups and variation in underlying data. If performed appropriately, IPD meta-analyses will promote tailored healthcare for all types of interventions and across all medical fields. Our proposal will lead to better use of data from existing and future studies to the benefit of individual patients and healthcare in general.

Onderwerpen

Kenmerken

Projectnummer:
91215058
Looptijd: 100%
Looptijd: 100 %
2016
2022
Onderdeel van programma:
Gerelateerde subsidieronde:
Projectleider en penvoerder:
Prof. dr. M.M. Rovers
Verantwoordelijke organisatie:
Universitair Medisch Centrum Utrecht