Retrospective Case-Control Study
Abstract
<h2>Cover Page</h2> <p>Retrospective Case-Control Study</p> <p>Student</p> <p>Institution</p> <p>Course</p> <p>Instructor</p> <p>Date</p> <h2>Research Context and Characteristics of the Retrospective COVID-19 Study</h2> <p>The case is a retrospective case-control study about COVID-19. A retrospective study is an experimental study design in which past events are studied and assessed (Frost, 2023). When the project begins, the researchers are already aware of the results for every subject. These studies use respondents’ recollections and data collected for non-project-related purposes rather than collecting data as events occur. Usually, patients are not followed up in these studies. According to the study, the coronavirus pandemic of 2019 (COVID-19) spread throughout the world, and as of 2021, there had been more positive than fatal cases. Consequently, it remained crucial to determine the risk factors for mortality among individuals in critical condition.</p> <p>The multicentre retrospective case-control study was conducted in four government-designated care centres for individuals with COVID-19 in three cities in China (Gao et al., 2021). All people identified as having severe COVID-19, described as severe and critical according to the Chinese Standards, were examined. Those who died or were released from care between January 2020 and March 2020 fulfilled the requirements for inclusion. The research was authorised by the Research Ethics Commission of the General Hospital of Southern Theater Command of PLA, and the Ethics Commission waived the informed consent requirements.</p> <p>For every severe COVID-19 inpatient, researchers gathered clinical and demographic information. To ascertain the independent risk factors associated with the probability of 28-day and 60-day survival, they conducted a survival curve analysis using univariate and multivariable Cox regression approaches. The study results showed that, before March 2020, 338 patients who met the research inclusion criteria were identified. The final analysis comprised 325 participants, of whom 270 recovered and 55 died, after excluding 13 individuals for whom no clinical data were available (Gao et al., 2021).</p> <p>Two hundred and twenty-two cases were classified as severe, while 103 were classified as critical. Six severe cases and 49 critical cases involved non-survivors. A greater likelihood of death was correlated with older age, more comorbidities, and more severe illness, as demonstrated by substantial differences in age, underlying conditions, and clinical classification between survivors and non-survivors.</p> <h2>Methodological Weaknesses and Strategies for Improving Retrospective Research</h2> <p>Unfortunately, this retrospective study has several drawbacks, including an increased risk of bias, discrepancies, and inaccuracies. The retrospective study used data obtained for other purposes. There could be discrepancies in the data because it was recorded using different methods, tools, and personnel involved in the care of the study participants (Frost, 2023). Conditions other than the pandemic may also have been present during the measurements but were not considered during the pandemic data collection.</p> <p>Confounding may result from the absence of measurements for control variables because the study included no controlled variables. The retrospective study also demonstrated possible bias in the inclusion criteria for participants. This bias occurs when the research population is not randomly drawn from the population of interest, when data are lost because of follow-up problems, withdrawal, or death, and when controlling confounders that may affect the results is difficult (Dziadkowiec et al., 2020).</p> <p>An additional significant limitation of this retrospective study design is the extent of the previously gathered data, which may not be accurate or may not adequately cover the intended target population. The retrospective study design could be improved by minimising bias. To reduce the likelihood of bias in retrospective research, it is essential to consider the sample size and statistical power required to prevent random errors, define a clear hypothesis, identify appropriate populations or groups, select clinically relevant data, maintain strict inclusion and exclusion criteria, determine and describe the outcomes that will be measured, and prepare an appropriate study plan in advance (Dziadkowiec et al., 2020).</p> <p>In addition, including patients who may have self-diagnosed and those who received a medical diagnosis could improve and enhance the design of this retrospective cohort study. Suppose a patient believed that they had COVID-19 after observing several signs and symptoms of the disease. In that case, their medical records could be used as a starting point to determine whether they satisfied all eligibility requirements for inclusion in the study. Considering this group of patients as study participants would increase the study sample and potentially improve the effectiveness of the retrospective study design.</p> <h2>Proposed Prospective Cohort Design for Monitoring COVID-19 Outcomes</h2> <p>The proposed new study design would be a prospective cohort study. A prospective cohort design follows patients for a predetermined period after they are first exposed to a disease or its risk factors, either in the past or currently. In the hierarchy of evidence, prospective designs are typically ranked higher than retrospective designs. Since follow-up studies are conducted in the future, some epidemiologists view all follow-up studies as inherently prospective. In addition, studies are frequently given the arbitrary label “prospective” to create a more significant impression.</p> <p>The accuracy and precision of data collection concerning exposures, confounding variables, outcomes, and endpoints are among the main advantages of prospective cohort studies (Wang & Kattan, 2020). The proposed prospective cohort study would track COVID-19 patients from three Chinese cities who were diagnosed in four government-designated treatment centres, had a positive COVID-19 test recorded in the Health Information System, and either survived or died between January 2020 and March 2020.</p> <p>Telephone interviews should be conducted on days 7, 21, 42, and 67 to identify any lingering symptoms experienced by the patients. Every adult patient aged 18 years or older with a confirmed COVID-19 infection would be enrolled in the study, which would run until the end of March 2020. Patients unable to complete the questionnaires or lost to follow-up should be excluded. When confirming every COVID-19 case in China among the participants, real-time polymerase chain reaction (RT-PCR) should be employed.</p> <h2>Significance of Comorbidity Data in Interpreting Patient Outcomes</h2> <p>Any distinct clinical entity that arises during the development of another disease or condition or coexists with it is referred to as a comorbidity. It describes the simultaneous presence of two or more different diseases, conditions, or disorders in one person. Specific disorders are more likely to occur together than in isolation because of comorbidity (Sanyaolu et al., 2020). Gathering data on comorbidities in cohort studies is primarily undertaken to compare the range of cases encountered by different clinicians while preserving a more comprehensive understanding of the complexity of co-occurring illnesses. Alternatively, diagnoses can be linked to their influence on healthcare resources for particular diseases and related comorbidities.</p> <p>The interpretation of the cohort and the study results should also be discussed based on the Chinese Guidelines and the risk factors for mortality in COVID-19. The cohort findings showed that 338 persons who met the study’s inclusion criteria were identified before March 2020. Three hundred and twenty-five individuals were included in the final analysis after 13 individuals who lacked clinical data were excluded.</p> <p>In contrast to the median period from the first symptoms to death, which was 28 days, the median duration from hospitalisation to discharge was 20 days (Gao et al., 2021). Non-survivors had a median hospital stay of 15 days, while the median duration of the infection was 30 days. Compared with survivors, non-survivors had higher organ dysfunction and inflammation indices. These results indicate that COVID-19 could last approximately 30 days and that a person could recover or die within approximately 20 and 15 days, respectively.</p> <p>In this retrospective cohort study, the alert value was suggested as a point of reference, and the factors associated with the likelihood of in-hospital mortality among critically ill COVID-19 patients at high risk of death were identified. The study findings demonstrated that increased rates of in-hospital death were associated with concurrent conditions, higher Sequential Organ Failure Assessment scores, lymphopenia, and critical classification (Gao et al., 2021).</p> <p>Compared with survivors, those who died had more severe inflammatory markers and organ damage. Approximately half of the patients had a comorbid condition, most frequently hypertension, followed by diabetes and cardiovascular diseases. These findings are notably similar to those reported among comparable patients in other studies.</p> <h2>Comparison of Observational and Experimental Research Approaches</h2> <p>Researchers can observe, record, and collect data through observational studies without actively intervening. These studies are useful when manipulating variables is impractical or unethical because they provide a means of investigating real-life situations and patterns (Rolfe, 2020). Observational studies provide insight into existing conditions and relationships through surveys and careful observations.</p> <p>Experimental studies, on the other hand, provide researchers with greater control. They involve manipulating variables to determine their influence on specific outcomes (Santos Research Center, 2023). Experimental studies directly address questions of causality by establishing causal relationships through controlled conditions.</p> <p>The findings of this retrospective study are conclusive regarding associations between COVID-19 and mortality-related factors, although the observational design limits definitive causal conclusions. The study classified the patients as critical or severe depending on their clinical condition. It also concluded that COVID-19 patients with a comorbidity, lymphopenia, a high Sequential Organ Failure Assessment score, and critical classification were at increased risk of death.</p> <h2>Academic Sources Supporting the Research Evaluation</h2> <p>Dziadkowiec, O., Durbin, J., Muralidharan, V. J., Novak, M., & Cornett, B. (2020). Improving the quality and design of retrospective clinical outcome studies that utilize electronic health records. HCA Healthcare Journal of Medicine, 1(3), 131.</p> <p>Frost, J. (2023, November 29). Retrospective study: Definition & examples. Statistics By Jim. https://statisticsbyjim.com/basics/retrospective-study/#:~:text=A%20retrospective%20study%20uses%20data%20measured%20for%20other,not%20be%20measured%2C%20leading%20to%20confounding.%20Recall%20bias.</p> <p>Gao, J., Zhong, L., Wu, M., Ji, J., Liu, Z., Wang, C., ... & Liu, Z. (2021). Risk factors for mortality in critically ill patients with COVID-19: A multicenter retrospective case-control study. BMC Infectious Diseases, 21(1), 1–8.</p> <p>Rolfe, S. A. (2020). Direct observation. In Doing early childhood research (pp. 224–239). Routledge.</p> <p>Santos Research Center. (2023, December 6). Observational vs. experimental study: A comprehensive guide. Santos Research Center Corp. https://www.santosresearch.com/observational-vs-experimental-study/</p> <p>Sanyaolu, A., Okorie, C., Marinkovic, A., Patidar, R., Younis, K., Desai, P., ... & Altaf, M. (2020). Comorbidity and its impact on patients with COVID-19. SN Comprehensive Clinical Medicine, 2, 1069–1076.</p> <p>Wang, X., & Kattan, M. W. (2020). Cohort studies: Design, analysis, and reporting. Chest, 158(1), S72–S78.</p>