Risk Factors for In-hospital Mortality of Patients with COVID-19 in Iran: A Cross-sectional Study from Six Pandemic Waves

Introduction

Coronavirus disease (COVID-19), followed by acute respiratory syndrome-coronavirus 2 (SARS-CoV-2), has infected millions of people worldwide since the end of 2019 (1, 2). Initially, acute respiratory failure caused by SARS-CoV-2 and the rapid incidence of coronavirus disease (COVID-19) has emerged as a hot topic in medicine (3). However, the prevalence became widespread and mortality rate increased. Governments and health policymakers faced a serious crisis with COVID-19 death (4). To date, more than 6.5 million people have died from the severity of COVID-19-related symptoms (5). Also in Iran, since the definitive diagnosis of the first infected person on February 19, 2020 (6, 7), the number of infected cases and COVID-19 deaths peaked in six time periods (waves) from March 5, 2020 to March 6, 2022 (8).

One of the most common ways to control from the mortality rate of the disease and more effective health policies is to estimate the risk factors for death associated with the disease (9). Several risk factors for COVID-19 death are known, such as age (10, 11), gender (12, 13), respiratory problems (14, 15), diabetes (16-18), cardiovascular (19, 20), blood pressure (21, 22), kidney diseases (23), and other comorbidities (11, 24, 25). Parohan et al. (2020) analyzed 14 retrospective cohort studies and concluded that older age (>65 years), male gender, hypertension, cardiovascular diseases (CVDs), diabetes, and chronic obstructive pulmonary disease (COPD) were associated with a higher risk of death from COVID-19 (26). In another meta-analysis, 58 cohort studies were analyzed and the findings showed that older age, male gender, hospitalization in the intensive care unit (ICU), and comorbidities (older age, gender, obesity, hypertension, CVDs, cancer, and diabetes) were associated with mortality in hospitalized patients with COVID-19 (27). Also, a multi-center and cross-sectional study which was conducted in 19 public hospitals in Tehran (the capital and most populated city of Iran) reported that older age, male gender, hospitalization in the ICU or critical care unit (CCU) and underlying comorbidities (e.g. CVDs and diabetes) were associated with mortality rate in COVID-19 patients (28). Despite the importance of detection of risk factors for mortality in COVID-19 patients, there remains a paucity of evidence on COVID-19 death in different pandemic waves. The present study was performed with aim to investigate the risk and protective factors (i.e., demographics factors, hospital status, and comorbidities) associated with COVID-19 death in six pandemic waves in northeastern Iran.

Methods

This retrospective cross-sectional study was performed on the hospitalized COVID-19 patients from March 5, 2021 to March 6, 2022 in Mashhad city. Data were collected from medical care monitoring center (MCMC) database. Individuals with symptomatic COVID-19 and/or with quantitative positive reverse transcript polymerase chain reaction (qRT-PCR) SARS-CoV-2 test were included in this study. The number of patients for the first wave was 4634 patients (from March 5 to April 6, 2020), the second wave was 7647 (from June 27 to August 6, 2020), the third wave was 8429 (from October 8 to December 3, 2020), the fourth wave was 11841 (from April 1 to June 11, 2021), the fifth wave was 19307 (from June 22 to September 23, 2021), and the sixth wave was 7241 (from January 15 to March 21, 2022). Electronic records were compared between survivors and non-survivors, which survivors were discharged from hospital and non-survivors died at the hospital followed by COVID-19 infection.

Demographic data, such as age and gender were included in this study. Respiratory distress, loss of consciousness, and oxygen saturation levels (SpO₂) were also collected as hospital status parameters. Comorbidities, such as cancer, diabetes, respiratory diseases, kidney, liver, cardiovascular, blood, immunodeficiency (acquired immunodeficiency syndrome and other immunodeficiencies), neurological, and other diseases were included in this analysis.

Respiratory distress was defined as the acute onset of oedema and non-cardiogenic pulmonary hypoxemia (29). Loss of consciousness was defined as a temporary and transient period of inability to maintain posture and environmental awareness (30). SpO₂ (oxygen saturation levels) was defined as the change in intracellular oxygen pressure in millimeters of mercury (mm­Hg), and less than 93 mm­Hg was considered the critical cut-off point (31). All comorbidities and disorders were diagnosed by specialist physicians based on valid medical guidelines and medical history.

Data were analyzed by SPSS (version 22.0). The continuous variables were presented as mean ± standard deviation, while the categorical data were expressed as numbers and percentages. The variables that presented significant differences in the univariate analysis were considered for analysis using binary logistic regression to verify the association between the independent demographic data, hospital status, and comorbidities risk factors and mortality as the dependent variable. The results were expressed as a 95% confidence interval (CI) and odds ratio (OR). P<0.05 was considered statistically significant.

Results

A total of 59099 patients (27728 women) in all waves were included in this study. The mean age of participants was 54.47±22.5 years (median age of 58 years). Table 1 shows descriptive information related to deceased and living patients separately for each wave according to the studied variables.

Table 1. Demographic data, hospital status and comorbidities of patients in different waves based on being alive and dead

Figure 1. Sensitivity, specificity and accuracy of binary logistic regression

Table 2. Binary logistic regression results to investigate the role of demographic data, hospital status and comorbidities in the death of patients

N/A: Not applicable; Due to the large difference in the number of patients who survived and died in these statistics, the effect size was not significant

Before binary logistic regression analysis, sensitivity, specificity and accuracy of the model were measured by ROC curve (Figure 1) and the results showed that this model can be efficient for the purpose of this study.

The results of the binary logistic regression analysis are shown in Table 2. Accordingly in the demographic data, older age and masculinity (in the gender variable) were significant in almost all waves [the sixth wave is marginally significant] (OR>1, P<0.05). In hospital status, respiratory distress and loss of consciousness were detected as risk factors (OR>1, P<0.05), but high SpO₂ (>93) was identified as a protective factor (OR<1, P<0.05). In comorbidities, diabetes in the third and fifth waves, cancer in the first to fourth waves, respiratory diseases in the sixth wave, kidney diseases in the second wave, and immunodeficiency diseases in the first and fourth waves (marginally significant in the fourth wave) were the risk factors for death (OR>1, P<0.05). Blood pressure, neurological, cardiovascular, and liver diseases had no significant results to increased risk of death in the six waves (P>0.05).

Discussion

Finding out the risk factors associated with COVID-19 death in different sinus waves in Iran can help treat effectively and make preventive health policies for the next outbreaks. Accordingly, the findings of the present study showed that several factors including age, masculinity, respiratory distress, loss of consciousness, diabetes, cancer, respiratory, kidney and immunodeficiency diseases had a risk effect on COVID-19 death in different pandemic waves (i.e., six pandemic waves). On the other hand, high SpO₂ rate (>93) was detected as a protective factor for death followed by COVID-19 infection. Also, blood pressure, neurological disorders, cardiovascular, and liver diseases were included in the model of the present study; however, there was no significant result between these factors and COVID-19 death. As a result, the present retrospective study demonstrated that oxygen level is a main factor for reducing the mortality rate in COVID-19 patients. Furthermore, immunodeficiency was detected as a risk factor for COVID-19 death that this might be due to a prolonged viral phase in immunocompromised patients (32); therefore, managing these patients is a critical step for future viral pandemics. Among above-mentioned risk factors, policymakers and physicians should pay attention to diabetes because the prevalence of Diabetes Mellitus (DM) has risen rapidly in Iran (age-standardized prevalence: 8.3%) (33). Moreover, the prevalence of pre-diabetes was estimated about 25.8%, which is a serious warning for next pandemics. It is suggested that diabetes prevention program must be noticed as one of the main priorities for countries with the highest diabetes rates.

In accordance with the present results, previous studies have demonstrated that older age and masculinity were found as the risk factors for COVID-19 death (10-12). As an example, the risk of COVID-19 death was significantly increased by older age (over 70 years old) and masculinity as reported in Peru (Latin America), India, and Africa population (34, 35). Furthermore, a doubled mortality rate was observed for male patients and those over 90 years of age infected with COVID-19 (36, 37). Besides, in the present study, data from hospital status (e.g. consciousness, oxygen level, and respiratory status) seem to be consistent with other research which found hypoxia in COVID-19 patients over time can be exacerbated by leukocytosis and impair seriously the recovery process of patients (38). The present study confirms that SpO₂ is associated with recovery and survival. Thus, early detection and medication in terms of hypoxia are very important to prevent the worsening situation and death followed by COVID-19.

The most obvious finding emerged from the present study is that the impact of hospital status and comorbidities as risk factors on COVID-19 death were evaluated during different waves of pandemic in northeastern of Iran for the first time. The results of the present study revealed that cancer in four waves, diabetes and immunodeficiency diseases in two waves, as well as respiratory and chronic kidney diseases in one wave of pandemic were detected as a risk factor for COVID-19 death. However, other comorbidities including blood pressure, neurological, cardiovascular, and liver diseases had no significant correlation with death in COVID-19 patients. Based on data of the present study, it can be concluded that the diversity of comorbidities, mass vaccination, and increasing awareness of protecting against virus affected the behavior of disease that should be evaluated in future studies. Besides that, updated guidelines and specific care for patients with disability can improve the survival rate in COVID-19 patients. In the interpretation of the present data, the pathogenesis of SARS-CoV-2 plays important roles in different contexts. For example, the expression of the enzymes angiotensin converting enzyme 2 (ACE2) and dipeptidyl peptidase-IV (DPP4) have been changed in diabetes and the activity of these enzymes are associated with severity of disease and outcomes (39, 40). However, more studies are needed to clarify the proportion of each risk factor in the pathogenesis and clinical outcomes of COVID-19.

The limitations of the present study include the missing data and recall biases. For this reason, the results of the present study should be carefully interpreted in claiming cause-effect relationship in data.

To sum up, different factors (i.e., demographic data, hospital status, and comorbidities) were analyzed on in-hospital mortality from COVID-19 infection during several pandemic waves in Iran. Data obtained from the present study showed that some parameters, such as age, masculinity, respiratory distress, loss of consciousness, diabetes, cancer, respiratory, chronic kidney disease and immunodeficiency disorders were associated with in-hospital mortality from COVID-19 in patients. Conversely, SpO2 was detected as a protective factor for COVID-19 death in all pandemic waves in Iran.

Implications for practice

The most obvious finding emerged from this study is that national prevention program and referral system in healthcare should be redesigned based on the most important risk factors for COVID-19 death to decrease in-hospital mortality rate. It is suggested that more researches be done to determine the relationship between risk factors and viral pandemics.

Acknowledgments

This manuscript was the result of a research project with the code of 4002144 and approved by the Vice Chancellor for Research of Mashhad University of Medical Sciences. Also, this study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Mashhad University of Medical Sciences with the code of IR.MUMS.REC.1401.209. The authors would like to thank all the patients and medical staff who helped to conduct this research. We would like to thank the Clinical Research Development Unit, Imam Reza Hospital, Mashhad University of Medical Sciences, for their assistance in this manuscript.

Conflicts of interest

The authors declared no conflict of interest.