Modeling National Trends on Health in the Philippines using ARIMA

Florence Jean B Talirongan^*, Hidear Talirongan and Markdy Y Orong
^*Correspondence: Florence Jean B Talirongan, Misamis University, Ozamiz City, Philippines, Tel: +09054137874, Email:

Keywords

ARIMA • Health • Diseases • Forecasting • Trend analysis

Introduction

People weigh high value on health since it is a core element in people’s well-being and happiness [1] to the extent of setting it as a priority in the governmental and societal agenda [2]. Health is considered as an important prerequisite in reaching person’s goals and aspirations, and contributory factor to development of societal undertakings [1,3,4].

The World Health Organization (WHO) battled underfunded diseases like AIDS/HIV and tuberculosis and issued International Health Regulations to make medical standards consistent and stop epidemics in their tracks [5]. In the United States, leading causes of infant, neonatal and postneonatal death in rank order which include diseases of heart; malignant neoplasms; chronic lower respiratory diseases; cerebrovascular diseases; accidents (unintentional injuries); alzheimer's disease; diabetes mellitus; influenza and pneumonia; nephritis, nephrotic syndrome and nephrosis; and intentional self-harm (suicide) [6]. In Asia Pacific regions like Thailand and China account for a significant burden of global poverty and disease. Indeed, approximately one-third of the world's intestinal helminthiases, most of the food-borne trematode infections, one-half of the active trachoma infections and a significant number of cases of lymphatic filariasis (LF), schistosomiasis and arboviral infections occur in the region [7]. Urban healths in developing countries were taken care of by WHO in terms of global overview on health and the cities, policies and health status and health environment [4,5,8]. However, it was an extraordinary opportunity to create a sustained global movement against premature death and preventable morbidity and disability from NCDs, mainly heart disease, stroke, cancer, diabetes, and chronic respiratory disease. The increasing global crisis in NCDs is a barrier to development goals including health [8,9].

Reflecting the information that is most readily available in the Philippines have placed on diseases like diabetes, tropical diseases, gastroduodenal diseases, dengue, lung disease, foot and mouth disease and allergenic rhinitis [10-13].

It was observed that several studies were more focused on the trend of individual disease however no further studies concentrated on the trends of health data. The objective of this research is to visualize the national trends of health data particularly in the Philippines. This motivates the researcher to identify the most and least vulnerable disease in the country. The trend analysis utilized time series and data forecasting using autoregressive integrated moving average (ARIMA).

Theoretical framework

Review of related literature: Several literature and studies support the use of ARIMA model as a forecasting tool in predicting diseases. Li et al., [14] applied ARIMA in the incidence of hemorrhagic fever with renal Syndrome (HFRS) in China during 1986 to 2009 which fit to the given study and can be applied to future forecasting on prevention and control. Yu et al., [15] constructed ARIMA model to forecast the number of HIV infections from 2013 to 2017 in Korea. Research done by Midekisa et al., [16] and Zhang et al., [17] in Ethiopia and China used ARIMA to quantify the relationship between malaria cases. Another study of Naiman et al., [18] used ARIMA model to test the relation between smoking bans and admission rates that lead to cardiovascular and respiratory conditions. Baker-Austin et al., [19] illustrated associations between environmental changes to the emergence of vibrio infections and forecasted the risk of infections. Time series analysis of dengue incidence was studied by Gharbi et al., [20] in Guadeloupe, French West Indies. ARIMA model was used to predict the occurrence of dengue epidemics which was also studied Wongkoon [21] in Thailand and Johansson et al., [22] in Mexico. The impact of climate change on dengue transmission in the Asia-Pacific region was also examined by Banu [23]. Soebiyanto [24] analyzed the role of climatic variables as input series on influenza transmission in two regions: Hong Kong (China) and Maricopa County (Arizona, USA) where the influenza cases depend on its past values and error signal. Razvodovsky [25] found out that alcohol is an important contributor to the liver cirrhosis mortality rate in Russia.

Materials and Methods

Materials

The data used in the study are the statistical data taken from the Philippines in Figures of the Philippine Statistics Authority (PSA) from the year 2012 up to 2016. There were several areas presented in the statistics but this study will highlight the Health comprising three areas of data on ten leading causes of death, leading causes of morbidity and leading causes of infant's deaths. These data were taken from the sources like PSA, Family Planning Survey, National Demographic and Health Survey, Department of Health and Department of Social Welfare and Development. The data were taken from the books of Philippines in Figures 2017 to 2018. However, the book only reflected data from 2012 up to 2016 which will be used for trend analysis and data forecasting on the three areas.

Methods

The study utilized Autoregressive Integrated Moving Average (ARIMA) model employed in many fields to construct models for forecasting time series [14,26]. ARIMA (p,d,q) algorithm is used to forecast the data pattern of diseases for the next fourteen years. Time series predictions are based on changes over time in historical data sets and can produce mathematical models by using statistical data that can be extrapolated [27,28]. The ARIMA (p,d,q) model is defined as follows:

(1)

Where, Φ’s (phis) represents the autoregressive parameters to be estimated, Θ's (thetas) are the moving average parameters to be determined, the original series is represented by X's, and the a’s are the unknown random errors which are assumed to follow the normal probability distribution.

Three steps were performed to predict the incidence of leading causes of death, morbidity and infants by using the ARIMA-related modules. Model identification used autocorrelation analysis and partial autocorrelation analysis methods to analyze any random, stationary, and seasonal effects on the time series data. The researcher prepared a stationary time series by considering the differences and then determined plausible models on the basis of an autocorrelogram and a partial autocorrelogram. Lastly, the parameter estimation and model testing were used to compare the plausible models obtained, and we selected the most appropriate model. Finally, we conducted predictive analysis. The study used GRETL (Gnu Regression, Econometrics and Time-series Library) software for plotting the graphs and analysis of the data sets. Figure 1 presents the architectural design in predicting incidences of leading causes of death, morbidity and infant’s death.

Conclusion and Recommendations

Having trend analysis and forecasting on each disease per area is very useful in foreseeing future incidences. Moreover, predicting its occurrences can give insights to the health administration in designing preventive measures against the possible spread of diseases. In the study, ARIMA (1,0,1) model helps in predicting the increasing and decreasing pattern of diseases in the community.

In the ten leading causes of death, heart disease was found as leading among others in terms of its forecasted data. On the other hand, in terms of morbidity, the acute respiratory infection was found as leading disease. Further, in the cause of disease of infant it is evident that all identified causes were found as the possible reasons of infant’s death.

Future research work will focus on the measles and chickenpox trends and predictions utilizing another algorithm for trend analysis and forecasting.

References

1. Leppo, Kimmo, Eeva Ollila, Sebastián Peña, and Matthias Wismar, et al. “Seizing Opportunities, Implementing Policies. Helsinki.” Finland: Ministry of Social Affairs and Health, 2013.
2. Kickbusch Ilona, and David Gleicher. “Governance for Health in the 21st Century.” Geneva: World Health Organization, 2012.
3. Elena, Mogilyansky, and Isidore Rigoutsos. “The miR-17/92 Cluster: A Comprehensive Update on its Genomics, Genetics, Functions and Increasingly Important and Numerous Roles in Health and Disease.” Cell Death & Differentiation 20 (2013): 1603-1614.
4. Tanner, Marcel, and Trudy Harpham. “Urban Health in Developing Countries: Progress and Prospects.” Routledge, 2014.
5. Reisman, David A. “Health tourism: Social welfare through international trade.” Edward Elgar Publishing, 2010.
6. Heron, Melonie. “Deaths: Leading Causes for 2008.” National Vital Statistics Reports: From the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System 60 (2012): 1-94.
7. Hotez, Peter J, and John, P Ehrenberg. “Escalating the Global Fight against Neglected Tropical Diseases through Interventions in the Asia Pacific Region.” Advances in Parasitology 72 (2010): 31-53.
8. Beaglehole, Robert, Ruth Bonita, Richard Horton, and Cary Adams, et al. “Priority Actions for the Non-Communicable Disease Crisis.” The Lancet, 377 (2011): 1438-1447.
9. Alwan, Ala, David R Maclean, Leanne M Riley LM, and Colin Douglas Mathers, et al. “Monitoring and Surveillance of Chronic Non-Communicable Diseases: Progress and Capacity in High-Burden Countries.” The Lancet 376 (2010): 1861-1868.
10. Higuchi, Michiyo. “Access to diabetes care and medicines in the Philippines.” Asia Pac J Public Health 22 (2010): 96S-102S.
11. Idolor, Luisito F, Teresita S. De Guia, Norberto A. Francisco, and Camilo C. Roa, et al. “Burden of Obstructive Lung Disease in A Rural Setting in the Philippines”. Respirology 16 (2011): 1111-1118.
12. Gaw, Albert C, Byron Lee, Giselle Gervacio-Domingo, and Charles Antzelevitch, et al. “Unraveling the Enigma of Bangungut: Is Sudden Unexplained Nocturnal Death Syndrome (SUNDS) in the Philippines a Disease Allelic to the Brugada Syndrome?” Philipp J Intern Med 49 (2011): 165-176.
13. Windsor, Peter Andrew, Freeman Paul, R Abila, and Carolyn Benigno. “Foot‐and‐Mouth Disease Control and Eradication in the Bicol Surveillance Buffer Zone of the Philippines.” Transbound Emerg Dis 58 (2011): 421-433.
14. Li, Qi, Na-Na Guo, Zhan-Ying Han, and Yan-Bo Zhang, et al. “Application of an Autoregressive Integrated Moving Average Model for Predicting the Incidence of Hemorrhagic Fever with Renal Syndrome.” Am J Trop Med Hyg 87 (2012): 364-370.
15. Yu, Hye-Kyung, Na-Young Kim, Sung Soon Kim, and Chaeshin Chu, et al. “Forecasting the Number of Human Immunodeficiency Virus Infections in the Korean Population Using the Autoregressive Integrated Moving Average Model.” Osong Public Health and Research Perspectives 4 (2013): 358-362.
16. Midekisa, Alemayehu, Gabriel Senay, Geoffrey M Henebry, and Paulos Semuniguse. “Remote Sensing-Based Time Series Models for Malaria Early Warning in the Highlands of Ethiopia.” Malar J 11 (2012): 165.
17. Zhang, Ying, Peng Bi, and Janet Hiller. “Meteorological Variables and Malaria in a Chinese Temperate City: A Twenty-Year Time-Series Data Analysis.” Environ Int 36 (2010): 439-445.
18. Naiman, Alisa, Richard H. Glazier and Rahim Moineddin. “Association of Anti-Smoking Legislation with Rates of Hospital Admission for Cardiovascular and Respiratory Conditions.” CMAJ 182 (2010): 761-767.
19. Baker-Austin, Craig, Joaquin A Trinanes, Nick G. H. Taylor, and Rachel Hartnell, et al. “Emerging Vibrio risk at high latitudes in response to ocean warming.” Nature Climate Change 3(2013): 73-77.
20. Gharbi, Myriam, Philippe Quénel, Joël Gustave, and Sylvie Cassadou, et al. Time Series Analysis of Dengue Incidence in Guadeloupe, French West Indies: Forecasting Models Using Climate Variables as Predictors. BMC Infect Dis 11 (2011): 166.
21. Wongkoon, Siriwan, Mullica Jaroensutasinee, and Krisanadej Jaroensutasinee. “Assessing the Temporal Modelling for Prediction of Dengue Infection in Northern and Northeastern, Thailand.” Tropical Biomed 29 (2012): 339-348.
22. Johansson, Michael, Nicholas G Reich, Aditi Hota, and John S Brownstein, et al. “Evaluating the Performance of Infectious Disease Forecasts: A Comparison of Climate-Driven and Seasonal Dengue Forecasts For Mexico”. Scientific Reports, 2016.
23. Banu, Shahera. Examining the impact of climate change on dengue transmission in the Asia-Pacific region (Doctoral dissertation, Queensland University of Technology), 2013.
24. Soebiyanto, Radina P, Farida Adimi, and Richard K. Kiang. “Modeling and Predicting Seasonal Influenza Transmission in Warm Regions Using Climatological Parameters.” PloS One 5 (2010): e9450.
25. Razvodovsky, Yury, Evgeny. “Alcohol Consumption and Liver Cirrhosis Mortality in Russia.” J Alcohol Drug Depend 2 (2014): 152.
26. Lee, Yi-Shian, and Tong Lee-Ing. “Forecasting Time Series Using A Methodology Based on Autoregressive Integrated Moving Average and Genetic Programming.” Knowledge-Based Systems 24 (2011): 66-72.
27. Kayacan, Erdal, Baris Ulutas, and Okyay Kaynak. “Grey System Theory-Based Models in Time Series Prediction.” Expert Systems with Applications 37 (2010): 1784-1789.
28. Orong, Markdy Y, Ariel M Sison, and Alexander A Hernandez. “Mitigating Vulnerabilities through Forecasting and Crime Trend Analysis”. 2018 5th International Conference on Business and Industrial Research (ICBIR) (2018): 57-62.
29. Institute for Health Metrics and Evaluation (IHME). “Global Burden of Disease Data on the Philippines.” Retrieved on November 15, 2019. Available from: http://www.healthdata.org/philippines.