"Spatial heterogeneity of environmental risk in randomized prevention trials: consequences and modeling".
Environmental factors
Integrated Nested Laplace Approximation
Randomized prevention trials
Spatial heterogeneity
Stochastic Partial Differential Equation
Journal
BMC medical research methodology
ISSN: 1471-2288
Titre abrégé: BMC Med Res Methodol
Pays: England
ID NLM: 100968545
Informations de publication
Date de publication:
15 07 2019
15 07 2019
Historique:
received:
26
11
2018
accepted:
21
05
2019
entrez:
17
7
2019
pubmed:
17
7
2019
medline:
11
6
2020
Statut:
epublish
Résumé
In the context of environmentally influenced communicable diseases, proximity to environmental sources results in spatial heterogeneity of risk, which is sometimes difficult to measure in the field. Most prevention trials use randomization to achieve comparability between groups, thus failing to account for heterogeneity. This study aimed to determine under what conditions spatial heterogeneity biases the results of randomized prevention trials, and to compare different approaches to modeling this heterogeneity. Using the example of a malaria prevention trial, simulations were performed to quantify the impact of spatial heterogeneity and to compare different models. Simulated scenarios combined variation in baseline risk, a continuous protective factor (age), a non-related factor (sex), and a binary protective factor (preventive treatment). Simulated spatial heterogeneity scenarios combined variation in breeding site density and effect, location, and population density. The performances of the following five statistical models were assessed: a non-spatial Cox Proportional Hazard (Cox-PH) model and four models accounting for spatial heterogeneity-i.e., a Data-Generating Model, a Generalized Additive Model (GAM), and two Stochastic Partial Differential Equation (SPDE) models, one modeling survival time and the other the number of events. Using a Bayesian approach, we estimated the SPDE models with an Integrated Nested Laplace Approximation algorithm. For each factor (age, sex, treatment), model performances were assessed by quantifying parameter estimation biases, mean square errors, confidence interval coverage rates (CRs), and significance rates. The four models were applied to data from a malaria transmission blocking vaccine candidate. The level of baseline risk did not affect our estimates. However, with a high breeding site density and a strong breeding site effect, the Cox-PH and GAM models underestimated the age and treatment effects (but not the sex effect) with a low CR. When population density was low, the Cox-SPDE model slightly overestimated the effect of related factors (age, treatment). The two SPDE models corrected the impact of spatial heterogeneity, thus providing the best estimates. Our results show that when spatial heterogeneity is important but not measured, randomization alone cannot achieve comparability between groups. In such cases, prevention trials should model spatial heterogeneity with an adapted method. The dataset used for the application example was extracted from Vaccine Trial #NCT02334462 ( ClinicalTrials.gov registry).
Sections du résumé
BACKGROUND
In the context of environmentally influenced communicable diseases, proximity to environmental sources results in spatial heterogeneity of risk, which is sometimes difficult to measure in the field. Most prevention trials use randomization to achieve comparability between groups, thus failing to account for heterogeneity. This study aimed to determine under what conditions spatial heterogeneity biases the results of randomized prevention trials, and to compare different approaches to modeling this heterogeneity.
METHODS
Using the example of a malaria prevention trial, simulations were performed to quantify the impact of spatial heterogeneity and to compare different models. Simulated scenarios combined variation in baseline risk, a continuous protective factor (age), a non-related factor (sex), and a binary protective factor (preventive treatment). Simulated spatial heterogeneity scenarios combined variation in breeding site density and effect, location, and population density. The performances of the following five statistical models were assessed: a non-spatial Cox Proportional Hazard (Cox-PH) model and four models accounting for spatial heterogeneity-i.e., a Data-Generating Model, a Generalized Additive Model (GAM), and two Stochastic Partial Differential Equation (SPDE) models, one modeling survival time and the other the number of events. Using a Bayesian approach, we estimated the SPDE models with an Integrated Nested Laplace Approximation algorithm. For each factor (age, sex, treatment), model performances were assessed by quantifying parameter estimation biases, mean square errors, confidence interval coverage rates (CRs), and significance rates. The four models were applied to data from a malaria transmission blocking vaccine candidate.
RESULTS
The level of baseline risk did not affect our estimates. However, with a high breeding site density and a strong breeding site effect, the Cox-PH and GAM models underestimated the age and treatment effects (but not the sex effect) with a low CR. When population density was low, the Cox-SPDE model slightly overestimated the effect of related factors (age, treatment). The two SPDE models corrected the impact of spatial heterogeneity, thus providing the best estimates.
CONCLUSION
Our results show that when spatial heterogeneity is important but not measured, randomization alone cannot achieve comparability between groups. In such cases, prevention trials should model spatial heterogeneity with an adapted method.
TRIAL REGISTRATION
The dataset used for the application example was extracted from Vaccine Trial #NCT02334462 ( ClinicalTrials.gov registry).
Identifiants
pubmed: 31307393
doi: 10.1186/s12874-019-0759-z
pii: 10.1186/s12874-019-0759-z
pmc: PMC6632226
doi:
Banques de données
ClinicalTrials.gov
['NCT02334462']
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
149Références
Biostatistics. 2003 Jan;4(1):123-42
pubmed: 12925334
PLoS One. 2018 Mar 5;13(3):e0193296
pubmed: 29505578
Malar J. 2018 Oct 5;17(1):347
pubmed: 30290808
Am J Trop Med Hyg. 2015 Oct;93(4):790-7
pubmed: 26324728
PLoS Med. 2016 Apr 12;13(4):e1001993
pubmed: 27071072
Lancet Infect Dis. 2017 May;17(5):498-509
pubmed: 28216244
J Vector Borne Dis. 2016 Oct-Dec;53(4):327-334
pubmed: 28035109
Infect Dis Poverty. 2018 Aug 1;7(1):68
pubmed: 30064504
Clin Vaccine Immunol. 2011 Aug;18(8):1351-7
pubmed: 21715576
Biometrics. 2002 Jun;58(2):287-97
pubmed: 12071401
Atmos Environ (1994). 2012 Aug 1;55:220-228
pubmed: 23439926
Stat Med. 2013 Aug 15;32(18):3224-33
pubmed: 23417920
BMC Med Res Methodol. 2012 Mar 23;12:34
pubmed: 22443286
PLoS One. 2018 Aug 9;13(8):e0201607
pubmed: 30092041
Trans R Soc Trop Med Hyg. 2014 Jan;108(1):13-21
pubmed: 24296325
Malar J. 2017 Sep 19;16(1):380
pubmed: 28927422
Malar J. 2016 May 11;15(1):274
pubmed: 27169513
Spat Spatiotemporal Epidemiol. 2012 Jun;3(2):163-71
pubmed: 22682442
Int J Health Geogr. 2010 Jul 19;9:37
pubmed: 20642827
Malar J. 2011 Dec 13;10:358
pubmed: 22166001
BMC Public Health. 2012 Aug 22;12:687
pubmed: 22914047
Spat Spatiotemporal Epidemiol. 2016 May;17:105-15
pubmed: 27246277
Kidney Int. 2008 Oct;74(8):994-7
pubmed: 18633346
Parasite Epidemiol Control. 2018 Feb 07;3(2):96-105
pubmed: 29988270
Stat Med. 2006 Dec 30;25(24):4279-92
pubmed: 16947139
J Med Entomol. 2007 Nov;44(6):923-9
pubmed: 18047189
Parasit Vectors. 2017 Jul 27;10(1):356
pubmed: 28750651
Environ Sci Pollut Res Int. 2016 Sep;23(17):17664-75
pubmed: 27239676
Malar J. 2011 Apr 10;10:81
pubmed: 21477340
Stat Med. 1994 May 30;13(10):1045-62
pubmed: 8073200
Annu Rev Public Health. 1999;20:145-57
pubmed: 10352854
Stat Med. 2003 Oct 30;22(20):3213-28
pubmed: 14518024
Stat Med. 2019 May 20;38(11):2074-2102
pubmed: 30652356
Malar J. 2016 Aug 30;15(1):442
pubmed: 27577237
Clin Infect Dis. 2016 Apr 15;62(8):964-971
pubmed: 26908796
Int J Appl Earth Obs Geoinf. 2013 Jun;22(100):86-98
pubmed: 24489526
Lancet Infect Dis. 2016 Jun;16(6):674-684
pubmed: 26906747