Estimation of Clinical Comorbidities in COVID-19 Patients: A Systematic Review and Meta-analysis

Background: The novel coronavirus or COVID-19 has become a catastrophic global crisis due to its accelerated upsurge in cases. Most hospital admissions and deaths are attributed to the underlying clinical conditions of the infected person. Objective: This study aims to estimate the proportion of comorbidities in patients with confirmed COVID-19 serological diagnosis. Methods: A systematic review was performed utilizing PubMed, EBSCO, and CINAHL databases prior to March 22, 2020. Results: A total of 2028 COVID-19 patients were analyzed from the 11 studies. Nearly 33% of the patients had some form of clinical comorbidities. Hypertension (15%, 95% CI, 9-21%), diabetes mellitus (11%, 95% CI, 8-14%), cardiovascular disease (6%, 95% CI: 3-8%), cerebrovascular disease (3%, 95% CI: 1-4%), chronic lung disease (2%, 95% CI: 1-3) were the most prevalent comorbid conditions. The studies had significant heterogeneity between 49% to 93%. Conclusion: Higher proportions of comorbidities in COVID-19 patients strengthen the conclusion that individuals with underlying chronic comorbidities could have a greater susceptibility to COVID-19 infections. A potential link between comorbid conditions and COVID-19 infection could protect vulnerable groups as well as help with risk assessment.


Introduction
A novel coronavirus referred to as coronavirus disease 2019 (COVID- 19), originated from Wuhan, China in December 2019, and the World Health Organization (WHO) declared it as a public health emergency on January 30, 2020 [1]. On March 11, 2020, it was declared a global pandemic owing to the massive surge in cases [2]. This is not the first time the world has suffered infectious disease outbreaks of this potential. However, an unrestricted movement of people and the highly contagious nature of this virus has facilitated its rapid spread across the globe, making it challenging to manage [3,4].
As of April 23, 2020 a total of 2,544,792 confirmed COVID-19 cases were recorded worldwide with 1,75,694 deaths [5]. Many of the infections are in the aging population; people aged 50 years or above contributing 35-45% of the total infections [6]. Furthermore, the case fatality rate is also highest amongst this age group. Most of these deaths are attributed to the pre-existing comorbid conditions [6][7][8].

Study selection
Study selection was performed by YA and AS independently. The studies were then assessed for final inclusion, followed by sorting of the duplicates, title and abstract screening, and full-text screening (Figure 1). Studies published as original articles, either observational or retrospective cohort in peer-reviewed journals from December 1, 2019 to March 20, 2020 were included in the study. Studies such as case studies or case series, editorial, review, or commentary and studies which did not have enough information on clinical morbidities were excluded. Similarly, studies conducted in the pediatric population and pregnant women were also excluded.

Data extraction
A data extraction form was developed in MS-excel, and

Methods
This systematic review was conducted following the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) guidelines [10]. The primary aim of the study is to report the associated clinical comorbidities in patients with confirmed COVID-19 serological diagnosis and estimate their pooled proportions.

Data sources and search strategy
We limited our search to three online databases: PubMed, EMBASE, and CINAHL. The last search was done on March 20, 2020. A systematic search strategy was employed without restrictions on language. Starting with a disease of interest, the search keywords included 'coronavirus disease 2019, SARS-CoV-2, 2019ncov, novel coronavirus 2019, Covid19, clinical characteristics, clinical outcomes, clinical comorbidity, comorbidity, clinical comorbidities, comorbidities, and chronic comorbidities. To avoid missing any relevant studies, we also

Quality assessment
Quality of the included studies was assessed using the standard checklist produced by the US National Health Lung and Blood Institute for observational cohort and cross-sectional studies [11]. The focus of the quality assessment was based on the objective clarity, population definition, inclusion and exclusion criteria, clearly defined intervention, appropriate use of statistical methods and the outcome of interest, well-description of results, and consideration of the potential confounding variables and statistical adjustment for their impact. The outcome of the quality was assessed into four categories: Yes, No, NA (not applicable to the study) or NR (not reported in the study) (Supplementary Table 1).

Data analysis and statistical methods
We analyzed pooled estimates of the total clinical comorbidities and each specific morbidity where applicable. The pooled estimates were determined using a random effect model (DerSimonian-Laird). The pooled proportion below one percent was not presented in the table. Any inconsistencies in the study were assessed using the chi-square test for heterogeneity (I 2 ). Higher the value of I 2 , the higher the inconsistency in the included studies. All statistical analyses and calculations were conducted using Minitab © , while the meta-analysis was performed in 'Open Meta Analyst © .' Data were presented in percentages, proportions, range, and confidence intervals as required. A p-value < 0.05 was considered statistically significant.

Study selection
A total of 11 studies were included for the evidence synthesis out of the 145 extracted from the databases (Figure 1). During the title/abstract screening, 82 studies were excluded resulting in 54 studies for full-text screening. Of these, 37 were excluded based on our exclusion criteria; case studies/ case series (10), studies in children and pregnant women (n = 6), studies without adequate data on comorbidities (n = 8), and studies either review, editorial, or commentary (n = 13). Table 1 [12][13][14][15][16][17][18][19][20][21][22] describes the characteristic of the studies included in the study. A total of 2028 COVID-19 patients were included from 11 studies with the mean age of 50 years (range: 39 to 60 years). More than half (59%) of the patients were males. Similarly, more than half of the patients needed hospitalization. The overall case fatality rate (CFR) was 8.14% among all the patients.

Chronic clinical co-morbidities
Out of 11 included studies, 10 studies reported Hypertension (HTN) and diabetes mellitus (DM), 9 cardiovascular disease (CVD), 8 each chronic lung diseases and malignancy, 5 each cerebrovascular diseases and chronic kidney diseases, and 4 chronic liver disease (Figure 2).
Comorbidities can be seen at any age. However, it incy and chronic liver disease; however, the pooled proportion was not calculated due to the small sample size. There was significant heterogeneity amongst the studies (48% to 93%).

Discussion
This systematic review and meta-analysis estimates the proportion of overall and specific morbidities in COVID-19 patients with a higher number of studies and patients pop-ulation (11 studies, 2028 patients) than previously published reviews [9,23].
At present, there is no definitive treatment of COVID-19. The associated clinical comorbidities further complicate this absence of therapeutic alternatives in COVID-19. Our me-ta-analysis reported that 33% of the patient infected with COVID-19 had some forms of clinical comorbidities. Among  tious diseases reported similar comorbidities. For example, 51% of the MERS patients had diabetes, 48% hypertension, and 30% cardiac diseases [32]. Similarly, of all the comorbidities in SARS patients, 16% had diabetes, and 12% had cardiac disease [33]. These identified clinical comorbidities are proven to be associated with the deprived prognosis of the disease [8]. Our findings suggest that, like other severe acute respiratory outbreak comorbidities such as hypertension, diabetes, and CVD may predispose the adverse clinical outcome of COVID-19 patients in terms of disease severity and death [34]. Hence, the management of existing morbidities should be part of the COVID-19 treatment plan for a better outcome.

Study Limitations
We did not perform a comparative analysis to explore the association between comorbidities and disease severity or death, as many of the studies included were case series without an internal control group. Furthermore, the study was not able to stratify the proportion of comorbidities according to patient's characteristics such as age and gender, which would have provided valuable information on target populations for the intervention. Sex disaggregated pattern has been suggested in COVID-19, however, our study did not analyze the role of comorbidities in disease severity and deaths between males and females. Besides, our study has significant heterogeneity in study design and sample size, which needs to be considered while interpreting the results.