RESEARCH ARTICLE
Applicability of internet search index for asthma admission forecast using machine learning
Li Luo, Chengcheng Liao, Fengyi Zhang,
Wei Zhang,
Chunyang Li,
Zhixin Qiu,
Debin Huang,
Corresponding Author
Fengyi Zhang
Business School, Sichuan University, China
Correspondence
Fengyi Zhang, Business School, Sichuan University, China.
Email: [email protected]
Search for more papers by this authorWei Zhang
West China Biomedical Big Data Center, West China Hospital, Sichuan University, China
Search for more papers by this authorChunyang Li
West China Biomedical Big Data Center, West China Hospital, Sichuan University, China
Search for more papers by this authorZhixin Qiu
Respiratory and Critical Care Medicine, West China Hospital, Sichuan University, China
Search for more papers by this authorLi Luo, Chengcheng Liao, Fengyi Zhang,
Wei Zhang,
Chunyang Li,
Zhixin Qiu,
Debin Huang,
Corresponding Author
Fengyi Zhang
Business School, Sichuan University, China
Correspondence
Fengyi Zhang, Business School, Sichuan University, China.
Email: [email protected]
Search for more papers by this authorWei Zhang
West China Biomedical Big Data Center, West China Hospital, Sichuan University, China
Search for more papers by this authorChunyang Li
West China Biomedical Big Data Center, West China Hospital, Sichuan University, China
Search for more papers by this authorZhixin Qiu
Respiratory and Critical Care Medicine, West China Hospital, Sichuan University, China
Search for more papers by this authorSummary
Objective
This study aimed to determine whether a search index could provide insight into trends in asthma admission in China. An Internet search index is a powerful tool to monitor and predict epidemic outbreaks. However, whether using an internet search index can significantly improve asthma admissions forecasts remains unknown. The long-term goal is to develop a surveillance system to help early detection and interventions for asthma and to avoid asthma health care resource shortages in advance.
Methods
In this study, we used a search index combined with air pollution data, weather data, and historical admissions data to forecast asthma admissions using machine learning.
Results
Results demonstrated that the best area under the curve in the test set that can be achieved is 0.832, using all predictors mentioned earlier.
Conclusion
A search index is a powerful predictor in asthma admissions forecast, and a recent search index can reflect current asthma admissions with a lag-effect to a certain extent. The addition of a real-time, easily accessible search index improves forecasting capabilities and demonstrates the predictive potential of search index.
CONFLICT OF INTERESTS
None declared.
REFERENCES
- 1Akinbami LJ, Moorman JE, Liu X. Asthma prevalence, health care use, and mortality: United States, 2005-2009.[J]. Natl Health Stat Rep. 2011; 12(32): 1.
- 2Zahran HS, Bailey C, Garbe P. Vital signs: asthma prevalence, disease characteristics, and self-management education: United States, 2001-2009[J]. MMWR Morb Mortal Wkly Rep. 2011; 60(17): 547.
- 3Chen ZH, Wang PL, Shen HH. Asthma research in China: a five-year review[J]. Respirology. 2013; 18(S3): 10-19.
- 4Li G, Jing H, Xu G, et al. The short term burden of ambient fine particulate matter on chronic obstructive pulmonary disease in Ningbo, China[J]. Environ Health, 2017, 16(1): 54.
- 5Bryden C, Bird W, Titley HA, Halpin DMG, Levy ML. Stratification of COPD patients by previous admission for targeting of preventative care.[J]. Respir Med. 2009; 103(4): 558-565.
- 6Rodó X, Ballester J, Dan C, et al. Association of Kawasaki disease with tropospheric wind patterns.[J]. Sci Rep. 2011; 1(6057): 152.
- 7Tsai PF, Lin FM. An application of multi-attribute value theory to patient-bed assignment in hospital admission management: an empirical study.[J]. J Healthcare Eng. 2014; 5(4): 439-456.
- 8Moustris KP, Douros K, Nastos PT, et al. Seven-days-ahead forecasting of childhood asthma admissions using artificial neural networks in Athens, Greece[J]. Int J Environ Health Res. 2012; 22(2): 93-104.
- 9Soyiri IN, Reidpath DD, Sarran C. Forecasting asthma-related hospital admissions in London using negative binomial models.[J]. Chron Respir Dis. 2013; 10(2): 85-94.
- 10Soyiri IN, Reidpath DD, Sarran C. Forecasting peak asthma admissions in London: an application of quantile regression models[J]. Int J Biometeorol. 2013; 57(4): 569-578.
- 11Restrepo CE, Simonoff JS, Thurston GD, et al. Asthma hospital admissions and ambient air pollutant concentrations in New York City[J]. J Environ Prot. 2012; 17(6).
- 12Zhang Y, Wang S, Liu Y. Application of ARIMA model on predicting monthly hospital admissions and hospitalization expenses for respiratory diseases[J]. Chin J Health Stat. 2015.
- 13Gouveia N, Fletcher T. Respiratory diseases in children and outdoor air pollution in S?O Paulo, Brazil: a time series analysis.[J]. Occup Environ Med. 2000; 57(7): 477-483.
- 14Hua J, Yin Y, Peng L, du L, Geng F, Zhu L. Acute effects of black carbon and PM₂.₅ on children asthma admissions: a time-series study in a Chinese city[J]. Sci Total Environ. 2014; 481(1): 433-438.
- 15Anderson HR, Butland BK, Donkelaar AV, et al. Satellite-based estimates of ambient air pollution and global variations in childhood asthma prevalence[J]. Environ Health Perspect. 2012; 120(9): 1333-1339.
- 16Gerke AK, Yang M, Tang F, Foster ED, Cavanaugh JE, Polgreen PM. Association of hospitalizations for asthma with seasonal and pandemic influenza[J]. Respirology. 2014; 19(1): 116-121.
- 17Mclaren J, Williams ID. The impact of communicating information about air pollution events on public health[J]. Sci Total Environ. 2015; 538: 478-491.
- 18Makra L, Puskás J, Matyasovszky I, et al. Weather elements, chemical air pollutants and airborne pollen influencing asthma emergency room visits in Szeged, Hungary: performance of two objective weather classifications[J]. Int J Biometeorol. 2015; 59(9): 1269-1289.
- 19Spickett JT, Brown HL, Rumchev K. Climate change and air quality: the potential impact on health. Asia-Pacific journal of public health. Asia Pac J Public Health. 2011; 23(2_suppl): 37S-45S.
- 20Sheffield PE, Knowlton K, Carr JL, Kinney PL. Modeling of regional climate change effects on ground-level ozone and childhood asthma.[J]. Am J Prev Med. 2011; 41(3): 251-257.
- 21Deschamps KM. Associations between daily asthma hospital admissions and ambient air pollutants in Montreal, 1992 to 1999[J]. Geography Planning & Environment, 2003.
- 22AAbe T, Tokuda Y, Ohde S, et al. The relationship of short-term air pollution and weather to ED visits for asthma in Japan.[J]. Am J Emerg Med. 2009; 27(2): 153-159.
- 23Cakmak S, Dales RE, Coates F. Does air pollution increase the effect of aeroallergens on hospitalization for asthma?[J]. J Allergy Clin Immunol. 2012; 129(1): 228-231.
- 24Miller MA, Levsky ME, Enslow W, Rosin A. Randomized evaluation of octreotide vs prochlorperazine for ED treatment of migraine headache[J]. Am J Emerg Med. 2009; 27(2): 160-164.
- 25Sun HL, Chou MC, Lue KH. The relationship of air pollution to ed visits for asthma differ between children and adults. Am J Emerg Med. 2006; 24(6): 709-713.
- 26Eken C. The relationship of air pollution to ED visits for asthma differs between children and adults. Am J Emerg Med. 2007; 25(7): 852.
- 27Tecer LH, Alagha O, Karaca F, Tuncel G, Eldes N. Particulate matter (PM(2.5), PM(10-2.5), and PM(10)) and children's hospital admissions for asthma and respiratory diseases: a bidirectional case-crossover study[J]. J Toxicol Environ Health A. 2008; 71(8): 512-520.
- 28Mar TF, Jansen K, Shepherd K, Lumley T, Larson TV, Koenig JQ. Exhaled nitric oxide in children with asthma and short-term PM2.5 exposure in Seattle[J]. Environ Health Perspect. 2005; 113(12): 1791-1794.
- 29Yun RS, Jin SU, Hong LY, et al. Effects of environmental pm2.5 pollution on the morbidity of asthma and lung cancer. J Labour Med. 2002; 23(4): 367-378.
- 30Weinmayr G, Romeo E, Sario MD, et al. Short-term effects of PM₁₀ and NO₂ on respiratory health among children with asthma or asthma-like symptoms: a systematic review and meta-analysis[J]. Environ Health Perspect. 2010; 118(4): 449-457.
- 31 the 39th China Statistical Report on Internet Development. available:http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201701/P020170123364672657408.pdf
- 32Artola C, Pinto F, García PDP. Can internet searches forecast tourism inflows?[J]. Int J Manpow. 2015; 36(1): 103-116.
- 33D'Amuri F, Marcucci J. ‘Google it!’ Forecasting the US Unemployment Rate with a Google Job Search Index[J]. Social Science Electronic Publishing; 2009.
- 34Jun SP, Park DH, Yeom J. The possibility of using search traffic information to explore consumer product attitudes and forecast consumer preference[J]. Technol Forecast Soc Change. 2014; 86(340): 237-253.
- 35Li X, Pan B, Law R, Huang X. Forecasting tourism demand with composite search index[J]. Tour Manag. 2017; 59: 57-66.
- 36Penna ND, Huang H. Constructing a consumer confidence index for the US using web search volume[J]. 2009.
- 37Fatima I, Halder S, Saleem MA, et al. Smart CDSS: integration of social media and interaction engine (SMIE) in healthcare for chronic disease patients[J]. Multimed Tool Appl. 2015; 74(14): 5109-5129.
- 38Li S, Peng G, Wang J. Search queries based sales forecasting on Taobao[C]. Ieee/acis, International Conference on Computer and Information Science. IEEE, 2014:223–228.
- 39Webb GK. Forecasting U.S. home foreclosures with an index of Internet keyword searches. Value Creation in E-Business Management. Americas Conference on Information Systems, Amcis 2009, Sigebiz Track, San Francisco, Ca, Usa, August 6–9, 2009. Selected Papers (Vol.36, pp.196–203). DBLP; 2009.
- 40Petroliagis TI, Pinson P. Early warnings of extreme winds using the ECMWF extreme forecast index[J]. Meteorol Appl. 2014; 21(2): 171-185.
- 41Teodoro MF. Industrial production index forecast: methods comparison[J]. 2012; 1504(1): 1103-1107.
- 42Xu W, Zheng T, Li Z. A neural network based forecasting method for the unemployment rate prediction using the search engine query data[C]. IEEE, international conference on E-business engineering. IEEE Computer Society, 2011:9–15.
- 43Xu W, Li Z, Cheng C, Zheng T. Data mining for unemployment rate prediction using search engine query data[J]. SOCA. 2013; 7(1): 33-42.
10.1007/s11761-012-0122-2 Google Scholar
- 44Hassani H, Silva ES. Forecasting energy data with a time lag into the future and Google trends[J]. Int J Energy Stat. 2017; 4(4). 1650020
- 45Askitas N. Predicting road conditions with internet search[J]. PLos One. 2016; 11(8):e0162080.
- 46Rivera R. A dynamic linear model to forecast hotel registrations in Puerto Rico using Google trends data[J]. Tour Manag. 2016; 57: 12-20.
- 47Fantazzini D, Toktamysova Z. Forecasting German car sales using google data and multivariate models. Int J Prod Econ. 2015; 170: 97-135.
- 48Smith P. Google's MIDAS touch: predicting UK unemployment with Internet search data[J]. J Forecast. 2016; 35(3):n/a-n/a): 263-284.
- 49Vicente MR, López-Menéndez AJ, Pérez R. Forecasting unemployment with internet search data: does it help to improve predictions when job destruction is skyrocketing?[J]. Technol Forecast Soc Change. 2015; 92(92): 132-139.
- 50Zou X, Zhu W, Yang L, Shu Y. [Google flu trends—the initial application of big data in public health][J]. Zhonghua Yu Fang Yi Xue Za Zhi. 2015; 49(6): 581-584.
- 51Moss R, Zarebski A, Dawson P, McCaw JM. Forecasting influenza outbreak dynamics in Melbourne from Internet search query surveillance data[J]. Influenza Other Respi Viruses. 2016; 10(4): 314-323.
- 52Araz OM, Bentley D, Muelleman RL. Using Google flu trends data in forecasting influenza-like-illness related ED visits in Omaha, Nebraska.[J]. Am J Emerg Med. 2014; 32(9): 1016-1023.
- 53Moss R, Zarebski A, Dawson P, et al. Retrospective forecasting of the 2010-2014 Melbourne influenza seasons using multiple surveillance systems[J]. Epidemiol Infect. 2017; 145(1): 156-169.
- 54Xu Q, Gel YR, Ramirez Ramirez LL, Nezafati K, Zhang Q, Tsui KL. Forecasting influenza in Hong Kong with Google search queries and statistical model fusion[J]. PLos One. 2017; 12(5):e0176690.
- 55Bardak B, Tan M. Prediction of influenza outbreaks by integrating Wikipedia article access logs and Google flu trend data[C]. IEEE, International Conference on Bioinformatics and Bioengineering. IEEE, 2015:1–6.
- 56Nsoesie E, Mararthe M, Brownstein J. Forecasting peaks of seasonal influenza epidemics.[J]. PLOS Currents. 2013; 5.
- 57Preis T, Moat HS. Adaptive nowcasting of influenza outbreaks using Google searches[J]. R Soc Open Sci. 2014; 1(2):140095.
- 58Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L. Detecting influenza epidemics using search engine query data. Nature. 2009; 457(7232): 1012-1014.
- 59Li Z, Liu T, Zhu G, et al. Dengue Baidu search index data can improve the prediction of local dengue epidemic: a case study in Guangzhou, China[J]. PLoS Negl Trop Dis. 2017; 11(3):e0005354.
- 60Wang HW, Chen DR, Yu HW, Chen YM. Forecasting the incidence of dementia and dementia-related outpatient visits with Google trends: evidence from Taiwan[J]. J Med Internet Res. 2015; 17(11):e264.
- 61Ekström A, Kurland L, Farrokhnia N, Castrén M, Nordberg M. Forecasting emergency department visits using internet data.[J]. Ann Emerg Med. 2015; 65(4): 436-442.e1.
- 62Lazer D, Kennedy R, King G, Vespignani A. The parable of Google flu: traps in big data analysis[J]. Science. 2014; 343(6176): 1203-1205.
- 63 Chengdu Environment Monitoring Center (EMC). Available:database,http://www.cdemc.org
- 64 Chengdu Meteorological Bureau. Available: http://www.cdmb.gov.cn
- 65 Search Engine Market Share in China. 2017 Available: http://gs.statcounter.com/search-engine-market-share/all/china/#monthly-201704-201704-bar
- 66Yuan Q, Nsoesie EO, Lv B, Peng G, Chunara R, Brownstein JS. Monitoring influenza epidemics in China with search query from Baidu. PLoS One. 2013; 8(5):e64323. doi: https://doi.org/10.1371/journal.pone.0064323 PMID: 23750192.
- 67Gu Y, Chen F, Liu T, et al. Early detection of an epidemic erythromelalgia outbreak using Baidu search data[J]. Sci Rep. 2015; 5(1):12649.
- 68Olson DR, Konty KJ, Paladini M, Viboud C, Simonsen L. Reassessing Google flu trends data for detection of seasonal and pandemic influenza: a comparative epidemiological study at three geographic scales[J]. PLoS Comput Biol. 2013; 9(10):e1003256.
- 69Milinovich GJ, Williams GM, Clements ACA, Hu W. Internet-based surveillance systems for monitoring emerging infectious diseases[J]. Lancet Infect Dis. 2014; 14(2): 160-168.
- 70Althouse BM, Scarpino SV, Meyers LA, et al. Enhancing disease surveillance with novel data streams: challenges and opportunities[J]. EpJ Data Sci. 2015; 4(1): 17.
