Underreported Data Research Articles

Estimation of Soil Carbon Balance Based on СО<sub>2</sub> Emission Determination

The increased interest nowadays in quantitative assessment of soil respiration is largely due to studies of the role of various terrestrial ecosystems in changing the concentration of the most important greenhouse gas, CO2, in the atmosphere. The review considers methodological aspects of determining the actual CO2 emission from soils using chamber and absorption methods, as well as the use of the obtained data to assess the carbon balance in soils. Successful development of this topic will allow to promptly get an answer to the main question of this pressing environmental issue: what is the soil of this or that ecosystem for atmospheric CO2 – a net source or a net sink? The article analyzes the results of works devoted to comparative determination of CO2 emission from soils by these methods. It is shown that the widespread opinion about obtaining underreported data by absorption method is often based on studies in which the basic principles of the method were violated. It is concluded that it is necessary to carry out in-depth comparative studies on determination of average daily indicators of CO2 emission from soil by chamber and absorption methods. It is recognized that, regardless of the method, the main problem in using CO2 production data to estimate soil C balance is the adequate partitioning of total respiration into heterotrophic and autotrophic components, the ratio between which varies widely depending on soil and vegetation conditions. Due to imperfection and labor intensity of existing methods, this division can cause significant errors in determining the annual mineralization of soil organic matter. To reduce them, the approach to determination of mineralization losses of CO2 in bare fallow soils is considered. It is natural for soils of agrocenoses, but as a methodological technique can probably be used in natural grass ecosystems as well. Reduction of errors can be ensured due to the fact that differences in actual mineralization of organic matter in bare fallow and plant-occupied soils are several times less than changes in the ratio between heterotrophic and autotrophic components of soil respiration in different biogeocenoses.

Pathological response to neoadjuvant chemoimmunotherapy in resectable non-small cell lung cancer (NSCLC): A systematic review and meta-analysis.

e20582 Background: The advantage of combining chemotherapy and immune checkpoint inhibitors (ICIs) in resectable non-small cell lung cancer (NSCLC) patients has recently been studied in clinical trial settings. We aim to systematically revise the evidence of efficacy and safety of neoadjuvant chemoimmunotherapy treatment. Methods: We carried out a systematic search on MEDLINE, Scopus, Cochrane, and ASCO meetings library. Screening was done to include clinical trials of stage I-IIIA NSCLC patients who underwent any combination of ICIs and chemotherapy. The primary outcomes were major pathological response (MPR) and pathological complete response (pCR), and the secondary outcomes were 24-month overall survival (OS) and safety measured by adverse events (AEs). A meta-analysis was conducted for intention-to-treat MPR and pCR outcomes, in addition to a mixed-effects meta-regression model to investigate the effect of PD-L1 expression, tumor mutational burden (TMB), N stage (AJCC 8th), tumor size and tumor histology on MPR and pCR. PD-L1+ patients were defined to have a tumor proportion score of more or equal to 1%. Results: We included 9 phase II clinical trials and 1 phase III trial, all consisting of 471 patients. Trials administered neoadjuvant chemotherapy agents like carboplatin (N = 6 studies), cisplatin (N = 5), nab-paclitaxel (N = 4), pemetrexed (N = 2), and others, either as a doublet- or a triplet-agent regimen. ICIs used were anti-PD-1 (N = 5), anti-PD-L1 (N = 3), or anti-CTLA-4 agents (N = 2). Meta-analysis of 447 patients from 9 trials showed a pooled MPR of 49.2% (95% CI: 34.8 to 63.7%) with high heterogeneity (I2 = 87%, p < 0.01). Pooling the pCR from 10 trials including 471 patients results in an estimate of 27.5% (95% CI: 17.2 to 38.2%) with significant heterogeneity (I2 = 80%, p < 0.01). Meta-regression has found a significant association between being PD-L1+ and achieving MPR (odds ratio = 3.82, 95% CI: 1.55 to 9.58). However, this effect diminished in the multiple meta-regression that included age, tumor histology, in addition to PD-L1 status. None of these factors was correlated to achieving a pCR. Investigating the effect of tumor size, T stage, N stage, or TMB on MPR or pCR was not possible due to underreported data. Three trials of 137 patients reported a 24-month OS of 84.4% (95% CI: 77.6 to 90.2%) with low heterogeneity (I2 = 11%, p = 0.33). The most prevalent grade 3/4 AEs from the neoadjuvant treatment were neutropenia in 7.2% of patients, diarrhea in 3.2%, fatigue in 3.0%, and anemia in 2.1% of patients. The commonest any-grade AEs were fatigue (43.9%), alopecia (38.4%), diarrhea (31.6%), anemia (23.8%) and neutropenia (17.0%). Conclusions: The addition of ICIs to chemotherapy in neoadjuvant settings is an effective and feasible option. Effect of tumor size, PD-L1 status and TMB on MPR and pCR should be further investigated for better selection of treatment candidates.

Omnibus Tests for Multiple Binomial Proportions via Doubly Sampled Framework with Under-Reported Data

Previously, Rahardja (2020) paper (in the first reference list) developed a (pairwise) multiple comparison procedure (MCP) to determine which (proportions) pairs of Multiple Binomial Proportions (with under-reported data), the significant differences came from. Generally, such an MCP test (developed by Rahardja, 2020) is the second part of a two-stage sequential test. In this paper, we derived two omnibus tests (i.e., the overall equality of multiple proportions test) as the first part of the above two-stage sequential test (with under-reported data), in general. Using two likelihood-based approaches, we acquire two Wald-type (Omnibus) tests to compare Multiple Binomial Proportions (in the presence of under-reported data). Our closed-form algorithm is easy to implement and not computationally burdensome. We applied our algorithm to a vehicle-accident data example.

Ten-Year Mortality Trends and Natural Causes of Death in the Iraqi Kurdistan

Background: Mortality and causes of death are among the most important statistics used in assessing the effectiveness of a country’s health system. Several countries do not have information systems for collecting these data, and they must therefore be estimated from surveys. Objective: This study analyzes mortality data retrieved from official government databases in Iraqi Kurdistan to describe ten-year trends in natural causes of death. Methods: Data for natural causes of death, reported from 2009 to 2018, were extracted from the databases of the Registration Bureau of Births and Deaths and of the Forensic Medicine of the Province of Sulaymaniyah. A sample of 16,433 causes of death was analyzed. Results: Causes of death were coded according to the ICD-10 classification. Overall, cardiovascular diseases were the leading cause of mortality (52.6%), followed by neoplasms (17.7%), infectious and parasitic diseases (8.9%), and genitourinary diseases (6.3%). Neonatal conditions, congenital anomalies, and neurological conditions each accounted for less than 1% each. Numbers of natural deaths by cause and cause-specific mortality rates have been estimated for the entire Region of Iraqi Kurdistan. Comparisons with other sources suggest that there is a substantial amount of underreporting, especially in relation to deaths of infants and under-five children. Conclusion: Our findings confirm that the region is facing a burden of non-communicable diseases, coupled with high proportions of infectious diseases. However, the lack of effective vital statistics with combined under-reported data collection highlights the need for implementation of health monitoring systems. Advancements in generating high-quality data are essential in improving health and reducing preventable deaths. The establishment of a novel Health Information System is discussed.

Survival Analysis of Patients With COVID-19 in India by Demographic Factors: Quantitative Study.

BackgroundStudies of the transmission dynamics of COVID-19 have depicted the rate, patterns, and predictions of cases of this pandemic disease. To combat transmission of the disease in India, the government declared a lockdown on March 25, 2020. Even after this strict lockdown was enacted nationwide, the number of COVID-19 cases increased and surpassed 450,000. A positive point to note is that the number of recovered cases began to slowly exceed that of active cases. The survival of patients, taking death as the event that varies by age group and sex, is noteworthy.ObjectiveThe aim of this study was to conduct a survival analysis to establish the variability in survivorship of patients with COVID-19 in India by age group and sex at different levels, that is, the national, state, and district levels.MethodsThe study period was taken from the date of the first reported case of COVID-19 in India, which was January 30, 2020, up to June 30, 2020. Due to the amount of underreported data and removal of missing columns, a total sample of 26,815 patients was considered. Kaplan-Meier survival estimation, the Cox proportional hazard model, and the multilevel survival model were used to perform the survival analysis.ResultsThe Kaplan-Meier survival function showed that the probability of survival of patients with COVID-19 declined during the study period of 5 months, which was supplemented by the log rank test (P<.001) and Wilcoxon test (P<.001) to compare the survival functions. Significant variability was observed in the age groups, as evident from all the survival estimates; with increasing age, the risk of dying of COVID-19 increased. The Cox proportional hazard model reiterated that male patients with COVID-19 had a 1.14 times higher risk of dying than female patients (hazard ratio 1.14; SE 0.11; 95% CI 0.93-1.38). Western and Central India showed decreasing survival rates in the framed time period, while Eastern, North Eastern, and Southern India showed slightly better results in terms of survival.ConclusionsThis study depicts a grave scenario of decreasing survival rates in various regions of India and shows variability in these rates by age and sex. In essence, we can safely conclude that the critical appraisal of the survival rate and thorough analysis of patient data in this study equipped us to identify risk groups and perform comparative studies of various segments in India.International Registered Report Identifier (IRRID)RR2-10.1101/2020.08.01.20162115

The COVID-19 (SARS-CoV-2) uncertainty tripod in Brazil: Assessments on model-based predictions with large under-reporting

The COVID-19 pandemic (SARS-CoV-2 virus) is the global crisis of our time. The absence of mass testing and the relevant presence of asymptomatic individuals causes the available data of the COVID-19 pandemic in Brazil to be largely under-reported regarding the number of infected individuals and deaths. We develop an adapted Susceptible-Infected-Recovered (SIR) model, which explicitly incorporates the under-reporting and the response of the population to public health policies (confinement measures, widespread use of masks, etc). Large amounts of uncertainty could provide misleading predictions of the COVID-19 spread. In this paper, we discuss the role of uncertainty in these model-based predictions, which is illustrated regarding three key aspects: (i) Assuming that the number of infected individuals is under-reported, we demonstrate anticipation regarding the infection peak. Furthermore, while a model with a single class of infected individuals yields forecasts with increased peaks, a model that considers both symptomatic and asymptomatic infected individuals suggests a decrease of the peak of symptomatic cases. (ii) Considering that the actual amount of deaths is larger than what is being registered, we demonstrate an increase of the mortality rates. (iii) When we consider generally under-reported data, we demonstrate how the transmission and recovery rate model parameters change qualitatively and quantitatively. We also investigate the “the uncertainty tripod”: under-reporting level in terms of cases, deaths, and the true mortality rate of the disease. We demonstrate that if two of these factors are known, the remainder can be inferred, as long as proportions are kept constant. The proposed approach allows one to determine the margins of uncertainty by assessments on the observed and true mortality rates.

An integrated‐likelihood‐ratio confidence interval for a proportion based on underreported and infallible data

We derive and examine the interval width and coverage properties of an integrated‐likelihood‐ratio confidence interval for the binomial parameter p using a double‐sampling scheme. The data consist of a relatively large fallible sample containing underreported data and a relatively small infallible subsample. Via Monte Carlo simulations, we determine that the new integrated‐likelihood‐ratio interval estimator displays slightly conservative to moderately conservative coverage properties for small to medium sample sizes and can have shorter average‐interval width than two previously proposed confidence intervals when p &lt; 0.10 or p &gt; 0.90. We also apply the integrated‐likelihood‐ratio confidence interval to a real‐data set and determine that the integrated‐likelihood‐ratio interval has superior performance when contrasted to two properties of two competing confidence intervals.

Description of Data Breaches Notifications in France and Lessons Learned for the Healthcare Stakeholders.

Although the consequences of the General Data Protection Regulation (GDPR) have been widely discussed, the violations have not been described in medical literature. In this study, we focus our analyses on the data breach notifications, in France, defined in the article 4 of GDPR as "a breach of security resulting, accidentally or unlawfully, in the destruction, loss, alteration, unauthorized disclosure of personal data transmitted, stored or otherwise processed, or unauthorized access to such data." Among 3,824 data breach notifications reported between May 2018 and February 2020, 244 (6.4%) is related to the health sector. Loss of confidentiality is the most important breach (80.7%) in this sector, followed by the loss of availability (27.5%). Malicious cause occurred in 58.2% of them. We hypothesized a phenomenon of underreported data breach incidents in health due to a mismatch between cybersecurity and data privacy issues.

Bootstrap-Tobit model for maritime accident economic loss considering underreporting issues

The underreporting of maritime accident data is a major concern in maritime transportation safety analysis. In order to avoid biased effects caused by the underreporting issue, this study aims to propose a Bootstrap-Tobit regression model for the estimation of economic loss resulting from maritime accidents. The questionnaire survey method is employed to determine the possible underreporting rates of maritime accident data and their occurrence probabilities, while the bootstrap sampling method is adopted to supplement the underreported data by generating a number of bootstrap samples. With ten year's archived maritime accident data, a case study is created to calibrate the proposed model. The proposed Bootstrap-Tobit model has captured the impacts of underreporting issues on the marginal effects of the economic loss contributory factors. The proposed model could supplement sufficient underreported data so that the results are more reliable and convincible.

A marginal moment matching approach for fitting endemic-epidemic models to underreported disease surveillance counts.

Count data are often subject to underreporting, especially in infectious disease surveillance. We propose an approximate maximum likelihood method to fit count time series models from the endemic-epidemic class to underreported data. The approach is based on marginal moment matching where underreported processes are approximated through completely observed processes from the same class. Moreover, the form of the bias when underreporting is ignored or taken into account via multiplication factors is analyzed. Notably, we show that this leads to a downward bias in model-based estimates of the effective reproductive number. A marginal moment matching approach can also be used to account for reporting intervals which are longer than the mean serial interval of a disease. The good performance of the proposed methodology is demonstrated in simulation studies. An extension to time-varying parameters and reporting probabilities is discussed and applied in a case study on weekly rotavirus gastroenteritis counts in Berlin,Germany.

Bias Correction in Clustered Underreported Data

Data quality from poor and socially deprived regions have given rise to many statistical challenges. One of them is the underreporting of vital events leading to biased estimates for the associated risks. To deal with underreported count data, models based on compound Poisson distributions have been commonly assumed. To be identifiable, such models usually require extra and strong information about the probability of reporting the event in all areas of interest, which is not always available. We introduce a novel approach for the compound Poisson model assuming that the areas are clustered according to their data quality. We leverage these clusters to create a hierarchical structure in which the reporting probabilities decrease as we move from the best group to the worst ones. We obtain constraints for model identifiability and prove that only prior information about the reporting probability in areas experiencing the best data quality is required. Several approaches to model the uncertainty about the reporting probabilities are presented, including reference priors. Different features regarding the proposed methodology are studied through simulation. We apply our model to map the early neonatal mortality risks in Minas Gerais, a Brazilian state that presents heterogeneous characteristics and a relevant socio-economical inequality.

A Hierarchical Framework for Correcting Under-Reporting in Count Data

Tuberculosis poses a global health risk and Brazil is among the top 20 countries by absolute mortality. However, this epidemiological burden is masked by under-reporting, which impairs planning for effective intervention. We present a comprehensive investigation and application of a Bayesian hierarchical approach to modeling and correcting under-reporting in tuberculosis counts, a general problem arising in observational count data. The framework is applicable to fully under-reported data, relying only on an informative prior distribution for the mean reporting rate to supplement the partial information in the data. Covariates are used to inform both the true count-generating process and the under-reporting mechanism, while also allowing for complex spatio-temporal structures. We present several sensitivity analyses based on simulation experiments to aid the elicitation of the prior distribution for the mean reporting rate and decisions relating to the inclusion of covariates. Both prior and posterior predictive model checking are presented, as well as a critical evaluation of the approach. Supplementary materials for this article, including a standardized description of the materials available for reproducing the work, are available as an online supplement.

Analysis of past and future continental crash mortality rates employing long-term data

ABSTRACTThe global and continental crash death rate trends, adjusted for under-reporting, death definition, and missing data, are analyzed, modelled, and compared with international findings considering data for 215 countries for 35 years. Although global reported mortality rates required 25% to 44% adjustments, that estimated by World Health Organization (WHO) and World Bank showed far greater rates. The model developed for global mortality rates per population though indicates a peak threshold, the current rate is expected to continue for some years before descending. In contradiction with WHO, Transport Research Laboratory (TRL), and World Bank (WB), the model indicates a drop of 22% in mortality rate by 2025. The rate for Europe, Oceania, and Asia drop to one eighth, one half, and one sixth, respectively. The rates for Africa doesn't change, and that for South America increases by 15%. The current global mortality per 10,000 vehicles drops exponentially by 38% toward 2025. The mortality rate per 10,000 vehicles for the various continents are expected to drop by 34% to 68%. The earlier rates showed a much faster drop than that per population. The observant reason could be the faster vehicle growth compared with population. The findings call for extensive traffic safety plans and tense implementation efforts in Africa, South America, and Asia to have faster global descending trends in mortality rates. More research that considers under-reporting of fatalities is necessary.

Disease biogeography: spatial and temporal analyses of infectious disease burden at the country-level scale provides new insights and challenges

In a recent study, Wood et al. (2017 Phil. Trans. Roy. Soc. B 372, 20160122) utilized a novel set of spatial and temporal analyses to identify which factors were most strongly correlated with changes in human infectious disease burdens from 1990 to 2010 in 60 countries. Using the statistical analyses and findings of this article as a framework, I have identified three important insights and challenges that this research presents for disease biogeography moving forward. First, the main factor still limiting disease biogeography research progress is underreported or absent data – particularly in the case of neglected tropical diseases. Second, the use of disability-adjusted life years instead of indirect measures of disease burden should be a focal point of disease biogeography research since it allows for comparisons of lethal and non-lethal diseases. Finally, disease biogeography studies that utilize country-level statistical analyses may be better at identifying demographic and economic drivers than environmental or biological drivers.

Multi-sectoral road traffic injury surveillance and road safety campaign in Kaduna Metropolis, 2016 - Pilot study

Introduction : road traffic injuries affect 50 million people globally with 90% of the 1.25 million deaths occuring in low and middle income countries (LMIC). LMIC’S often suffer from under-reporting and poor data quality contingent on solely tasking RTI surveillance to lead agency. We developed and tested the practicability of a multisectoral RTI surveillance in Kaduna metropolis Nigeria, where Federal Road Safety Corps (FRSC) collate and report RTI data. Methods : we adapted Injury surveillance guidelines from WHO and CDC. Stakeholders defined a case of RTI as any person injured or killed within 30 days of exposure to RTI from a crash on public road in Kaduna Metropolis and selected five RTI reporting sites (three health facilities, Police and FRSC). Data collected with a pretested tool, linked by deterministic method to get a single dataset analysed with Epi-Info 7.1. Results : from February to July 2016, 667 road traffic crashes and 1062 injured people compared to 173 (26%) crashes and 183 (17%) injured reported by FRSC. Overall case fatality rate, 10.4%. Male preponderance was 84.0%, age group 25 - 34 years was 35.2%. Of the injured, 17.0% pedestrians, only 15.2% used helmet or safety belt and 44.2% evacuated by other road users. RTI injuries per week reduced from 59 to 14 (76.3%) during seat belt use sensitization. Associated factors with moderate to fatal injury were travelling between 00:00 - 11:59 hours (Odd Ratio (OR) 1.9, 95% Confidence Interval (C.I) = 1.4 - 2.5, 18:00 - 23:59 hours (OR 1.6, 95% C.I = 1.2 - 2.1) and being a driver at time of crash (OR 1.7, 95% C.I = 1.3 - 2.2). Conclusion : a multi-sectoral RTI surveillance is practicable and yielded a robust evidence – based information rather than solely relying on FRSC data for action to reduce RTI in Kaduna metropolis. This was recommended in keeping with global push to achieve sustainable development goal (SDG) 3.6.

Road characteristics and environment factors associated with motorcycle fatal crashes in Malaysia

This study aims to determine risk factors contributing to traffic crashes in 9,176 fatal cases involving motorcycle in Malaysia between 2010 and 2012. For this purpose, both multinomial and mixed models of motorcycle fatal crash outcome based on the number of vehicle involved are estimated. The corresponding model predicts the probability of three fatal crash outcomes: motorcycle single-vehicle fatal crash, motorcycle fatal crash involving another vehicle and motorcycle fatal crash involving two or more vehicles. Several road characteristic and environmental factors are considered including type of road in the hierarchy, location, road geometry, posted speed limit, road marking type, lighting, time of day and weather conditions during the fatal crash. The estimation results suggest that curve road sections, no road marking, smooth, rut and corrugation of road surface and wee hours, i.e. between 00.00am to 6am, increase the probability of motorcycle single-vehicle fatal crashes. As for the motorcycle fatal crashes involving multiple vehicles, factors such as expressway, primary and secondary roads, speed limit more than 70km/h, roads with non-permissible marking, i.e. double lane line and daylight condition are found to cause an increase the probability of their occurrence. The estimation results also suggest that time of day (between 7pm to 12pm) has an increasing impact on the probability of motorcycle single-vehicle fatal crashes and motorcycle fatal crashes involving two or more vehicles. Whilst the multinomial logit model was found as more parsimonious, the mixed logit model is likely to capture the unobserved heterogeneity in fatal motorcycle crashes based on the number of vehicles involved due to the underreporting data with two random effect parameters including 70km/h speed limit and double lane line road marking.

Under-reported data analysis with INAR-hidden Markov chains.

In this work, we deal with correlated under-reported data through INAR(1)-hidden Markov chain models. These models are very flexible and can be identified through its autocorrelation function, which has a very simple form. A naïve method of parameter estimation is proposed, jointly with the maximum likelihood method based on a revised version of the forward algorithm. The most-probable unobserved time series is reconstructed by means of the Viterbi algorithm. Several examples of application in the field of public health are discussed illustrating the utility of the models. Copyright © 2016 John Wiley & Sons, Ltd.

The tricky charm of the radial access.

The tricky charm of the radial access.

Abstract 349: Comparisons of Self-reported and Measured Body Mass Index in a large Employee Population: The Baptist Health South Florida Employee Study

Background: Self reporting of health information by participants in employer sponsored programs may result in under reporting of this data. Since policies are formulated on the basis of analysis of available data, underreported data may adversely inform healthcare policy decisions in workplace and other relevant settings. In this study, we examine the disparity in measured versus self reported body mass index (BMI) in a large, diverse working population participating in a work-based Health Risk Assessment (HRA). Methods: Baptist Health South Florida, a large not-for-profit health care organization, conducts a HRA annually for its employees. The HRA consists of an online health questionnaire and an in-person assessment of biometric measures. Participants report their height and weight in the online questionnaire, and height and weight is measured during collection of biometric information, from which BMI is calculated. BMI was categorized as: not obese (&lt;25kg/m 2 ), obese (30 - 34.9kg/m 2 ) and morbidly obese (≥35kg/m 2 ). Results: A total of 9536 persons self reported their BMI and subsequently had BMI measured. Mean age of the participants was 43.1 years (± 12.1). Mean BMI was 28.6 (± 6.2) and 27.8 (± 5.9) respectively for measured and self reported BMI respectively. Among all participants, 90% self reported accurately in their respective weight categories, while 9% underreported their weight. However, among obese participants (n=3168), 72% correctly self-reported their BMI. Approximately 29% of obese class I participants reported BMIs in the non-obese category, while about 22% of those with BMI ≥35 kg/m 2 reported as being in Obese class I (&lt;35kg/m 2 ). There was a high degree of agreement overall observed between measured and self reported BMI (κ = 0.84; Intra Class Coefficient = 0.89). Conclusion: Despite of a high degree of overall agreement between measured and self reported BMI, obese participants considerably underreport BMI. Thus, underreporting of self-reported data should be taken into account during program planning and formulation of healthcare policy decisions in the workplace.

Estimating the incidence reporting rates of new influenza pandemics at an early stage using travel data from the source country

During the surveillance of influenza pandemics, underreported data are a public health challenge that complicates the understanding of pandemic threats and can undermine mitigation efforts. We propose a method to estimate incidence reporting rates at early stages of new influenza pandemics using 2009 pandemic H1N1 as an example. Routine surveillance data and statistics of travellers arriving from Mexico were used. Our method incorporates changes in reporting rates such as linearly increasing trends due to the enhanced surveillance. From our results, the reporting rate was estimated at 0·46% during early stages of the pandemic in Mexico. We estimated cumulative incidence in the Mexican population to be 0·7% compared to 0·003% reported by officials in Mexico at the end of April. This method could be useful in estimation of actual cases during new influenza pandemics for policy makers to better determine appropriate control measures.