Aug 21, 2024

Public workspacePerformance of five dynamic models in predicting tuberculosis incidence in three prisons in Thailand

DOI
dx.doi.org/10.17504/protocols.io.n92ld8z8xv5b/v1
  • Nithinan Mahawan1,
  • Wiroj Jiamjarasrangsi2,
  • Thanapoom Rattananupong2,
  • Puchong Sri-Uam3
  • 1School of Nursing, Walailak University, Nakhon Si Thammarat, Thailand;
  • 2Faculty of Medicine, Chulalongkorn University;
  • 3Center for Safety, Health and Environment of Chulalongkorn University
Nithinan Mahawan
Icon indicating open access to content
QR code linking to this content
DOI: dx.doi.org/10.17504/protocols.io.n92ld8z8xv5b/v1
Protocol Citation: Nithinan Mahawan, Wiroj Jiamjarasrangsi, Thanapoom Rattananupong, Puchong Sri-Uam 2024. Performance of five dynamic models in predicting tuberculosis incidence in three prisons in Thailand. protocols.io https://dx.doi.org/10.17504/protocols.io.n92ld8z8xv5b/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: In development
We are still developing and optimizing this protocol
Created: May 05, 2024
Last Modified: August 21, 2024
Protocol Integer ID: 99259
Keywords: five dynamic models, tuberculosis incidence, three prisons, Thailand, Performance
Funders Acknowledgement:
The 90th Anniversary of Chulalongkorn University Fund
Grant ID: no. 035, 1/2021
the FY2021 Thesis Grant for Doctoral Degree Study of the National Research Council of Thailand
Grant ID: N41D640002, 2021
Disclaimer
Download validity-of-five-dynamic-models-of-predicting-tube-dc632zgn.pdfvalidity-of-five-dynamic-models-of-predicting-tube-dc632zgn.pdf

Abstract
This study examined the ability of the following five dynamic models for predicting pulmonary tuberculosis (PTB) incidence in a prison setting. This 1-year prospective cohort study employed 985 cells from three Thai prisons. The results show that a 1% increase in baseline TB transmission probability was associated with a 3%–7% increase in future PTB incidence rate, depending on the dynamic model. The Wells–Riley model exhibited the best performance in terms of both internal and external validity.
Attachments
Icon for validity-of-five-dynamic-models-of-predicting-tube-dc632zgn.pdf
validity-of-five-dyn...
1.2MB
Guidelines
1. The literature review.
2. A walk through survey to assess the suitability of prisons.
Materials
1. The absolute ventilation rate, always used as a surrogate for exhaled air, was assessed by steady-state carbon dioxide (CO2) in parts per million using the Kimo HQ210 with SCOH 112 probe (Sauermann Industries, ZA Bernard Moulinet, Montpon, France).
2. ACH (Q; ACH), as a rule of thumb, is classically used as a metric for assessing infection control risk. It is the total air volume in a room or space that is completely removed and replaced in an hour. In this method, the wind speed (meters/second) in each cell was measured at the location of the opening facing prevailing winds using a hot wire thermo-anemometer and datalogger (Model SDL350, Extech Instruments, Waltham, MA).
Before start
1. The literature review comprises the following topics
1.1 The incidence and prevalence of tuberculosis in prisons globally and specifically in Thailand.
1.2 Evidence indicating that tuberculosis in prisons may serve as a reservoir for tuberculosis in the general population.
1.3 Humanitarian issues concerning prisoners related to tuberculosis or overall health problems.
1.4 Factors associated with the transmission of tuberculosis/drug-resistant tuberculosis in prisons
1.5 Current dynamic models of tuberculosis transmission in prisons, including (1) the Wells-Riley equation (2) Rudnick & Milton-proposed (air change per hour: ACH) model (3) Rudnick & Milton-proposed (litre/second/person: l/s/p) model (4) the Applied Susceptible-Exposed-Infected-Recovered (SEIR) tuberculosis transmission model and (5) Issarow et al model.
2. Contact the academic department of the Department of Corrections for a walk through survey to assess the suitability of three sample prisons for research study (June, 2019).

Exposure assessment
Exposure assessment
Baseline risk assessment of TB transmission was conducted for each cell based on each of the five dynamic models: the classic Wells–Riley equation, Rudnick & Milton’s, and Issarow et al.’s modified Wells–Riley equations, and the applied SEIR model. The detailed procedure has been described previously "Assessment of tuberculosis transmission probability in three Thai prisons based on five dynamic models" [Mahawan N, Rattananupong T, Sri-Uam P, Jiamjarasrangsi W. Assessment of tuberculosis transmission probability in three Thai prisons based on five dynamic models. PLoS One. 2024 Jul 19;19(7):e0305264. https://doi.org/0.1371/journal.pone.0305264 PMID: 39028741; PMCID: PMC11259261.].
Outcome assessment
Outcome assessment
The outcome of interest was PTB occurrence. The diagnostic procedure adhered to accepted clinical practice and included a physical examination, chest radiograph, sputum acid-fast bacilli smear, and nucleic acid amplification testing using the XpertÒ MTB/RIF assay (Cepheid, Sunnyvale, USA) [22]. The incidence of PTB outcome was regarded as a repeated outcome. The data sources used for determining this outcome included documents and computer databases of the central administrative and medical facilities in the prisons, while information regarding the cell location of the PTB inmateswas determined by surveys and interviews with zonal staff and inmates who were prison health volunteers. The data included the number of TB-infected patients in each cell and zone, which represented the incidence at the cell and zonal levels. The magnitude of PTB was reflected by the number of PTB cases per cell and the incidence rate (IR) of PTB, which was calculated as the number of new PTB cases divided by the person-years of observation and reported as the number of new PTB cases per 1,000 person-years.
Statistical analysis
Statistical analysis
A four-step procedure, which encompassed model specification tests, in-sample model fitting, internal validation, and external validation, was implemented to evaluate the performance of five dynamic models. The first three steps were performed using prison B data (referred to as in-sample, n=652), and the final step was conducted with prison A and C data (referred to as out-of-sample, n=333). The outcome variable was the number of PTB cases, while the predictors were the TB transmission probabilities estimated by the five dynamic models.
Results and Conclusions
Results and Conclusions
As determined by a standard procedure for evaluatingthe performance of the five dynamic models in predicting future PTB incidence in prison settings, our study demonstrated that the classic Wells–Riley model exhibited the best performance both in the internal and external validations. Given its requirement for fewer parameters, the Wells–Riley model is therefore recommended as the most appropriate dynamic model, especially for large-scale investigations. However, all the assessed dynamic models still exhibited poor fit in their ability to explain the variation of future PTB incidence in the analytical models. Consequently, additional research is required to confirm our findings and to acquire more detailed information that can be used as input to enhance the performance of these dynamic models.
Protocol references
1.         Cords O, Martinez L, Warren JL, O'Marr JM, Walter KS, Cohen T, et al. Incidence and prevalence of tuberculosis in incarcerated populations: a systematic review and meta-analysis. Lancet Public Health. 2021;6(5):e300-e8. https://doi.org/10.1016/S2468-2667(21)00025-6PMID: 33765455; PMCID: PMC8168455
2.         Velen K, Charalambous S. Tuberculosis in prisons: an unintended sentence? Lancet Public Health. 2021;6(5):e263-e4. https://doi.org/10.1016/S2468-2667(21)00049-9 PMID: 33765454.
3.         Woodman M, Grandjean L. Detecting tuberculosis in prisons: switching off the disease at Its source. Clin Infect Dis. 2021;72(5):778-9. https://doi.org/10.1093/cid/ciaa139 PMID: 32064517; PMCID: PMC7935381.
4.         Walter KS, Martinez L, Arakaki-Sanchez D, Sequera VG, Estigarribia Sanabria G, Cohen T, et al. The escalating tuberculosis crisis in central and South American prisons. Lancet. 2021; 397(10284):1591-6. https://doi.org/10.1016/S0140-6736(20)32578-2 PMID: 33838724; PMCID: PMC9393884.
5.         World prison population list. Institute for Criminal Policy Research. [Cited 2024 March 4]. Available from: https://www.prisonstudies.org/sites/default/files/resources/downloads/wppl_12.pdf.
6.         Haeusler IL, Torres-Ortiz A, Grandjean L. A systematic review of tuberculosis detection and prevention studies in prisons. Global Public Health. 2022;17(2):194-209. https://doi.org/10.1080/17441692.2020.1864753 PMID: 33427099.
7.         World Health Organization. Global Tuberculosis report. 2023. [Cited 2024 March 1]. Available from: https://books.google.co.th/bookshl=th&lr=&id=DTfxEAAAQBAJ&oi=fnd&pg=PA40&dq=Global+tuberculosis+report+2023&ots=o_5vJCq6Nf&sig=FV5PptWFmAEcL7m3qgXA1wp6s64&redir_esc=y#v=onepage&q=Global%20tuberculosis%20report%202023&f=false
8.         Su Y, Garcia Baena I, Harle AC, Crosby SW, Micah AE, Siroka A, et al. Tracking total spending on tuberculosis by source and function in 135 low-income and middle-income countries, 2000-17: a financial modelling study. Lancet Infect Dis. 2020;20(8):929-42. https://doi.org/ 10.1016/S1473-3099(20)30124-9 PMID: 32334658; PMCID: PMC7649746.
9.         Reid MJA, Arinaminpathy N, Bloom A, Bloom BR, Boehme C, Chaisson R, et al. Building a tuberculosis-free world: The Lancet Commission on tuberculosis. Lancet (London, England). 2019;393(10178):1331-84. https://doi.org/10.1016/S0140-6736(19)30024-8 PMID: 30904263.
10.       Urrego J, Ko AI, da Silva Santos Carbone A, Paiao DS, Sgarbi RV, Yeckel CW, et al. The impact of ventilation and early diagnosis on tuberculosis transmission in Brazilian Prisons. Am J Trop Med Hyg. 2015; 93(4):739–46. https://doi.org/10.4269/ajtmh.15-0166 PMID: 26195459; PMCID: PMC4596592.
11.       Cooper-Arnold K, Morse T, Hodgson M, Pettigrew C, Wallace R, Clive J, et al. Occupational tuberculosis among deputy sheriffs in Connecticut: a risk model of transmission. Appl Occup Environ Hyg. 1999;14(11):768-76. https://doi.org/10.1080/104732299302198 PMID: 10590550.
12.       Johnstone-Robertson S, Lawn SD, Welte A, Bekker LG, Wood R. Tuberculosis in a South African prison—a transmission modelling analysis. S Afr Med J. 2011; 101(11):809–13. PMID: 22272961; PMCID: PMC4538692.
13.       Rudnick SN, Milton DK. Risk of indoor airborne infection transmission estimated from carbon dioxide concentration. Indoor Air. 2003;13(3):237-45. https://doi.org/10.1034/j.1600-0668.2003.00189.x PMID: 12950586.
14.       Issarow CM, Mulder N, Wood R. Modelling the risk of airborne infectious disease using exhaled air. J Theor Biol. 2015;372:100-6. https://doi.org/10.1016/j.jtbi.2015.02.010 PMID: 25702940.
15.       Naning H, Al-Darraji HAA, McDonald S, Ismail NA, Kamarulzaman A. Modelling the impact of different tuberculosis control interventions on the prevalence of tuberculosis in an overcrowded prison. Asia Pac J Public Health. 2018;30(3):235-43. https://doi.org/10.1177/1010539518757229 PMID: 29502429.
16.       Sze To GN, Chao CY. Review and comparison between the Wells-Riley and dose-response approaches to risk assessment of infectious respiratory diseases. Indoor Air. 2010;20(1):2-16. https://doi.org/10.1111/j.1600-0668.2009.00621.x PMID: 19874402; PMCID: PMC7202094.
17.       Riley RL. Airborne infection. Am J Med. 1974;57(3):466-75. https://doi.org/10.1016/0002 9343(74)90140-5 PMID: 4212915.
18.       Issarow CM, Mulder N, Wood R. Environmental and social factors impacting on epidemic and endemic tuberculosis: a modelling analysis. R Soc Open Sci. 2018;5(1):170726. https://doi.org/10.1098/rsos.170726 PMID: 29410796; PMCID: PMC5792873.
19.       Noakes CJ, Beggs CB, Sleigh PA, Kerr KG. Modelling the transmission of airborne infections in enclosed spaces. Epidemiol Infect. 2006;134(5):1082-91. https://doi.org/10.1017/S0950268806005875 PMID: 16476170; PMCID: PMC2870476.
20.       Gabriela MGM, Rodrigues P, Hilker FM, Mantilla-Beniers NB, Muehlen M, Cristina Paulo A, et al. Implications of partial immunity on the prospects for tuberculosis control by post-exposure interventions. J Theor Biol. 2007;248(4):608-17. https://doi.org/ 10.1016/j.jtbi.2007.06.005 PMID: 17669435.
21.       Mahawan N, Rattananupong T, Sri-Uam P, Jiamjarasrangsi W. Assessment of tuberculosis transmission probability in three Thai prisons based on five dynamic models. PLoS One. 2024 Jul 19;19(7):e0305264. https://doi.org/0.1371/journal.pone.0305264 PMID: 39028741; PMCID: PMC11259261.
22.       Lewinsohn DM, Leonard MK, LoBue PA, Cohn DL, Daley CL, Desmond E, et al. Official American Thoracic Society/Infectious Diseases Society of America/Centers for Disease Control and Prevention Clinical Practice Guidelines: Diagnosis of Tuberculosis in Adults and Children. Clin Infect Dis. 2017;64(2):111-5. https://doi.org/ 10.1093/cid/ciw778 PMID: 28052967; PMCID: PMC5504475.
23.       Deb P, Norton EC, Manning WG. Health econometrics using Stata. College Station: Stata Press; 2017.
24.       Holodinsky JK, Yu AYX, Kapral MK, Austin PC. Comparing regression modeling strategies for predicting hometime. BMC Med Res Methodol. 2021;21(1):138. https://doi.org/10.1186/s12874-021-01331-9 PMID: 34233616; PMCID: PMC8261957.
25.       Le DD, Gonzalez RL, Matola JU. Modeling count data for health care utilization: an empirical study of outpatient visits among Vietnamese older people. BMC Med Inform Decis Mak. 2021;21(1):265. https://doi.org/ 10.1186/s12911-021-01619-2 PMID: 34525986; PMCID: PMC8442353.
26.       Manjón M, Martínez O. The Chi-Squared Goodness-of-Fit Test for Count-Data Models. The Stata Journal. 2014;14(4):798-816.
27.       Lewis F, Butler A, Gilbert L. A unified approach to model selection using the likelihood ratio test. Methods in Ecology and Evolution. 2011;2:155–62. https://doi.org/ 10.1111/j.2041-210X.2010.00063.x.
28.       Lorah J, Womack A. Value of sample size for computation of the Bayesian information criterion (BIC) in multilevel modeling. Behav Res Methods. 2019;51(1):440-50. https://doi.org/ 10.3758/s13428-018-1188-3 PMID: 30684229.
29.       Raftery AE. Bayesian Model Selection in Social Research. Sociological Methodology. 1995;25:111-63.
30.       Van Calster B, McLernon DJ, van Smeden M, Wynants L, Steyerberg EW. Calibration: the Achilles heel of predictive analytics. BMC Med. 2019;17(1):230. https://doi.org/10.1186/s12916-019-1466-7 PMID: 31842878; PMCID: PMC6912996.
31.       Matthay EC, Glymour MM. A Graphical Catalog of Threats to Validity: Linking Social Science with Epidemiology. Epidemiology. 2020;31(3):376-84. https://doi.org/ 10.1097/EDE.0000000000001161 PMID: 31977593; PMCID: PMC7144753.
32.       Iwagami M, Matsui H. Introduction to Clinical Prediction Models. Ann Clin Epidemiol. 2022;4(3):72-80. https://doi.org/ 10.37737/ace.22010 PMID: 38504943; PMCID: PMC10760493.