Benjamin A Goldstein1,2, Matthew Phelan2, Neha J Pagidipati2,3, Rury R Holman4, Michael J Pencina1,2, Elizabeth A Stuart5,6. 1. Department of Biostatistics & Bioinformatics, Duke University, Durham, North Carolina, USA. 2. Center for Predictive Medicine, Duke Clinical Research Institute, Durham, North Carolina, USA. 3. Department of Medicine, Duke Clinical Research Institute, Center for Predictive Medicine, Duke University, Durham, North Carolina, USA. 4. Endocrinology and Metabolism, University of Oxford, Oxford, United Kingdom. 5. Department of Biostatistics John Hopkins University, Baltimore, Maryland, USA. 6. Department of Mental Health, John Hopkins University, Baltimore, Maryland, USA.
Abstract
OBJECTIVE: Participants enrolled into randomized controlled trials (RCTs) often do not reflect real-world populations. Previous research in how best to transport RCT results to target populations has focused on weighting RCT data to look like the target data. Simulation work, however, has suggested that an outcome model approach may be preferable. Here, we describe such an approach using source data from the 2 × 2 factorial NAVIGATOR (Nateglinide And Valsartan in Impaired Glucose Tolerance Outcomes Research) trial, which evaluated the impact of valsartan and nateglinide on cardiovascular outcomes and new-onset diabetes in a prediabetic population. MATERIALS AND METHODS: Our target data consisted of people with prediabetes serviced at the Duke University Health System. We used random survival forests to develop separate outcome models for each of the 4 treatments, estimating the 5-year risk difference for progression to diabetes, and estimated the treatment effect in our local patient populations, as well as subpopulations, and compared the results with the traditional weighting approach. RESULTS: Our models suggested that the treatment effect for valsartan in our patient population was the same as in the trial, whereas for nateglinide treatment effect was stronger than observed in the original trial. Our effect estimates were more efficient than the weighting approach and we effectively estimated subgroup differences. CONCLUSIONS: The described method represents a straightforward approach to efficiently transporting an RCT result to any target population.
OBJECTIVE:Participants enrolled into randomized controlled trials (RCTs) often do not reflect real-world populations. Previous research in how best to transport RCT results to target populations has focused on weighting RCT data to look like the target data. Simulation work, however, has suggested that an outcome model approach may be preferable. Here, we describe such an approach using source data from the 2 × 2 factorial NAVIGATOR (Nateglinide And Valsartan in Impaired Glucose Tolerance Outcomes Research) trial, which evaluated the impact of valsartan and nateglinide on cardiovascular outcomes and new-onset diabetes in a prediabetic population. MATERIALS AND METHODS: Our target data consisted of people with prediabetes serviced at the Duke University Health System. We used random survival forests to develop separate outcome models for each of the 4 treatments, estimating the 5-year risk difference for progression to diabetes, and estimated the treatment effect in our local patient populations, as well as subpopulations, and compared the results with the traditional weighting approach. RESULTS: Our models suggested that the treatment effect for valsartan in our patient population was the same as in the trial, whereas for nateglinide treatment effect was stronger than observed in the original trial. Our effect estimates were more efficient than the weighting approach and we effectively estimated subgroup differences. CONCLUSIONS: The described method represents a straightforward approach to efficiently transporting an RCT result to any target population.
Authors: Issa J Dahabreh; Sarah E Robertson; Eric J Tchetgen; Elizabeth A Stuart; Miguel A Hernán Journal: Biometrics Date: 2019-06-21 Impact factor: 2.571
Authors: John J McMurray; Rury R Holman; Steven M Haffner; M Angelyn Bethel; Björn Holzhauer; Tsushung A Hua; Yuri Belenkov; Mitradev Boolell; John B Buse; Brendan M Buckley; Antonio R Chacra; Fu-Tien Chiang; Bernard Charbonnel; Chun-Chung Chow; Melanie J Davies; Prakash Deedwania; Peter Diem; Daniel Einhorn; Vivian Fonseca; Gregory R Fulcher; Zbigniew Gaciong; Sonia Gaztambide; Thomas Giles; Edward Horton; Hasan Ilkova; Trond Jenssen; Steven E Kahn; Henry Krum; Markku Laakso; Lawrence A Leiter; Naomi S Levitt; Viacheslav Mareev; Felipe Martinez; Chantal Masson; Theodore Mazzone; Eduardo Meaney; Richard Nesto; Changyu Pan; Rudolf Prager; Sotirios A Raptis; Guy E H M Rutten; Herbert Sandstroem; Frank Schaper; Andre Scheen; Ole Schmitz; Isaac Sinay; Vladimir Soska; Steen Stender; Gyula Tamás; Gianni Tognoni; Jaako Tuomilehto; Alberto S Villamil; Juraj Vozár; Robert M Califf Journal: N Engl J Med Date: 2010-03-14 Impact factor: 91.245
Authors: Ashley L Buchanan; Michael G Hudgens; Stephen R Cole; Katie R Mollan; Paul E Sax; Eric S Daar; Adaora A Adimora; Joseph J Eron; Michael J Mugavero Journal: J R Stat Soc Ser A Stat Soc Date: 2018-02-26 Impact factor: 2.483