Download:
-
In December 2022, the Joint Prevention and Control Mechanism of the State Council issued the Notice on Further Optimizing the Implementation of COVID-19 Prevention and Control Measures. The notice emphasized that the virulence of the Omicron strain had decreased, with clinical observations indicating that most infections resulted in asymptomatic or mild symptoms (1). Concurrently, the government devised targeted prevention, control, and treatment strategies for specific populations (2). Following the adjustment of these policies, the number of infected individuals experienced fluctuations, reaching a peak shortly after the new measures were implemented. As of December 23, 2022, fever clinics across the nation recorded a cumulative total of 2.867 million consultations and treatments (3). Consequently, the effective allocation of medical resources and the provision of high-quality, tiered diagnosis and treatment for coronavirus disease 2019 (COVID-19) patients have emerged as significant public health concerns warranting attention from relevant departments.
Supported by the Beijing Municipal Health Commission, a cross-sectional questionnaire survey was conducted at 354 community health service centers across 16 districts in Beijing. A random sampling method was employed, which was based on the population proportion within each district. From December 26 to December 31, 2022, a total of 33,968 infected individuals participated in the study. Eligible participants met the following inclusion criteria: 1) confirmed COVID-19 diagnosis through nucleic acid or antigen tests, or reported symptoms of COVID-19; 2) aged 18 years or older; and 3) willingness to participate in the survey and provide informed consent. Those who refused participation were not included in the study.
The study obtained data on fundamental demographic characteristics, post-infection symptoms, comorbidities, treatment modalities, and healthcare pathway selection. To delineate the treatment landscape, the 33,968 patients were classified into 2 cohorts: those who underwent treatment and those who did not. Treatment types were further segregated into 3 categories: primary care facilities, hospitals, and a combination of both. R software (version 4.2.2, R Foundation for Statistical Computing, Vienna, Austria) was employed for data analysis and visualization. Count data are expressed as composition ratios or percentages. Statistical analyses were conducted to investigate the factors influencing consultation rates and treatment preferences. Moreover, variables exhibiting statistical significance in univariate analysis were incorporated into a multivariate logistic regression model to compute odds ratios (OR) and 95% confidence intervals (CI). P<0.05 was considered the threshold for statistical significance.
In this study, it was found that out of 33,968 COVID-19 patients, 81.7% (n=27,741) did not seek medical treatment. Among the 6,227 patients who did receive treatment, 74.0% (n=4,609) chose primary medical institutions, 19.7% (n=1,228) opted for hospitals, and 6.3% (n=390) utilized both primary healthcare facilities and hospitals for their treatment.
The results of the statistical analysis demonstrated that the mean time required to visit primary medical institutions was the lowest at 1.67±0.04 hours, while the mean time to visit hospitals was the highest at 3.75±0.11 hours. The overall mean time to visit both hospitals and primary healthcare facilities was 3.04±0.17 hours.
Univariate analysis demonstrated that the consultation rate of elderly patients (24.2%, 2,134) exceeded that of other age groups, which suggested a preference for treatment among older patients. Additionally, patients with higher education were more inclined (83.9%, 16,867) to select home quarantine following infection. A greater outpatient rate was observed among rural patients (20.4%, 1,845) compared to their urban counterparts, while individuals with co-infections exhibited an outpatient rate of 23.2% (3,777) (Table 1).
Variable Attendances,
n (%)Non-attendences,
n (%)χ2 P Hospitals,
n (%)Primary care facilities,
n (%)Hospitals and primary care facilities,
n (%)χ2 P Gender Male 2,018 (18.4) 8,976 (81.6) 0.004 0.95 408 (20.2) 1,475 (73.1) 135 (6.7) 1.59 0.45 Female 4,209 (18.3) 18,765 (81.7) 820 (19.5) 3,134 (74.5) 255 (6.1) Age (years) 18–59 4,093 (16.3) 21,058 (83.7) 273.60 <0.001 794 (19.4) 3,062 (74.8) 237 (5.8) 5.90 0.05 >60 2,134 (24.2) 6,683 (75.8) 434 (20.3) 1,547 (72.5) 153 (7.2) Education level Elementary school/below 377 (26.7) 1,035 (73.3) 195.67 <0.001 65 (17.2) 285 (75.6) 27 (7.2) 37.66 <0.001 Junior high school/
high school/junior College/
technical school2,621 (21.0) 9,839 (79.0) 439 (16.8) 2,036 (77.7) 146 (5.6) Undergraduate/
postgraduate or above3,229 (16.1) 16,867 (83.9) 724 (22.4) 2,288 (70.9) 217 (6.7) Employment status Service trade staff 503 (18.4) 2,231 (81.6) 2.96 0.23 109 (21.7) 369 (73.4) 25 (5.0) 10.57 0.03 Medical industry staff 1,720 (17.8) 7,962 (82.2) 314 (18.3) 1,314 (76.4) 92 (5.4) Others 4,004 (18.6) 17,548 (81.4) 805 (20.1) 2,926 (73.1) 273 (6.8) Area type Urban 4,382 (17.6) 20,521 (82.4) 33.55 <0.001 983 (22.4) 3,087 (70.5) 312 (7.1) 97.96 <0.001 Rural 1,845 (20.4) 7,220 (79.7) 245 (13.3) 1,522 (82.5) 78 (4.2) History of physical illness Chronic disease 2,450 (13.8) 15,254 (86.2) 498.00 <0.001 730 (19.3) 2,784 (73.7) 263 (7.0) 8.39 0.02 Healthy 3,777 (23.2) 12,487 (76.8) 498 (20.3) 1,825 (74.5) 127 (5.2) Table 1. Medical counselling status, and medical care route selection of 33,968 patients with different demographic characteristics infected with COVID-19 — Beijing, China, December 2022.
A multivariate binary logistic regression analysis was conducted to identify the factors influencing patients’ inclination for treatment. Seeking medical treatment was the dependent variable, and gender, age, education level, occupation, residence, and COVID-19-related symptoms were independent variables (
Supplementary Table S1 ). After adjusting for other confounding factors, the results of multivariate logistic regression analysis indicated that individuals who were infected with COVID-19 tended to seek medical care at healthcare facilities if they were elderly (OR=1.538, 95% CI: 1.428–1.657), had a low level of education (OR=0.697, 95% CI: 0.605–0.805), worked in the service industry (OR=0.894, 95% CI: 0.803–0.996), or resided in rural areas (OR=1.159, 95% CI: 1.085–1.237). Fever (OR=0.913, 95% CI: 0.844–0.989), headache (OR=0.893, 95% CI: 0.835–0.954), fatigue (OR=0.884, 95% CI: 0.825-0.948), loss of taste (OR=0.929, 95% CI: 0.869–0.993), decreased appetite (OR=0.995,95% CI: 0.929–1.065), diarrhea (OR=0.998, 95% CI: 0.922–1.081), and constipation (OR=0.919, 95% CI: 0.811–1.038) were inversely associated with seeking medical care. Coughing (OR=1.379, 95% CI: 1.267–1.502), dryness of the pharynx or sore throat (OR=1.218, 95% CI: 1.138–1.304), runny or stuffy nose (OR=1.012, 95% CI: 0.950–1.079), muscle pain (OR=1.006, 95% CI: 0.939–1.078) and joint pain (OR=1.033, 95% CI: 0.967–1.104), conjunctivitis (OR=1.125, 95% CI: 0.973–1.297), chest tightness (OR=1.602, 95% CI: 1.480–1.734), nausea or vomiting (OR=1.073, 95% CI: 0.990–1.162), difficulty breathing (OR=1.347, 95% CI: 1.190–1.522), and tachypnea (OR=1.384, 95% CI: 1.238–1.546) were positively associated with seeking medical care (Table 2). Variable cOR P 95% CI aOR P 95% CI AOR P 95% CI Gender: (ref. Male) Female 0.998 0.938 0.941−1.058 1.028 0.373 0.968−1.091 0.973 0.389 0.915−1.035 Age: (ref. 18–59 years) ≥60 years 1.643 <0.001 1.549−1.743 1.584 <0.001 1.473−1.704 1.538 <0.001 1.428−1.657 Education level: (ref. primary or below) Junior high school/high school/junior college/technical school 0.731 <0.001 0.646−0.830 0.857 0.019 0.754−0.976 0.897 0.103 0.788−1.023 Undergraduate/postgraduate or above 0.526 <0.001 0.465−0.595 0.693 <0.001 0.603−0.798 0.697 <0.001 0.605−0.805 Employment status: (ref. service trade staff) Medical industry staff 0.958 0.446 0.859−1.070 1.124 0.050 1.001−1.264 1.065 0.300 0.946−1.199 Others 1.012 0.819 0.914−1.122 0.880 0.019 0.792−0.980 0.894 0.040 0.803−0.996 Residence area: (ref. urban) Rural 1.197 <0.001 1.126−1.272 1.128 <0.001 1.057−1.203 1.159 <0.001 1.085−1.237 Fever: (ref. no) Yes 0.904 0.007 0.839−0.973 0.842 <0.001 0.779−0.910 0.913 0.025 0.844−0.989 Cough: (ref. no) Yes 1.477 <0.001 1.365−1.598 1.400 <0.001 1.287−1.525 1.379 <0.001 1.267−1.502 Dry throat/sore throat: (ref. no) Yes 1.305 <0.001 1.227−1.388 1.214 <0.001 1.135−1.300 1.218 <0.001 1.138−1.304 Stuffy/runny nose: (ref. no) Yes 1.097 0.001 1.038−1.160 0.967 0.293 0.908−1.030 1.012 0.706 0.950−1.079 Painful muscles: (ref. no) Yes 1.048 0.117 0.988−1.111 0.991 0.801 0.925−1.062 1.006 0.867 0.939−1.078 Arthralgia: (ref. no) Yes 1.114 <0.001 1.054−1.178 1.048 0.163 0.981−1.119 1.033 0.332 0.967−1.104 Headaches: (ref. no) Yes 0.968 0.244 0.916−1.023 0.841 <0.001 0.788−0.898 0.893 <0.001 0.835−0.954 Conjunctivitis: (ref. no) Yes 1.400 <0.001 1.223−1.602 1.105 0.168 0.957−1.273 1.125 0.107 0.973−1.297 Physical weakness: (ref. no) Yes 0.998 0.954 0.943−1.057 0.870 <0.001 0.812−0.933 0.884 <0.001 0.825−0.948 Chest tightness: (ref. no) Yes 1.734 <0.001 1.624−1.851 1.599 <0.001 1.478−1.730 1.602 <0.001 1.480−1.734 Decreased or absent sense of taste and smell: (ref. no) Yes 1.008 0.780 0.952−1.067 0.891 <0.001 0.834−0.951 0.929 0.031 0.869−0.993 Nausea/vomiting: (ref. no) Yes 1.209 <0.001 1.129−1.296 1.067 0.110 0.985−1.155 1.073 0.085 0.990−1.162 Poor appetite: (ref. no) Yes 1.097 0.001 1.036−1.160 1.014 0.679 0.948−1.085 0.995 0.875 0.929−1.065 Diarrhea: (ref. no) Yes 1.094 0.014 1.019−1.175 0.954 0.242 0.881−1.032 0.998 0.968 0.922−1.081 Constipation: (ref. no) Yes 1.089 0.151 0.969−1.224 0.904 0.108 0.799−1.021 0.919 0.179 0.811−1.038 Breathing difficulties: (ref. no) Yes 1.989 <0.001 1.787–2.214 1.397 <0.001 1.236−1.577 1.347 <0.001 1.190−1.522 Increased respiratory rate: (ref. no) Yes 1.798 <0.001 1.632−1.980 1.369 <0.001 1.226−1.528 1.384 <0.001 1.238−1.546 Note: cOR, crude odds ratio, which was a single factor logistic regression coefficient; aOR, odds ratio, which was a logistic regression coefficient after adjusting for demographic confounding factors and symptom confounding factors, respectively; AOR was the odds ratio adjusted for sex, age, residence, occupation, education and confounding factors such as fever, cough or sputum.
Abbreviation: CI=confidence interval.Table 2. Binomial logistic regression analysis of medical counseling status influenced by demographics and symptoms among coronavirus disease 2019 (COVID-19) patients — Beijing, China, December 2022.
Patients experiencing symptoms such as dyspnea, elevated respiratory rate, chest tightness, conjunctivitis, and nausea or vomiting were more likely to seek medical consultation. Moreover, individuals with conditions including stroke, cerebrovascular diseases, bronchitis, emphysema, asthma, pneumonia, hepatitis, chronic kidney disease, and cardiac diseases demonstrated a high consultation rate (
Supplementary Table S2 ).Individuals with pre-existing comorbidities displayed the following number of symptoms after infection: hepatitis (7.9±3.7), gastritis/gastric ulcer (7.9±3.3), immunodeficiency diseases (7.8±3.8), bronchitis/emphysema/asthma/pneumonia (7.7±3.5), and tuberculosis (7.7±3.4) (
Supplementary Table S3 ). The Kolmogorov-Smirnov (K-S) normality test revealed that the number of symptoms following COVID-19 infection did not adhere to a normal distribution (P<0.001), with a median value of 2 symptoms (Figure 1).
Figure 1.The number of chronic diseases in patients with coronavirus disease 2019 (COVID-19) — Beijing, China, December 2022.
For the analysis, the number of pre-existing conditions was categorized using a threshold of 2 symptoms (<2 and ≥2). The results indicated that patients with <2 underlying conditions exhibited 6.2±3.3 post-infection symptoms, whereas patients with ≥2 underlying conditions displayed 6.6±3.3 symptoms. Additionally, the number of symptoms among patients who visited primary care institutions, hospitals, or both was 6.7±3.4, 7.5±3.6, and 7.8±3.7, respectively (Figure 2). The heatmap for visiting institutions suggested that the median number of symptoms was 8, 9, and 10 for primary medical institutions, hospitals, and both, respectively (Figure 3).
Figure 2.Phase diagram illustrating the relationship between the number of coronavirus disease 2019 (COVID-19) symptoms, presence of pre-existing comorbidities, and various medical counseling pathways.
Note: 1=Hypertension; 2=Diabetes; 3=Dyslipidemia; 4=Cardiac disease; 5=Stroke/cerebrovascular disease; 6=Bronchitis/emphysema, asthma/pneumonia; 7=Tuberculosis; 8=Gastritis/gastric ulcer; 9=Immunodeficiency; 10=Arthritis/rheumatism/rheumatoid disease; 11=Chronic kidney disease; 12=Hepatitis; and 13=Cancer. <2 indicates that the number of comorbidities was less than 2; ≥2 indicates that the number of comorbidities was more than 2; a=Primary care facilities; b=Hospitals; c=Hospitals and primary care facilities.
Figure 3.Heatmap depicting the prevalence of coronavirus disease 2019 (COVID-19) symptoms according to various medical consultation pathways.
-
In December 2022, China updated its epidemic prevention and control policy in response to the rising number of COVID-19 infections and medical visits. Government departments collaborated to allocate medical resources and establish a hierarchical diagnosis and treatment strategy. Considering this context, a cross-sectional questionnaire survey was conducted among 354 community health centers in 16 districts of Beijing between December 26 and December 31, 2022. The findings revealed that more than 80% of patients did not seek medical attention, while over 70% of patients opted for primary healthcare facilities for treatment. Higher consultation rates were observed among elderly individuals, those with lower education levels, residents of rural areas, and individuals with comorbidities. Additionally, patients with varying pre-existing comorbidities or COVID-19 symptoms demonstrated different consultation rates and preferences.
The implementation of graded diagnosis and treatment strategies, in conjunction with pre-issued guidelines from government departments (4), provides rehabilitation guidance for individuals with COVID-19 in home isolation and directs those with mild cases to primary medical institutions. This approach aids in maintaining a stable and orderly system for diagnosis and treatment, while minimizing the waste of medical resources. The effectiveness of this policy was confirmed in the study, with over 80% of patients opting for self-medication and home-based treatment. This preference can be attributed to the mild clinical manifestations of COVID-19 (1) and patient conditions favorable for home-based recovery.
Elderly individuals with COVID-19, due to their underlying health conditions and weakened immune systems, are more likely to seek medical consultation compared to younger patients. The outpatient rate of infected individuals in rural areas was higher than that of COVID-19 patients in urban areas. Given the weak medical infrastructure in rural areas of China, the capacity for primary medical care and health services was severely tested during the epidemic. In response, the government issued an emergency plan aimed at strengthening the graded and stratified treatment and referral of patients with COVID-19 by strictly implementing the first diagnosis responsibility system and emergency treatment system (5).
The implementation of stratified treatment has been vital in combating the COVID-19 pandemic. Primary healthcare institutions categorize health risks into three distinct levels, taking into consideration factors such as age, pre-existing comorbidities, and vaccination status. These institutions offer tiered medical services for individuals infected with COVID-19, ranging from community screening to diagnosis and treatment at community health centers (6). The findings of this study indicate that over 70% of infected patients sought care at primary medical institutions due to their expedited waiting time and convenient access to medical services. As a result, these institutions serve essential roles as both “sentinels” and “network bottoms” in controlling the epidemic (7).
This study revealed that patients exhibiting various symptoms opted for different routes when seeking treatment. Those presenting severe symptoms, such as dyspnea, tachypnea, nausea/vomiting, chest tightness, and diarrhea, were more inclined to seek treatment at specialized hospitals. The quantity of symptoms reported post-infection diverged among patients with distinct comorbidities. In accordance with prior research, the mean number of COVID-19 symptoms escalated in correlation to the number of comorbidities. This is consistent with previous studies indicating that pre-existing comorbidities’ presence can heighten the risk of complications (8) and severe adverse outcomes (9). These findings carry significant implications for risk stratification and future strategizing.
Historically, large-scale infectious disease epidemics have led to significant strain on medical resources (10). Consequently, effective management of medical resources has emerged as an essential public health concern during such epidemics. The findings of this study suggest that, in the context of large-scale epidemics characterized by high infectivity but low morbidity and mortality, governmental agencies should adopt proactive measures to guide residents in seeking treatment in a graded and stratified manner. This approach would accommodate patient needs while simultaneously reducing the burden on medical resources. Furthermore, enhancing the diagnostic and treatment capabilities of primary healthcare institutions can decrease the influx of patients with minor illnesses at specialized referral hospitals. Consequently, this strategy can alleviate the workload on referral hospitals and establish early warning and referral systems for elderly patients and those with underlying comorbidities.
This study presents several limitations. Despite utilizing a multi-center survey encompassing 33,968 COVID-19 patients from 16 districts in Beijing, potential selection bias must be acknowledged, and extrapolation of the findings should be approached cautiously. Moreover, the study did not specifically examine medications employed by patients during home-based treatment. Consequently, the confounding effects of medication were not accounted for in the multivariate logistic regression model, which could potentially impact the results.
-
No conflicts of interest.
HTML
| Citation: |
|