Big Data-Planetary Health approach for evaluating the Brazilian Dengue Control Program

Xavier, Fernando; Barbosa, Gerson Laurindo; Marques, Cristiano Corrêa de Azevedo; Saraiva, Antonio Mauro

doi:10.11606/s1518-8787.2024058005491

Acessibilidade / Reportar erro

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Original Article • Rev. Saúde Pública 58 • 2024 • https://doi.org/10.11606/s1518-8787.2024058005491 copy

Big Data-Planetary Health approach for evaluating the Brazilian Dengue Control Program

Authorship SCIMAGO INSTITUTIONS RANKINGS

ABSTRACT

OBJECTIVE

This study aims to integrate the concepts of planetary health and big data into the Donabedian model to evaluate the Brazilian dengue control program in the state of São Paulo.

METHODS

Data science methods were used to integrate and analyze dengue-related data, adding context to the structure and outcome components of the Donabedian model. This data, considering the period from 2010 to 2019, was collected from sources such as Department of Informatics of the Unified Health System (DATASUS), the Brazilian Institute of Geography and Statistics (IBGE), WorldClim, and MapBiomas. These data were integrated into a Data Warehouse. K-means algorithm was used to identify groups with similar contexts. Then, statistical analyses and spatial visualizations of the groups were performed, considering socioeconomic and demographic variables, soil, health structure, and dengue cases.

OUTCOMES

Using climate variables, the K-means algorithm identified four groups of municipalities with similar characteristics. The comparison of their indicators revealed certain patterns in the municipalities with the worst performance in terms of dengue case outcomes. Although presenting better economic conditions, these municipalities held a lower average number of community healthcare agents and basic health units per inhabitant. Thus, economic conditions did not reflect better health structure among the three studied indicators. Another characteristic of these municipalities is urbanization. The worst performing municipalities presented a higher rate of urban population and human activity related to urbanization.

CONCLUSIONS

This methodology identified important deficiencies in the implementation of the dengue control program in the state of São Paulo. The integration of several databases and the use of Data Science methods allowed the evaluation of the program on a large scale, considering the context in which activities are conducted. These data can be used by the public administration to plan actions and invest according to the deficiencies of each location.

Epidemiology; Dengue; Outcome and Process Assessment, Health Care; Data Science

Indicator	Description	Type	Source
Context
Tmax	Maximum mean annual temperature	Climate	WorldClim
Tmin	Minimum mean annual temperature	Climate	WorldClim
Prec	Annual precipitation volume	Climate	WorldClim
GDP	Gross domestic product per capita	Socioeconomic	IBGE
Gini	Gini Index	Socioeconomic	DATASUS
HDI	Human development index	Socioeconomic	IPEA
Infant mortality	Infant mortality rate per 1000 live births	Health	DATASUS
Water	Percentage of population with access to water services	Sanitation	SNIS
Sewage	Percentage of population with access to sewage services	Sanitation	SNIS
Urban population	Percentage of urban population	Demographic	SNIS
NAgri	Percentage of total area modified for agricultural activities	Soil use	MAPBIOMAS.
NPast	Percentage of total area modified for pasture	Soil use	MAPBIOMAS.
NUrb	Percentage of total area modified for urbanization	Soil use	MAPBIOMAS.
NAct	Percentage of total area modified by human activities	Soil use	MAPBIOMAS.
Structure
CHA	Number of community healthcare agents per inhabitant	Health	DATASUS
BHU	Number of basic health units per inhabitant	Health	DATASUS
Expenses	Health expenditure per inhabitant	Health	SIOPS
Outcome
Coefficient100	If the municipality recorded less than five years of incidence below 100% of the coefficient by population size (1 (Best) = less than five years and 2 (Worst) = five years or more)	Health	DATASUS
Coeficient20	If the municipality recorded less than five years of incidence below 20% of the coefficient by population size (1 (Best) = less than five years and 2 (Worst) = five years or more)	Health	DATASUS

Indicator	Coefficient100		Coeficient20
Indicator	Class 1	Class 2	Class 1	Class 2
Total population	29,160,653	14,654,207	5,574,468	38,240,392
Incidence	464.44	1,199.38	230.05	872.21
HDI	0.73	0.75	0.72	0.75
Gini	0.46	0.46	0.46	0.46
Infant mortality	11.82	10.79	12.14	11.26
GDP per capita	26,738.83	30,332.25	22,090.17	30,387.45
% urban population	81.80	91.02	76.69	87.94
% population with water serv.	79.75	86.71	74.81	84.90
% population with sewage serv.	71.88	82.40	63.96	79.88
Maximum Temperature (°C)	27.11	28.59	26.17	28.15
Minimum temperature (°C)	15.88	17.07	15.09	16.73
Annual precipitation (mm³)	1,406.48	1,408.43	1,442.33	1,390.32
% Agricultural area	24.48	35.50	18.35	32.03
% Pasture area	21.87	18.07	21.84	20.35
% Urban area	5.21	5.47	3.39	6.17
% area modified by human activity	51.79	59.03	43.64	58.57
CHA per inhabitant.	1.47	1.31	1.46	1.40
BHU per inhabitant.	0.32	0.25	0.33	0.29
Expenditure per inhabitant	794.67	775.48	754.19	804.04

Indicator	Group
Indicator	0	1	2	3
Number of municipalities	236	274	111	24
Total population	11,092,927	6,115,605	23,502,689	3,103,638
Mean incidence	504.75	994.04	180.2	756.61
% of municipalities with the worst performance (Coefficient100 = 2)	16.95	44.16	3.60	50.00
% of municipalities with the worst performance (Coefficient20 = 2)	55.51	90.51	34.23	87.50
HDI	0.74	0.74	0.73	0.74
Gini	0.46	0.43	0.50	0.51
Infant mortality	11.60	10.96	12.41	13.45
GDP per capita	29,409.25	24,682.13	29,371.26	38,277.96
% urban population	84.91	85.22	79.88	89.04
% population with water serv.	83.19	82.56	76.43	80.58
% population with sewage serv.	77.88	80.13	59.23	54.78
Maximum Temperature (°C)	27.17	29.38	24.04	25.73
Temperature Minimum Temperature (°C)	15.72	17.37	14.02	17.78
Annual precipitation (mm³)	1,382.38	1,268.38	1,594.74	2,563.74
% Agricultural area	28.37	37.73	6.31	3.97
% Pasture area	18.44	27.02	14.87	1.14
% Urban area	4.46	1.45	15.76	8.93
% area modified by human activity	51.29	66.19	37.01	14.11
CHA per inhabitant.	1.21	1.74	1.07	1.50
BHU per inhabitant.	0.29	0.35	0.22	0.21
Expenditure per inhabitant	764.63	853.72	661.2	891.67

Group	0		1		2		3
Class (2 = worst)	1	2	1	2	1	2	1	2
Number of municipalities	105	131	26	248	73	38	3	21
Incidence	225.83	728.31	531.74	1,042.50	118.89	297.99	467.8	797.86
HDI	0.72	0.75	0.73	0.74	0.72	0.76	0.72	0.74
Gini	0.46	0.47	0.41	0.44	0.49	0.51	0.48	0.51
Infant mortality	12.23	11.10	9.69	11.1	12.79	11.68	14.12	13.36
GDP per capita (x1000 Brazilian reais)	20.95	36.19	23.81	24.77	23.24	41.15	33.25	43.30
% urban population	79.08	89.59	78.31	85.95	72.02	94.34	85.00	89.62
% population with water serv.	78.82	86.70	77.46	83.09	68.12	92.47	70.48	82.02
% population with sewage serv.	70.72	83.62	74.72	80.70	51.36	75.68	43.41	56.41
Maximum Temperature (°C)	27.02	27.30	29.22	29.39	23.91	24.27	24.86	25.86
Minimum temperature (°C)	15.39	15.98	17.28	17.38	13.82	14.42	16.76	17.93
Annual precipitation (mm³)	1,353.30	1,405.70	1,264.10	1,268.80	1,605.00	1,575.10	2,144.40	2,395.10
% Agricultural area	22.57	33.03	32.77	38.26	7.73	3.59	4.43	3.91
% Pasture area	22.90	14.87	36.74	26.00	15.88	12.92	0.53	1.22
% Urban area	1.50	6.83	0.38	1.53	7.09	32.40	5.80	9.37
% area modified by human activity	47.00	54.73	70,01	65.79	30.77	48.98	10.73	14.59
CHA per inhabitant.	1.45	1.01	2.21	1.69	1.21	0.81	1.69	1.48
BHU per inhabitant.	0.33	0.25	0.48	0.33	0.27	0.14	0.28	0.21
Expenditure per inhabitant	762.93	765.99	1,043.7	833.8	646.7	689.05	555.2	939.75

Faculdade de Saúde Pública da Universidade de São Paulo Avenida Dr. Arnaldo, 715, 01246-904 São Paulo SP Brazil, Tel./Fax: +55 11 3061-7985 - São Paulo - SP - Brazil
E-mail: revsp@usp.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] Correspondence: Fernando Xavier. Universidade de São Paulo. Escola Politécnica. Av. Prof. Luciano Gualberto, 380. 05508-010 São Paulo, SP, Brazil. E-mail: fernando.xavier@gmail.com