Journal of Pharmaceutical Care & Health Systems
Open Access
ISSN: 2376-0419
ISSN: 2376-0419
Mini Review - (2021)
Health records involve unique data which are highly subjected to privacy right security checks and its disclosure may lead to violation of this right and therefore cannot be implemented without additional consideration. The health care community has long recognized the potential for health information technology systems in its management, thus improving clinical and health care while reducing costs and this has synergized the access to health care services and information. The mobility in health care provision demands the need for sharing of patient data and this require interoperable health information Technology infrastructure, privacy and security of the resource and this will enhance stakeholders trust and promotes health information interoperability diffusion. It has been noted that the major challenge in the integration of Health Information Record Management system is interoperability and practitioners in a private practice may have difficulty obtaining complete information about a patient who is currently being hospitalized. The study established that interoperability and privacy issues still stands out as the major hindrance to sharing of health care records. There is a need for closer collaboration and trust between the major stakeholders in the industry focusing on their inclusivity in working toward the achievement of interoperability and privacy concerns.
COVID-19; GLARMA; Growth models; Model risk; Natural language processing; Sentiment ana
NYT: New York Times; ECDC: European Centre for Disease Prevention and Control; USCDC: Disease Control and Prevention; WHO: World Health Organization; SDE: Stochastic Differential Equation.
A newly surfaced coronavirus strain has led, over the past year, to the onset of a global devastating pandemic that still continues into 2021, and which became widely known as the COVID-19 respiratory disease. COVID-19 has had an immense impact on society in multiple ways, including significant mortality and morbidity, long term health effects (long-COVID) and significant impact on the economies globally. Therefore, joining the global battle against COVID-19, we considered crucial that we study the statistical properties of the evolution of this disease so that we address statistical questions such as “Why were so many disease growth rate projections so significantly wrong in the early stages of the pandemic?”
As scientists feeling the need to support our community and to contribute to the fight against the pandemic, we investigated this question from a statistical perspective based on a statistical analysis of model risk, to assist policymakers and public health officials in managing the severe consequences of the pandemic and of the imposed countermeasures. We also aimed to obtain a greater understanding of how the communication of public health announcements affected the populations’ behavior, and if this had a noticeable effect on “flattening the curve” of infection cases; in other words, whether governments successfully communicated how critical it is that the public follow the protective policies taken. We quantified this effect through daily changes in infection rates as a result of public health policy and information announcements, whose impact we extracted over time via text mining methodologies from a variety of press releases and news articles from authoritative news agencies and public health authorities that included, among others, the New York Times (NYT), the Guardian, the Telegraph, the European Centre for Disease Prevention and Control (ECDC), the US Centre for Disease Control and Prevention (USCDC) and the World Health Organization (WHO).
Our research results gave us significant insight on two additional questions, namely “What is the most reliable and accurate way to build an epidemic growth model for this disease?” and “Can one assess the influence of public policy and public health reporting on the dynamics of the COVID-19 pandemic spread over time?”
We need to remark that there are many ways to model epidemics, through for instance compartmental epidemic models that capture individuals in groups of stages of, for example, healthy, infected, recovered or deceased [1]. These models are primarily useful for a detailed epidemiological analysis and analysis of vaccine response, however, they often utilize a few key parameters in the estimation that strongly affect the model outputs, such as the reproductive number which has been hotly debated in the epidemiological studies of COVID-19 [2]. Alternative approaches to epidemic modelling include stochastic epidemic models that focus more on models based on a Stochastic Differential Equation (S.D.E.) formulation, usually under certain assumptions to obtain tractability, which, however, may be too simple to capture the pandemic evolution effectively [3].
Given all the different modelling approaches, it is therefore critical to understand that each modelling approach is suitable for specific purposes, and although the aforementioned models are required for sophisticated epidemiological disease modelling at the individual demographic level or to better understand vaccine program responses, they are not the models that public health policymakers resort to. Often, throughout the COVID-19 pandemic crisis, the public health decision-makers in governments had to abstract from such detailed models and instead focus on simpler infection growth rate models that are better suited to national-level epidemiological analysis. This class of models is the one we focus on in our analysis. The most popular of these models used during the COVID-19 pandemic was the Gompertz model, as discussed in which forms the baseline in our analysis [4].
The evolution of the pandemic may be captured through a number of different proxies which form the datasets that researchers can access. These include time-series of daily death counts, new infections, total cumulative infections, number of hospitalizations, number of patients in Intensive Care Units, virus presence levels in public sewers etc. We have selected to study the daily number of infections at the national level, as the number of deaths is often under-reported or prone to reporting errors, as it is not only based on the exact daily counts but also depends on the governmental reporting policy. This is often affected by overloaded hospital and coroner reporting systems during the pandemic, hence making the use of this data less reliable for the models we want to calibrate.
In addition, we rely on data aggregated at the national level since we seek the policymakers’ perspective for the prediction of the infection spread at a large-scale population level. Our research contributes to the epidemiological growth rate model literature by extending the basic macro infection growth rate models in four major ways: i) we develop advanced, flexible time-series models that incorporate stochastic population growth models; ii) we modify our observation distribution to better capture the different phases of COVID-19 spread in the community; iii) we develop a Bayesian estimation procedure for these models that readily offers uncertainty quantification; iv) we incorporate a novel exposure adjustment feature that involves a custom public health news announcement sentiment index. This feature allows public health officials to quantify statistically the impact of public health announcements and news releases in terms of their influence on the public’s behavior, as reflected by daily changes in infections as people learned more about COVID-19 through health and news releases.
The incorporation of public sentiment information in the models allows us to obtain an insight into the effectiveness of public information and health campaigns. Our method constitutes a novel way to measure how well the pandemic protective measures are received by the public, and whether or not people adhere to the measures taken. These concepts are inherently included in the number of infected cases, and are therefore very useful to incorporate in epidemic models, as has been demonstrated in related works [5,6]. Moreover, our results demonstrate that textual data obtained from public news is a suitable proxy to public sentiment, owing to the effect of the pandemic on linguistic expression in the time of COVID-19 [7-12].
The seven countries we analyzed were selected due to the varying epidemic spread profiles (United Kingdom, Germany, Spain, Italy, United States, Japan, Australia) they exhibit. We set up our analysis to include pre- and post-vaccination phases, and showed that the baseline Gompertz model is not able to adequately describe data that span both phases, due to the fact that the COVID-19 pandemic exhibited significantly different epidemiological profiles in the two periods under study, such as, in particular, the rapid growth rate in the second wave of the pandemic starting in Autumn 2020. These results make the model risk component explicitly clear, and illustrate that there may be significant repercussions if policymakers are not aware of it.
Therefore, policy-makers ought to pay particular attention to the characteristics of the growth curves of the number of infections, and if necessary, employ more advanced population growth models such that they include the identified curve features. In this way, model risk will be reduced, provided also those decisions on the selected models are reviewed frequently while continuously incorporating new data that express the pandemic’s evolution. Furthermore, we contributed a novel sentiment index that was obtained from news articles and institutional announcements about the COVID-19 pandemic, as communicated by institutions and national Centres for Disease Control. The inclusion of the sentiment index in the population growth models via an exposure adjustment, revealed that at the onset of the pandemic the in-sample model fit is much better, namely the sentiment adjustment significantly helps the model capture the growth rate of infected cases. This is especially important for model assessment, and assessment of the effectiveness of the applied pandemic countermeasures, protective policies, and the way they were communicated to the public. Our results prove that this research work is particularly useful for designing, communicating and evaluating protective policies during extreme events, as well as scenario generation to better prepare for crises in the future.
Citation: Chalkiadakis I, Yan KH, Peters GW, Shevchenko PV (2021) Public Health Policymaking using Insights from COVID-19 Modelling with News Sentiment. J Pharma Care Health Sys S7:237.
Received: 10-Aug-2021 Accepted: 24-Aug-2021 Published: 31-Aug-2021
Copyright: © Chalkiadakis I, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.