An Evaluation of Non-Communicable Diseases and Risk Factors Associated with COVID-19 Disease Severity in Dubai, United Arab Emirates: An Observational Retrospective Study

The aim of this study was to determine the prevalence of non-communicable diseases and their correlation with COVID-19 disease severity among patients infected in Dubai. Clinical and demographic data were extracted from hospital records of 34,687 COVID-19 patients who visited or were admitted into Dubai hospitals between 28 January 2020 and 30 September 2020. Prevalence of co-morbidities in COVID-19 patients were assessed. The main risk factors associated with COVID-19 disease severity were also identified using three regression models. All co-morbidities were significantly associated with COVID-19 severity in the bivariate analysis (p-value ≤ 0.05) except for vitamin-D deficiency and chronic lower respiratory diseases. Patients with ischemic heart diseases (AOR: 2.08; 95% CI: 1.37, 3.15), pulmonary and other heart diseases (AOR: 2.13; 95% CI: 1.36, 3.32) and chronic kidney diseases (AOR: 1.81; 95% CI: 1.01, 3.25) had higher odds of severe COVID-19 symptoms. Suffering from multiple co-morbidities increased the odds of developing severe COVID-19 symptoms substantially in comparison to having only one co-morbidity i.e., (AOR: 1.52; 95% CI 1.76–2.60) to (AOR: 2.33; 95% CI: 1.37, 3.97). Identifying these risk factors could assist in the early recognition of high-risk populations and ensure the most appropriate preventive measures and required medical management during the pandemic.


Introduction
Since its first detection in Wuhan, China in December 2019, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), later known as coronavirus disease 2019 (COVID-19) has spread worldwide to over 223 countries and territories including the United Arab Emirates (UAE) [1]. In the UAE there have been over 892,000 confirmed COVID-19 cases and 2302 total deaths as of 3rd April 2022 [1]. Research during the early phases of the pandemic indicates that demographic and behavioral factors such as sex, age and smoking may impact the severity of infections with COVID-19 [2][3][4][5][6].

Study Design and Setting
We conducted a retrospective observational study using confirmed real-time polymerase chain reaction (RT-PCR) positive COVID-19 cases in the Emirate of Dubai, UAE. Data were extracted from the DHA health information system section. DHA is the governmental health authority responsible for providing medical services to the population of Dubai and governs four hospitals comprising approximately 1500 beds for adults and one 200 bed hospital for children. The EPIC electronic health record system is utilized throughout DHA facilities to document all medical information related to the patient's hospital admission including patient information, symptoms, treatments, medical history and all relevant clinical data.

Participants and Study Size
Data from a total of 57,292 patients of all ages and both sexes that were clinically diagnosed with COVID-19 and had tested positive in PCR laboratories for hospitalization, or were in isolation in Dubai DHA government hospitals, or quarantine centers under DHA between 28th January 2020 to 30th September 2020 were extracted from the DHA health information system. Inclusion criteria included, (i) adults aged 20 years and above, (ii) COVID-19 positive test results that met the case definition, and (iii) visited a DHA facility during the study period. A total of 34,687 patient data that fulfilled these criteria were included in the bivariate analysis. A detailed description and breakdown of included, missing, and repeated cases are presented in Figure 1.

Defining Dataset Variables, Co-Morbidities, and Disease Severity
The dataset included: demographic variables (age, sex, residence, and smoking status); Body Mass Index (BMI) categorized as: underweight (<18.5), normal (18.5-24.9), overweight (25.0-29.9), and obese (>29.9) [24]; and pre-existing co-morbidities (grouped into 14 broad categories as per the International Classification of Diseases, Tenth Revision (ICD-10) (Table A1). Vitamin-D deficiency was classified as a serum 25-hydroxyvitamin D (25OHD) value less than 30 nmol/L as defined by the Institute of Medicine [25]. There were very few cases of tuberculosis and viral hepatitis, so these were excluded from further statistical analyses.

Defining Dataset Variables, Co-Morbidities, and Disease Severity
The dataset included: demographic variables (age, sex, residence, and smoking status); Body Mass Index (BMI) categorized as: underweight (<18.5), normal (18.5-24.9), overweight (25.0-29.9), and obese (>29.9) [24]; and pre-existing co-morbidities (grouped into 14 broad categories as per the International Classification of Diseases, Tenth Revision (ICD-10) (Table A1). Vitamin-D deficiency was classified as a serum 25-hydroxyvitamin D (25OHD) value less than 30 nmol/L as defined by the Institute of Medicine [25]. There were very few cases of tuberculosis and viral hepatitis, so these were excluded from further statistical analyses.
COVID-19 severity was classified into four categories following the WHO ISARIC classification [26] as; (i) Mild: symptomatic patients meeting the case definition for COVID-19 without evidence of viral pneumonia or hypoxia; (ii) Moderate: A Patients with clinical signs of pneumonia (fever, cough, dyspnea, fast breathing) but no signs of severe pneumonia, including SpO2 ≥ 90% in room air; (iii) Severe: A Patients with clinical signs of pneumonia (fever, cough, dyspnea, fast breathing) plus one of the following: respiratory rate > 30 breaths/min, severe respiratory distress, or SpO2 < 90% on room air); and COVID-19 severity was classified into four categories following the WHO ISARIC classification [26] as; (i) Mild: symptomatic patients meeting the case definition for COVID-19 without evidence of viral pneumonia or hypoxia; (ii) Moderate: A Patients with clinical signs of pneumonia (fever, cough, dyspnea, fast breathing) but no signs of severe pneumonia, including SpO2 ≥ 90% in room air; (iii) Severe: A Patients with clinical signs of pneumonia (fever, cough, dyspnea, fast breathing) plus one of the following: respiratory rate > 30 breaths/min, severe respiratory distress, or SpO2 < 90% on room air); and (iv) Critical: Individuals who have respiratory failure, septic shock, and/or multiple organ dysfunction.

Statistical Analyses
Data were analyzed using Stata (16) [27]. Descriptive data are displayed in counts and percentages. Pearson Chi-Square (χ 2 ) analysis identified associations between co-morbidities and COVID-19 severity. Only those comorbidities that showed significant association in Chi-square analysis were further included in the binomial logistic regression analysis. COVID-19 severity was recoded into two groups for the regression model with mild and moderate severity classified as non-severe and severe and critical classified as severe. Three models were used for the multivariable logistic regression to identify predictors of COVID-19 severity. The first model controlled for comorbidity, the second model controlled for both comorbidity and sociodemographic characteristics, while the third model controlled for comorbidity, sociodemographic characteristics, and number of co-morbidities. The results are presented as adjusted odds ratios (aORs) with 95% confidence intervals (CI) after adjustment for the effects of potential confounders. A two-tailed p < 0.05 was considered statistically significant.

Ethical Approval
The study was approved by the DHA research ethics committee (DSREC-07/2020_18) on 22 July 2020 and the Emirates Institutional Review Board for COVID-19 Research DOH/CVDC/2020/1576 on 9 August 2020. Both committees waived the need for obtaining patient consent.

Results
The majority of COVID-19 patients were male (80%), aged 30-39 years (37%), and expatriates (90%) ( Table 1). Almost two thirds of COVID-19 patients were either overweight (41%) or obese (25%) while 3% were current smokers and 1% former smokers. Most patients had mild COVID-19 severity (92%) followed by moderate (5%), and only 2% of patients were classified as severe and admitted to ICU (Table 1). A higher percentage of elderly, i.e., aged 60 years and above (7.2%) and male patients (2.1%) were admitted into the ICU ( Figure 2). Also, patients who did not maintain normal weight (i.e., underweight (23%), obese (13.4%), or overweight (16.2%)) were more likely to be admitted into the ICU. A higher percentage of elderly, i.e., aged 60 years and above (7.2%) and male patients (2.1%) were admitted into the ICU ( Figure 2). Also, patients who did not maintain normal weight (i.e., underweight (23%), obese (13.4%), or overweight (16.2%)) were more likely to be admitted into the ICU. The most prevalent pre-existing comorbidities among COVID-19 patients were diabetes (5.27%) and hypertensive disorders (4.40%) followed by vitamin-D deficiency (2.78%), nervous system diseases (1.2%) and thyroid gland disorders (1.07%). Less than 1% of patients had other co-morbidities as shown in Figure 3. The most prevalent pre-existing comorbidities among COVID-19 patients were diabetes (5.27%) and hypertensive disorders (4.40%) followed by vitamin-D deficiency (2.78%), nervous system diseases (1.2%) and thyroid gland disorders (1.07%). Less than 1% of patients had other co-morbidities as shown in Figure 3. A significantly higher percentage of patients who had pulmonary and other heart diseases (27.64%), chronic kidney disease (27.08%), cerebrovascular diseases (25.95%), ischemic heart diseases (19.57%), and hypertensive disorders (7.87%) were reported as being critical COVID-19 cases (Table 2). There was significant association found between the A significantly higher percentage of patients who had pulmonary and other heart diseases (27.64%), chronic kidney disease (27.08%), cerebrovascular diseases (25.95%), ischemic heart diseases (19.57%), and hypertensive disorders (7.87%) were reported as being critical COVID-19 cases (Table 2). There was significant association found between the number of existing co-morbidities and COVID-19 severity with patients reporting multiple co-morbidities having worse (critical) severity (11.46%). The Chi-square test for association between comorbidities and COVID-19 severity for the 4220 patients included in the multivariable regression analysis are displayed in Table A2. Table 3 presents results from the multivariable logistic regression analysis. Among the three regression models, the Pseudo R2 values according to (McFadden estimates) were 0.062 for model one, 0.087 for model two, and 0.090 for model three, respectively, indicating that model three provided the best goodness of fit to the data.

Discussion
Almost two years have passed since the COVID-19 outbreak was declared a pandemic [28], and research continues to advance knowledge of disease prognosis, treatment and the long-term effects. Although COVID-19 treatment protocols have been proposed since the early stages of the pandemic [29][30][31][32], none have been widely adopted. In Dubai, hospitals are accredited by the Joint Commission International (JCI) accreditation body and JCI patient standards of care were maintained throughout the pandemic. However, in the absence of specific treatment protocols and the emergence of new variants, the identification of high-risk patients prone to clinical deterioration is critical for ensuring adequate and timely access to intensive care treatment and resource allocation.
The results from this study demonstrate that although most patients who were admitted to DHA facilities had mild COVID-19 severity, a significant number had moderateto-critical health outcomes and required ICU admission. These results are consistent with prior reports concerning COVID-19 infections in China [33], which reported severe (14%) and critical (5%) symptoms and more recently in Abu Dhabi, UAE with similar reports of severe (1.8%) and critical (0.5%) outcomes [34]. COVID-19 infection was also observed to be highest among men and adults aged 20-50 years. Researchers from Abu Dhabi (UAE) who reported similar findings in 2021, indicated that one of the factors supporting these findings is that most expatriate men are of working age (20-50 years). This population is largely employed in high-risk occupations such as construction, hospitality, domestic work, and retail, and they live in shared accommodation where infections are quite likely to occur [35,36].
Compared with younger adults, older patients aged 60 years and over were more likely to experience severe COVID-19 symptoms. Several studies have shown similar results [6,37]. Interestingly, this association has also been observed in previous respiratory pandemics such as influenza (H7N9) [38,39]. Ageing as a risk factor for COVID-19 severity is particularly important given the rapid increase in the number and proportion of older people globally. The United Nations estimates that by 2050, one in six people worldwide will be 65 years of age or older, compared to one in eleven in 2019 [40]. Additionally, older patients with infections often present with atypical symptoms, making the diagnosis and treatment challenging [41]. With a median age of 30.3 years reported for Dubai [18], the UAE has a relatively young population. The disease trends observed in our study may be explained by these factors.
Another interesting finding in our study was the effect of weight on COVID-19 severity. Underweight patients had a 2.24 times higher likelihood of experiencing severe health outcomes than normal weight patients. Furthermore, obese patients had a 58% higher risk of experiencing poorer COVID-19 health outcomes. Studies from the MENA region have reported similar results including Kuwait [42], the Kingdom of Saudi Arabia (KSA) [5], and Iran [43]. In fact, the study of interactions between obesity and infectious respiratory diseases began even before the COVID-19 pandemic. Earlier research has shown that obesity can restrict ventilation and damage the immune system, both of which lead to an increased susceptibility to severe respiratory illness and adverse outcomes [44][45][46]. The exact mechanism by which obesity contributes to severe outcomes for COVID-19 is still unclear; however, one possible explanation might be the increased expression of the functional angiotensin-converting enzyme 2 (ACE2) receptor for SARS-CoV2 in obese individuals [47]. Further, obesity is associated with dysregulated lipid synthesis and clearance which results in increased pulmonary inflammation and injury [47,48]. The high prevalence of overweight and obesity in the UAE and other countries of the Gulf Cooperation Council (GCC) [20,49], emphasizes the need for future interventions and policies aimed at improving the health of underweight and obese patients.
Similar to other studies from the UAE, we also found that diabetes was the most common comorbidity among COVID-19 patients [34,50]. Higher mortality and worse disease outcomes have previously been reported in patients with diabetes [5,51]. Given the high prevalence of diabetes in the UAE in comparison to other countries [16,19], it is of critical importance to highlight the impact of COVID-19 on diabetes. Diabetes is known to have a negative impact on the innate immune response through insulin resistance and β-cell damage. Additionally, diabetes may impair the immune response by impairing macrophage and lymphocyte function, thus making patients more susceptible to disease complications and poor outcomes [52]. Based on our current observations and the literature cited previously, diabetes continues to be a major risk factor for poor prognosis among patients with COVID-19.
Although a relatively small percentage of patients in our study suffered from ischemic, pulmonary, and other heart diseases, those who did were more likely to develop severe COVID-19 outcomes, consistent with the literature [51,53]. The risk associated with COVID-19 severity among cardiac patients is relevant because cardiovascular patients are at an increased risk of cell damage due to direct viral damage, systemic inflammatory responses, destabilized coronary plaques, and aggravated hypoxia [54]. Moreover, cardiovascular patients often have increased ACE2 receptor expression, which facilitates the entry of viruses into the host body [53].
Chronic kidney disease has long been reported as a major public health problem and a significant risk factor for poor health outcomes [55]. Our finding that chronic kidney disease is a significant predictor of COVID-19 severity is consistent with previous observations in Spain and Mexico where studies found that kidney disease is significantly associated with severe COVID-19 symptoms and higher mortality compared with those without chronic kidney disorders [56,57]. Chronic kidney disease has become more prevalent over the past decade, with 13.4% of the population affected worldwide [58]. Among UAE nationals, the prevalence of kidney disease has been estimated to range from 2.8% to 4.6% [59]. These results may be underestimated due to the high prevalence of known kidney disease risk factors such as smoking and hypertension in the country [60,61]. There are several mechanisms responsible for kidney involvement during COVID-19 infection including cytokine damage (an exaggerated immune response), organ crosstalk (a mutual biological communication between distant organs mediated by signaling factors) and systemic effects (fluid expansion and hemodynamic instability) [62,63].
Our study also found that COVID-19 severity is strongly influenced by the presence of multiple comorbidities such that once these comorbidities were controlled in the model, the effects of diabetes, thyroid disorders, mental disorders, and behavioral disorders no longer appeared. These findings suggest that an increase in comorbidities is a significant predictor of COVID-19 severity, ICU admission, ventilator assistance, and even mortality-consistent with previous studies [64,65]. Additionally, multiple comorbidities are typically seen in older individuals. Therefore, aging and multiple comorbidities increase the likelihood of developing critical complications and outcomes from a novel coronavirus disease.
This study is particularly noteworthy for its comprehensive coverage of all DHA hospitals in Dubai and a large sample size of 34,687 patients collected over a period of nine months. The study provides further evidence to support the previous worldwide literature concerning the impact of age and pre-existing comorbidities on COVID-19 severity. Moreover, this data is from the first wave of the pandemic, which has two advantages. The first is that these results are not confounded by vaccination status, since vaccines were not available during 2020 [66] and secondly, these results are not confounded by serious second COVID-19 reinfection, because natural immunity held up throughout 2020. Nevertheless, one disadvantage of the first wave data is that the virus has since mutated substantially, so there can be some quantitative variability in the reported associations for the viral strains that are now circulating during 2022. However, qualitatively speaking, the associations can be reasonably expected to persist as reported. It is also important to note that the present study can only examine associations and cannot infer causality since it is retrospective. Additionally, the prevalence of NCDs such as diabetes, hypertension and obesity in the present study was lower than in previous studies conducted in the UAE. This may be due to the majority of our study population being relatively young and from Dubai, which comprises mostly expatriate males aged 20 to 40 years old. Moreover, since the comorbidity data were obtained from hospital records during admission, it is possible that some comorbidities were not detected and underreported.

Conclusions
The results of this study, in which data were extracted from a large dataset of patient medical records collected during the peak of the COVID-19 outbreak in Dubai, demonstrate that obese patients, those suffering from ischemic, pulmonary heart and kidney diseases and those afflicted with multiple co-morbidities are more likely to experience worse COVID-19 outcomes. Continually monitoring COVID-19 patients with underlying comorbidities will reduce the risk of severe outcomes. Due to the nature of the pandemic and the possibility of new variants emerging, it is imperative to improve healthcare providers' awareness of vulnerable patients including expatriates and the elderly, so that specific strategies for their management of COVID-19 can be developed. Understanding the risk factors associated with COVID-19 infection will facilitate the identification of those at high risk, prompt their referral to hospitals at the earliest onset of symptoms, and prevent COVID-19 from progressing to severe levels. Furthermore, we recommend enhancing the systematic and real-time availability of epidemiological data for future more precise longitudinal and interventional studies.  Data Availability Statement: Data used in this study were retrieved from hospital inpatient records and due to data protection policies cannot be shared at individual level.

Conflicts of Interest:
The authors declare no conflict of interest.