Predicting the incidence of infectious diarrhea with symptom surveillance data using a stacking-based ensembled model.

BACKGROUND: Infectious diarrhea remains a major public health problem worldwide. This study used stacking ensemble to developed a predictive model for the incidence of infectious diarrhea, aiming to achieve better prediction performance. METHODS: Based on the surveillance data of infectious diarrhea cases, relevant symptoms and meteorological factors of Guangzhou from 2016 to 2021, we developed four base prediction models using artificial neural networks (ANN), Long Short-Term Memory networks (LSTM), support vector regression (SVR) and extreme gradient boosting regression trees (XGBoost), which were then ensembled using stacking to obtain the final prediction model. All the models were evaluated with three metrics: mean absolute percentage error (MAPE), root mean square error (RMSE), and mean absolute error (MAE). RESULTS: Base models that incorporated symptom surveillance data and weekly number of infectious diarrhea cases were able to achieve lower RMSEs, MAEs, and MAPEs than models that added meteorological data and weekly number of infectious diarrhea cases. The LSTM had the best prediction performance among the four base models, and its RMSE, MAE, and MAPE were: 84.85, 57.50 and 15.92%, respectively. The stacking ensembled model outperformed the four base models, whose RMSE, MAE, and MAPE were 75.82, 55.93, and 15.70%, respectively. CONCLUSIONS: The incorporation of symptom surveillance data could improve the predictive accuracy of infectious diarrhea prediction models, and symptom surveillance data was more effective than meteorological data in enhancing model performance. Using stacking to combine multiple prediction models were able to alleviate the difficulty in selecting the optimal model, and could obtain a model with better performance than base models.

Implementation of an Automated Text Message-Based System for Tracking Patient-Reported Outcomes in Spine Surgery: An Overview of the Concept and Our Early Experience.

OBJECTIVE: Text message-based interventions have been demonstrated to be a valuable monitoring tool across various conditions. Here, we aimed to describe our early experience using a newly developed text message-based platform designed to track symptoms in spine surgery patients. METHODS: We used the Informed Mindset Medical (IMM) platform to automatically send text messages with secure and encrypted hyperlinks to enrolled patients. Patient symptoms were monitored using well-standardized functional assessments. Limited patient data and responses were stored on a Health Insurance Portability and Accountability Act-compliant SQL cloud-based server database. RESULTS: In 3 months, 101 patients scheduled for elective spine surgery accepted participation in our pilot study. Overall, 71.2% of the enrolled patients responded to at least 1 preoperative baseline questionnaire. The response rates were similar across attendings, questionnaire bundles (cervical vs. thoracolumbar), genders, and age groups. The overall preoperative IMM pain scores were found to correlate positively with the preoperative electronic medical record pain rates. Similarly, the overall preoperative IMM and electronic medical record pain scores correlated positively with the IMM-collected Neck Disability Index/Oswestry Disability Index scores. From an initial 71.2%, the response rate decreased to 54.9% for the 6-week follow-up questionnaires. CONCLUSIONS: Our preliminary findings support the reliability of this text message-based strategy to monitor symptoms in spine surgery patients. Further studies are warranted to explore strategies to increase the response rate and expand this platform's clinical and research applicability.

Miscommunication in the age of communication: A crowdsourcing framework for symptom surveillance at the time of pandemics.

OBJECTIVE: There was a significant delay in compiling a complete list of the symptoms of COVID-19 during the 2020 outbreak of the disease. When there is little information about the symptoms of a novel disease, interventions to contain the spread of the disease would be suboptimal because people experiencing symptoms that are not yet known to be related to the disease may not limit their social activities. Our goal was to understand whether users' social media postings about the symptoms of novel diseases could be used to develop a complete list of the disease symptoms in a shorter time. MATERIALS AND METHODS: We used the Twitter API to download tweets that contained 'coronavirus', 'COVID-19', and 'symptom'. After data cleaning, the resulting dataset consisted of over 95,000 unique, English tweets posted between January 17, 2020 and March 15, 2020 that contained references to the symptoms of COVID-19. We analyzed this data using network and time series methods. RESULTS: We found that a complete list of the symptoms of COVID-19 could have been compiled by mid-March 2020, before most states in the U.S. announced a lockdown and about 75 days earlier than the list was completed on CDC's website. DISCUSSION & CONCLUSION: We conclude that national and international health agencies should use the crowd-sourced intelligence obtained from social media to develop effective symptom surveillance systems in the early stages of pandemics. We propose a high-level framework that facilitates the collection, analysis, and dissemination of information that are posted in various languages and on different social media platforms about the symptoms of novel diseases.

[Seasonality of clustering of fever and diarrhea in Beijing, 2009-2015].

Objective: To understand the seasonal distribution of the clustering of fever and diarrhea. Methods: Concentration degree and circular distribution methods were used to analyze the seasonal distribution of the clustering of fever and diarrhea in Beijing from 2009 to 2015. The information were collected from the Infectious Disease Surveillance Information System of Beijing. Results: The M values of the clustering of fever and diarrhea were 0.57 and 0.47. Circular distribution results showed that the clustering of fever and diarrhea angle dispersion index R values were 0.57 and 0.46 respectively, the sample average angle of Rayleigh' s test Z values were 414.14, 148.09 respectively (all P<0.01). The clustering of fever and diarrhea had seasonality. The incidence peak of fever was on October 13, and the epidemic period was during August 13-December 14. The incidence peak of diarrhea was on July 31, and the epidemic period was during May 20-October 11. Conclusion: The clustering of fever had obvious seasonality in Beijing, which mainly occurred in autumn and winter. The cluster of diarrhea had certain seasonality, which mainly occurred in summer and autumn.

Feasibility and acceptability of electronic symptom surveillance with clinician feedback using the Patient-Reported Outcomes version of Common Terminology Criteria for Adverse Events (PRO-CTCAE) in Danish prostate cancer patients.

BACKGROUND: The aim was to examine the feasibility, acceptability and clinical utility of electronic symptom surveillance with clinician feedback using a subset of items drawn from the Patient-Reported Outcomes version of Common Terminology Criteria for Adverse Events (PRO-CTCAE) in a cancer treatment setting. METHODS: Danish-speaking men with castration-resistant metastatic prostate cancer receiving treatment at the Department of Oncology, Rigshospitalet, Copenhagen between March 9, 2015 and June 8, 2015 were invited to participate (n = 63 eligible). Participants completed the PRO-CTCAE questionnaire on tablet computers using AmbuFlex software at each treatment visit in the outpatient clinic. In total, 22 symptomatic toxicities (41 PRO-CTCAE items), corresponding to the symptomatic adverse-events profile associated with the regimens commonly used for prostate cancer treatment (Docetaxel, Cabazitaxel, Abiraterone, Alpharadin), were selected. Participants' PRO-CTCAE responses were presented graphically to their treating oncologists via an AmbuFlex dashboard, for real-time use to enhance the patient-clinician dialogue that occurs during the consultation prior to each treatment cycle. Technical and clinical barriers and acceptability were evaluated through semi-structured interviews with both patients and oncologists. Patients receiving active treatment at the end of the study period completed an evaluation questionnaire. RESULTS: Fifty-four out of sixty-three (86%) eligible patients were enrolled. The PRO-CTCAE questionnaire was completed a total of 168 times by 54 participants (median number per patient was 3, range 1-5). Eight surveys were missed, resulting in a compliance rate of 97%. At the end of the study period, 35 patients (65%) were still receiving active treatment and completed the evaluation questionnaire. Patients reported that their PRO-CTCAE responses served as a communication tool. Oncologists stated that the availability of the PRO-CTCAE self-reports during the consultation improved patient-clinician communication about side effects. CONCLUSION: Electronic capture of symptomatic toxicities using PRO-CTCAE and the submission of self-reports to clinicians prior to consultation were feasible among metastatic prostate cancer patients receiving chemotherapy in an outpatient setting, and this procedure was acceptable to both patients and clinicians. Continued research, including a cluster-randomized trial, will evaluate the effects of submitting patients' PRO-CTCAE results to clinicians prior to consultation on the quality of side-effects management and resultant clinical outcomes.

Seasonality of clustering of fever and diarrhea in Beijing, 2009-2015 / 中华流行病学杂志

Objective To understand the seasonal distribution of the clustering of fever and diarrhea.Methods Concentration degree and circular distribution methods were used to analyze the seasonal distribution of the clustering of fever and diarrhea in Beijing from 2009 to 2015.The information were collected from the Infectious Disease Surveillance Information System of Beijing.Results The M values of the clustering of fever and diarrhea were 0.57 and 0.47.Circular distribution results showed that the clustering of fever and diarrhea angle dispersion index R values were 0.57 and 0.46 respectively,the sample average angle of Rayleigh's test Z values were 414.14,148.09 respectively (all P＜0.01).The clustering of fever and diarrhea had seasonality.The incidence peak of fever was on October 13,and the epidemic period was during August 13-December 14.The incidence peak of diarrhea was on July 31,and the epidemic period was during May 20-October 11.Conclusion The clustering of fever had obvious seasonality in Beijing,which mainly occurred in auttmn and winter.The cluster of diarrhea had certain seasonality,which mainly occurred in summer and autumn.

Seasonality of clustering of fever and diarrhea in Beijing, 2009-2015 / 中华流行病学杂志

Objective To understand the seasonal distribution of the clustering of fever and diarrhea.Methods Concentration degree and circular distribution methods were used to analyze the seasonal distribution of the clustering of fever and diarrhea in Beijing from 2009 to 2015.The information were collected from the Infectious Disease Surveillance Information System of Beijing.Results The M values of the clustering of fever and diarrhea were 0.57 and 0.47.Circular distribution results showed that the clustering of fever and diarrhea angle dispersion index R values were 0.57 and 0.46 respectively,the sample average angle of Rayleigh's test Z values were 414.14,148.09 respectively (all P＜0.01).The clustering of fever and diarrhea had seasonality.The incidence peak of fever was on October 13,and the epidemic period was during August 13-December 14.The incidence peak of diarrhea was on July 31,and the epidemic period was during May 20-October 11.Conclusion The clustering of fever had obvious seasonality in Beijing,which mainly occurred in auttmn and winter.The cluster of diarrhea had certain seasonality,which mainly occurred in summer and autumn.

Etiological epidemiology of viral diarrhea on the basis of sentinel surveillance in children younger than 5 years in Gansu, northwest China, 2009-2013.

To explore the etiological spectrum of diarrhea and its epidemiological characteristics in diarrhea symptoms surveillance cases younger than 5 years from 2009 to 2013 in Gansu province, northwest China. Systematic diarrhea symptoms surveillance were conducted in 27 sentinel sites in Gansu province and outpatients with three or more loose, watery, or sticky pus stools per day were defined as surveillance cases. All stool specimens were tested for Rotavirus, Human calicivirus, Adenovirus, and Astrovirus. Totally, 1,119 cases (51.54%) were identified as any enteric virus. The average isolation rate of Rotavirus was 51.13%, Astrovirus was 10.84%, Adenovirus was 6.94%, and Human calicivirus was 6.60% (P < 0.01). Rotavirus was identified with the highest frequency among these enteric pathogens except in 2011, with a notable downward trend over time (P < 0.01). Rotavirus A was the most proportion in rotavirus, G3P[8] and G9P[8] were the most common combination. Rotavirus mixed Human calicivirus infections was the most common mixed infected patterns. Viral-positive rate was higher among children aged group of 0-12 and 13-24 months (P < 0.01, respectively). The isolation rates of four enteric viral pathogens showed a similar distinct seasonal variation with a higher rate in spring, autumn, and winter months. Rotavirus was the major epidemiological viral pathogen in diarrhea symptom surveillance cases in Gansu province, northwest China, during period 2009-2013. Seasonal and age-related variations were observed in enteric viral pathogen isolation rate. The comprehensive and continuous surveillance is needed to identify the prevalence of different enteric viral pathogens.

Epidemiological features of pertussis resurgence based on community populations with high vaccination coverage in China.

Active symptom surveillance was applied to three selected communities ( 160,147 persons) in Tianjin from 2010 to 2012. We examined 1089 individuals showing pertussis-like symptoms, of which 1022 nasopharyngeal specimens were tested for pertussis by polymerase chain reaction and 802 sera for anti-pertussis toxin antibodies. Of the total cases tested, 113 were confirmed, and their demographic, clinical, and vaccination-related data were collected. The annual incidence was 23.52 cases/100,000 persons among communities, which was 16.22 times that obtained via hospital reports for the same period (P < 0.001). The actual incidence in the 15-69 years age group was most significantly underestimated by hospitals, given that it was 43.08 times that of the reported hospital rate. Among the cases aged <15 years, 84.5% were individuals who had been fully vaccinated. The misdiagnosis rate was as high as 94.69%, and only 5.31% of the confirmed pertussis cases were properly diagnosed as pertussis at their first medical visit. Pertussis incidence in China has been severely underestimated and this was in part due to a high misdiagnosis rate. Adolescents and adults have become new high-risk populations. Future work should focus on reinforcing immunization programmes, especially among adolescents and adults.

Application of Fever Symptom Surveillance in Hospital Infection Prospective Monitoring / 中华医院感染学杂志

OBJECTIVE To explore the effect of fever symptom surveillance in hospital infection prospective monitoring.METHODS The material were divided into tow groups.A group was adopted retrospective method and B group was used prospective study method.A practical electronic function modules was applied in B group.The patients information of temperature ≥37.5 ℃ was collected to the data-base.Hospital information systems were utilized to estimate hospital infection.RESULTS There were significant difference(P