Clinical Outcomes of COVID-19 Cases and Inuencing Factors of Severe Cases in Qingdao City: A Retrospective Cohort Study

Background To analyze the clinical outcomes of COVID-2019 cases and the inuencing factors of severe cases in Qingdao City and provide theoretical reference basis for optimizing medical treatment and the strategies of epidemic prevention and control. Methods The demographical, epidemiological, clinical data of 81 conrmed COVID-2019 cases in Qingdao City were collected via epidemiological investigation and clinical process tracking. The status of cure, discharge, clinical outcome and inuencing factors were analyzed in our study. Results

The typical symptoms were fever, sore throat, fatigue, cough or dyspnea, and so on [5][6][7]. The transmission of COVID-19 is potent. As of September 27, 2020, the outbreak of COVID-19 has spread to 235 countries, areas or territories, generated 32,730,945 con rmed cases, including 991,224 deaths worldwide, with rapid daily increases in some countries.
Since December 2019, some cases had been linked to Huanan market, a seafood and wild animal market, suggesting animal-to-human transmission [8,9]. At the end of 2019, the WHO was informed about an outbreak of pneumonia of unknown etiology in Wuhan, Hubei Province, China. Current evidence indicated that SARS-CoV-2 spread by human-to-human. It was mainly transmitted by symptomatic cases [10][11][12]. China had used two strategies, containment and suppression, to promote prevention and control [13]. The strategies effort has reduced morbidity and case-fatality ratio of COVID-19. As of September 27, 2020, a total of 85,372 cases of the COVID-19 were reported in 31 provinces (districts and cities) and Xinjiang Production and Construction Corps in China. Especially recently, the number of con rmed cases nationwide was few, and most of them were imported cases.
On 21 January 2020, the rst COVID-19 case of Qingdao was con rmed. Although we control COVID-19 transmission effectively in the domestic, we are facing additional waves from imported cases. To be able to prevent future outbreaks, we studied the factors in uencing the outcome of COVID-19 con rmed cases, based on the previous epidemic situation in Qingdao, Shandong Province, China.

Data Collection
The data of COVID-19 con rmed cases was obtained from the National Noti able Disease Surveillance System. The demographics and epidemiological information (time of onset, admission, hospitalization; History of underlying diseases; clinical outcome, and so on) was included in the investigation report. The epidemiological questionnaire was collected onto standardized forms through interviews of cases. All epidemiologic information was collected during eld investigations.

Case and related de nitions
The criteria was as follows. A con rmed case was de ned by positive respiratory specimens and clinical symptoms. The respiratory specimens was tested by real-time reverse-transcription-polymerase-chainreaction (RT-PCR) assay for SARS-CoV-2 or a genetic sequence that matches SARS-CoV-2. Clinical classi cation was based on the most serious clinical manifestations, which was divided into mild, ordinary, severe and critical. Among this clinical classi cation, mild and ordinary were de ned as mild cases; severe, critical and fetal were de ned as severe cases.
The duration of illness was from onset of symptoms to hospital discharge. The length of hospitalization was from hospital admission to discharge. The cure rate was de ned as the probability of being cured from onset of symptoms to hospital discharge on the n days of onset. The hospital discharge rate was de ned as the probability of being cured from hospital admission to discharge on the n days of onset.

In uencing factors assignment rule
The assignment rule was as follows. Gender: Gender was controlled by women. Age: 0 to 39 years old was assigned as 1, above 40 years old was assigned as 2. Occupation: Retired and unemployment was assigned as 1, students and children was assigned as 2, in-service staff and the others were assigned as 3 and 4. Time interval between onset and hospital admission: On the day from onset to hospital admission was assigned as 0. 1-2d, 3-4d, and more than 5d from onset to hospital admission was assigned as 1, 2, 3, respectively. Imported cases from abroad, cluster cases, chest computed tomography (CT), body temperature, and close contacts were dichotomous variables.

Laboratory Testing
Upper or lower respiratory tract specimens were sampled from patients. The samples were divided and tested in biosafety level 2 facilities. Real-time RT-PCR with SARS-CoV-2 -speci c primers and probes was used to test RNA which was extracted from the samples. The laboratory-con rmed case was diagnosed based on two targets (open reading frame 1a or 1b, nucleocap-sid protein) positive test by speci c realtime RT-PCR. If a cycle threshold value (Ct-value) was less than 37, the result was positive, and a Ct-value of 40 or more was de ned as negative. If the Ct-value was 37 to less than 40, retested was required. A repeated Ct-value, tested less than 40 and observed an obvious peak, was de ned as a positive result.

Statistical analysis
We carried out descriptive analysis for the categories. Continuous variables were reported as means with standard deviations or medians with ranges. The proportion was calculated for categorical variable. Pearson's chi-square test or Fisher's exact test was used to comparison analysis. Logistic regression analysis was used to analyze the in uencing factors of severe cases. A P-value<0.05 was considered to indicate statistical signi cance. All analysis was performed with SAS software (SAS version 9.4) for Windows.

Demographical epidemiologic and clinical characteristics
As the rst con rmed case of COVID-19 in Qingdao was reported on Jan 22, 2020, the cumulative number of con rmed cases gradually increased (Fig 1). A total of 81 con rmed cases had been reported in Qingdao City until to May 2, 2020. Among these cases, 80 cases had been discharged from hospitals and 1 had been deceased.
Of 81 con rmed cases, 38 were males and 43 were females, with an average male-to-female sex ratio 0.88:1. The median age of cases was 42 years (range, 1 to 90). 82.72% of the cases were under 60 years old. A total of 24.69% of the cases had one or more coexisting underlying medical conditions (20 cases, including 7 with hypertension, 4 with gallstones, 4 with diabetes, 3 with coronary heart disease and 2 with other diseases). The majority of cases were retired and unemployment, which accounted for 33.33%, following by in-service staffs (24.68%) and students and children (23.46%).

Status of cure, discharge and clinical outcome
The median time from onset of symptoms to hospital admission were 3.67 days (IQR, 1.75 to 6.71). The median duration of illness were 21.00 days (IQR, 16.00 to 26.00) and the median length of hospitalization were 15.63 days (IQR, 11.60 to 20.50), mild and severe cases of which were showed in Table 2, respectively. The cure rate of mild cases within 20, 30, and 40 days were respectively 50.75%, 85.07%, and 98.51%, while 42.86%, 64.29% and 85.71% for severe cases. The difference of the cure rate between mild and severe cases was statistically signi cant by chi-square test (P<0.05). The hospital discharge rate of mild cases within 20, 30, and 40 days were respectively 77.61%, 97.01% and 98.51%, while 57.14%, 85.71%, and 100.00% for severe cases. The difference of the hospital discharge rate between mild and severe cases was statistically signi cant by chi-square test (P<0.05). Fig. 2.
At admission, 12(14.81%) and 55(67.90%) were mild and ordinary, 9(11.11%) and 4(4.94%) were severe and critical among the 81 cases, and 1 critical cases (1.23%) developed into fatal. The median time for progression to severe cases was 6.00 days after onset (IQR, 5.00-10.00). The median duration of severe cases was 8.00 days (IQR, 6.25-14.00) . Fig 3. Table 2 summarized the results of univariate analysis. We evaluated the effect of each factor on severe cases by chi-square test or Fisher's exact test. Age, occupation, rst chest CT and body temperature were statistically signi cant.

Multivariate analysis of the in uencing factors
Multivariate Logistic Regression was performed with the signi cant factors detected by univariate analysis. Table 3 showed that age older than 40 years old (OR=5.797, 95%CI: 1.064~31.568) and rst chest CT abnormal (OR=0.1140, 95%CI: 0.014~0.923) were the in uencing factors of COVID-19 severe cases, suggesting that older age and rst chest CT normal would be more prone to develop to severe cases of COVID-19.
Sensitivity analysis was conducted to ensure the reliability of the results. A general liner model (GLM) was performed to analyze the association between four signi cant factors selected by univariate analysis and severe cases. The results were largely consistent, showing older age and rst chest CT normal were positively correlated with severe cases (Table 4).

Discussion
This study retrospectively analyzed the demographical, epidemiological, clinical data in a cohort of COVID-19 cases in Qingdao City, Shandong Province. In this study, the median age of cases was 42 years, ranging from 1 to 90 years, with the majority (53.09%) being female, which was different from other studies [14,15]. These differences may be related to exposure person, 43.21% (35/81) of which were close contacts and most belonging to attendants in our study. Unlike the SARS with hyperpyrexia and dyspnea, many studies including our results, had showed that the most common clinical symptoms were fever, cough, fatigue, and myalgia, which was suggestive of common cold [16]. It had been reported that SARS-CoV-2 appeared to replicate e ciently in the upper respiratory tract and might cause less abrupt onset of symptoms [17]. However, Kim et al had found that fever was absent in 75.00% of cases at the time of admission and they did not timely realize that they had been infected with SARS-CoV-2, which could result in the virus spreading inadvertently [18].
Compared to SARS and MERS, the fatality rate of COVID-19 was greatly lower than SARS and MERS (10.90% and 34.40%, respectively) [19][20][21]. The fatality rate of COVID-19 in Qingdao City was 1.23%, which was lower than that in Hubei and Wuhan in the early phase [22]. The emergence of this situation might be associated with the rapid and effective response of epidemic prevention and control in Qingdao City. 'Four early' measures, namely early detection, early isolation, early diagnosis and early treatment, had vastly reduced the occurrence of cases and deaths.
Studies had found that data for COVID-19 available to date revealed a wide clinical spectrum of disease consisting of mild, ordinary, severe and critical disease. The severe rate was 17.28% in Qingdao City, appreciably lower than Guangdong Province (16.40%). Wen et al found that the proportion of severe cases with older age and underlying diseases was much higher in Beijing [23]. Likewise, 78.57% (11/14) were over 50 years old and 42.86%(6/14) had underlying disease of the 14 severe case in our study. Chen et al also showed that older adult males with chronic comorbidities was highly susceptible to SARS-CoV-2 as a result of the weaker immune functions of these patients [24].In addition, some studies suggested that a substantial leucopentia and lymphopenia might cause damage to T lymphocytes leading to exacerbation of cases [25,26].
This study provided a comprehensive description of the three intervals associated with the clinical course in COVID-19 cases in Qingdao City. The average time of onset of symptoms to hospital admission were 3.67 days, which was slightly shorter than SARS (3.8 days) [19]. The average time of onset of symptoms to hospital discharge (the duration of illness, 21.00 days) was in line with other studies in MERS, but shorter than SARS (23 days) [20,21]. Zhang et al showed that the duration of illness (23 days) and the length of hospitalization (20 days) in COVID-19 cases in Guangdong Province were all longer than our results [27].The discrepancies between these studies might attribute to the physical and clinical condition of individual.
Multivariate analysis revealed that older age and rst chest CT abnormal were the in uencing factors of COVID-19 severe cases. This result provided some references for early identi cation of severe risk and reduction of severe and fatality rate. Older age had been prone to develop to poor clinical outcome [28].
The severity of COVID-19 was positively correlated with age [29]. The reason may be that, as the age got older, T-cell decreased in numbers and functions of controlling viral replication and host in ammatory response which made more di cult for the host cell immunity to eradicate the invasive pathogen [30]. The rst chest CT abnormal was the protective factors for severe cases. This might be because su cient attention were paid by doctors and patients due to abnormal CT and more targeted treatment and nursing were given, which reduced the likelihood of developing severe disease. YH H et al had pointed out that some patients presented normally in CT and the appearance of pulmonary in ltrates may be delayed and the absence of pulmonary changes on initial imaging did not mean that pneumonia will not develop [31].
Many studies had found that 79.4% patients were in ltrated bilaterally from chest CT, which was similar with our results [32,33].
Our ndings further veri ed the results that older age and rst chest CT normal would be more prone to develop to severe cases of COVID-19. Therefore, age and CT should be the focus of diagnosis and treatment of COVID-19 cases. In conclusion, the course and clinical severity of con rmed COVID-19 cases were under control in Qingdao City, which thanked to the timeliness and effectiveness of prevention, control and treatment measures. During the epidemic period, it was necessary to classify and manage cases according to the needs of prevention and control in order to ensure the rational allocation of medical resources.
There were several limitations of this study. First, our sample size is relatively small in comparison with other studies. There were 81 con rmed COVID-19 cases in Qingdao City during the study period. Second, only con rmed cases were analyzed and asymptomatic infected persons were not included in our study.
Third, only the clinical outcome of the con rmed cases was analyzed. At present, each con rmed case had been admitted to hospital and isolated for treatment in Qingdao, regardless of clinical classi cation.
Hence, conclusion extrapolation should be cautious.

Conclusions
Older age and rst chest CT normal would be more prone to develop to severe cases of COVID-2019.
During the epidemic period, it was necessary to classify and manage cases according to the needs of prevention and control in order to ensure the rational allocation of medical resources.

Declarations Funding
This work was nancially supported by The study on epidemiological characteristics and Transmission mechanism of COVID-19 in Shandong Province (grant number: 2020SFXGFY02-1).

Availability of data and materials
The datasets used and/or analyzed during the current study are available on request to the corresponding author.
Ethics approval and consent to participate None.

Consent for publication
Yes.

Authors' contributions
Jing Jia, Xiaoqi Dai and Xiaolin Jiang are co-authors of this study. These authores contributed equally to this work.