Estimating spatial disease rates from health statistics without geographic identifiers ====================================================================================== * Javier Cortes-Ramirez * Juan D. Wilches-Vega * Ruby N. Michael * Vishal Singh * Olga M. Paris-Pineda ## Abstract Acute respiratory infections (ARI) statistics in Cúcuta, Colombia are reported for each health service or health care providers rather than for residence area. Although official statistics are important sources of data to support evidence-based decisions for at-risk communities, these are of limited use if the geographical distribution of diseases cannot be identified. This study aims to calculate the rate of ARI in each of the Cúcuta’s urban sections using a spatial analysis of the distribution of the HCP. The spatial scope (geographical area of influence) of the health care providers was established from their spatial distribution and the population accessing their services. Three spatial aggregation levels were established considering the spatial scope of the primary, intermediate and tertiary health care providers. The ARI cases per urban section were calculated according to the spatial distribution of health care providers and the proportion of population, per urban section in each level. A further spatial analysis included the calculation of spatial rates and hotspots of ARI. There were 97 health care providers providing health services in 31, 20 and 47 urban sections in levels 1, 2 and 3 respectively. A higher spatial rate of ARI was found in urban sections in central south; central west; north and northwest; northeast; central east; and central Cúcuta. Hotspots of higher risk were in clustered in urban sections in central south and west Cúcuta and three isolated urban sections in central and northwest Cúcuta. The spatial distribution of health services can be used to calculate health indicators at the census district level. This methodology can be used in socioeconomic contexts where geographic identifiers are not attached to health statistics. Key words * Census district boundaries * spatial epidemiology * Local Indicators of Spatial Autocorrelation -LISA * morbidity * Cucuta -North Santander * hotspots analysis ## 1. Introduction Effective decision making in public health is underpinned by using the best available evidence and research on risk factors determinants of disease. An important condition to conduct effective analysis in research is the availability and quality of data that can be used in various quantitative methodological frameworks [1]. Statistics and health reports are especially important for epidemiological studies to identify the association of risk factors with health outcomes in the general population and support potential prevention strategies. However, statistical reports can be inadequate to provide the best evidence unless they are designed for specific analyses for at-risk populations [2]. Decisions related to diagnostic and therapeutic interventions are usually based on high-quality research studies such as clinical trials. In contrast, public health decisions, especially by local authorities, are often based on the review of basic statistics with limited academic rigour [3]. The lack of high-quality data can favour the role that politics and political ideology have on shaping local governments initiatives in public health which can impact effective evidence-based decision making [4]. This can be accentuated in low and middle-income countries where the public health sector has limited access to resources for population health research [5]. Public health departments usually collect statistics from health providers such as hospitals, general practices and clinics and health insurance providers to monitor public health. The utility of these data depends on the inclusion of individual identifiers or categories such as socioeconomic status or residential areas. Public health data linked to small geographical areas such as counties or local government areas can be used to identify the distribution of diseases or mortality in cities or larger regions [6]. In addition, census information for geographical areas can be linked to the public health data to calculate proportional rates of cases by population and the association of these proportions with other indicators such as socioeconomic and education indexes. This comes with challenges, in that basic health statistics may include unformatted and incomplete data and a lack of information at the individual level. This paucity of information is more evident in resource-constrained countries with up to 152 low-income and middle-income countries having no data or poor-quality health data [7]. The North Santander region in Colombia has an important burden of disease due to respiratory infections with the capital city, Cúcuta, accounting for more than 80% of the respiratory morbidity in the region [8, 9]. Cúcuta has a population of 711,715 inhabitants in an area of 1,119 Km2 that access a complex network of public and private health services of primary, intermediate and tertiary healthcare. The population is distributed in 460 geographical areas defined by census district boundaries called urban sections (USEC). Despite the large number of USEC and the socioeconomic diversity between these districts, the statistics of acute respiratory infections (ARI) can be obtained only at the health centre and hospital level. The usability of the statistics of ARI in Cúcuta for analyses and assessment to support evidence-based decision-making in public health is very limited because it is not linked to the distribution of the population across the city. This prevents the calculation of useful health indicators such as the rate of ARI per USEC. Although the statistics of ARI in Cúcuta are not provided at a geographical level, the geographical distribution of the health care providers (HCP) can be used to make estimations of the number of cases occurring in each USEC. Spatial statistics can be used to estimate the distribution of observations in well-defined geographic settings using assumptions based on known characteristics of the population and health services to interpolate missing observations and compensate for the lack of data [10, 11]. A similar approach can be used to estimate the cases of ARI in Cúcuta’s USEC, considering the spatial distribution of the HCP and the population in the USEC. The aim of this study is to design a method to calculate and map the spatial rate of ARI per USEC in Cúcuta, using statistics of cases reported at the HCP level, and to identify the hotspots of higher risk of ARI. ## 2. Materials and Methods Data on the number of medical consultations with a diagnosis of ARI in the period 01-January-2018 to 31-December-2018 categorised by HCP, were obtained from the Public Health Department. As single ARI cases were not linked to geographical areas, the number of ARI cases per USEC in Cúcuta was estimated by categorising HCP according to their spatial distribution. This was done by identifying the USEC around each HCP according to the spatial extent of areas receiving their services (i.e., spatial scope). Three types of HCP were identified: primary public HCP that provide primary health services to population located in USEC in their proximity; intermediate public HCP that provide intermediate health care to people in large areas of the city; and tertiary hospitals; and nonselective private HCP and general practitioners (GPs) that provide health services to people from any area in the city, located mostly near the central business district (Figure 1). ![Figure 1.](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2022/06/06/2022.04.18.22274002/F1.medium.gif) [Figure 1.](http://medrxiv.org/content/early/2022/06/06/2022.04.18.22274002/F1) Figure 1. Spatial distribution of the Cúcuta Urban Sections and the primary (purple dots) and intermediate (squares) public health care providers (HCP), and hospitals and hospitals and nonselective private HCP and GPs (green dots) Zones for three independent levels of spatial aggregation were established based on the characteristics and spatial scope of the public HCP: Level 1 zones were the most tightly defined and included the USEC within the spatial scope of each single public primary HCP. Level 2 zones included the USEC in the spatial scope of the three public intermediate HCP. The Level 3 zone included all Cúcuta’s USEC, defined as the spatial scope of the hospitals and nonselective private HCPs (Figure 2). The spatial scope for level 1 and 2 zones was established in ARCmap (v.10.6) using the nearest neighbour algorithm (i.e., identifies the USEC that is geographically closest to the HCP). Some level 2 zones were edited manually to adjust for the USEC connectivity according to the bus routes and the Pamplonita river shape. For each level, the total ARI cases per zone were calculated as the sum of cases in the HCP within the zone. ![Figure 2.](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2022/06/06/2022.04.18.22274002/F2.medium.gif) [Figure 2.](http://medrxiv.org/content/early/2022/06/06/2022.04.18.22274002/F2) Figure 2. Spatial distribution of the geographic zones for levels of aggregation 2 (a) and 1 (b). Dots represent primary and intermediate public health care providers respectively. To estimate the number of ARI cases per USEC, a weight value was assigned according to the proportion of population of each USEC, for each zone, for each level. For example, the weight value of a USEC with 500 people in a level 1 zone with a total zone population of 5000 people is 500 ÷ 5000 = 0.1. For each level, the total ARI cases per zone were multiplied by the weight value to obtain the ARI cases per USEC. Each USEC would have ARI cases calculated for each of the three levels, therefore, the total ARI cases per USEC was calculated as the sum of ARI cases in the three levels. Once the total ARI cases were calculated for each USEC, the incidence rate of ARI per USEC can be calculated as the total ARI divided by the USEC population. However, the ARI cases per USEC calculated on the assumption of an even distribution of cases per population can produce distortion of the estimated ARI cases in addition to difficulties to compare the rates between USEC due to underlying differences between the USECs population. To overcome these issues, the properties of the ARI rates can be improved by spatial smoothing the rate estimates in each USEC by borrowing information from neighbour USEC, which is conventionally implemented using Bayesian principles, particularly the Empirical Bayes (EB) approach [12]. To calculate the spatial rate of ARI per USEC (ARI-sr) in Cúcuta, spatial EB rates were estimated with Geoda (v. 1.18). A similar approach can be used to assess the spatial trend ARI in the Cúcuta’s USEC to identify statistically significant hotspots of higher risk of ARI [13]. The hotspots of ARI in Cucuta were identified using the local Moran’s I test with EB rate in Geoda. Maps were drawn in Rstudio ver. 1.2 using the R-package Tmap [14]. This study was approved by the ethics committee of the University of Santander, record VII-FT-025-UDES, 021 25/06/2019. ## 3. Results There were 118,469 cases of ARI in Cúcuta over the study period. Of these, 38,236 (32.3%) were reported by hospitals; 43,173 (36.4%) were reported by intermediate HCP; and 37,060 (31.3%) reported by primary health centres or GPs. Table 1 shows the number of ARI cases for each spatial level of aggregation. View this table: [Table 1.](http://medrxiv.org/content/early/2022/06/06/2022.04.18.22274002/T1) Table 1. Cases of acute respiratory infections per Health Care Provider and spatial level of aggregation Figure 2 shows the zones for levels 1 and 2 of spatial aggregation. Three level 2 zones were identified from the area of influence of the intermediate public HCP (Figure 2a) and 26 level 1 zones were identified from the area of influence of the public primary HCP (Figure 2b). There were 97 HCP that included 5 hospitals that provided health care to the population in all USEC (level 3); 19 intermediate HCP of which 14 provided services to the population in all USEC and 5 provided services to the population in USEC in level 2 zones; 74 Primary HCP or GPs of which 28 provided services to population in all USEC; 15 provided services to population in USEC in level 2 zones, and 31 provided services to population in USEC in level 1 zones (details in the appendix). Figure 3.a shows the distribution of the spatial rate of ARI (ARI-sr). The ARI-sr ranged from 0.16 to 3.65 per inhabitant with three USEC having more than one ARI case per inhabitant (i.e., ARI-sr >1). Higher ARI-sr were found in USEC in central south; central west; north and northwest; northeast; central east; and central Cúcuta, compared to the whole of Cúcuta. Figure 3.b shows the location of USEC with significantly higher risk of ARI that have neighbour USEC also with significantly higher ARI-sr (i.e., statistically significant hotspots of higher risk). Hotspots of ARI were found in central south and west Cúcuta and central and northwest Cúcuta. ![Figure 3.](http://medrxiv.org/http://medrxiv.stage.highwire.org/content/medrxiv/early/2022/06/06/2022.04.18.22274002/F3.medium.gif) [Figure 3.](http://medrxiv.org/content/early/2022/06/06/2022.04.18.22274002/F3) Figure 3. Spatial rates of Acute Respiratory Infections (a) and statistically significant hotspots of risk of ARI (b) in Cúcuta, 2018. ## 4. Discussion In this study, we calculated and mapped the highest spatial rates of Acute Respiratory Infections (ARI) in multiple urban sections (USEC) in central and north regions of Cúcuta in 2018 and identified the statistically significant hotspots of higher risk of ARI. We used a GIS to identify the USECs within the area of influence of each health care provider, implementing a novel methodology to assign the number of ARI cases to each USEC according to three geographical levels of aggregation in the city. Our method shows that health data from health care providers can be used to estimate the rate of health outcomes such as ARI, per small geographical areas, using the census district boundaries when the spatial scope of the HCP is considered. This is the first study to estimate the rate of respiratory infections in census district areas in a major city in the Colombia northeast region. An important advantage of having indicators at the census districts level is the possibility of assessing their relationship with other indicators such as indexes of socioeconomic development and demographics characteristics including ethnicity, gender and age groups. This can increase the utility of the morbidity rates as calculated in this study to produce more specific and complex analyses. For example, correlation tests can be used to explore predictive relationships between health outcomes and some risk factors of interest [15]. These measures are often considered to better understanding the potential effect of environmental exposures on specific diseases and their relationship with public health interventions [16, 17]. For example, the spatial rate of ARI can be correlated with important factors such as the index of multidimensional poverty and education levels that are measured at the USEC level in Cúcuta. Having data of multiple factors can also be used for multivariate analysis using a regression model to estimate the relative influence of one or more predictors on the ARI rates and to identify outliers or anomalies [18]. This is particularly important in studies at a geographical level because of the risk of ecological bias, where confounding factors at the group level can produce spurious associations. The use of spatial statistic techniques such as the design of mixed-effects spatial regressions using the spatial units as the random effects can reduce the risk of ecological bias [19]. The calculation of rates presented in this study can be extended to identify potentially associated risk factors by designing further multivariate analyses. The availability of relevant and adequate data is a basic condition for public health departments to deal with many challenges including efficient health care provision and planning. The analysis of data using the characteristics of their spatial distribution is an important approach to address public health challenges with the help of increasingly used tools such as GIS and geostatistical software [20]. We used GIS techniques and spatial data science software in combination with assumptions on the Cúcuta health care institutions and population characteristics to estimate morbidity indicators at a small geographical level. Spatial analyses of health data have several advantages that can be used to support decision making including the identification of spatial autocorrelation and detection of spatial clusters [21]. The potential distortion introduced by having assumptions about the distribution of cases in geographical urban areas can be reduced by the smoothing effect that is incorporated in spatial rate calculations to increase the robustness of the estimates [22]. Whereas the health department of cities such as Cúcuta might not have data on morbidity per census districts levels, their calculation using the method used in this study can be incorporated as an instrument to support planning strategies or prioritise resources in areas of interest. Although we used census districts to build a health indicator such as the spatial rate of ARI, this methodology can be applied to other types of data such as community participation and political or electoral involvement. This extends to measuring community participation to support decision-making in public health which is an significant challenge for the public health and academic sectors that can bring important benefits because communities have a key role in the planning and organising of health care [23, 24]. On the other hand, in some socioeconomic contexts, especially middle and low-income countries, electoral and administrative districts do not overlap geographically or there are significant gaps between rural and urban areas, making it harder to identify patterns that align with the sociodemographic distribution of the population [25]. The use of GIS techniques to improve the resolution of health statistics used in this study is a methodological approach that can be implemented to do analyses of data in the health and social sciences. This can also facilitate the implementation of other emerging tools to identify and measure socio-political determinants in public health such as System Dynamics Modelling [26] and Structural Equation Modelling [27]. Our study had some limitations, especially regarding the data for specific individual markers such as age or gender that would have allowed the estimation of age and sex-adjusted spatial rates. Age-adjusted rates are used because crude rates of populations with different frequencies in specific age groups (i.e., age structure) can have great variability [28]. To reduce the potential effect of the age structure of the population, we calculated spatial rates to smooth the estimates in the spatial units (i.e., USEC), removing the potential effect of outliers and taking into account the spatial trend or the crude rate to increase robustness of the estimates [12]. However, the standardisation of rates by age and or sex should be considered in further analyses using these data. We had access to health data of 2018 and could match these data with the 2018 census districts although the inclusion of data matching more than one census year could have allowed identifying a temporal trend, producing stronger estimates. However, this would be effective in contexts with a periodic census which is not the case in Colombia where the previous census was in 2005, and the spatial trend of the census districts is less reliable. ## 5. Conclusions Basic health statistics can be used to estimate health indicators at the same geographical level of the census districts areas including when these data do not include geographical identifiers. This study established geographical levels of aggregation considering the spatial scope of services from the health care providers to calculate the spatial rate of acute respiratory infections in a major capital city in Colombia. This approach also allowed the identification of hotspot or areas of higher risk to support public health decision making. The methodology developed in this study can be used to enhance the usability of public health statistics where geographic identifiers are not included at the individual level, especially in low- and middle-income countries where restrictions due to limited resources affect the availability of adequate and quality health data. ## Data Availability All data produced in the study are available upon reasonable request to the authors. The raw data provided by the Department of Health is subject to their authorisation to be published. ## Additional information ### Authors contribution JC-R: conceptualisation, formal analysis and writing (original draft). JDWV: Data curation, writing (review & editing). RNM: Validation, writing (review & editing). VS: Validation, writing (review & editing). OMPP: Data curation, writing (review & editing). ### Data availability statement All data produced in the study are available upon reasonable request to the authors. The raw data provided by the Department of Health is subjected to their authorisation to be published. ### Ethical approval This study was approved by the ethics committee of the University of Santander, record VII-FT-025-UDES, 021 25/06/2019. ### Informed consent N/A ### Registry and registration no. of the study/trial N/A ### Conflict of interest The authors declare no conflict of interest. This study did not receive funding. ## Appendix List of health care providers by category and spatial scope View this table: [Table2](http://medrxiv.org/content/early/2022/06/06/2022.04.18.22274002/T2) ## Footnotes * Some typos and grammar issues were identified and corrected in the introduction. Additional clarification on the methodology was added in the methods section. Figures 1 and 2 were updated to include the location of hospitals and GP practices, and the location of public health providers in level 1 and 2 zones. There are no changes in the analysis or the results. * Received April 18, 2022. * Revision received June 2, 2022. * Accepted June 6, 2022. * © 2022, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution-NonCommercial-NoDerivs 4.0 International), CC BY-NC-ND 4.0, as described at [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/) ## 6. References 1. 1.European Centre for Disease Prevention and Control, The use of evidence in decision-making during public health emergencies. Stockholm: ECDC, 2019. 2. 2.Kneale, D., A. Rojas-García, and J. Thomas, Exploring the importance of evidence in local health and wellbeing strategies. Journal of Public Health, 2018. 40(suppl_1): p. i13–i23. 3. 3.McGill, E., et al., Trading quality for relevance: non-health decision-makers’ use of evidence on the social determinants of health. BMJ Open, 2015. 5(4): p. e007053. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiYm1qb3BlbiI7czo1OiJyZXNpZCI7czoxMToiNS80L2UwMDcwNTMiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMi8wNi8wNi8yMDIyLjA0LjE4LjIyMjc0MDAyLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 4. 4.Kneale, D., A. Rojas-García, and J. Thomas, Obstacles and opportunities to using research evidence in local public health decision-making in England. Health Research Policy and Systems, 2019. 17(1): p. 61. 5. 5.Chopra, M., Inequalities in health in developing countries: Challenges for public health research. Critical Public Health, 2005. 15(1): p. 19–26. 6. 6.Cortes-Ramirez, J., et al., Environmental and sociodemographic risk factors associated with environmentally transmitted zoonoses hospitalisations in Queensland, Australia. One health (Amsterdam, Netherlands), 2020. 12: p. 100206–100206. 7. 7.Boerma, J.T. and S.K. Stansfield, Health statistics now: are we making the right investments? The Lancet, 2007. 369(9563): p. 779–786. 8. 8.Gobierno de Colombia, Morbilidad consulta externa. Medicina General. 2019. Sistema de Datos Abiertos, 2021. Retrieved from https://www.datos.gov.co/browse?Informaci%;C3%;B3n-de-la-Entidad\_Departamento=Norte+de+Santander&Informaci%;C3%;B3n-de-la-Entidad\_Municipio=C%;C3%;BAcuta&category=Salud+y+Protecci%;C3%;B3n+Social&page=2. 9. 9.Instituto Nacional de Salud, Infección respiratoria aguda. Semana epidemiológica 40. Boletin Epidemiologico Semanal, 2019. retrieved from: [https://www.ins.gov.co/buscador-eventos/BoletinEpidemiologico/2019\_Boletin\_epidemiologico\_semana\_40.pdf](https://www.ins.gov.co/buscador-eventos/BoletinEpidemiologico/2019\_Boletin_epidemiologico_semana_40.pdf). 10. 10.Griffith, D.A., R.J. Bennett, and R.P. Haining, Statistical Analysis of Spatial Data in the Presence of Missing Observations: A Methodological Guide and an Application to Urban Census Data. Environment and Planning A: Economy and Space, 1989. 21(11): p. 1511–1523. 11. 11.Boakes, E.H., et al., Uncertainty in identifying local extinctions: the distribution of missing data and its effects on biodiversity measures. Biology Letters, 2016. 12(3): p. 20150824. 12. 12.Anselin, L., N. Lozano, and J. Koschinsky, Rate transformations and smoothing. Urbana, 2006. 51: p. 61801. 13. 13.Anselin, L., Local Indicators of Spatial Association—LISA. Geographical Analysis, 1995. 27(2): p. 93–115. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=https://doi.org/10.1111/j.1538-4632.1995.tb00338.x&link_type=DOI) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1995QU19800001&link_type=ISI) 14. 14.Tennekes, M., tmap: Thematic Maps in R. Journal of Statistical Software, 2018. 84(1): p. 1–39. 15. 15.Franco, F. and A. Di Napoli, Measures of Association in Medicine and Epidemiology. Giornale di Tecniche Nefrologiche e Dialitiche, 2017. 29(2): p. 127–128. 16. 16.Zartarian, V., et al., Children’s Lead Exposure: A Multimedia Modeling Analysis to Guide Public Health Decision-Making. Environ Health Perspect, 2017. 125(9): p. 097009. 17. 17.Linka, K., M. Peirlinck, and E. Kuhl, The reproduction number of COVID-19 and its correlation with public health interventions. Computational Mechanics, 2020. 66(4): p. 1035–1050. 18. 18.Hu, Y., et al., An overview of multiple linear regression model and its application. Zhonghua yu Fang yi xue za zhi [Chinese Journal of Preventive Medicine], 2019. 53(6): p. 653–656. 19. 19.Wakefield, J., Multi-level modelling, the ecologic fallacy, and hybrid study designs. Int J Epidemiol, 2009. 38(2): p. 330–6. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/ije/dyp179&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=19339258&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F06%2F06%2F2022.04.18.22274002.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000264890300002&link_type=ISI) 20. 20.Chinnaswamy, A., et al., Big data visualisation, geographic information systems and decision making in healthcare management. Management Decision, 2019. 21. 21.Kirby, R.S., E. Delmelle, and J.M. Eberth, Advances in spatial epidemiology and geographic information systems. Annals of Epidemiology, 2017. 27(1): p. 1–9. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.annepidem.2016.12.001&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F06%2F06%2F2022.04.18.22274002.atom) 22. 22.Blangiardo, M. and M. Cameletti, Spatial and spatio-temporal Bayesian models with R-INLA. 2015: John Wiley & Sons. 23. 23.World Health Organization. Primary health care: report of the International Conference on primary health care. in Alma-Ata, USSR, 6-12 September 1978/jointly sponsored by the World Health Organization and the United Nations Children’s Fund. 1978. 24. 24.Haldane, V., et al., Community participation in health services development, implementation, and evaluation: A systematic review of empowerment, health, community, and process outcomes. PLOS ONE, 2019. 14(5): p. e0216112. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F06%2F06%2F2022.04.18.22274002.atom) 25. 25.Nathan, N.L., Electoral politics and Africa’s Urban transition: class and ethnicity in Ghana. 2019: Cambridge University Press. 26. 26.Currie, D.J., C. Smith, and P. Jagals, The application of system dynamics modelling to environmental health decision-making and policy - a scoping review. BMC Public Health, 2018. 18(1): p. 402. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F06%2F06%2F2022.04.18.22274002.atom) 27. 27.Factor, R. and M. Kang, Corruption and population health outcomes: an analysis of data from 133 countries using structural equation modeling. International journal of public health, 2015. 60(6): p. 633–641. 28. 28.Israëls, A., Methods of standardisation. The Hague/Heerlen, The Netherlands: Statistics Netherlands, 2013.