An ontology-based approach to patient follow-up assessment for continuous and personalized chronic disease management.

OBJECTIVE: Chronic diseases are complex and persistent clinical conditions that require close collaboration among patients and health care providers in the implementation of long-term and integrated care programs. However, current solutions focus partially on intensive interventions at hospitals rather than on continuous and personalized chronic disease management. This study aims to fill this gap by providing computerized clinical decision support during follow-up assessments of chronically ill patients at home. METHODS: We proposed an ontology-based framework to integrate patient data, medical domain knowledge, and patient assessment criteria for chronic disease patient follow-up assessments. A clinical decision support system was developed to implement this framework for automatic selection and adaptation of standard assessment protocols to suit patient personal conditions. We evaluated our method in the case study of type 2 diabetic patient follow-up assessments. RESULTS: The proposed framework was instantiated using real data from 115,477 follow-up assessment records of 36,162 type 2 diabetic patients. Standard evaluation criteria were automatically selected and adapted to the particularities of each patient. Assessment results were generated as a general typing of patient overall condition and detailed scoring for each criterion, providing important indicators to the case manager about possible inappropriate judgments, in addition to raising patient awareness of their disease control outcomes. Using historical data as the gold standard, our system achieved a rate of accuracy of 99.93% and completeness of 95.00%. CONCLUSIONS: This study contributes to improving the accessibility, efficiency and quality of current patient follow-up services. It also provides a generic approach to knowledge sharing and reuse for patient-centered chronic disease management.

Accurate Prediction of Coronary Heart Disease for Patients With Hypertension From Electronic Health Records With Big Data and Machine-Learning Methods: Model Development and Performance Evaluation.

BACKGROUND: Predictions of cardiovascular disease risks based on health records have long attracted broad research interests. Despite extensive efforts, the prediction accuracy has remained unsatisfactory. This raises the question as to whether the data insufficiency, statistical and machine-learning methods, or intrinsic noise have hindered the performance of previous approaches, and how these issues can be alleviated. OBJECTIVE: Based on a large population of patients with hypertension in Shenzhen, China, we aimed to establish a high-precision coronary heart disease (CHD) prediction model through big data and machine-learning. METHODS: Data from a large cohort of 42,676 patients with hypertension, including 20,156 patients with CHD onset, were investigated from electronic health records (EHRs) 1-3 years prior to CHD onset (for CHD-positive cases) or during a disease-free follow-up period of more than 3 years (for CHD-negative cases). The population was divided evenly into independent training and test datasets. Various machine-learning methods were adopted on the training set to achieve high-accuracy prediction models and the results were compared with traditional statistical methods and well-known risk scales. Comparison analyses were performed to investigate the effects of training sample size, factor sets, and modeling approaches on the prediction performance. RESULTS: An ensemble method, XGBoost, achieved high accuracy in predicting 3-year CHD onset for the independent test dataset with an area under the receiver operating characteristic curve (AUC) value of 0.943. Comparison analysis showed that nonlinear models (K-nearest neighbor AUC 0.908, random forest AUC 0.938) outperform linear models (logistic regression AUC 0.865) on the same datasets, and machine-learning methods significantly surpassed traditional risk scales or fixed models (eg, Framingham cardiovascular disease risk models). Further analyses revealed that using time-dependent features obtained from multiple records, including both statistical variables and changing-trend variables, helped to improve the performance compared to using only static features. Subpopulation analysis showed that the impact of feature design had a more significant effect on model accuracy than the population size. Marginal effect analysis showed that both traditional and EHR factors exhibited highly nonlinear characteristics with respect to the risk scores. CONCLUSIONS: We demonstrated that accurate risk prediction of CHD from EHRs is possible given a sufficiently large population of training data. Sophisticated machine-learning methods played an important role in tackling the heterogeneity and nonlinear nature of disease prediction. Moreover, accumulated EHR data over multiple time points provided additional features that were valuable for risk prediction. Our study highlights the importance of accumulating big data from EHRs for accurate disease predictions.

Forecasting hand, foot, and mouth disease in Shenzhen based on daily level clinical data and multiple environmental factors.

Hand, foot, and mouth disease (HFMD) is caused by a group of enteroviruses. It infects millions of children in the Southeast Asian area. An accurate forecasting of outbreaks of HFMD could facilitate public health officials to suggest public health actions earlier. Many researchers tried to develop an early warning system for HFMD to lower the damage caused by a HFMD outbreak. The research data based on daily level could help figure out the relationship between HFMD and environmental factors, but nevertheless is difficult to collect. In this study, we collected the daily clinical data from the Shenzhen Health Information Center and multiple environmental factors to analyze the outbreaks of HFMD. Considering the incubation period of HFMD, we fed the previous 60 days' HFMD rates, 7 days' temperature factors and 7 days' air-quality factors into the tree model, XGBoost. The following conclusions were drawn in this study: i) Compared with the model only using the previous HFMD rate and temperature factors, the addition of the air-quality factors could make the model better, improving MAE nearly 16.7%. ii) By analyzing the Pearson correlation, we found that the temperature showed a positive correlation and the air quality showed a negative correlation for the HFMD outbreaks. Improving the air quality, especially decreasing PM2.5 and PM10 could decrease the risk of HFMD outbreaks.

The Association between Air Pollution and Outpatient and Inpatient Visits in Shenzhen, China.

Nowadays, air pollution is a severe environmental problem in China. To investigate the effects of ambient air pollution on health, a time series analysis of daily outpatient and inpatient visits in 2015 were conducted in Shenzhen (China). Generalized additive model was employed to analyze associations between six air pollutants (namely SO2, CO, NO2, O3, PM10, and PM2.5) and daily outpatient and inpatient visits after adjusting confounding meteorological factors, time and day of the week effects. Significant associations between air pollutants and two types of hospital visits were observed. The estimated increase in overall outpatient visits associated with each 10 µg/m³ increase in air pollutant concentration ranged from 0.48% (O3 at lag 2) to 11.48% (SO2 with 2-day moving average); for overall inpatient visits ranged from 0.73% (O3 at lag 7) to 17.13% (SO2 with 8-day moving average). Our results also suggested a heterogeneity of the health effects across different outcomes and in different populations. The findings in present study indicate that even in Shenzhen, a less polluted area in China, significant associations exist between air pollution and daily number of overall outpatient and inpatient visits.

Time series analysis of weekly influenza-like illness rate using a one-year period of factors in random forest regression.

Influenza, a disease caused by a respiratory virus, sickened over 5,043,127 citizens in Shenzhen, China, from January 2014 to April 2016. An accurate forecasting of outbreaks of influenza-like illness (ILI, here we refer to ILI as the upper respiratory infection) could facilitate public health officials to suggest public health actions earlier. In this study, a random forest regression constructed with a one-year period of factors was adopted to forecast the weekly ILI rate using the clinical data from Shenzhen Health Information Center. The following conclusions were drawn based on this method: i) Compared to the predication with 52 (one-year) history observations, the accuracy of the predication was improved by adding another 52 first-order difference variables: mean absolute percentage error (MAPE) decreased from 5.04% to 4.35% and mean squared error (MSE) decreased from 2.85E-04 to 1.97E-04. ii) The variables with the first-order difference seemed more significant than the original history observations during the predication. In addition, both the recent observations and the later observations seemed important in the predicating procedure. iii) Analysis using the Pearson correlation concluded that weather conditions, the influence of which could have been implied by history observations and seemed insignificant for the predication, showed correlation to the weekly average temperature and maximum temperature. The correlation coefficients were -0.3656 and -0.3583, respectively.

[Evaluation on quitting intervention in health professionals of six cities].

OBJECTIVE: To evaluate the impact of quitting intervention in health professionals of six cities, developing smoking cessation models in China. METHODS: All community directors and health professionals in Seven communities were selected in six cities of Beijing, Shanghai, Tianjin, Changsha, Shenzhen and Puyang were surveyed for smoking cessation using cross-sectional study. RESULTS: There were 25 hospitals that kept on providing smoking cessation service after the intervention. The percent of the awareness about "the harm of smoking is public health problem" has improved 12.8%. The percent of getting some smoking cessation methods and actively providing smoking cessation have improved 9.2% and 7.3% respectively. CONCLUSION: Training the health professionals can not only increase their knowledge but also provide them smoking cessation service actively. It is a effective way to obtain methods and skills of quitting.

[Analysis on main factors for successful quitting--study on the one-year follow-up for Chinese 'Quit and Win' in 2002].

OBJECTIVE: China took part in the fifth 'Quit & Win in 2002 which was held by WHO and KTL in Finland. 27398 participants from eleven cities in China took part in the contest. Then we had a one-year follow-up research from May to June in 2003. METHODS: A random sample of 1298 was surveyed by mailing, telephone and visiting. RESULTS: One year were 27.71%. Main factors for successful quitting in one year included "intention to stop smoking completely by participating the contest", "age", "marriage", et al. But some factors had no effect on the successful quitting, such as cessation measures, helping from others, et al. CONCLUSION: It is important for smokers to make a decision and perseverance to stop smoking.

Spatial analysis of the distribution, risk factors and access to medical resources of patients with hepatitis B in Shenzhen, China.

Considering the high morbidity of hepatitis B in China, many epidemiological studies based on classic medical statistical analysis have been started but lack spatial information. However, spatial information such as the spatial distribution, autocorrelation and risk factors of the disease is of great help in studying patients with hepatitis B. This study examined 2851 cases of hepatitis B that were hospitalized in Shenzhen in 2010 and studied the spatial distribution, risk factors and spatial access to health services using spatial interpolation, Pearson correlation analysis and the improved two-step floating catchment area method. The results showed that the spatial distribution of hepatitis B, along with risk factors as well as spatial access to the regional medical resources, was uneven and mainly concentrated in the south and southwest of Shenzhen in 2010. In addition, the distribution characteristics of hepatitis B revealed a positive correlation between four types of service establishments and risk factors for the disease. The Pearson correlation coefficients are 0.566, 0.515, 0.626, 0.538 corresponding to bath centres, beauty salons, massage parlours and pedicure parlours (p < 0.05). Additionally, the allocation of medical resources for hepatitis B is adequate, as most patients could be treated at nearby hospitals.

Spatial analysis of the home addresses of hospital patients with hepatitis B infection or hepatoma in Shenzhen, China from 2010 to 2012.

BACKGROUND: Hepatoma associated with hepatitis B infection is a major public health problem in Shenzhen (China) and worldwide. China has the largest number of people infected with the hepatitis B virus (HBV), and many studies have demonstrated that HBV infection is associated with hepatoma development. Shenzhen officials have been attempting to monitor and control these diseases for many years. The methodology and the results of this study may be useful in developing a system to monitor, prevent and control these diseases. METHODS: The aim of the study was to analyze HBV infection and hepatoma distribution characteristics and patterns in Shenzhen by combining geographic information system (GIS) technology and spatial analysis. The study used data from patients at the district level from the 2010-2012 population censuses. RESULTS: Only one-third of the patients were female, and 70.7% of all cases were 20-50 years of age. There was no global spatial correlation of the distribution of hepatitis B infections and hepatomas; however, there was a local spatial correlation, and certain sub-districts of the Nanshan district had significant agglomeration effects. Based on incidence density and rate maps, we can conclude that the Shenzhen special zone had a higher incidence density and rate of hepatitis B infections and hepatomas compared with the area outside of the Shenzhen special zone during 2010-2012. CONCLUSIONS: This study demonstrated substantial geographic variation in the incidence of hepatitis B infection and hepatoma in Shenzhen. The prediction and control of hepatitis B infections and hepatoma development and interventions for these diseases should focus on disadvantaged areas to reduce disparities. GIS and spatial analysis play an important role in public health risk-reduction programs and may become integral components in the epidemiologic description, analysis and risk assessment of hepatitis B and hepatoma.

Analysis of the spatial variation of hospitalization admissions for hypertension disease in Shenzhen, China.

In China, awareness about hypertension, the treatment rate and the control rate are low compared to developed countries, even though China's aging population has grown, especially in those areas with a high degree of urbanization. However, limited epidemiological studies have attempted to describe the spatial variation of the geo-referenced data on hypertension disease over an urban area of China. In this study, we applied hierarchical Bayesian models to explore the spatial heterogeneity of the relative risk for hypertension admissions throughout Shenzhen in 2011. The final model specification includes an intercept and spatial components (structured and unstructured). Although the road density could be used as a covariate in modeling, it is an indirect factor on the relative risk. In addition, spatial scan statistics and spatial analysis were utilized to identify the spatial pattern and to map the clusters. The results showed that the relative risk for hospital admission for hypertension has high-value clusters in the south and southeastern Shenzhen. This study aimed to identify some specific regions with high relative risk, and this information is useful for the health administrators. Further research should address more-detailed data collection and an explanation of the spatial patterns.

Spatio-temporal variation and prediction of ischemic heart disease hospitalizations in Shenzhen, China.

Ischemic heart disease (IHD) is a leading cause of death worldwide. Urban public health and medical management in Shenzhen, an international city in the developing country of China, is challenged by an increasing burden of IHD. This study analyzed the spatio-temporal variation of IHD hospital admissions from 2003 to 2012 utilizing spatial statistics, spatial analysis, and space-time scan statistics. The spatial statistics and spatial analysis measured the incidence rate (hospital admissions per 1,000 residents) and the standardized rate (the observed cases standardized by the expected cases) of IHD at the district level to determine the spatio-temporal distribution and identify patterns of change. The space-time scan statistics was used to identify spatio-temporal clusters of IHD hospital admissions at the district level. The other objective of this study was to forecast the IHD hospital admissions over the next three years (2013-2015) to predict the IHD incidence rates and the varying burdens of IHD-related medical services among the districts in Shenzhen. The results show that the highest hospital admissions, incidence rates, and standardized rates of IHD are in Futian. From 2003 to 2012, the IHD hospital admissions exhibited similar mean centers and directional distributions, with a slight increase in admissions toward the north in accordance with the movement of the total population. The incidence rates of IHD exhibited a gradual increase from 2003 to 2012 for all districts in Shenzhen, which may be the result of the rapid development of the economy and the increasing traffic pollution. In addition, some neighboring areas exhibited similar temporal change patterns, which were also detected by the spatio-temporal cluster analysis. Futian and Dapeng would have the highest and the lowest hospital admissions, respectively, although these districts have the highest incidence rates among all of the districts from 2013 to 2015 based on the prediction using the GM (1,1). In addition, the combined analysis of the prediction of IHD hospital admissions and the general hospital distributions shows that Pingshan and Longgang might experience the most serious burden of IHD hospital services in the near future, although Futian would still have the greatest number and the highest incidence rate of hospital admissions for IHD.