On time series analysis of public health and biomedical data

Salzburg Intensive Care database (SICdb): a detailed exploration and comparative analysis with MIMIC-IV

The utilization of artificial intelligence (AI) in healthcare is on the rise, demanding increased accessibility to (public) medical data for benchmarking. The digitization of healthcare in recent years has facilitated medical data scientists' access to extensive hospital data, fostering AI-based research. A notable addition to this trend is the Salzburg Intensive Care database (SICdb), made publicly available in early 2023. Covering over 27 thousand intensive care admissions at the University Hospital Salzburg from 2013 to 2021, this dataset presents a valuable resource for AI-driven investigations. This article explores the SICdb and conducts a comparative analysis with the widely recognized Medical Information Mart for Intensive Care - version IV (MIMIC-IV) database. The comparison focuses on key aspects, emphasizing the availability and granularity of data provided by the SICdb, particularly vital signs and laboratory measurements. The analysis demonstrates that the SICdb offers more detailed information with higher data availability and temporal resolution for signal data, especially for vital signs, compared to the MIMIC-IV. This is advantageous for longitudinal studies of patients' health conditions in the intensive care unit. The SICdb provides a valuable resource for medical data scientists and researchers. The database offers comprehensive and diverse healthcare data in a European country, making it well suited for benchmarking and enhancing AI-based healthcare research. The importance of ongoing efforts to expand and make public datasets available for advancing AI applications in the healthcare domain is emphasized by the findings.

Forecasting seasonal influenza activity in Canada-Comparing seasonal Auto-Regressive integrated moving average and artificial neural network approaches for public health preparedness

INTRODUCTION

Public health preparedness is based on timely and accurate information. Time series forecasting using disease surveillance data is an important aspect of preparedness. This study compared two approaches of time series forecasting: seasonal auto-regressive integrated moving average (SARIMA) modelling and the artificial neural network (ANN) algorithm. The goal was to model weekly seasonal influenza activity in Canada using SARIMA and compares its predictive accuracy, based on root mean square prediction error (RMSE) and mean absolute prediction error (MAE), to that of an ANN.

METHODS

An initial SARIMA model was fit using automated model selection by minimizing the Akaike information criterion (AIC). Further inspection of the autocorrelation function and partial autocorrelation function led to 'manual' model improvements. ANNs were trained iteratively, using an automated process to minimize the RMSE and MAE.

RESULTS

A total of 378, 462 cases of influenza was reported in Canada from the 2010-2011 influenza season to the end of the 2019-2020 influenza season, with an average yearly incidence risk of 20.02 per 100,000 population. Automated SARIMA modelling was the better method in terms of forecasting accuracy (per RMSE and MAE). However, the ANN correctly predicted the peak week of disease incidence while the other models did not.

CONCLUSION

Both the ANN and SARIMA models have shown to be capable tools in forecasting seasonal influenza activity in Canada. It was shown that applying both in tandem is beneficial, SARIMA better forecasted overall incidence while ANN correctly predicted the peak week.

Analyzing the COVID-19 Transmission Dynamics in Acre, Brazil: An Ecological Study

The north region of Brazil is characterized by significant vulnerabilities, notably surpassing national poverty indicators. These disparities exacerbated the impact of respiratory illnesses on the healthcare system during the COVID-19 pandemic, particularly in areas with limited healthcare resources, inadequate infrastructure, and barriers to healthcare access. The crisis was further influenced by multiple lineages that emerged as significant virus variants associated with increased transmissibility. Within this context, our ecological study focused on analyzing the epidemiological evolution of COVID-19 in the state of Acre. We constructed time-series trends in incidence, lethality, and mortality from March 2020 to December 2022 using the Prais-Winsten regression model. Our findings revealed that in 2020, there was an increasing trend in incidence, while mortality and lethality continued to decrease (p < 0.05). In the following year, both incidence and mortality decreased, while lethality increased at a rate of 1.02% per day. By the end of 2022, trends remained stationary across all rates. These results underscore the importance of ongoing surveillance and adaptive public health measures to bolster the resilience of healthcare systems in remote and vulnerable regions. Indeed, continuous monitoring of the most predominant SARS-CoV-2 lineages and their dynamics is imperative. Such proactive actions are essential for addressing emerging challenges and ensuring effective responses to adverse situations.

Higher-order Granger reservoir computing: simultaneously achieving scalable complex structures inference and accurate dynamics prediction

Recently, machine learning methods, including reservoir computing (RC), have been tremendously successful in predicting complex dynamics in many fields. However, a present challenge lies in pushing for the limit of prediction accuracy while maintaining the low complexity of the model. Here, we design a data-driven, model-free framework named higher-order Granger reservoir computing (HoGRC), which owns two major missions: The first is to infer the higher-order structures incorporating the idea of Granger causality with the RC, and, simultaneously, the second is to realize multi-step prediction by feeding the time series and the inferred higher-order information into HoGRC. We demonstrate the efficacy and robustness of the HoGRC using several representative systems, including the classical chaotic systems, the network dynamical systems, and the UK power grid system. In the era of machine learning and complex systems, we anticipate a broad application of the HoGRC framework in structure inference and dynamics prediction.

Applications of time series analysis in epidemiology: Literature review and our experience during COVID-19 pandemic

Time series analysis is a valuable tool in epidemiology that complements the classical epidemiological models in two different ways: Prediction and forecast. Prediction is related to explaining past and current data based on various internal and external influences that may or may not have a causative role. Forecasting is an exploration of the possible future values based on the predictive ability of the model and hypothesized future values of the external and/or internal influences. The time series analysis approach has the advantage of being easier to use (in the cases of more straightforward and linear models such as Auto-Regressive Integrated Moving Average). Still, it is limited in forecasting time, unlike the classical models such as Susceptible-Exposed-Infectious-Removed. Its applicability in forecasting comes from its better accuracy for short-term prediction. In its basic form, it does not assume much theoretical knowledge of the mechanisms of spreading and mutating pathogens or the reaction of people and regulatory structures (governments, companies, etc.). Instead, it estimates from the data directly. Its predictive ability allows testing hypotheses for different factors that positively or negatively contribute to the pandemic spread; be it school closures, emerging variants, etc. It can be used in mortality or hospital risk estimation from new cases, seroprevalence studies, assessing properties of emerging variants, and estimating excess mortality and its relationship with a pandemic.

Incorporating intraoperative blood pressure time-series variables to assist in prediction of acute kidney injury after type a acute aortic dissection repair: an interpretable machine learning model

BACKGROUND

Acute kidney injury (AKI) is a common and serious complication after the repair of Type A acute aortic dissection (TA-AAD). However, previous models have failed to account for the impact of blood pressure fluctuations on predictive performance. This study aims to develop machine learning (ML) models combined with intraoperative medicine and blood pressure time-series data to improve the accuracy of early prediction for postoperative AKI risk.

METHODS

Indicators reflecting the duration and depth of hypotension were obtained by analyzing continuous mean arterial pressure (MAP) monitored intraoperatively with multiple thresholds (<65, 60, 55, 50) set in the study. The predictive features were selected by logistic regression and the least absolute shrinkage and selection operator (LASSO), and 4 ML models were built based on the above features. The performance of the models was evaluated by area under receiver operating characteristic curve (AUROC), calibration curve and decision curve analysis (DCA). Shapley additive interpretation (SHAP) was used to explain the prediction models.

RESULTS

Among the indicators reflecting intraoperative hypotension, 65 mmHg showed a statistically superior difference to other thresholds in patients with or without AKI (p < .001). Among 4 models, the extreme gradient boosting (XGBoost) model demonstrated the highest AUROC: 0.800 (95% 0.683-0.917) and sensitivity: 0.717 in the testing set and was verified the best-performing model. The SHAP summary plot indicated that intraoperative urine output, cumulative time of mean arterial pressure lower than 65 mmHg outside cardiopulmonary bypass (OUT_CPB_MAP_65 time), autologous blood transfusion, and smoking were the top 4 features that contributed to the prediction model.

CONCLUSION

With the introduction of intraoperative blood pressure time-series variables, we have developed an interpretable XGBoost model that successfully achieve high accuracy in predicting the risk of AKI after TA-AAD repair, which might aid in the perioperative management of high-risk patients, particularly for intraoperative hemodynamic regulation.

The impact of ambient air pollutants on childhood respiratory system disease and the resulting disease burden: a time-series study

PURPOSE

The effects of air pollution on human health have long been a hot topic of research. For respiratory diseases, a large number of studies have proved that air pollution is one of the main causes. The purpose of this study was to investigate the risk of hospitalization of children with respiratory system diseases (CRSD) caused by six pollutants (PM2.5, PM10, NO2, SO2, CO, and O3) in Hefei City, and further calculate the disease burden.

METHOD

In the first stage, the generalized additive models were combined with the distributed lag non-linear models to evaluate the impact of air pollution on the inpatients for CRSD in Hefei. In the second stage, this study used the cost-of-illness approach to calculate the attributable number of hospitalizations and the extra disease burden.

RESULT

Overall, all the six kinds of pollutants had the strongest effects on CRSD inpatients within lag10 days. SO2 and CO caused the highest and lowest harm, respectively, and the RR values were SO2 (lag0-5): 1.1 20 (1.053, 1.191), and CO (lag0-6): 1.002 (1.001, 1.003). During the study period (January 1, 2014 to December 30, 2020), the 7-year cumulative burden of disease was 36.19 million CNY under the WHO air pollution standards.

CONCLUSION

In general, we found that six air pollutants were risk factors for CRSD in Hefei City, and create a huge burden of disease.

State Space Modeling of Event Count Time Series

This paper proposes a class of algorithms for analyzing event count time series, based on state space modeling and Kalman filtering. While the dynamics of the state space model is kept Gaussian and linear, a nonlinear observation function is chosen. In order to estimate the states, an iterated extended Kalman filter is employed. Positive definiteness of covariance matrices is preserved by a square-root filtering approach, based on singular value decomposition. Non-negativity of the count data is ensured, either by an exponential observation function, or by a newly introduced "affinely distorted hyperbolic" observation function. The resulting algorithm is applied to time series of the daily number of seizures of drug-resistant epilepsy patients. This number may depend on dosages of simultaneously administered anti-epileptic drugs, their superposition effects, delay effects, and unknown factors, making the objective analysis of seizure counts time series arduous. For the purpose of validation, a simulation study is performed. The results of the time series analysis by state space modeling, using the dosages of the anti-epileptic drugs as external control inputs, provide a decision on the effect of the drugs in a particular patient, with respect to reducing or increasing the number of seizures.

Digital Transformation of Cancer Care in the Era of Big Data, Artificial Intelligence and Data-Driven Interventions: Navigating the Field

OBJECTIVES

To navigate the field of digital cancer care and define and discuss key aspects and applications of big data analytics, artificial intelligence (AI), and data-driven interventions.

DATA SOURCES

Peer-reviewed scientific publications and expert opinion.

CONCLUSION

The digital transformation of cancer care, enabled by big data analytics, AI, and data-driven interventions, presents a significant opportunity to revolutionize the field. An increased understanding of the lifecycle and ethics of data-driven interventions will enhance development of innovative and applicable products to advance digital cancer care services.

IMPLICATIONS FOR NURSING PRACTICE

As digital technologies become integrated into cancer care, nurse practitioners and scientists will be required to increase their knowledge and skills to effectively use these tools to the patient's benefit. An enhanced understanding of the core concepts of AI and big data, confident use of digital health platforms, and ability to interpret the outputs of data-driven interventions are key competencies. Nurses in oncology will play a crucial role in patient education around big data and AI, with a focus on addressing any arising questions, concerns, or misconceptions to foster trust in these technologies. Successful integration of data-driven innovations into oncology nursing practice will empower practitioners to deliver more personalized, effective, and evidence-based care.

Medical Applications of Nonadditive Entropies

The Boltzmann-Gibbs additive entropy SBG=-k∑ipilnpi and associated statistical mechanics were generalized in 1988 into nonadditive entropy Sq=k1-∑ipiqq-1 and nonextensive statistical mechanics, respectively. Since then, a plethora of medical applications have emerged. In the present review, we illustrate them by briefly presenting image and signal processings, tissue radiation responses, and modeling of disease kinetics, such as for the COVID-19 pandemic.

A fast kernel independence test for cluster-correlated data

Cluster-correlated data receives a lot of attention in biomedical and longitudinal studies and it is of interest to assess the generalized dependence between two multivariate variables under the cluster-correlated structure. The Hilbert-Schmidt independence criterion (HSIC) is a powerful kernel-based test statistic that captures various dependence between two random vectors and can be applied to an arbitrary non-Euclidean domain. However, the existing HSIC is not directly applicable to cluster-correlated data. Therefore, we propose a HSIC-based test of independence for cluster-correlated data. The new test statistic combines kernel information so that the dependence structure in each cluster is fully considered and exhibits good performance under high dimensions. Moreover, a rapid p value approximation makes the new test fast applicable to large datasets. Numerical studies show that the new approach performs well in both synthetic and real world data.

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.

Prediction of low cardiac output syndrome in patients following cardiac surgery using machine learning

Background

This study aimed to develop machine learning models to predict Low Cardiac Output Syndrome (LCOS) in patients following cardiac surgery using machine learning algorithms.

Methods

The clinical data of cardiac surgery patients in Nanjing First Hospital between June 2019 and November 2020 were retrospectively extracted from the electronic medical records. Six conventional machine learning algorithms, including logistic regression, support vector machine, decision tree, random forest, extreme gradient boosting and light gradient boosting machine, were employed to construct the LCOS predictive models with all predictive features (full models) and selected predictive features (reduced models). The discrimination of these models was evaluated by the area under the receiver operating characteristic curve (AUC) and the calibration of the models was assessed by the calibration curve. Shapley Additive explanation (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) were used to interpret the predictive models.

Results

Data from 1,585 patients [982 (62.0%) were male, aged 18 to 88, 212 (13.4%) with LCOS] were employed to train and validate the LCOS models. Among the full models, the RF model (AUC: 0.909, 95% CI: 0.875–0.943; Sensitivity: 0.849, 95% CI: 0.724–0.933; Specificity: 0.835, 95% CI: 0.796–0.869) and the XGB model (AUC: 0.897, 95% CI: 0.859–0.935; Sensitivity: 0.830, 95% CI: 0.702–0.919; Specificity: 0.809, 95% CI: 0.768–0.845) exhibited well predictive power for LCOS. Eleven predictive features including left ventricular ejection fraction (LVEF), first post-operative blood lactate (Lac), left ventricular diastolic diameter (LVDd), cumulative time of mean artery blood pressure (MABP) lower than 65 mmHg (MABP &lt; 65 time), hypertension history, platelets level (PLT), age, blood creatinine (Cr), total area under curve above threshold central venous pressure (CVP) 12 mmHg and 16 mmHg, and blood loss during operation were used to build the reduced models. Among the reduced models, RF model (AUC: 0.895, 95% CI: 0.857–0.933; Sensitivity: 0.830, 95% CI: 0.702–0.919; Specificity: 0.806, 95% CI: 0.765–0.843) revealed the best performance. SHAP and LIME plot showed that LVEF, Lac, LVDd and MABP &lt; 65 time significantly contributed to the prediction model.

Conclusion

In this study, we successfully developed several machine learning models to predict LCOS after surgery, which may avail to risk stratification, early detection and management of LCOS after cardiac surgery.

Recent Techniques in Determining the Effects of Climate Change on Depressive Patients: A Systematic Review

Climate change is amongst the most serious issues nowadays. Climate change has become a concern for the scientific community as it could affect human health. Researchers have found that climate change potentially impacts human mental health, especially among depressive patients. However, the relationship is still unclear and needs further investigation. The purpose of this systematic review is to systematically evaluate the evidence of the association between climate change effects on depressive patients, investigate the effects of environmental exposure related to climate change on mental health outcomes for depressive patients, analyze the current technique used to determine the relationship, and provide the guidance for future research. Articles were identified by searching specified keywords in six electronic databases (Google Scholar, PubMed, Scopus, Springer, ScienceDirect, and IEEE Digital Library) from 2012 until 2021. Initially, 1823 articles were assessed based on inclusion criteria. After being analyzed, only 15 studies fit the eligibility criteria. The result from included studies showed that there appears to be strong evidence of the association of environmental exposure related to climate change in depressive patients. Temperature and air pollution are consistently associated with increased hospital admission of depressive patients; age and gender became the most frequently considered vulnerability factors. However, the current evidence is limited, and the output finding between each study is still varied and does not achieve a reasonable and mature conclusion regarding the relationship between the variables. Therefore, more evidence is needed in this domain study. Some variables might have complex patterns, and hard to identify the relationship. Thus, technique used to analyze the relationship should be strengthened to identify the relevant relationship.

Quantitative methods for climate change and mental health research: current trends and future directions

The quantitative literature on climate change and mental health is growing rapidly. However, the methodological quality of the evidence is heterogeneous, and there is scope for methodological improvement and innovation. The first section of this Personal View provides a snapshot of current methodological trends and issues in the quantitative literature on climate change and mental health, drawing on literature collected through a previous scoping review. The second part of this Personal View outlines opportunities for methodological innovation concerning the assessment of the relationship between climate change and mental health. We then highlight possible methodological innovations in intervention research and in the measurement of climate change and mental health-related variables. This section draws upon methods from public mental health, environmental epidemiology, and other fields. The objective is not to provide a detailed description of different methodological techniques, but rather to highlight opportunities to use diverse methods, collaborate across disciplines, and inspire methodological innovation. The reader will be referred to practical guidance on different methods when available. We hope this Personal View will constitute a roadmap and launching pad for methodological innovation for researchers interested in investigating a rapidly growing area of research.

Prediction of Red Blood Cell Demand for Pediatric Patients Using a Time-Series Model: A Single-Center Study in China

Background

Red blood cells (RBCs) are an essential factor to consider for modern medicine, but planning the future collection of RBCs and supply efforts for coping with fluctuating demands is still a major challenge.

Objectives

This study aimed to explore the feasibility of the time-series model in predicting the clinical demand of RBCs for pediatric patients each month.

Methods

Our study collected clinical RBC transfusion data from years 2014 to 2019 in the National Center for Children's Health (Beijing) in China, with the goal of constructing a time-series, autoregressive integrated moving average (ARIMA) model by fitting the monthly usage of RBCs from 2014 to 2018. Furthermore, the optimal model was used to forecast the monthly usage of RBCs in 2019, and we subsequently compared the data with actual values to verify the validity of the model.

Results

The seasonal multiplicative model SARIMA (0, 1, 1) (1, 1, 0)₁₂ (normalized BIC = 8.740, R² = 0.730) was the best prediction model and could better fit and predict the monthly usage of RBCs for pediatric patients in this medical center in 2019. The model residual sequence was white noise (Ljung-Box Q₍₁₈₎ = 15.127, P > 0.05), and its autocorrelation function (ACF) and partial autocorrelation function (PACF) coefficients also fell within the 95% confidence intervals (CIs). The parameter test results were statistically significant (P < 0.05). 91.67% of the actual values were within the 95% CIs of the forecasted values of the model, and the average relative error of the forecasted and actual values was 6.44%, within 10%.

Conclusions

The SARIMA model can simulate the changing trend in monthly usage of RBCs of pediatric patients in a time-series aspect, which represents a short-term prediction model with high accuracy. The continuously revised SARIMA model may better serve the clinical environments and aid with planning for RBC demand. A clinical study including more data on blood use should be conducted in the future to confirm these results.

Technology investigation on time series classification and prediction

Time series appear in many scientific fields and are an important type of data. The use of time series analysis techniques is an essential means of discovering the knowledge hidden in this type of data. In recent years, many scholars have achieved fruitful results in the study of time series. A statistical analysis of 120,000 literatures published between 2017 and 2021 reveals that the topical research about time series is mostly focused on their classification and prediction. Therefore, in this study, we focus on analyzing the technical development routes of time series classification and prediction algorithms. 87 literatures with high relevance and high citation are selected for analysis, aiming to provide a more comprehensive reference base for interested researchers. For time series classification, it is divided into supervised methods, semi-supervised methods, and early classification of time series, which are key extensions of time series classification tasks. For time series prediction, from classical statistical methods, to neural network methods, and then to fuzzy modeling and transfer learning methods, the performance and applications of these different methods are discussed. We hope this article can help aid the understanding of the current development status and discover possible future research directions, such as exploring interpretability of time series analysis and online learning modeling.

Research on prediction of daily admissions of respiratory diseases with comorbid diabetes in Beijing based on long short-term memory recurrent neural network

To compare the performance of generalized additive model (GAM) and long short-term memory recurrent neural network (LSTM-RNN) on the prediction of daily admissions of respiratory diseases with comorbid diabetes. Daily data on air pollutants, meteorological factors and hospital admissions for respiratory diseases from Jan 1st, 2014 to Dec 31st, 2019 in Beijing were collected. LSTM-RNN was used to predict the daily admissions of respiratory diseases with comorbid diabetes, and the results were compared with those of GAM. The evaluation indexes were calculated by five-fold cross validation. Compared with the GAM, the prediction errors of LSTM-RNN were significantly lower [root mean squared error (RMSE): 21.21±3.30 vs. 46.13±7.60, <0.01; mean absolute error (MAE): 14.64±1.99 vs. 36.08±6.20, <0.01], and the value was significantly higher (0.79±0.06 vs. 0.57±0.12, <0.01). In gender stratification, RMSE, MAE and values of LSTM-RNN were better than those of GAM in predicting female admission (all <0.05), but there were no significant difference in predicting male admission between two models (all >0.05). In seasonal stratification, RMSE and MAE of LSTM-RNN were lower than those of GAM in predicting warm season admission (all <0.05), but there was no significant difference in value (>0.05). There were no significant difference in RMSE, MAE and between the two models in predicting cold season admission (all >0.05). In the stratification of functional areas, the RMSE, MAE and values of LSTM-RNN were better than those of GAM in predicting core area admission (all <0.05). has lower prediction errors and better fitting than the GAM, which can provide scientific basis for precise allocation of medical resources in polluted weather in advance.

The seasonal behaviour of COVID-19 and its galectin-like culprit of the viral spike

Seasonal behaviour is an attribute of many viral diseases. Like other 'winter' RNA viruses, infections caused by the causative agent of COVID-19, SARS-CoV-2, appear to exhibit significant seasonal changes. Here we discuss the seasonal behaviour of COVID-19, emerging viral phenotypes, viral evolution, and how the mutational landscape of the virus affects the seasonal attributes of the disease. We propose that the multiple seasonal drivers behind infectious disease spread (and the spread of COVID-19 specifically) are in 'trade-off' relationships and can be better described within a framework of a 'triangle of viral persistence' modulated by the environment, physiology, and behaviour. This 'trade-off' exists as one trait cannot increase without a decrease in another. We also propose that molecular components of the virus can act as sensors of environment and physiology, and could represent molecular culprits of seasonality. We searched for flexible protein structures capable of being modulated by the environment and identified a galectin-like fold within the N-terminal domain of the spike protein of SARS-CoV-2 as a potential candidate. Tracking the prevalence of mutations in this structure resulted in the identification of a hemisphere-dependent seasonal pattern driven by mutational bursts. We propose that the galectin-like structure is a frequent target of mutations because it helps the virus evade or modulate the physiological responses of the host to further its spread and survival. The flexible regions of the N-terminal domain should now become a focus for mitigation through vaccines and therapeutics and for prediction and informed public health decision making.

Estimating the incidence of cocaine use and mortality with music lyrics about cocaine

In the United States, cocaine use and mortality have surged in the past 5 years. Considering cocaine’s reputation as a fashionable social drug, the rise of cocaine mentions in popular music may provide a signal of epidemiological trends of cocaine use. We characterized the relationship between mentions of cocaine in song lyrics and incidence of cocaine use and mortality in the US. Incidence of cocaine use from 2002 to 2017 was obtained from the National Survey on Drug Use and Health and cocaine overdose mortality rate from 2000 to 2017 was obtained from the Centers for Disease Control. Distributed lag models were fit using ordinary least squares on the first difference to identify associations between changes in cocaine lyric mentions and changes in incidence of cocaine use and mortality. A total of 5955 song lyrics with cocaine mentions were obtained from Lyrics.com. Cocaine mentions in song lyrics were stable from 2000 to 2010 then increased by 190% from 2010 to 2017. The first-order distributed lag model estimated that a 0.01 increase in mentions of cocaine in song lyrics is associated with an 11% increase in incidence of cocaine use within the same year and a 14% increase in cocaine mortality with a 2-year lag. Lag-times were confirmed with cross-correlation analyses and the association remained after accounting for street pricing of cocaine. Mentions of cocaine in song lyrics are associated with the rise of incidence of cocaine use and cocaine overdose mortality. Popular music trends are a potentially valuable tool for understanding cocaine epidemiology trends.

Impact of short-term air pollution exposure on acute coronary syndrome in two cohorts of industrial and non-industrial areas: A time series regression with 6,000,000 person-years of follow-up (ACS - Air Pollution Study)

BACKGROUND

There is a lack of studies directly comparing the effect of air pollution on acute coronary syndrome (ACS) occurrence in industrial and non-industrial areas.

OBJECTIVES

A comparison of association of air pollution exposure with ACS in two cohorts of industrially different areas.

MATERIALS AND METHODS

The study covered 6,000,000 person-years of follow-up and five pollutants between 2008 and 2017. A time series regression analysis with 7-lag was used to assess the effects air pollution on ACS.

RESULTS

A total of 9046 patients with ACS were included in the analysis, of whom 3895 (43.06%) had ST-elevation myocardial infarction (STEMI) - 45.39% from non-industrial area, and 42.37% from industrial area; and 5151 (56.94%) had non-ST-elevation myocardial infarction (NSTEMI) - 54.61% from non-industrial area and 57.63% from industrial area. The daily concentrations of PM_2.5, PM₁₀, NO₂, SO₂, CO were higher in industrial than in non-industrial area (P < 0.001). In non-industrial area, an increase of 10 μg/m³ of NO₂ concentration (Odds Ratio (OR) = 1.126, 95%CI = 1.009-1.257; P = 0.034, lag-0) and an increase of 1 mg/m³ in CO concentration (RR = 1.055, 95%CI = 1.010-1.103; P = 0.017, lag-0) were associated with an increase in the number of hospitalization due to NSTEMI (for industrial area increase of 10 μg/m³ in NO₂ (OR = 1.062, 95%CI = 1.020-1.094; P = 0.005, lag-0), SO₂ (OR = 1.061, 95%CI = 1.010-1.116; P = 0.018, lag-4), PM₁₀ (OR = 1.010, 95%CI = 1.001-1.030; P = 0.047, lag-6). In STEMI patients in industrial area, an increased hospitalization was found to be associated with an increase of 10 μg/m³ in SO₂ (OR = 1.094, 95%CI = 1.030-1.162; P = 0.002, lag-1), PM_2.5 (OR = 1.041, 95%CI = 1.020-1.073; P < 0.001, lag-1), PM₁₀ (OR = 1.030, 95%CI = 1.010-1.051; P < 0.001, lag-1). No effects of air pollution on the number of hospitalization due to STEMI were noted from non-industrial area.

CONCLUSION

The risk of air pollution-related ACS was higher in industrial over non-industrial area. The effect of NO₂ on the incidence of NSTEMI was observed in both areas. In industrial area, the effect of PMs and SO₂ on NSTEMI and STEMI were also observed. A clinical effect was more delayed in time in patients with NSTEMI, especially after exposure to PM₁₀. Chronic exposure to air pollution may underlie the differences in the short-term effect between particulate air pollution impact on the incidence of STEMI.

Changes in the Trend of Sexually Acquired Chlamydia Infections in Sweden and the Role of Testing: A Time Series Analysis

BACKGROUND

We investigated the notification trends of sexually acquired chlamydia (chlamydia) and its association with testing in Sweden before (1992-2004) and after (2009-2018) the discovery of a new variant of Chlamydia trachomatis (nvCT).

METHODS

We applied monthly time series analysis to study chlamydia trends and annual time series to study chlamydia rates adjusted for testing. We analyzed incidence nationally and by county group (based on able and unable to detect nvCT at time of discovery).

RESULTS

We present data on 606,000 cases of chlamydia and 9.9 million persons tested. We found a U-shaped chlamydia trend during the period 1992-2004, with an overall increase of 83.7% from 1996 onward. The period 2009-2018 began with a stable trend at a high incidence level followed by a decrease of 19.7% during the period 2015-2018. Peaks were seen in autumn and through during winter and summer. Similar results were observed by groups of county, although with varying levels of increase and decrease in both periods. Furthermore, increased testing volume was associated with increased chlamydia rates during the first period (P = 0.019) but not the second period.

CONCLUSIONS

Our results showed that chlamydia trends during the period 2009-2018 were not driven by testing, as they were during the period 1992-2004. This suggests less biased notified chlamydia rates and thus possibly a true decrease in chlamydia incidence rates. It is important to adjust case rates for testing intensity, and future research should target other potential factors influencing chlamydia rates.

The effect of nitrogen dioxide and atmospheric pressure on hospitalization risk for chronic obstructive pulmonary disease in Guangzhou, China

BACKGROUND

The relationship between air pollution and meteorological factors on diseases has become a research hotspot recently. Nevertheless, few studies have touched the inferences of nitrogen dioxide (NO₂) and atmospheric pressure (AP) on hospitalization risk for chronic obstructive pulmonary disease (COPD).

OBJECTIVES

To investigate the short-term impact of particulate air pollutants and meteorology factors on hospitalizations for COPD and quantify the corresponding risk burden of hospital admission.

METHODS

In our study, COPD cases were collected from Guangzhou Panyu Central Hospital (n = 11,979) from Dec of 2013 to Jun 2019. The 24-h average temperature, relative humidity (RH), wind speed (V), AP and other meteorological data were obtained from Guangzhou Meteorological Bureau. Air pollution data were collected from Guangzhou Air Monitoring Station. The influence of different NO₂ and AP values on COPD risk was quantified by a distributed lag nonlinear model (DLNM) combined with Poisson Regression and Time Series analysis.

RESULTS

We found that NO₂ had a non-linear relationship with the incidence of COPD, with an approximate "M" type, appearing at the peaks of 126 μg/m³ (RR = 1.32, 95%CI, 1.07 to 1.64) and 168 μg/m³ (RR = 1.21, 95%CI, 0.94 to 1.55), respectively. And the association between AP and COPD incidence exhibited an approximate J-shape with a peak occurring at 1035 hPa (RR = 1.16, 95% CI, 1.02 to 1.31).

CONCLUSIONS

The nonlinear relationship of NO₂ and AP on COPD admission risk in different periods of lag can be used to establish an early warning system for diseases and reduce the possible outbreaks and burdens of COPD in a sensitive population.

Methodological approach for measuring the effects of organisational-level interventions on employee withdrawal behaviour

Background

Theoretical frameworks have recommended organisational-level interventions to decrease employee withdrawal behaviours such as sickness absence and employee turnover. However, evaluation of such interventions has produced inconclusive results. The aim of this study was to investigate if mixed-effects models in combination with time series analysis, process evaluation, and reference group comparisons could be used for evaluating the effects of an organisational-level intervention on employee withdrawal behaviour.

Methods

Monthly data on employee withdrawal behaviours (sickness absence, employee turnover, employment rate, and unpaid leave) were collected for 58 consecutive months (before and after the intervention) for intervention and reference groups. In total, eight intervention groups with a total of 1600 employees participated in the intervention. Process evaluation data were collected by process facilitators from the intervention team. Overall intervention effects were assessed using mixed-effects models with an AR (1) covariance structure for the repeated measurements and time as fixed effect. Intervention effects for each intervention group were assessed using time series analysis. Finally, results were compared descriptively with data from process evaluation and reference groups to disentangle the organisational-level intervention effects from other simultaneous effects.

Results

All measures of employee withdrawal behaviour indicated statistically significant time trends and seasonal variability. Applying these methods to an organisational-level intervention resulted in an overall decrease in employee withdrawal behaviour. Meanwhile, the intervention effects varied greatly between intervention groups, highlighting the need to perform analyses at multiple levels to obtain a full understanding. Results also indicated that possible delayed intervention effects must be considered and that data from process evaluation and reference group comparisons were vital for disentangling the intervention effects from other simultaneous effects.

Conclusions

When analysing the effects of an intervention, time trends, seasonal variability, and other changes in the work environment must be considered. The use of mixed-effects models in combination with time series analysis, process evaluation, and reference groups is a promising way to improve the evaluation of organisational-level interventions that can easily be adopted by others.

Trends, structural changes, and assessment of time series models for forecasting hospital discharge due to death at a Mexican tertiary care hospital

BACKGROUND

Data on hospital discharges can be used as a valuable instrument for hospital planning and management. The quantification of deaths can be considered a measure of the effectiveness of hospital intervention, and a high percentage of hospital discharges due to death can be associated with deficiencies in the quality of hospital care.

OBJECTIVE

To determine the overall percentage of hospital discharges due to death in a Mexican tertiary care hospital from its opening, to describe the characteristics of the time series generated from the monthly percentage of hospital discharges due to death and to make and evaluate predictions.

METHODS

This was a retrospective study involving the medical records of 81,083 patients who were discharged from a tertiary care hospital from April 2007 to December 2019 (first 153 months of operation). The records of the first 129 months (April 2007 to December 2017) were used for the analysis and construction of the models (training dataset). In addition, the records of the last 24 months (January 2018 to December 2019) were used to evaluate the predictions made (test dataset). Structural change was identified (Chow test), ARIMA models were adjusted, predictions were estimated with and without considering the structural change, and predictions were evaluated using error indices (MAE, RMSE, MAPE, and MASE).

RESULTS

The total percentage of discharges due to death was 3.41%. A structural change was observed in the time series (March 2009, p>0.001), and ARIMA(0,0,0)(1,1,2)12 with drift models were adjusted with and without consideration of the structural change. The error metrics favored the model that did not consider the structural change (MAE = 0.63, RMSE = 0.81, MAPE = 25.89%, and MASE = 0.65).

CONCLUSION

Our study suggests that the ARIMA models are an adequate tool for future monitoring of the monthly percentage of hospital discharges due to death, allowing us to detect observations that depart from the described trend and identify future structural changes.

Three Statistical Approaches for Assessment of Intervention Effects: A Primer for Practitioners

INTRODUCTION

Statistical methods to assess the impact of an intervention are increasingly used in clinical research settings. However, a comprehensive review of the methods geared toward practitioners is not yet available.

METHODS AND MATERIALS

We provide a comprehensive review of three methods to assess the impact of an intervention: difference-in-differences (DID), segmented regression of interrupted time series (ITS), and interventional autoregressive integrated moving average (ARIMA). We also compare the methods, and provide illustration of their use through three important healthcare-related applications.

RESULTS

In the first example, the DID estimate of the difference in health insurance coverage rates between expanded states and unexpanded states in the post-Medicaid expansion period compared to the pre-expansion period was 5.93 (95% CI, 3.99 to 7.89) percentage points. In the second example, a comparative segmented regression of ITS analysis showed that the mean imaging order appropriateness score in the emergency department at a tertiary care hospital exceeded that of the inpatient setting with a level change difference of 0.63 (95% CI, 0.53 to 0.73) and a trend change difference of 0.02 (95% CI, 0.01 to 0.03) after the introduction of a clinical decision support tool. In the third example, the results from an interventional ARIMA analysis show that numbers of creatinine clearance tests decreased significantly within months of the start of eGFR reporting, with a magnitude of drop equal to -0.93 (95% CI, -1.22 to -0.64) tests per 100,000 adults and a rate of drop equal to 0.97 (95% CI, 0.95 to 0.99) tests per 100,000 per adults per month.

DISCUSSION

When choosing the appropriate method to model the intervention effect, it is necessary to consider the structure of the data, the study design, availability of an appropriate comparison group, sample size requirements, whether other interventions occur during the study window, and patterns in the data.

Alternative adjustment for seasonality and long-term time-trend in time-series analysis for long-term environmental exposures and disease counts

Background

Time-series analysis with case-only data is a prominent method for the effect of environmental determinants on disease events in environmental epidemiology. In this analysis, adjustment for seasonality and long-term time-trend is crucial to obtain valid findings. When applying this analysis for long-term exposure (e.g., months, years) of which effects are usually studied via survival analysis with individual-level longitudinal data, unlike its application for short-term exposure (e.g., days, weeks), a standard adjustment method for seasonality and long-term time-trend can extremely inflate standard error of coefficient estimates of the effects. Given that individual-level longitudinal data are difficult to construct and often available to limited populations, if this inflation of standard error can be solved, rich case-only data over regions and countries would be very useful to test a variety of research hypotheses considering unique local contexts.

Methods

We discuss adjustment methods for seasonality and time-trend used in time-series analysis in environmental epidemiology and explain why standard errors can be inflated. We suggest alternative methods to solve this problem. We conduct simulation analyses based on real data for Seoul, South Korea, 2002–2013, and time-series analysis using real data for seven major South Korean cities, 2006–2013 to identify whether the association between long-term exposure and health outcomes can be estimated via time-series analysis with alternative adjustment methods.

Results

Simulation analyses and real-data analysis confirmed that frequently used adjustment methods such as a spline function of a variable representing time extremely inflate standard errors of estimates for associations between long-term exposure and health outcomes. Instead, alternative methods such as a combination of functions of variables representing time can make sufficient adjustment with efficiency.

Conclusions

Our findings suggest that time-series analysis with case-only data can be applied for estimating long-term exposure effects. Rich case-only data such as death certificates and hospitalization records combined with repeated measurements of environmental determinants across countries would have high potentials for investigating the effects of long-term exposure on health outcomes allowing for unique contexts of local populations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-020-01199-1.

Mitigating heat-related mortality risk in Shanghai, China: system dynamics modeling simulations

Numerous studies in epidemiology, meteorology, and climate change research have demonstrated a significant association between abnormal ambient temperature and mortality. However, there is a shortage of research attention to a systematic assessment of potential mitigation measures which could effectively reduce the heat-related morbidity and mortality risks. This study first illustrates a conceptualization of a systems analysis version of urban framework for climate service (UFCS). It then constructs a system dynamics (SD) model for the UFCS and employs this model to quantify the impacts of heat waves on public health system in Shanghai and to evaluate the performances of two mitigation measures in the context of a real heat wave event in July 2013 in the city. Simulation results show that in comparison with the baseline without mitigation measures, if the hospital system could prepare 20% of beds available for emergency response to heat waves once receiving the warning in advance, the number of daily deaths could be reduced by 40-60 (15.8-19.5%) on the 2 days of day 7 and day 8; if increasing the minimum living allowance of 790 RMB/month in 2013 by 20%, the number of daily deaths could be reduced by 50-70 (17.7-21.9%) on the 2 days of day 8 and day 12. This tool can help policy makers systematically evaluate adaptation and mitigation options based on performance assessment, thus strengthening urban resilience to changing climate.

The Interactive Effects between Particulate Matter and Heat Waves on Circulatory Mortality in Fuzhou, China

The interactive effects between particulate matter (PM) and heat waves on circulatory mortality are under-researched in the context of global climate change. We aimed to investigate the interaction between heat waves and PM on circulatory mortality in Fuzhou, a city characterized by a humid subtropical climate and low level of air pollution in China. We collected data on deaths, pollutants, and meteorology in Fuzhou between January 2016 and December 2019. Generalized additive models were used to examine the effect of PM on circulatory mortality during the heat waves, and to explore the interaction between different PM levels and heat waves on the circulatory mortality. During heat waves, circulatory mortality was estimated to increase by 8.21% (95% confidence intervals (CI): 0.32–16.72) and 3.84% (95% CI: 0.28–7.54) per 10 μg/m³ increase of PM_2.5 and PM₁₀, respectively, compared to non-heat waves. Compared with low-level PM_2.5 concentration on non-heat waves layer, the high level of PM_2.5 concentration on heat waves layer has a significant effect on the cardiovascular mortality, and the effect value was 48.35% (95% CI: 6.37–106.89). Overall, we found some evidence to suggest that heat waves can significantly enhance the impact of PM on circulatory mortality.

Effects of Data Aggregation on Time Series Analysis of Seasonal Infections

Time series analysis in epidemiological studies is typically conducted on aggregated counts, although data tend to be collected at finer temporal resolutions. The decision to aggregate data is rarely discussed in epidemiological literature although it has been shown to impact model results. We present a critical thinking process for making decisions about data aggregation in time series analysis of seasonal infections. We systematically build a harmonic regression model to characterize peak timing and amplitude of three respiratory and enteric infections that have different seasonal patterns and incidence. We show that irregularities introduced when aggregating data must be controlled during modeling to prevent erroneous results. Aggregation irregularities had a minimal impact on the estimates of trend, amplitude, and peak timing for daily and weekly data regardless of the disease. However, estimates of peak timing of the more common infections changed by as much as 2.5 months when controlling for monthly data irregularities. Building a systematic model that controls for data irregularities is essential to accurately characterize temporal patterns of infections. With the urgent need to characterize temporal patterns of novel infections, such as COVID-19, this tutorial is timely and highly valuable for experts in many disciplines.

Application of Systems Engineering Principles and Techniques in Biological Big Data Analytics: A Review

In the past few decades, we have witnessed tremendous advancements in biology, life sciences and healthcare. These advancements are due in no small part to the big data made available by various high-throughput technologies, the ever-advancing computing power, and the algorithmic advancements in machine learning. Specifically, big data analytics such as statistical and machine learning has become an essential tool in these rapidly developing fields. As a result, the subject has drawn increased attention and many review papers have been published in just the past few years on the subject. Different from all existing reviews, this work focuses on the application of systems, engineering principles and techniques in addressing some of the common challenges in big data analytics for biological, biomedical and healthcare applications. Specifically, this review focuses on the following three key areas in biological big data analytics where systems engineering principles and techniques have been playing important roles: the principle of parsimony in addressing overfitting, the dynamic analysis of biological data, and the role of domain knowledge in biological data analytics.

Total TauProtein as Investigated by Cerebral Microdialysis Increases in Hypothermic Cardiac Arrest: A Pig Study

The understanding and neurological prognostication of hypoxic ischemic encephalopathy (HIE) after hypothermic cardiac arrest (CA) is limited. Recent data suggest that the protein tau (total tau) might be a useful marker for outcome in patients with HIE. This translational porcine study aimed to analyze brain physiology in relation to total tau protein release during hypothermic CA. Eight domestic pigs were studied as part of a prospective porcine study using cerebral microdialysis (CMD). CMD samples for tau analysis were collected at baseline, after reaching the targeted core temperature of 28°C (hypothermia), after hypoxic hypercapnia (partial asphyxia), and finally 20 minutes after cardiopulmonary resuscitation. CMD-total tau-protein was analyzed using enzyme-linked immunosorbent essay. Cerebral tau protein was slightly elevated at baseline most likely due to an insertion trauma, remained stable during hypercapnic hypoxia, and significantly (p = 0.009) increased in 8/8 pigs during resuscitation to 1335 pg/mL (interquartile range: 705-2100). CMD-tau release was associated with lower levels of brain tissue oxygen tension (p = 0.011), higher CMD-lactate/pyruvate ratio, higher CMD-lactate, CMD-glutamate, and CMD-glycerol levels (p < 0.001, respectively), but not with cerebral perfusion pressure, intracranial pressure, or CMD-glucose levels. This study demonstrates an immediate tau protein release accompanied by deranged cerebral metabolism and decreased brain tissue oxygen tension during mechanical resuscitation in hypothermic CA. Understanding tau physiology and release kinetics is important for the design and interpretation of studies investigating tau as a biomarker of HIE.

Vitamin D Serum Levels in the UK Population, including a Mathematical Approach to Evaluate the Impact of Vitamin D Fortified Ready-to-Eat Breakfast Cereals: Application of the NDNS Database

Vitamin D status is relatively poor in the general population, potentially leading to various conditions. The present study evaluates the relationship between vitamin D status and intake in the UK population and the impact of vitamin D fortified ready-to-eat cereals (RTEC) on this status via data from the National Diet and Nutrition Survey (NDNS: 2008–2012). Four cohorts were addressed: ages 4–10 (n = 803), ages 11–18 (n = 884), ages 19–64 (n = 1655) and ages 65 and higher (n = 428). The impact of fortification by 4.2 μg vitamin D per 100 g of RTEC on vitamin D intake and status was mathematically modelled. Average vitamin D daily intake was age-dependent, ranging from ~2.6 (age range 4–18 years) to ~5.0 μg (older than 64 years). Average 25(OH)D concentration ranged from 43 to 51 nmol/L, the highest in children. The relationship between vitamin D intake and status followed an asymptotic curve with a predicted plateau concentration ranging from 52 in children to 83 nmol/L in elderly. The fortification model showed that serum concentrations increased with ~1.0 in children to ~6.5 nmol/L in the elderly. This study revealed that vitamin D intake in the UK population is low with 25(OH)D concentrations being suboptimal for general health. Fortification of breakfast cereals can contribute to improve overall vitamin D status.

Emergency department safety assessment and follow-up evaluation 2: An implementation trial to improve suicide prevention

BACKGROUND

Emergency departments (EDs) are important for preventing suicide. Historically, many patients with suicide risk are not detected during routine clinical care, and those who are often do not receive suicide-specific intervention. The original Emergency Department Safety Assessment and Follow-up Evaluation (ED-SAFE 1) study examined the implementation of universal suicide risk screening and a multi-component ED-initiated suicide prevention intervention.

PURPOSE

The ED-SAFE 2 aims to study the impact of using a continuous quality improvement approach (CQI) to improve suicide related care, with a focus on improving universal suicide risk screening in adult ED patients and evaluating implementation of a new brief intervention called the Safety Planning Intervention (SPI) into routine clinical practice. CQI is a quality management process that uses data and collaboration to drive incremental, iterative improvements. The SPI is a personalized approach that focuses on early identification of warning signs and execution of systematic steps to manage suicidal thoughts. ED-SAFE 2 will provide data on the effectiveness of CQI procedures in improving suicide-related care processes, as well as the impact of these improvements on reducing suicide-related outcomes.

METHODS

Using a stepped wedge design, eight EDs collected data cross three study phases: Baseline (retrospective), Implementation (12 months), and Maintenance (12 months). Lean methods, a specific approach to pursuing CQI which focuses on increasing value and eliminating waste, were used to evaluate and improve suicide-related care.

CONCLUSIONS

The results will build upon the success of the ED-SAFE 1 and will have a broad public health impact through promoting better suicide-related care processes and improved suicide prevention.

Pairwise likelihood estimation of latent autoregressive count models

Latent autoregressive models are useful time series models for the analysis of infectious disease data. Evaluation of the likelihood function of latent autoregressive models is intractable and its approximation through simulation-based methods appears as a standard practice. Although simulation methods may make the inferential problem feasible, they are often computationally intensive and the quality of the numerical approximation may be difficult to assess. We consider instead a weighted pairwise likelihood approach and explore several computational and methodological aspects including estimation of robust standard errors and the role of numerical integration. The suggested approach is illustrated using monthly data on invasive meningococcal disease infection in Greece and Italy.

Applying time series modeling to assess the dynamics and forecast monthly reports of abuse, neglect and/or exploitation involving a vulnerable adult

Background

Application of time series modeling to predict reports related to maltreatment of vulnerable adults can be helpful for efficient early planning and resource allocation to handle a high volume of investigations. The goal of this study is to apply: (1) autoregressive integrated moving average (ARIMA) time series modeling to fit and forecast monthly maltreatment reports accepted for assessment reported to adult protective services (APS), and (2) interrupted time series analysis to test whether the implementation of intake hubs have a significant impact in the number of maltreatment reports after the implementation period.

Methods

A time series analysis on monthly APS intake reports was conducted using administrative data from SC Child and Adult Protective Services between January 2014 and June 2018. Monthly APS data were subjected to ARIMA modeling adjusting for the time period when intake hubs were implemented. The coefficient of determination, normalized SBC, AIC, MSE, and Ljung-Box Q-test were used to evaluate the goodness-of-fit of constructed models. The most parsimonious model was selected to predict the monthly APS intakes from July to December 2018. Poisson regression was fit to examine the association of the implementation of the hubs and the number of intake reports received to APS, adjusting for confounders.

Results

Over 24,000 APS intakes accepted for investigation were identified over a period of four calendar years with an increase in the monthly average of APS intakes between 2014 and 2017. An increase in the number of monthly APS intakes was found after the intake hubs were implemented in 2015 (Phase-1) and 2017 (Phase-2). Of all the models tested, an ARIMA (12), 1, 1 model was found to work best after evaluating all fit measures for both models. For Phase-1, the optimum model predicted an average of 488 APS intake reports between July and December 2018, representing a 9% increase from January–June 2018 (median = 445). For Phase-2, the percent increase was 32%.

Conclusions

The implementation of the intake hubs has a significant impact in the number of reports received after the implementation period. ARIMA time series is a valuable tool to predict future reports of maltreatment of vulnerable adults, which could be used to allow appropriate planning and resource allocation to handle a high volume of monthly intake reports.

Age-Specific Associations of Ozone and Fine Particulate Matter with Respiratory Emergency Department Visits in the United States

RATIONALE

Whereas associations between air pollution and respiratory morbidity for adults 65 years and older are well documented in the United States, the evidence for people under 65 is less extensive. To address this gap, the Centers for Disease Control and Prevention's National Environmental Public Health Tracking Program collected respiratory emergency department (ED) data from 17 states.

OBJECTIVES

To estimate age-specific acute effects of ozone and fine particulate matter (particulate matter ≤2.5 mm in aerodynamic diameter [PM_2.5]) on respiratory ED visits.

METHODS

We conducted time-series analyses in 894 counties by linking daily respiratory ED visits with estimated ozone and PM_2.5 concentrations during the week before the date of the visit. Overall effect estimates were obtained with a Bayesian hierarchical model to combine county estimates for each pollutant by age group (children, 0-18; adults, 19-64; adults ≥ 65, and all ages) and by outcome group (acute respiratory infection, asthma, chronic obstructive pulmonary disease, pneumonia, and all respiratory ED visits).

MEASUREMENTS AND MAIN RESULTS

Rate ratios (95% credible interval) per 10-μg/m³ increase in PM_2.5 and all respiratory ED visits were 1.024 (1.018-1.029) among children, 1.008 (1.004-1.012) among adults younger than 65 years, and 1.002 (0.996-1.007) among adults 65 and older. Per 20-ppb increase in ozone, rate ratios were 1.017 (1.011-1.023) among children, 1.051 (1.046-1.056) among adults younger than 65, and 1.033 (1.026-1.040) among adults 65 and older. Associations varied in magnitude by age group for each outcome group.

CONCLUSIONS

These results address a gap in the evidence used to ensure adequate public health protection under national air pollution policies.

Understanding and using time series analyses in addiction research

Time series analyses are statistical methods used to assess trends in repeated measurements taken at regular intervals and their associations with other trends or events, taking account of the temporal structure of such data. Addiction research often involves assessing associations between trends in target variables (e.g. population cigarette smoking prevalence) and predictor variables (e.g. average price of a cigarette), known as a multiple time series design, or interventions or events (e.g. introduction of an indoor smoking ban), known as an interrupted time series design. There are many analytical tools available, each with its own strengths and limitations. This paper provides addiction researchers with an overview of many of the methods available (GLM, GLMM, GLS, GAMM, ARIMA, ARIMAX, VAR, SVAR, VECM) and guidance on when and how they should be used, sample size det ermination, reporting and interpretation. The aim is to provide increased clarity for researchers proposing to undertake these analyses concerning what is likely to be acceptable for publication in journals such as Addiction. Given the large number of choices that need to be made when setting up time series models, the guidance emphasizes the importance of pre-registering hypotheses and analysis plans before the analyses are undertaken.

Period of Measurement in Time-Series Predictions of Disease Counts from 2007 to 2017 in Northern Nevada: Analytics Experiment

BACKGROUND

The literature in statistics presents methods by which autocorrelation can identify the best period of measurement to improve the performance of a time-series prediction. The period of measurement plays an important role in improving the performance of disease-count predictions. However, from the operational perspective in public health surveillance, there is a limitation to the length of the measurement period that can offer meaningful and valuable predictions.

OBJECTIVE

This study aimed to establish a method that identifies the shortest period of measurement without significantly decreasing the prediction performance for time-series analysis of disease counts.

METHODS

The data used in this evaluation include disease counts from 2007 to 2017 in northern Nevada. The disease counts for chlamydia, salmonella, respiratory syncytial virus, gonorrhea, viral meningitis, and influenza A were predicted.

RESULTS

Our results showed that autocorrelation could not guarantee the best performance for prediction of disease counts. However, the proposed method with the change-point analysis suggests a period of measurement that is operationally acceptable and performance that is not significantly different from the best prediction.

CONCLUSIONS

The use of change-point analysis with autocorrelation provides the best and most practical period of measurement.

The association between seasonal influenza-like illness cases and foetal death: a time series analysis

It has been reported that foetal death follows a seasonal pattern. Influenza virus infection has been postulated as one possible contributor to this seasonal variation. This ecological study explored the temporal association between the influenza activity and the frequency of foetal death. Time series analysis was conducted using weekly influenza-like illness consultation proportions from the Danish sentinel surveillance system and weekly proportions of spontaneous abortions and stillbirths from hospital registers from 1994 to 2009. The association was examined in an autoregressive (AR) integrated (I) moving average (MA) model and subsequently analysed with cross-correlation functions. Our findings confirmed the well-known seasonality in influenza, but also seasonality in spontaneous abortion. No clear pattern of seasonality was found for stillbirths, although the analysis exposed dependency between observations. One final AR integrated MA model was identified for the influenza-like illness (ILI) series. We found no statistically significant relationship between weekly influenza-like illness consultation proportions and weekly spontaneous abortion proportions (five lags: P = 0.52; 11 lags: P = 0.91) or weekly stillbirths (five lags: P = 0.93; 11 lags: P = 0.40). Exposure to circulating influenza during pregnancy was not associated with rates of spontaneous abortions or stillbirths. Seasonal variations in spontaneous abortion were confirmed and this phenomenon needs further investigation.

Natality Decline and Spatial Variation in Excess Death Rates During the 1918-1920 Influenza Pandemic in Arizona, United States

A large body of epidemiologic research has concentrated on the 1918 influenza pandemic, but more work is needed to understand spatial variation in pandemic mortality and its effects on natality. We collected and analyzed 35,151 death records from Arizona for 1915–1921 and 21,334 birth records from Maricopa county for 1915–1925. We estimated the number of excess deaths and births before, during, and after the pandemic period, and we found a significant decline in the number of births occurring 9–11 months after peak pandemic mortality. Moreover, excess mortality rates were highest in northern Arizona counties, where Native Americans were historically concentrated, suggesting a link between ethnic and/or sociodemographic factors and risk of pandemic-related death. The relationship between birth patterns and pandemic mortality risk should be further studied at different spatial scales and in different ethnic groups.

Short-term association between outdoor air pollution and osteoporotic hip fracture

This study examines the association of the levels of different airborne pollutants on the incidence of osteoporotic hip fracture in a southern European region. Association was detected between SO₂ and NO₂ and hospital admissions due to hip fracture.

INTRODUCTION

To examine the short-term effects of outdoor air pollution on the incidence of osteoporotic hip fracture in a southern European region.

METHODS

This is an ecological retrospective cohort study based on data obtained from three databases. In a time-series analysis, we examined the association between hip fracture incidence and different outdoor air pollutants (sulfur dioxide (SO₂), monoxide (NO), nitrogen dioxide (NO₂), ozone (O₃), and particulate matter in suspension < 2.5 (PM_2.5) and < 10-μm (PM₁₀) conditions by using general additive models (Poisson distribution). The incidence rate ratio (IRR), crude and adjusted by season and different weather conditions, was estimated for all parameters. Hip incidence was later analyzed by sex and age (under or over age 75) subgroups. The main outcome measure was daily hospital admissions due to fracture.

RESULTS

Hip fracture incidence showed association with SO₂ (IRR 1.11 (95% CI 1.04-1.18)), NO (IRR 1.01 (95% CI 1.01-1.02)), and NO₂ (IRR 1.02 (95% CI 1.01-1.04)). For O₃ levels, this association was negative (IRR 0.97 (95% CI 0.95-0.99)). The association persisted for SO₂ and NO₂ when the models were adjusted by season. After adjusting by season and weather conditions, the association persisted for NO₂. When participants were stratified by age and sex, associations persisted only in women older than 75 years.

CONCLUSIONS

A short-term association was observed with several indicators of air pollution on hip fracture incidence. This is the first study that shows these associations.

Association between Oklahoma earthquakes and anxiety-related Google search episodes

BACKGROUND

Oklahoma has experienced a rise in seismicity since 2010, with many earthquakes induced by wastewater injection. While large single earthquakes have documented mental health repercussions, health implications of these new, frequent earthquakes remain unknown. We aimed to examine associations between Oklahoma earthquakes and statewide anxiety measured by Google queries.

METHODS

The U.S. Geologic Survey's Advanced National Seismic System Comprehensive Catalog supplied earthquake dates and magnitudes. We used the Google Health application programming interface to compile the proportion of weekly Oklahoma-based health-related search episodes for anxiety. A quasi-experimental time-series analysis from January 2010 to May 2017 evaluated monthly counts of earthquakes ≥ magnitude 4 (a level felt by most people) in relation to anxiety, controlling for US-wide anxiety search episodes and Oklahoma-specific health-related queries.

RESULTS

Oklahoma experienced an average of two (SD = 2) earthquakes ≥ magnitude 4 per month during the study period. For each additional earthquake ≥ magnitude 4, the proportion of Google search episodes for anxiety increased by 1.3% (95% confidence interval = 0.1%, 2.4%); 60% of this increase persisted for the following month. In months with 2 or more ≥ magnitude 4 earthquakes, the proportion of Google search episodes focused on anxiety increased by 5.8% (95% confidence interval = 2.3%, 9.3%). In a sub-analysis, Google search episodes for anxiety peaked about 3 weeks after ≥ magnitude 4 quakes.

CONCLUSIONS

These findings suggest that the recent increase in Oklahoma earthquakes has elicited a psychological response that may have implications for public health and regulatory policy.

The Effect of Seasonal Floods on Health: Analysis of Six Years of National Health Data and Flood Maps

There is limited knowledge on the effect of seasonal flooding on health over time. We quantified the short- and long-term effects of floods on selected health indicators at public healthcare facilities in 11 districts in Cambodia, a flood-prone setting. Counts of inpatient discharge diagnoses and outpatient consultations for diarrhea, acute respiratory infections, skin infections, injuries, noncommunicable diseases and vector-borne diseases were retrieved from public healthcare facilities for each month between January 2008 and December 2013. Flood water was mapped by month, in square kilometers, from satellite data. Poisson regression models with three lag months were constructed for the health problems in each district, controlled for seasonality and long-term trends. During times of flooding and three months after, there were small to moderate increases in visits to healthcare facilities for skin infections, acute respiratory infections, and diarrhea, while no association was seen at one to two months. The associations were small to moderate, and a few of our results were significant. We observed increases in care seeking for diarrhea, skin infections, and acute respiratory infections following floods, but the associations are uncertain. Additional research on previous exposure to flooding, using community- and facility-based data, would help identify expected health risks after floods in flood-prone settings.

Long-Acting Reversible Contraception Free of Charge, Method Initiation, and Abortion Rates in Finland

OBJECTIVES

To evaluate whether a public program providing long-acting reversible contraceptive (LARC) methods free of charge increases the LARC initiation rate and reduces the unintended pregnancy rate in the general population.

METHODS

Since 2013, all women in Vantaa, Finland, have been entitled to 1 LARC method free of charge. With time-series analysis between 2000 and 2015, we assessed whether this public program was associated with changes in steady-state mean rates of LARC initiation and abortions.

RESULTS

The initiation rate of LARCs (1/1000 women) increased 2.2-fold from 1.9 to 4.2 after the intervention (P < .001). Concomitantly, the abortion rate (1/1000 women) declined by 16% from 1.1 to 0.9 in the total sample (P < .001), by 36% from 1.3 to 0.8 among those aged 15 to 19 years (P < .001), and by 14% from 2.0 to 1.7 among those aged 20 to 24 years (P = .01).

CONCLUSIONS

The LARC program was associated with increased uptake of LARC methods and fewer abortions in the population. Public Health Implications. Entitling the population to LARC methods free of charge is an effective means to reduce the unmet need of contraception and the need for abortion, especially among women younger than 25 years.

Methods for computational disease surveillance in infection prevention and control: Statistical process control versus Twitter's anomaly and breakout detection algorithms

BACKGROUND

Although not all health care-associated infections (HAIs) are preventable, reducing HAIs through targeted intervention is key to a successful infection prevention program. To identify areas in need of targeted intervention, robust statistical methods must be used when analyzing surveillance data. The objective of this study was to compare and contrast statistical process control (SPC) charts with Twitter's anomaly and breakout detection algorithms.

METHODS

SPC and anomaly/breakout detection (ABD) charts were created for vancomycin-resistant Enterococcus, Acinetobacter baumannii, catheter-associated urinary tract infection, and central line-associated bloodstream infection data.

RESULTS

Both SPC and ABD charts detected similar data points as anomalous/out of control on most charts. The vancomycin-resistant Enterococcus ABD chart detected an extra anomalous point that appeared to be higher than the same time period in prior years. Using a small subset of the central line-associated bloodstream infection data, the ABD chart was able to detect anomalies where the SPC chart was not.

DISCUSSION

SPC charts and ABD charts both performed well, although ABD charts appeared to work better in the context of seasonal variation and autocorrelation.

CONCLUSIONS

Because they account for common statistical issues in HAI data, ABD charts may be useful for practitioners for analysis of HAI surveillance data.