Pub Date : 2022-10-10eCollection Date: 2022-01-01DOI: 10.1515/em-2021-0033
Caroline L Gaglio, Mohammed F Islam, Joseph Cotler, Leonard A Jason
Objectives: The Institute of Medicine (IOM 2015. Beyond Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: Redefining an Illness. Washington: The National Academies Press) suggested new criteria for Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS), which requires an endorsement of either neurocognitive impairment or orthostatic intolerance (OI) in addition to other core symptoms. While some research supports the inclusion of OI as a core symptom, others argue that overlap with neurocognitive impairment does not justify the either/or option. The current study assessed methods of operationalizing OI using items from the DePaul Symptom Questionnaire (DSQ-1 and -2) as a part of the IOM criteria. Evaluating the relationship between OI and neurocognitive symptoms may lead to a better understanding of diagnostic criteria for ME/CFS.
Methods: Two-hundred and forty-two participants completed the DSQ. We examined how many participants met the IOM criteria while endorsing different frequencies and severities of various OI symptoms.
Results: Neurocognitive impairment was reported by 93.4% of respondents. OI without concurrent neurocognitive symptoms only allowed for an additional 1.7-4.5% of participants to meet IOM criteria.
Conclusions: Neurocognitive symptoms and OI overlap in ME/CFS, and our results do not support the IOM's inclusion of neurocognitive impairment and OI as interchangeable symptoms. Furthermore, our findings highlight the need for a uniform method of defining and measuring OI via self-report in order to accurately study OI as a symptom of ME/CFS.
{"title":"Orthostatic intolerance and neurocognitive impairment in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS).","authors":"Caroline L Gaglio, Mohammed F Islam, Joseph Cotler, Leonard A Jason","doi":"10.1515/em-2021-0033","DOIUrl":"https://doi.org/10.1515/em-2021-0033","url":null,"abstract":"<p><strong>Objectives: </strong>The Institute of Medicine (IOM 2015. Beyond Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: Redefining an Illness. Washington: The National Academies Press) suggested new criteria for Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS), which requires an endorsement of either neurocognitive impairment or orthostatic intolerance (OI) in addition to other core symptoms. While some research supports the inclusion of OI as a core symptom, others argue that overlap with neurocognitive impairment does not justify the either/or option. The current study assessed methods of operationalizing OI using items from the DePaul Symptom Questionnaire (DSQ-1 and -2) as a part of the IOM criteria. Evaluating the relationship between OI and neurocognitive symptoms may lead to a better understanding of diagnostic criteria for ME/CFS.</p><p><strong>Methods: </strong>Two-hundred and forty-two participants completed the DSQ. We examined how many participants met the IOM criteria while endorsing different frequencies and severities of various OI symptoms.</p><p><strong>Results: </strong>Neurocognitive impairment was reported by 93.4% of respondents. OI without concurrent neurocognitive symptoms only allowed for an additional 1.7-4.5% of participants to meet IOM criteria.</p><p><strong>Conclusions: </strong>Neurocognitive symptoms and OI overlap in ME/CFS, and our results do not support the IOM's inclusion of neurocognitive impairment and OI as interchangeable symptoms. Furthermore, our findings highlight the need for a uniform method of defining and measuring OI via self-report in order to accurately study OI as a symptom of ME/CFS.</p>","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"11 1","pages":"20210033"},"PeriodicalIF":0.0,"publicationDate":"2022-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9550273/pdf/em-11-1-em-2021-0033.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40655332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Objectives To introduce a novel way of measuring the spreading speed of an epidemic. Methods We propose to use the mean time between infections (MTBI) metric obtained from a recently introduced nonhomogeneous Markov stochastic model. Different types of parameter calibration are performed. We estimate the MTBI using data from different time windows and from the whole stage history and compare the results. In order to detect waves and stages in the input data, a preprocessing filtering technique is applied. Results The results of applying this indicator to the COVID-19 reported data of infections from Argentina, Germany and the United States are shown. We find that the MTBI behaves similarly with respect to the different data inputs, whereas the model parameters completely change their behaviour. Evolution over time of the parameters and the MTBI indicator is also shown. Conclusions We show evidence to support the claim that the MTBI is a rather good indicator in order to measure the spreading speed of an epidemic, having similar values whatever the input data size.
{"title":"Measuring COVID-19 spreading speed through the mean time between infections indicator","authors":"G. Pena, Ver'onica Moreno, N. R. Barraza","doi":"10.1515/em-2022-0106","DOIUrl":"https://doi.org/10.1515/em-2022-0106","url":null,"abstract":"Abstract Objectives To introduce a novel way of measuring the spreading speed of an epidemic. Methods We propose to use the mean time between infections (MTBI) metric obtained from a recently introduced nonhomogeneous Markov stochastic model. Different types of parameter calibration are performed. We estimate the MTBI using data from different time windows and from the whole stage history and compare the results. In order to detect waves and stages in the input data, a preprocessing filtering technique is applied. Results The results of applying this indicator to the COVID-19 reported data of infections from Argentina, Germany and the United States are shown. We find that the MTBI behaves similarly with respect to the different data inputs, whereas the model parameters completely change their behaviour. Evolution over time of the parameters and the MTBI indicator is also shown. Conclusions We show evidence to support the claim that the MTBI is a rather good indicator in order to measure the spreading speed of an epidemic, having similar values whatever the input data size.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83728773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Objectives To forecast the true growth of COVID-19 cases in Singapore after accounting for asymptomatic infections, we study and make modifications to the SEIR (Susceptible-Exposed-Infected-Recovered) epidemiological model by incorporating hospitalization dynamics and the presence of asymptomatic cases. We then compare the simulation results of our three epidemiological models of interest against the daily reported COVID-19 case counts during the time period from 23rd January to 6th April 2020. Finally, we compare and evaluate on the performance and accuracy of the aforementioned models’ simulations. Methods Three epidemiological models are used to forecast the true growth of COVID-19 case counts by accounting for asymptomatic infections in Singapore. They are the exponential model, SEIR model with hospitalization dynamics (SEIHRD), and the SEIHRD model with inclusion of asymptomatic cases (SEAIHRD). Results Simulation results of all three models reflect underestimation of COVID-19 cases in Singapore during the early stages of the pandemic. At a 40% asymptomatic proportion, we report basic reproduction number R 0 = 3.28 and 3.74 under the SEIHRD and SEAIHRD models respectively. At a 60% asymptomatic proportion, we report R 0 = 3.48 and 3.96 under the SEIHRD and SEAIHRD models respectively. Conclusions Based on the results of different simulation scenarios, we are highly confident that the number of COVID-19 cases in Singapore was underestimated during the early stages of the pandemic. This is supported by the exponential increase of COVID-19 cases in Singapore as the pandemic evolved.
{"title":"Accounting for the role of asymptomatic patients in understanding the dynamics of the COVID-19 pandemic: a case study from Singapore","authors":"Fu Teck Liew, P. Ghosh, Bibhas Chakraborty","doi":"10.1515/em-2021-0031","DOIUrl":"https://doi.org/10.1515/em-2021-0031","url":null,"abstract":"Abstract Objectives To forecast the true growth of COVID-19 cases in Singapore after accounting for asymptomatic infections, we study and make modifications to the SEIR (Susceptible-Exposed-Infected-Recovered) epidemiological model by incorporating hospitalization dynamics and the presence of asymptomatic cases. We then compare the simulation results of our three epidemiological models of interest against the daily reported COVID-19 case counts during the time period from 23rd January to 6th April 2020. Finally, we compare and evaluate on the performance and accuracy of the aforementioned models’ simulations. Methods Three epidemiological models are used to forecast the true growth of COVID-19 case counts by accounting for asymptomatic infections in Singapore. They are the exponential model, SEIR model with hospitalization dynamics (SEIHRD), and the SEIHRD model with inclusion of asymptomatic cases (SEAIHRD). Results Simulation results of all three models reflect underestimation of COVID-19 cases in Singapore during the early stages of the pandemic. At a 40% asymptomatic proportion, we report basic reproduction number R 0 = 3.28 and 3.74 under the SEIHRD and SEAIHRD models respectively. At a 60% asymptomatic proportion, we report R 0 = 3.48 and 3.96 under the SEIHRD and SEAIHRD models respectively. Conclusions Based on the results of different simulation scenarios, we are highly confident that the number of COVID-19 cases in Singapore was underestimated during the early stages of the pandemic. This is supported by the exponential increase of COVID-19 cases in Singapore as the pandemic evolved.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74053156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sulan Lin, C. Rattanapan, A. Mongkolchati, M. N. Aung, W. Ounsaneha, N. Sritoomma, O. Laosee
Abstract Objectives To determine the point prevalence of undergraduate students who are hesitant to accept COVID-19 vaccination and to identify the predictors of COVID-19 vaccine hesitancy in university students. Methods A cross-sectional study was conducted during June–July 2021. A total of 542 undergraduate students from universities in three central provinces of Thailand participated in an online survey via Google Form. We used a transculturally translated, Thai version of the Oxford Coronavirus Explanations, Attitudes, and Narratives Survey (OCEANS II). Results There were 217 undergraduate students (40%) who were hesitant to receive the COVID-19 vaccine and the significant predictors for this hesitancy were: being students in Year 2 and higher (AOR: 2.73; 95% CI: 1.55–4.84); having negative beliefs toward the COVID-19 vaccine (AOR: 10.99; 95% CI: 6.82–17.73); and having a perceived positive general vaccine conspiracy belief (AOR: 1.90; 95% CI: 1.02–3.52). Conclusions It is important to minimize vaccine hesitancy among Thai undergraduate students with a negative perception of vaccines by clarifying false information.
{"title":"COVID-19 vaccine hesitancy among undergraduate students in Thailand during the peak of the third wave of the coronavirus pandemic in 2021","authors":"Sulan Lin, C. Rattanapan, A. Mongkolchati, M. N. Aung, W. Ounsaneha, N. Sritoomma, O. Laosee","doi":"10.1515/em-2022-0109","DOIUrl":"https://doi.org/10.1515/em-2022-0109","url":null,"abstract":"Abstract Objectives To determine the point prevalence of undergraduate students who are hesitant to accept COVID-19 vaccination and to identify the predictors of COVID-19 vaccine hesitancy in university students. Methods A cross-sectional study was conducted during June–July 2021. A total of 542 undergraduate students from universities in three central provinces of Thailand participated in an online survey via Google Form. We used a transculturally translated, Thai version of the Oxford Coronavirus Explanations, Attitudes, and Narratives Survey (OCEANS II). Results There were 217 undergraduate students (40%) who were hesitant to receive the COVID-19 vaccine and the significant predictors for this hesitancy were: being students in Year 2 and higher (AOR: 2.73; 95% CI: 1.55–4.84); having negative beliefs toward the COVID-19 vaccine (AOR: 10.99; 95% CI: 6.82–17.73); and having a perceived positive general vaccine conspiracy belief (AOR: 1.90; 95% CI: 1.02–3.52). Conclusions It is important to minimize vaccine hesitancy among Thai undergraduate students with a negative perception of vaccines by clarifying false information.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73388770","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mauricio Canals L, Andrea Canals C, Cristóbal Cuadrado N
Abstract Objectives The ability to predict COVID-19 dynamic has been very low, reflected in unexpected changes in the number of cases in different settings. Here the objective was to study the temporal memory of the reported daily incidence time series and propose a simple model for short-term forecast of the incidence. Methods We propose a new concept called incidence moments that allows exploring the memory of the reported incidence time series, based on successive products of the incidence and the reproductive number that allow a short term forecast of the future incidence. We studied the correlation between the predictions of and the reported incidence determining the best predictor. We compared the predictions and observed COVID-19 incidences with the mean arctangent absolute percentage error (MAAPE) analyses for the world, 43 countries and for Chile and its regions. Results The best predictor was the third moment of incidence, determining a short temporal prediction window of 15 days. After 15 days the absolute percentage error of the prediction increases significantly. The method perform better for larger populations and presents distortions in contexts of abrupt changes in incidence. Conclusions The epidemic dynamics of COVID 19 had a very short prediction window, probably associated with an intrinsic chaotic behavior of its dynamics. The incident moment modeling approach could be useful as a tool whose simplicity is appealing, since it allows rapid implementation in different settings, even with limited epidemiological technical capabilities and without requiring a large amount of computational data.
{"title":"Incidence moments: a simple method to study the memory and short term forecast of the COVID-19 incidence time-series","authors":"Mauricio Canals L, Andrea Canals C, Cristóbal Cuadrado N","doi":"10.1515/em-2021-0029","DOIUrl":"https://doi.org/10.1515/em-2021-0029","url":null,"abstract":"Abstract Objectives The ability to predict COVID-19 dynamic has been very low, reflected in unexpected changes in the number of cases in different settings. Here the objective was to study the temporal memory of the reported daily incidence time series and propose a simple model for short-term forecast of the incidence. Methods We propose a new concept called incidence moments that allows exploring the memory of the reported incidence time series, based on successive products of the incidence and the reproductive number that allow a short term forecast of the future incidence. We studied the correlation between the predictions of and the reported incidence determining the best predictor. We compared the predictions and observed COVID-19 incidences with the mean arctangent absolute percentage error (MAAPE) analyses for the world, 43 countries and for Chile and its regions. Results The best predictor was the third moment of incidence, determining a short temporal prediction window of 15 days. After 15 days the absolute percentage error of the prediction increases significantly. The method perform better for larger populations and presents distortions in contexts of abrupt changes in incidence. Conclusions The epidemic dynamics of COVID 19 had a very short prediction window, probably associated with an intrinsic chaotic behavior of its dynamics. The incident moment modeling approach could be useful as a tool whose simplicity is appealing, since it allows rapid implementation in different settings, even with limited epidemiological technical capabilities and without requiring a large amount of computational data.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"17 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75699923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Objectives The main objective of this research is to demonstrate the effectiveness of non-pharmaceutical interventions (social isolation and quarantine) and of vaccination. Methods The SIR epidemiological numerical model has been revised to obtain a new model (SAIRDQ), which involves additional variables: the population that died due to the disease (D), the isolated (A), quarantined population (Q) and the effect of vaccination. We have obtained the epidemiological parameters from the data, which are not constant during the evolution of the pandemic, using an iterative approximation method. Results Analysis of the data of infected and deceased suggest that the evolution of the coronavirus epidemic in Peru has arrived at the end of the second wave (around October 2021). We have simulated the effect of quarantine and vaccination, which are effective measures to reduce the impact of the pandemic. For a variable infection and isolation rate, due to the end of the quarantine, the death toll would be around 200 thousand; if the isolation and quarantine were relaxed since March 01, 2021, there could be more than 280 thousand deaths. Conclusions Without non-pharmaceutical interventions and vaccination, the number of deaths would be much higher than 280 thousand.
{"title":"Numerical modelling of coronavirus pandemic in Peru","authors":"C. Jiménez, M. Merma","doi":"10.1515/em-2020-0026","DOIUrl":"https://doi.org/10.1515/em-2020-0026","url":null,"abstract":"Abstract Objectives The main objective of this research is to demonstrate the effectiveness of non-pharmaceutical interventions (social isolation and quarantine) and of vaccination. Methods The SIR epidemiological numerical model has been revised to obtain a new model (SAIRDQ), which involves additional variables: the population that died due to the disease (D), the isolated (A), quarantined population (Q) and the effect of vaccination. We have obtained the epidemiological parameters from the data, which are not constant during the evolution of the pandemic, using an iterative approximation method. Results Analysis of the data of infected and deceased suggest that the evolution of the coronavirus epidemic in Peru has arrived at the end of the second wave (around October 2021). We have simulated the effect of quarantine and vaccination, which are effective measures to reduce the impact of the pandemic. For a variable infection and isolation rate, due to the end of the quarantine, the death toll would be around 200 thousand; if the isolation and quarantine were relaxed since March 01, 2021, there could be more than 280 thousand deaths. Conclusions Without non-pharmaceutical interventions and vaccination, the number of deaths would be much higher than 280 thousand.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"31 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80880935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1101/2022.06.10.22276234
M. Gaspari
Abstract Objectives Recent studies show that Test Positivity Rate (TPR) gains a better correlation than incidence with the number of hospitalized patients in COVID-19 pandemic. Nevertheless, epidemiologists remain sceptical concerning the widespread use of this metric for surveillance, and indicators based on known cases like incidence rate are still preferred despite the large number of asymptomatic carriers, which remain unknown. Our aim is to compare TPR and incidence rate, to determine which of the two has the best characteristics to predict the trend of hospitalized patients in the COVID-19 pandemic. Methods We perform a retrospective study considering 60 outbreak cases, using global and local data from Italy in different waves of the pandemic, in order to detect peaks in TPR time series, and peaks in incidence rate, finding which of the two indicators has the best ability to anticipate peaks in patients admitted in hospitals. Results On average, the best TPR-based approach anticipates the incidence rate of about 4.6 days (95 % CI 2.8, 6.4), more precisely the average distance between TPR peaks and hospitalized peaks is 17.6 days (95 % CI 15.0, 20.4) with respect to 13.0 days (95 % CI 10.4, 15.8) obtained for incidence. Moreover, the average difference between TPR and incidence rate increased to more than 6 days in the Delta outbreak during summer 2021, where presumably the percentage of asymptomatic carriers was larger. Conclusions We conclude that TPR should be used as the primary indicator to enable early intervention, and for predicting hospital admissions in infectious diseases with asymptomatic carriers.
【摘要】目的近期研究表明，2019冠状病毒病(COVID-19)大流行期间，检测阳性率(TPR)与住院人数的相关性优于发病率。然而，流行病学家仍然对广泛使用这一指标进行监测持怀疑态度，尽管大量无症状携带者仍然未知，但基于发病率等已知病例的指标仍然是首选。我们的目的是比较TPR和发病率，确定两者中哪一个最能预测2019冠状病毒病大流行期间住院患者的趋势。方法对60例暴发病例进行回顾性研究，利用意大利在不同流行波中的全球和当地数据，以检测TPR时间序列的峰值和发病率的峰值，找出这两个指标中哪一个最能预测住院患者的峰值。结果平均而言，基于TPR的最佳方法预计发病率约为4.6天(95 % CI 2.8, 6.4)，更准确地说，TPR峰值与住院高峰之间的平均距离为17.6天(95 % CI 15.0, 20.4)，而发病率为13.0天(95 % CI 10.4, 15.8)。此外，2021年夏季三角洲疫情中，TPR和发病率之间的平均差异增加到6天以上，无症状携带者的比例可能更大。结论TPR应作为早期干预的主要指标，用于预测无症状感染者的住院率。
{"title":"The impact of test positivity on surveillance with asymptomatic carriers","authors":"M. Gaspari","doi":"10.1101/2022.06.10.22276234","DOIUrl":"https://doi.org/10.1101/2022.06.10.22276234","url":null,"abstract":"Abstract Objectives Recent studies show that Test Positivity Rate (TPR) gains a better correlation than incidence with the number of hospitalized patients in COVID-19 pandemic. Nevertheless, epidemiologists remain sceptical concerning the widespread use of this metric for surveillance, and indicators based on known cases like incidence rate are still preferred despite the large number of asymptomatic carriers, which remain unknown. Our aim is to compare TPR and incidence rate, to determine which of the two has the best characteristics to predict the trend of hospitalized patients in the COVID-19 pandemic. Methods We perform a retrospective study considering 60 outbreak cases, using global and local data from Italy in different waves of the pandemic, in order to detect peaks in TPR time series, and peaks in incidence rate, finding which of the two indicators has the best ability to anticipate peaks in patients admitted in hospitals. Results On average, the best TPR-based approach anticipates the incidence rate of about 4.6 days (95 % CI 2.8, 6.4), more precisely the average distance between TPR peaks and hospitalized peaks is 17.6 days (95 % CI 15.0, 20.4) with respect to 13.0 days (95 % CI 10.4, 15.8) obtained for incidence. Moreover, the average difference between TPR and incidence rate increased to more than 6 days in the Delta outbreak during summer 2021, where presumably the percentage of asymptomatic carriers was larger. Conclusions We conclude that TPR should be used as the primary indicator to enable early intervention, and for predicting hospital admissions in infectious diseases with asymptomatic carriers.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89581540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Objectives Spurious associations between an exposure and outcome not describing the causal estimand of interest can be the result of selection of the study population. Recently, sensitivity parameters and bounds have been proposed for selection bias, along the lines of sensitivity analysis previously proposed for bias due to unmeasured confounding. The basis for the bounds is that the researcher specifies values for sensitivity parameters describing associations under additional identifying assumptions. The sensitivity parameters describe aspects of the joint distribution of the outcome, the selection and a vector of unmeasured variables, for each treatment group respectively. In practice, selection of a study population is often made on the basis of several selection criteria, thereby affecting the proposed bounds. Methods We extend the previously proposed bounds to give additional guidance for practitioners to construct i) the sensitivity parameters for multiple selection variables and ii) an alternative assumption free bound, producing only logically feasible values. As a motivating example we derive the bounds for causal estimands in a study of perinatal risk factors for childhood onset Type 1 Diabetes Mellitus where selection of the study population was made by multiple inclusion criteria. To give further guidance for practitioners, we provide a data learner in R where both the sensitivity parameters and the assumption-free bounds are implemented. Results The assumption-free bounds can be both smaller and larger than the previously proposed bounds and can serve as an indicator of settings when the former bounds do not produce feasible values. The motivating example shows that the assumption-free bounds may not be appropriate when the outcome or treatment is rare. Conclusions Bounds can provide guidance in a sensitivity analysis to assess the magnitude of selection bias. Additional knowledge is used to produce values for sensitivity parameters under multiple selection criteria. The computation of values for the sensitivity parameters is complicated by the multiple inclusion/exclusion criteria, and a data learner in R is provided to facilitate their construction. For comparison and assessment of the feasibility of the bound an assumption free bound is provided using solely underlying assumptions in the framework of potential outcomes.
{"title":"Selection bias and multiple inclusion criteria in observational studies","authors":"Stina Zetterstrom, I. Waernbaum","doi":"10.1515/em-2022-0108","DOIUrl":"https://doi.org/10.1515/em-2022-0108","url":null,"abstract":"Abstract Objectives Spurious associations between an exposure and outcome not describing the causal estimand of interest can be the result of selection of the study population. Recently, sensitivity parameters and bounds have been proposed for selection bias, along the lines of sensitivity analysis previously proposed for bias due to unmeasured confounding. The basis for the bounds is that the researcher specifies values for sensitivity parameters describing associations under additional identifying assumptions. The sensitivity parameters describe aspects of the joint distribution of the outcome, the selection and a vector of unmeasured variables, for each treatment group respectively. In practice, selection of a study population is often made on the basis of several selection criteria, thereby affecting the proposed bounds. Methods We extend the previously proposed bounds to give additional guidance for practitioners to construct i) the sensitivity parameters for multiple selection variables and ii) an alternative assumption free bound, producing only logically feasible values. As a motivating example we derive the bounds for causal estimands in a study of perinatal risk factors for childhood onset Type 1 Diabetes Mellitus where selection of the study population was made by multiple inclusion criteria. To give further guidance for practitioners, we provide a data learner in R where both the sensitivity parameters and the assumption-free bounds are implemented. Results The assumption-free bounds can be both smaller and larger than the previously proposed bounds and can serve as an indicator of settings when the former bounds do not produce feasible values. The motivating example shows that the assumption-free bounds may not be appropriate when the outcome or treatment is rare. Conclusions Bounds can provide guidance in a sensitivity analysis to assess the magnitude of selection bias. Additional knowledge is used to produce values for sensitivity parameters under multiple selection criteria. The computation of values for the sensitivity parameters is complicated by the multiple inclusion/exclusion criteria, and a data learner in R is provided to facilitate their construction. For comparison and assessment of the feasibility of the bound an assumption free bound is provided using solely underlying assumptions in the framework of potential outcomes.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"231 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85579853","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
José Rafael Tovar Cuevas, Andrés Camilo Méndez Alzate, Diana María Caicedo Borrero, Juan David Díaz Mutis, Lizeth Fernanda Suárez Mensa, Lyda Elena Osorio Amaya
Abstract Objectives To develop a methodological proposal to build clinical classifiers using information about signs and symptoms reported by the patient in initial the consultation and laboratory test results. Methods The proposed methodology considers procedures typical of the Bayesian paradigm of statistics as predictive probabilities and the sequential use of the Bayes formula. Additionally, some procedures belonging to classical statistics, such as Youden’s index and ROC curves, are applied. The method assumes two possible scenarios; when the patient only reports the signs and symptoms and the physician does not have access to information from laboratory tests. The other one is when the physician, besides the patient’s information, knows the blood test results. The method is illustrated using data from patients diagnosed with dengue. Results The performance of the proposed method depends of the set of signs and symptoms and the laboratory tests considered by the doctor as good indicators of presence of the sick in the individual. Conclusions The classifier can be used as a screening tool in scenarios where there is no extensive experience treating sick individuals, or economic and social conditions do not allow laboratory methods or gold standard procedures to complete the diagnosis.
{"title":"Methodological proposal for constructing a classifier algorithm in clinical diagnostics of diseases using Bayesian methods","authors":"José Rafael Tovar Cuevas, Andrés Camilo Méndez Alzate, Diana María Caicedo Borrero, Juan David Díaz Mutis, Lizeth Fernanda Suárez Mensa, Lyda Elena Osorio Amaya","doi":"10.1515/em-2021-0020","DOIUrl":"https://doi.org/10.1515/em-2021-0020","url":null,"abstract":"Abstract Objectives To develop a methodological proposal to build clinical classifiers using information about signs and symptoms reported by the patient in initial the consultation and laboratory test results. Methods The proposed methodology considers procedures typical of the Bayesian paradigm of statistics as predictive probabilities and the sequential use of the Bayes formula. Additionally, some procedures belonging to classical statistics, such as Youden’s index and ROC curves, are applied. The method assumes two possible scenarios; when the patient only reports the signs and symptoms and the physician does not have access to information from laboratory tests. The other one is when the physician, besides the patient’s information, knows the blood test results. The method is illustrated using data from patients diagnosed with dengue. Results The performance of the proposed method depends of the set of signs and symptoms and the laboratory tests considered by the doctor as good indicators of presence of the sick in the individual. Conclusions The classifier can be used as a screening tool in scenarios where there is no extensive experience treating sick individuals, or economic and social conditions do not allow laboratory methods or gold standard procedures to complete the diagnosis.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89949650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Objective COVID-19 data released by public health authorities is subject to inherent time delays. Such delays have many causes, including delays in data reporting and the natural incubation period of the disease. We develop and introduce a numerical procedure to recover the distribution of these delays from data. Methods We extend a previously-introduced compartmental model with a nonlinear, distributed-delay term with a general distribution, obtaining an integrodifferential equation. We show this model can be approximated by a weighted-sum of constant time-delay terms, yielding a linear problem for the distribution weights. Standard optimization can then be used to recover the weights, approximating the distribution of the time delays. We demonstrate the viability of the approach against data from Italy and Austria. Results We find that the delay-distributions for both Italy and Austria follow a Gaussian-like profile, with a mean of around 11 to 14 days. However, we note that the delay does not appear constant across all data types, with infection, recovery, and mortality data showing slightly different trends, suggesting the presence of independent delays in each of these processes. We also found that the recovered delay-distribution is not sensitive to the discretization resolution. Conclusions These results establish the validity of the introduced procedure for the identification of time-delays in COVID-19 data. Our methods are not limited to COVID-19, and may be applied to other types of epidemiological data, or indeed any dynamical system with time-delay effects.
{"title":"Identification of time delays in COVID-19 data","authors":"N. Guglielmi, E. Iacomini, Alex Viguerie","doi":"10.1515/em-2022-0117","DOIUrl":"https://doi.org/10.1515/em-2022-0117","url":null,"abstract":"Abstract Objective COVID-19 data released by public health authorities is subject to inherent time delays. Such delays have many causes, including delays in data reporting and the natural incubation period of the disease. We develop and introduce a numerical procedure to recover the distribution of these delays from data. Methods We extend a previously-introduced compartmental model with a nonlinear, distributed-delay term with a general distribution, obtaining an integrodifferential equation. We show this model can be approximated by a weighted-sum of constant time-delay terms, yielding a linear problem for the distribution weights. Standard optimization can then be used to recover the weights, approximating the distribution of the time delays. We demonstrate the viability of the approach against data from Italy and Austria. Results We find that the delay-distributions for both Italy and Austria follow a Gaussian-like profile, with a mean of around 11 to 14 days. However, we note that the delay does not appear constant across all data types, with infection, recovery, and mortality data showing slightly different trends, suggesting the presence of independent delays in each of these processes. We also found that the recovered delay-distribution is not sensitive to the discretization resolution. Conclusions These results establish the validity of the introduced procedure for the identification of time-delays in COVID-19 data. Our methods are not limited to COVID-19, and may be applied to other types of epidemiological data, or indeed any dynamical system with time-delay effects.","PeriodicalId":37999,"journal":{"name":"Epidemiologic Methods","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90118089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号