Comparing the performance of time series models with or without meteorological factors in predicting incident pulmonary tuberculosis in eastern China

Forecasting tuberculosis incidence: a review of time series and machine learning models for prediction and eradication strategies

Despite efforts by the World Health Organization (WHO), tuberculosis (TB) remains a leading cause of fatalities globally. This study reviews time series and machine learning models for TB incidence prediction, identifies popular algorithms, and highlights the need for further research to improve accuracy and global scope. SCOPUS, PUBMED, IEEE, Web of Science, and PRISMA were used for search and article selection from 2012 to 2023. The results revealed that ARIMA, SARIMA, ETS, GRNN, BPNN, NARNN, NNAR, and RNN are popular time series and ML algorithms adopted for TB incidence rate predictions. The inaccurate TB incidence prediction and limited global scope of prior studies suggest a need for further research. This review serves as a roadmap for the WHO to focus on regions that require more attention for TB prevention and the need for more sophisticated models for TB incidence predictions.

Trajectories tracking of maternal and neonatal health in eastern China from 2010 to 2021: A multicentre cross-sectional study

Background

China's fertility policy has dramatically changed in the past decade with the successive promulgation of the partial two-child policy, universal two-child policy and three-child policy. The trajectories of maternal and neonatal health accompanied the changes in fertility policy are unknown.

Methods

We obtained data of 280 203 deliveries with six common pregnancy complications and thirteen perinatal outcomes between 2010 and 2021 in eastern China. The average annual percent change (AAPC) was calculated to evaluated the temporal trajectories of obstetric characteristics and adverse outcomes during this period. Then, the autoregressive integrated moving average (ARIMA) models were constructed to project future trend of obstetric characteristics and outcomes until 2027.

Results

The proportion of advanced maternal age (AMA), assisted reproduction technology (ART) treatment, gestational diabetes mellitus (GDM), anaemia, thrombocytopenia, thyroid dysfunction, oligohydramnios, placental abruption, small for gestational age (SGA) infants, and congenital malformation significantly increased from 2010 to 2021. However, the placenta previa, large for gestational age (LGA) infants and stillbirth significantly decreased during the same period. The AMA and ART treatment were identified as independent risk factors for the uptrends of pregnancy complications and adverse perinatal outcomes. The overall caesarean section rate remained above 40%. Importantly, among multiparas, a previous caesarean section was found to be associated with a significantly reduced risk of hypertensive disorders of pregnancy (HDP), premature rupture of membranes (PROM), placenta previa, placental abruption, perinatal asphyxia, LGA infants, stillbirths, and preterm births. In addition, the ARIMA time series models predicted increasing trends in the ART treatment, GDM, anaemia, thrombocytopenia, postpartum haemorrhage, congenital malformation, and caesarean section until 2027. Conversely, a decreasing trend was predicted for HDP, PROM, and placental abruption premature, LGA infants, SGA infants, perinatal asphyxia, and stillbirth.

Conclusions

Maternal and neonatal adverse outcomes became more prevalent from 2010 to 2021 in China. Maternal age and ART treatment were independent risk factors for adverse obstetric outcomes. The findings offered comprehensive trajectories for monitoring pregnancy complications and perinatal outcomes in China, and provided robust intervention targets in obstetric safety. The development of early prediction models and the implementation of prevention efforts for adverse obstetric events are necessary to enhance obstetric safety.

Predicting monthly hospital outpatient visits based on meteorological environmental factors using the ARIMA model

Accurate forecasting of hospital outpatient visits is beneficial to the rational planning and allocation of medical resources to meet medical needs. Several studies have suggested that outpatient visits are related to meteorological environmental factors. We aimed to use the autoregressive integrated moving average (ARIMA) model to analyze the relationship between meteorological environmental factors and outpatient visits. Also, outpatient visits can be forecast for the future period. Monthly outpatient visits and meteorological environmental factors were collected from January 2015 to July 2021. An ARIMAX model was constructed by incorporating meteorological environmental factors as covariates to the ARIMA model, by evaluating the stationary [Formula: see text], coefficient of determination [Formula: see text], mean absolute percentage error (MAPE), and normalized Bayesian information criterion (BIC). The ARIMA [Formula: see text] model with the covariates of [Formula: see text], [Formula: see text], and [Formula: see text] was the optimal model. Monthly outpatient visits in 2019 can be predicted using average data from past years. The relative error between the predicted and actual values for 2019 was 2.77%. Our study suggests that [Formula: see text], [Formula: see text], and [Formula: see text] concentration have a significant impact on outpatient visits. The model built has excellent predictive performance and can provide some references for the scientific management of hospitals to allocate staff and resources.

Estimating the tuberculosis incidence using a SARIMAX-NNARX hybrid model by integrating meteorological factors in Qinghai Province, China

Tuberculosis (TB) is recognized as being a major public health concern owing to its increase in Qinghai, China. In this study, we aimed to estimate the long-term effects of meteorological variables on TB incidence and construct an advanced hybrid model with seasonal autoregressive integrated moving average (SARIMA) and a neural network nonlinear autoregression (SARIMAX-NNARX) by integrating meteorological factors and evaluating the model fitting and prediction effect. During 2005-2017, TB experienced an upward trend with obvious periodic and seasonal characteristics, peaking in spring and winter. The results showed that TB incidence was positively correlated with average relative humidity (ARH) with a 2-month lag (β = 1.889, p = 0.003), but negatively correlated with average atmospheric pressure (AAP) with a 1-month lag (β =  - 1.633, p = 0.012), average temperature (AT) with a 2-month lag (β =  - 0.093, p = 0.027), and average wind speed (AWS) with a 0-month lag (β =  - 13.221, p = 0.033), respectively. The SARIMA (3,1,0)(1,1,1)_12, SARIMAX(3,1,0)(1,1,1)₁₂, and SARIMAX(3,1,0)(1,1,1)₁₂-NNARX(15,3) were considered preferred models based on the evaluation criteria. Of them, the SARIMAX-NNARX technique had smaller error values than the SARIMA and SARIMAX models in both fitting and forecasting aspects. The sensitivity analysis also revealed the robustness of the mixture forecasting model. Therefore, the SARIMAX-NNARX model by integrating meteorological variables can be used as an accurate method for forecasting the epidemic trends which would be great importance for TB prevention and control in the coming periods in Qinghai.

Predicting the Number of Reported Pulmonary Tuberculosis in Guiyang, China, Based on Time Series Analysis Techniques

Tuberculosis (TB) is one of the world's deadliest infectious disease killers today, and despite China's increasing efforts to prevent and control TB, the TB epidemic is still very serious. In the context of the COVID-19 pandemic, if reliable forecasts of TB epidemic trends can be made, they can help policymakers with early warning and contribute to the prevention and control of TB. In this study, we collected monthly reports of pulmonary tuberculosis (PTB) in Guiyang, China, from January 1, 2010 to December 31, 2020, and monthly meteorological data for the same period, and used LASSO regression to screen four meteorological factors that had an influence on the monthly reports of PTB in Guiyang, including sunshine hours, relative humidity, average atmospheric pressure, and annual highest temperature, of which relative humidity (6-month lag) and average atmospheric pressure (7-month lag) have a lagging effect with the number of TB reports in Guiyang. Based on these data, we constructed ARIMA, Holt-Winters (additive and multiplicative), ARIMAX (with meteorological factors), LSTM, and multivariable LSTM (with meteorological factors). We found that the addition of meteorological factors significantly improved the performance of the time series prediction model, which, after comprehensive consideration, included the ARIMAX (1,1,1) (0,1,2)₁₂ model with a lag of 7 months at the average atmospheric pressure, outperforms the other models in terms of both fit (RMSE = 37.570, MAPE = 10.164%, MAE = 28.511) and forecast sensitivity (RMSE = 20.724, MAPE = 6.901%, MAE = 17.306), so the ARIMAX (1,1,1) (0,1,2)₁₂ model with a lag of 7 months can be used as a predictor tool for predicting the number of monthly reports of PTB in Guiyang, China.

Contributions of ambient temperature and relative humidity to the risk of tuberculosis admissions: A multicity study in Central China

BACKGROUND

As a communicable disease and major public health issue, many studies have quantified the associations between tuberculosis (TB) and meteorological factors with inconsistent results. The purpose of this multicenter study was to characterize the associations between ambient temperature, humidity and the risk of TB hospitalizations and to investigate potential heterogeneity.

METHOD

Data on daily hospitalizations for TB, meteorological factors and ambient air pollutants for 16 cities in Anhui Province were collected from 2015 to 2020. A distributed lag nonlinear model (DLNM) was performed to obtain the estimates of meteorological-TB relationships by cities. Then, we used the multivariate meta-regression model to pool the city-specific estimates with air pollution, demographic indicators, medical resource and latitude as potential modifiers to explore the sources of heterogeneity. Finally, we divided the whole province into three regions to validate the meteorological-TB relationships by regions.

RESULTS

The overall pooled temperature-TB association presented an approximate S-shaped curve, with relative risk (RR) peaking at 5 °C (RR = 1.536, 95% CI: 1.303-1.811) compared to the reference temperature (27 °C). Lag-response curve suggested that low temperature exposure increased the risk of TB hospitalizations at lag 0 and 1 day (lag0 day: RR = 1.136, 95% CI: 1.048-1.231, lag1 day: RR = 1.052, 95% CI: 1.023-1.082). However, the overall exposure-response curve between relative humidity and TB showed almost horizontal line with reference relative humidity to 78%. The residual heterogeneity ranged from 27.1% to 36.9%, with air pollution, latitude and medical resource explained the largest proportion.

CONCLUSION

We found that low temperature exposure is associated with an acute increased risk of TB hospitalizations in Anhui Province. The association between temperature and TB admission varies depending on air pollution, latitude, and medical resources. Since the effect of short-term exposure to humidity is not significant, further studies are supposed to focus on the long-term effect of humidity.

Association between averaged meteorological factors and tuberculosis risk: A systematic review and meta-analysis

Inconsistencies were discovered in the findings regarding the effects of meteorological factors on tuberculosis (TB). This study conducted a systematic review of published studies on the relationship between TB and meteorological factors and used a meta-analysis to investigate the pooled effects in order to provide evidence for future research and policymakers. The literature search was completed by August 3rd, 2021, using three databases: PubMed, Web of Science and Embase. Relative risks (RRs) in included studies were extracted and all effect estimates were combined together using meta-analysis. Subgroup analyses were carried out based on the resolution of exposure time, regional climate, and national income level. A total of eight studies were included after screening for inclusion and exclusion criteria. Our results show that TB risk was positively correlated with precipitation (RR = 1.32, 95% CI: 1.14, 1.51), while temperature (RR = 1.15, 95% CI: 1.00, 1.32), humidity (RR = 1.05, 95% CI: 0.99, 1.10), air pressure (RR = 0.89, 95% CI: 0.69, 1.14) and sunshine duration (RR = 0.95, 95% CI: 0.80, 1.13) all had no statistically significant correlation. Subgroup analysis shows that quarterly measure resolution, low and middle Human Development Index (HDI) level and subtropical climate increase TB risk not only in precipitation, but also in temperature and humidity. Moreover, less heterogeneity was observed in "high and extremely high" HDI areas and subtropical areas than that in other subgroups (I² = 0%). Precipitation, a subtropical climate, and a low HDI level are all positive influence factors to tuberculosis. Therefore, residents and public health managers should take precautionary measures ahead of time, especially in extreme weather conditions.

Application of the ARIMA Model in Forecasting the Incidence of Tuberculosis in Anhui During COVID-19 Pandemic from 2021 to 2022

Objective

Forecasting the seasonality and trend of pulmonary tuberculosis is important for the rational allocation of health resources. In this study, we predict the incidence of pulmonary tuberculosis by establishing the autoregressive integrated moving average (ARIMA) model and providing support for pulmonary tuberculosis prevention and control during COVID-19 pandemic.

Methods

Registered tuberculosis(TB) cases from January 2013 to December 2020 in Anhui province were analysed using traditional descriptive epidemiological methods. Then we used the monthly incidence rate of TB from January 2013 through June 2020 to construct ARIMA model, and used the incidence rate from July 2020 to December 2020 to evaluate the forecasting accuracy. Ljung Box test, Akaike's information criterion(AICc), Bayesian information criterion(BIC) and Realtive error were used to evaluate the model fitting and forecasting effect, Finally, the optimal model was used to forecast the expected monthly incidence of tuberculosis for 2021 and 2022 to learn about the incidence trend.

Results

A total of 255,656 TB cases were registered. The reported rate of tuberculosis was highest in 2013 and lowest in 2020. The peak incidence was in March, Tongling (71.97/100,000), Chizhou (59.93/100,000), and Huainan (58.36/100,000) had the highest number of cases. The ratio of male to female incidence was 2.59:1, with the largest proportion of people being between 66 and 75 years old. The main occupation of patients was farmer. ARIMA (0, 1, 1) (0, 1, 1)₁₂ model was the optimal model to forecast the incidence trend of TB.

Conclusion

Tongling, Chizhou, and Huainan should strengthen measures for TB. In particular, the government should pay more attention on elderly people to prevent tuberculosis infections. The rate of TB patient registration and reporting has decreased under the pandemic of COVID-19. The ARIMA model can be a useful tool for predicting future TB cases.

Time trend prediction and spatial-temporal analysis of multidrug-resistant tuberculosis in Guizhou Province, China, during 2014-2020

Background

Guizhou is located in the southwest of China with high multidrug-resistant tuberculosis (MDR-TB) epidemic. To fight this disease, Guizhou provincial authorities have made efforts to establish MDR-TB service system and perform the strategies for active case finding since 2014. The expanded case finding starting from 2019 and COVID-19 pandemic may affect the cases distribution. Thus, this study aims to analyze MDR-TB epidemic status from 2014 to 2020 for the first time in Guizhou in order to guide control strategies.

Methods

Data of notified MDR-TB cases were extracted from the National TB Surveillance System correspond to population information for each county of Guizhou from 2014 to 2020. The percentage change was calculated to quantify the change of cases from 2014 to 2020. Time trend and seasonality of case series were analyzed by a seasonal autoregressive integrated moving average (SARIMA) model. Spatial–temporal distribution at county-level was explored by spatial autocorrelation analysis and spatial–temporal scan statistic.

Results

Guizhou has 9 prefectures and 88 counties. In this study, 1,666 notified MDR-TB cases were included from 2014–2020. The number of cases increased yearly. Between 2014 and 2019, the percentage increase ranged from 6.7 to 21.0%. From 2019 to 2020, the percentage increase was 62.1%. The seasonal trend illustrated that most cases were observed during the autumn with the trough in February. Only in 2020, a peak admission was observed in June. This may be caused by COVID-19 pandemic restrictions being lifted until May 2020. The spatial–temporal heterogeneity revealed that over the years, most MDR-TB cases stably aggregated over four prefectures in the northwest, covering Bijie, Guiyang, Liupanshui and Zunyi. Three prefectures (Anshun, Tongren and Qiandongnan) only exhibited case clusters in 2020.

Conclusion

This study identified the upward trend with seasonality and spatial−temporal clusters of MDR-TB cases in Guizhou from 2014 to 2020. The fast rising of cases and different distribution from the past in 2020 were affected by the expanded case finding from 2019 and COVID-19. The results suggest that control efforts should target at high-risk periods and areas by prioritizing resources allocation to increase cases detection capacity and better access to treatment.

Modeling and Predicting Pulmonary Tuberculosis Incidence and Its Association with Air Pollution and Meteorological Factors Using an ARIMAX Model: An Ecological Study in Ningbo of China

The autoregressive integrated moving average with exogenous regressors (ARIMAX) modeling studies of pulmonary tuberculosis (PTB) are still rare. This study aims to explore whether incorporating air pollution and meteorological factors can improve the performance of a time series model in predicting PTB. We collected the monthly incidence of PTB, records of six air pollutants and six meteorological factors in Ningbo of China from January 2015 to December 2019. Then, we constructed the ARIMA, univariate ARIMAX, and multivariate ARIMAX models. The ARIMAX model incorporated ambient factors, while the ARIMA model did not. After prewhitening, the cross-correlation analysis showed that PTB incidence was related to air pollution and meteorological factors with a lag effect. Air pollution and meteorological factors also had a correlation. We found that the multivariate ARIMAX model incorporating both the ozone with 0-month lag and the atmospheric pressure with 11-month lag had the best performance for predicting the incidence of PTB in 2019, with the lowest fitted mean absolute percentage error (MAPE) of 2.9097% and test MAPE of 9.2643%. However, ARIMAX has limited improvement in prediction accuracy compared with the ARIMA model. Our study also suggests the role of protecting the environment and reducing pollutants in controlling PTB and other infectious diseases.

Study on the Associations between Meteorological Factors and the Incidence of Pulmonary Tuberculosis in Xinjiang, China

Pulmonary tuberculosis (PTB) has been a major threat to global public health. The association between meteorological factors and the incidence of PTB has been widely investigated by the generalized additive model, auto-regressive integrated moving average model and the distributed lag model, etc. However, these models could not address a non-linear or lag correlation between them. In this paper, a penalized distributed lag non-linear model, as a generalized and improved one, was applied to explore the influence of meteorological factors (such as air temperature, relative humidity and wind speed) on the PTB incidence in Xinjiang from 2004 to 2019. Moreover, we firstly use a comprehensive index (apparent temperature, AT) to access the impact of multiple meteorological factors on the incidence of PTB. It was found that the relationships between air temperature, relative humidity, wind speed, AT and PTB incidence were nonlinear (showed “wave-type “, “invested U-type”, “U-type” and “wave-type”, respectively). When air temperature at the lowest value (−16.1 °C) could increase the risk of PTB incidence with the highest relative risk (RR = 1.63, 95% CI: 1.21–2.20). An assessment of relative humidity demonstrated an increased risk of PTB incidence between 44.5% and 71.8% with the largest relative risk (RR = 1.49, 95% CI: 1.32–1.67) occurring at 59.2%. Both high and low wind speeds increased the risk of PTB incidence, especially at the lowest wind speed 1.4 m/s (RR = 2.20, 95% CI: 1.95–2.51). In particular, the lag effects of low and high AT on PTB incidence were nonlinear. The lag effects of extreme cold AT (−18.5 °C, 1st percentile) on PTB incidence reached a relative risk peak (RR = 2.18, 95% CI: 2.06–2.31) at lag 1 month. Overall, it was indicated that the environment with low air temperature, suitable relative humidity and wind speed is more conducive to the transmission of PTB, and low AT is associated significantly with increased risk of PTB in Xinjiang.

Delayed Diagnosis and Treatment of Cancer Patients During the COVID-19 Pandemic in Henan, China: An Interrupted Time Series Analysis

Objective

To investigate the possible impact of lockdown policies on the diagnosis and treatment of cancer patients in Henan, China.

Design Setting and Participants

We collected data from the Henan Cancer Hospital, affiliated with Zhengzhou University. The monthly numbers of inpatient admissions from January 2014 to December 2019 were used to forecast the number of inpatient admissions in 2020, which was then compared to the actual number of patients admitted during the pandemic to evaluate how the actual number diverges from this forecast. We conducted an interrupted time series analysis using the autoregressive integrated moving average (ARIMA) model.

Main Outcomes and Measures

For specific diagnoses, treatment modalities, and age groups, we compared the changes in monthly admissions after the pandemic with the forecasted changes from the model.

Results

The observed overall monthly number of inpatient admissions decreased by 20.2% [95% confidence interval (CI), 11.7-27.2%], 78.9% (95% CI, 77.3-80.4%), and 40.9% (95% CI, 35.6-45.5%) in January, February, and March 2020, respectively, as compared with those predicted using the ARIMA model. After the lockdown, visits for all treatment modalities decreased sharply. However, apparent compensation and recovery of the backlog appeared in later surgeries. As a result, the number of patients who underwent surgery in 2020 (30,478) was close to the number forecasted by the ARIMA model (30,185). In the same period, patients who received other treatments or underwent examinations were 106,074 and 36,968, respectively; the respective numbers that were forecasted by ARIMA were 127,775 and 60,025, respectively. These findings depict a decrease of 16.9 and 38.4% in patients who received other treatments or underwent examinations only, respectively. Regarding diagnosis, the reported incidence of various cancers decreased dramatically in February, with varying extent and speed of recovery.

Conclusion and Relevance

The COVID-19 pandemic has significantly delayed the diagnosis and treatment of cancer in Henan, China. Long-term research should be conducted to assess the future effects of lockdown policies.

Meteorological factors contribute to the risk of pulmonary tuberculosis: A multicenter study in eastern China

BACKGROUND

Most studies on associations between meteorological factors and tuberculosis (TB) were conducted in a single city, used different lag times, or merely explored the qualitative associations between meteorological factors and TB. Thus, we performed a multicenter study to quantitatively evaluate the effects of meteorological factors on the risk of pulmonary tuberculosis (PTB).

METHODS

We collected data on newly diagnosed PTB cases in 13 study sites in Jiangsu Province between January 1, 2014, and December 31, 2019. Data on meteorological factors, air pollutants, and socioeconomic factors at these sites during the same period were also collected. We applied the generalized additive mixed model to estimate the associations between meteorological factors and PTB.

RESULTS

There were 20,472 newly diagnosed PTB cases reported in the 13 study sites between 2014 and 2019. The median (interquartile range) weekly average temperature, weekly average wind speed, and weekly average relative humidity of these sites were 17.3 °C (8.0-24.1), 2.2 m/s (1.8-2.7), and 75.1% (67.1-82.0), respectively. In the single-meteorological-factor models, for a unit increase in weekly average temperature, weekly average wind speed, and weekly average relative humidity, the risk of PTB decreased by 0.9% [lag 0-13 weeks, 95% confidence interval (CI): -1.5, -0.4], increased by 56.2% (lag 0-16 weeks, 95% CI: 32.6, 84.0) when average wind speed was <3 m/s, and decreased by 28.1% (lag 0-14 weeks, 95% CI: -39.2, -14.9) when average relative humidity was ≥72%, respectively. Moreover, the associations remained significant in the multi-meteorological-factor models.

CONCLUSIONS

Average temperature and average relative humidity (≥72%) are negatively associated with the risk of PTB. In contrast, average wind speed (<3 m/s) is positively related to the risk of PTB, suggesting that an environment with low temperature, relatively high wind speed, and low relative humidity is conducive to the transmission of PTB.

Forecasting the Tuberculosis Incidence Using a Novel Ensemble Empirical Mode Decomposition-Based Data-Driven Hybrid Model in Tibet, China

Objective

The purpose of this study is to develop a novel data-driven hybrid model by fusing ensemble empirical mode decomposition (EEMD), seasonal autoregressive integrated moving average (SARIMA), with nonlinear autoregressive artificial neural network (NARNN), called EEMD-ARIMA-NARNN model, to assess and forecast the epidemic patterns of TB in Tibet.

Methods

The TB incidence from January 2006 to December 2017 was obtained, and then the time series was partitioned into training subsamples (from January 2006 to December 2016) and testing subsamples (from January to December 2017). Among them, the training set was used to develop the EEMD-SARIMA-NARNN combined model, whereas the testing set was used to validate the forecasting performance of the model. Whilst the forecasting accuracy level of this novel method was compared with the basic SARIMA model, basic NARNN model, error-trend-seasonal (ETS) model, and traditional SARIMA-NARNN mixture model.

Results

By comparing the accuracy level of the forecasting measurements including root-mean-square error, mean absolute deviation, mean error rate, mean absolute percentage error, and root-mean-square percentage error, it was shown that the EEMD-SARIMA-NARNN combined method produced lower error rates than the others. The descriptive statistics suggested that TB was a seasonal disease, peaking in late winter and early spring and a trough in autumn and early winter, and the TB epidemic indicated a drastic increase by a factor of 1.7 from 2006 to 2017 in Tibet, with average annual percentage change of 5.8 (95% confidence intervals: 3.5–8.1).

Conclusion

This novel data-driven hybrid method can better consider both linear and nonlinear components in the TB incidence than the others used in this study, which is of great help to estimate and forecast the future epidemic trends of TB in Tibet. Besides, under present trends, strict precautionary measures are required to reduce the spread of TB in Tibet.