COVID-19 symptom-onset to diagnosis and diagnosis to treatment intervals are significant predictors of disease progression and hospitalization in high-risk patients: A real world analysis

Background Coronavirus disease (COVID-19) is overwhelming healthcare systems worldwide. This study aimed to elucidate factors that influence disease progression to pneumonia and hospitalization before and after antiviral treatment for COVID-19 in an outpatient setting. Methods A total of 206 high-risk patients with COVID-19 were treated with sotrovimab, remdesivir, and molnupiravir at the Toshiwakai clinic between January 4 and April 30, 2022. Of these, 49 patients visited the Toshiwakai clinic directly and were treated immediately after diagnosis (Toshiwakai-clinic study group). The remaining patients were diagnosed elsewhere, and of these, 102 patients were quarantined at home (health-center study group) and 55 at designated facilities (quarantine-facility study group) before being referred to Toshiwakai clinic. Patients were categorized into those with mild and moderate COVID-19, based on the presence of pneumonia at the initial visit to Toshiwakai clinic. Results The symptom-onset-to-diagnosis and diagnosis-to-treatment intervals were significant predictors of moderate disease. Age, dyspnea, and diagnosis-to-treatment interval at the first visit to Toshiwakai clinic were significant predictors for hospitalization even after antiviral treatment. Although the symptom-onset-to-diagnosis interval did not differ among the three study groups, the diagnosis-to-treatment and symptom-onset-to-treatment intervals were significantly longer in the health-center and quarantine-facility study groups than in the Toshiwakai-clinic study group. Conclusion The symptom-onset-to-diagnosis and diagnosis-to-treatment intervals reflect diagnostic and interventional delays, respectively, which are closely related to the current COVID-19 clinical management protocol. Easy access to the clinics and immediate antiviral treatment after diagnosis may be the best methods to prevent disease progression and hospitalization in high-risk patients.


Introduction
The Omicron variant of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is currently overwhelming healthcare systems globally [1,2]. It was detected in Nagoya city, Japan during the sixth wave of coronavirus disease (COVID-19) [3], which began in December 2021 (Fig. 1). The protocol for COVID-19 patient management based on the infectious disease law in Japan remained unchanged for 2.5 years until September 2022. All COVID-19 cases are mandatorily registered at the local health center [4] through an online registration system (HER-SYS) [5] or a specific form faxed by the diagnosing physician. Public health nurses at health centers document the patient's medical history, current condition, comorbidities, and past medical and contact histories, and monitor the patient's daily condition for 10 days, either via HER-SYS or direct telephonic interviews. A health center medical team evaluates the disease severity to determine the follow-up method, which can be performed at home, in a quarantine facility within a designated hotel, or at a hospital. Physicians at the health center or quarantine facility request inpatient antiviral or supportive therapy for patients with prolonged disease or potential exacerbation. As the number of new patients in Nagoya city increased dramatically at the beginning of the sixth COVID-19 wave from the end of 2021 (Fig. 1), a hospital-bed shortage was anticipated, and the Aichi provincial government requested designated COVID-19 hospitals to allocate more beds. Furthermore, the Nagoya health center lowered the antiviral treatment initiation threshold for high-risk patients (before clinical deterioration) at outpatient clinics, owing to the safety of monoclonal antibodies or antiviral medication use in outpatient clinics and efficacy in preventing disease progression and hospitalization [6e9]. Over the past 4 months, after the onset of the sixth wave, the Toshiwakai clinic actively cooperated with the Nagoya health center and quarantine facilities to provide antiviral therapy to high-risk patients, and doctors at the clinic noticed greater disease progression in referred cases than in those who directly visited Toshiwakai clinic.
This study investigated potential factors for disease progression and hospitalization by focusing on the symptomonset-to-treatment interval to facilitate risk-stratified clinical management of high-risk COVID-19 patients.

Ethical and safety considerations
Informed consent was obtained from the patients who were included in this study using opt-out method, which was explained via a letter and poster exhibited at the clinic [10]. This study was approved by the ethics committee of Daiyukai Health System (approval no 2022e006) and was conducted in accordance with the principles evinced in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Study design
This retrospective, observational study enrolled 206 patients who underwent antiviral treatment at Toshiwakai clinic. All patients in the study population were designated as having a high risk for severe COVID-19, owing to the presence of predisposing factors, such as cardiovascular disease, hypertension, hyperlipidemia, chronic obstructive pulmonary disease (COPD), bronchial asthma, chronic respiratory diseases, cerebrovascular diseases, obesity, chronic kidney disease, malignancies, immunosuppressive condition, liver disease, current smoker and age over 65 years, as reported in the Japanese COVID-19 management guideline [11]. These 206 high-risk patients were categorized by disease severity or hospitalization before and after treatment, and potential risk factors for disease progression and hospitalization at the first clinic visit were ascertained.

Study population
A total of 206 patients with high-risk comorbidities visited Toshiwakai clinic from January 4 to April 30, 2022 and underwent antiviral treatment (Fig. 2). Forty-nine symptomatic patients, who visited the Toshiwakai clinic directly and were treated immediately after testing positive for SARS-CoV-2 via an antigen test or the nicking enzyme amplification reaction, comprised the Toshiwakai-clinic study group; 102 patients, who were diagnosed with COVID-19 elsewhere and referred from three branches of Nagoya health center to the Toshiwakai clinic for further examination by chest computed tomography (CT) and antiviral treatment after daily monitoring at home via HER-SYS or direct telephonic interview by public health-nurses at health centers, comprised the health-center study group; and the remaining 55 patients, who were diagnosed elsewhere and admitted to the quarantine facilities before referral to Toshiwakai clinic for chest CT examination and antiviral treatment, comprised the quarantine-facility study group. These patients required isolation in private rooms and close observation at quarantine facilities (with daily meal provision). Their clinical condition was checked by a registered-nurse and a consulting physician at the quarantine facilities; all data were reported to the health center. Referral to Toshiwakai clinic for chest CT and antiviral treatment was provided by the consulting doctor at the quarantine facilities.

Pretreatment patient categorization of disease severity
The detailed medical history of symptoms; date of symptomonset; date of diagnosis; findings from physical examination, including chest auscultation, blood pressure measurement; and percutaneous oxygen saturation of hemoglobin (SpO 2 ), were collected. Most patients referred from health center and quarantine facilities (82.2% and 96%, respectively) underwent chest CT at Toshiwakai clinic. Approximately one-third of the Toshiwakai-clinic study group required chest CT (32.1%) based on their symptoms and physical examination findings at the first visit. Disease severity was categorized using NIH guidelines [12] into mild (pneumonia absent) and moderate (pneumonia present) disease based on the symptoms and findings of chest CT and physical examination.

Treatment regimen
Patients were administered three treatment regimens included sotrovimab, molnupiravir, and remdesivir. Sotrovimab 500 mg was infused intravenously over 0.5 h once at the first visit. Molnupiravir 800 mg (four 200 mg capsules) was administered orally twice daily for 5 days. Remdesivir 200 mg was infused intravenously over 0.5 h once at the first visit followed by intravenous administration of 100 mg once daily for 4 days from day 2.

Posttreatment outcome measure
The primary outcome was posttreatment hospitalization.

Statistical analysis
The normality of data distribution was ascertained using the Shapiro-Wilk test. Normally distributed data were expressed as the mean ± standard deviation and compared using the Student's t-test. Non-normally distributed data were expressed as the median and interquartile ranges and compared using the Mann-Whitney U test or Kruskal-Wallis test with multiple comparisons using the Steel-Dwas test; Nominal variables were expressed as percentages and compared using the chi-squared test. Univariate and multivariate logistic regression was used to investigate the possible predictors of disease severity and hospitalization; p-values<0.05 indicated statistical significance. All statistical analyses were performed using the free software EZR on R commander [13,14].

Results
At the first visit to Toshiwakai clinic, 123 of 206 participants with high-risk COVID-19 were diagnosed with mild disease and 83 were diagnosed with moderate disease (Table 1). In patients with mild disease (n ¼ 123), the possibility of pneumonia was eliminated on the basis of CT in 64 cases and symptoms/physical examination in 59 cases. In patients with moderate disease (n ¼ 83), the diagnosis of pneumonia was confirmed by chest CT in 82 cases; one participant was diagnosed clinically based on cough, dyspnea, and apparent coarse crackles on auscultation. In those with moderate disease, 5 out of 83 had SpO 2 93% and required oxygen therapy at home or at the quarantine facility (classifiable as severe disease according to the NIH guideline [12]); however, these participants were included in the moderate category as outpatient antiviral treatment was initiated at the clinic owing to the shortage of beds at the designated hospitals. Patient characteristics, high-risk comorbidities, study groups, and frequency of COVID-19 vaccination were compared between patients with mild and moderate disease (Table 1). Age, sex, obesity (BMI 30 kg/m 2 ), pre-obesity (30 kg/ m 2 > BMI 25 kg/m 2 ), high-risk comorbidities, smoking, and number of comorbidities did not differ significantly between patients with mild and moderate disease. The incidence of sore throat was significantly higher in patients with mild disease, whereas that of cough and dyspnea was significantly higher in patients with moderate disease (p < 0.001 for all). The proportion of patients with mild and moderate disease differed significantly among the study groups (p < 0.0001). The percentages of patients with moderate disease were 12.2%, 40.2%, and 65.5% for Toshiwakai-clinic, health-center, and quarantine-facility study groups, respectively. Although the frequency of vaccination did not differ significantly between patients with mild and moderate disease, the patients with moderate disease tended to receive a fewer frequency of vaccination ( Table 1).
The exact dates of symptom-onset, diagnosis, and treatment for each patient were obtained to investigate the manner in which the time course from symptom-onset to treatment affected COVID-19 progression and hospitalization. The interval between symptom-onset and treatment (symptom-treatment) was subdivided into sub-intervals between symptom-onset and diagnosis (symptom-diagnosis) and diagnosis and treatment (diagnosis-treatment). Symptom-diagnosis, diagnosis-treatment, and symptomtreatment intervals were significantly longer in patients with moderate disease than in those with mild disease (Fig. 3). Univariate logistic regression analysis (Table 2) revealed that symptom-diagnosis interval (p ¼ 0.019), diagnosis-treatment interval (p < 0.0001), frequency of vaccination (p ¼ 0.044), and study groups (health-center study group versus Toshiwakai-clinic study group, p < 0.001; quarantine-facility study group versus Toshiwakai-clinic study group, p < 0.0001) were significant predictors of moderate disease. Multivariate analysis revealed that only symptom-diagnosis and diagnosis-treatment intervals were significant independent predictors (p < 0.001 and p < 0.0001, respectively).
Overall, 211 antiviral medications were used (92, 71, and 48 instances of sotrovimab, molnupiravir, and remdesivir, respectively) for 206 patients. Four patients received a combination of sotrovimab and remdesivir and one patient received a combination of molnupiravir and remdesivir owing to severe symptoms and fewer frequency of vaccination (three Fig. 2 e The study population and timeline from symptom-onset to treatment. Interval from symptom-onset to treatment is indicated (left). In the Toshiwakai-clinic study group, patients were treated immediately after diagnosis. In the healthcenter study group, patients were diagnosed elsewhere and stayed at home under monitoring by a health center before treatment. In the quarantine-facility study group, patients were diagnosed elsewhere and underwent close monitoring at designated facilities before treatment.
patients were unvaccinated and two had received two vaccination doses each). Two other unvaccinated patients with moderate disease received dexamethasone and remdesivir because of severe symptoms.
Ten patients (3 and 7 with mild and moderate disease, respectively) required posttreatment hospitalization. Comparison between the clinical variables of hospitalized and non-hospitalized patients (Table 3) revealed a significantly higher rate of dyspnea (p ¼ 0.0086), greater age (p ¼ 0.0052), and greater likelihood of SpO 2 93% (p ¼ 0.024) in hospitalized than non-hospitalized patients. Disease severity (with or without pneumonia) was unrelated to hospitalization ( Table  3). The number of hospitalized patients differed significantly between the study groups (p ¼ 0.029); no patient in Toshiwakai-clinic study group was hospitalized whereas 4 and 6 patients in the health-center and quarantine-facility study groups, respectively, were hospitalized ( Table 3). The other symptoms, high-risk comorbidities, or the frequency of vaccination did not differ significantly between hospitalized and non-hospitalized patients. Although no significant difference was observed in symptom-diagnosis interval, both diagnosis-treatment and symptom-treatment intervals were significantly longer among hospitalized patients (Fig. 4). Univariate logistic regression analysis performed to investigate potential independent risk factors for post-treatment hospitalization (Table 4) revealed that age (p ¼ 0.0082), dyspnea (p ¼ 0.0081), SpO 2 93% (p ¼ 0.0094), and diagnosis-treatment interval (p ¼ 0.045) at the first visit to Toshiwakai clinic were significant predictors of hospitalization. However, only age (p ¼ 0.0041), dyspnea (p ¼ 0.015), and diagnosis-treatment interval (p ¼ 0.024) were significant independent predictors of hospitalization in the multivariate analysis (Table 4).

Discussion
Several studies have shown that early antiviral treatment prevents the incidence of severe COVID-19 and associated hospitalization [7e9,15]; however, these studies included patients who were treated within 5e7 days of symptom onset. r e s p i r a t o r y i n v e s t i g a t i o n 6 1 ( 2 0 2 3 ) 2 2 0 e2 2 9 No study has investigated the relation between disease progression and antiviral treatment immediately after symptom onset. We found that moderate disease was associated with only a 1-day longer median symptom-diagnosis interval compared to mild disease (Fig. 3). The symptom-diagnosis interval reflects a diagnostic delay and was a significant predictor of moderate disease (OR ¼ 1.74, 95% CI ¼ 1.27e2.37, p < 0.001), but did not differ significantly among the three study groups. Thus, the variation in the symptom-diagnosis interval could depend on the ease of access to the clinics for diagnostic confirmation, due to the rapid rise in incidence (Fig. 1). A genomic survey in Aichi prefecture revealed that the sixth wave was attributable to the rapid spread of Omicron BA.1 and the subsequent surge of the BA.2 lineage [3]. Since the infectivity of the Omicron variant was reported to be considerably greater than that of the wild-type or Delta variants [16], airborne spread could have been a possibility, in addition to droplet or direct contact infection. Since the average incubation time of the Omicron variant (3.2 days) is shorter than that of the Delta variant (4.4 days) [17], the prediagnosis incubation phase of SARS-CoV-2 may be the primary factor for disease progression, because viral replication and shedding are extremely high during this period [18,19]. Therefore, limiting the diagnostic delay by improving medical infrastructure to facilitate consultation, especially for symptomatic high-risk patients, may considerably hinder disease progression. The 2-day longer median diagnosis-treatment interval in patients with moderate disease compared to those with mild disease reflects an interventional delay (Fig. 3), which was a significant independent predictor of moderate disease (OR ¼ 2.27, 95% CI ¼ 1.58e3.24, p < 0.0001)) ( Table 2). This delay may be attributed to the patient management system, where the condition of all patients at home or in quarantine facilities was monitored by the health center following a COVID-19 diagnosis, and the decision to implement antiviral treatment or hospitalization depended on exacerbation of the patients' condition or prolongation of illness. Due to the rapid rise in cases during the sixth COVID-19 wave, the Nagoya health center shortened the observation time and requested earlier treatment of high-risk patients at the clinics. Even after a switch to this policy, a delay of a few days persisted before patient referral to the clinics for treatment; this interventional delay could have increased disease progression substantially. Fig. 3 e Sub-analysis of time from symptom-onset to treatment (mild vs moderate). Interval from symptom-onset and treatment (symptom-treatment) was subdivided into intervals between symptom-onset and diagnosis (symptom-diagnosis) and between diagnosis and treatment (diagnosis-treatment). The differences in these time frames were ascertained between patients with mild and moderate diseases. There was no multicollinearity as the variation inflation factor for each predictor was less than 2.5. OR, odds ratio; CI; confidence interval.
r e s p i r a t o r y i n v e s t i g a t i o n 6 1 ( 2 0 2 3 ) 2 2 0 e2 2 9 The Toshiwakai clinic actively treated high-risk outpatients from the beginning of the sixth COVID-19 wave and initiated antiviral treatment immediately after diagnosis, in contrast to the other patients referred from the health center and quarantine facilities. Patients in the Toshiwakai-clinic study group had a lower chance of developing moderate disease (12.2%) than those in the health-center and quarantine-facility study groups (40.2 and 65.5%, respectively). Moreover, no patients in the Toshiwakai-clinic study group required hospitalization, whereas 4 and 6 patients in the health-center and quarantinefacility study groups, respectively, required hospitalization (Table 3). A median interventional delay in diagnosistreatment interval of only 1.5 days (Fig. 4) was a significant predictor of hospitalization (OR ¼ 1.57, 95% CI ¼ 1.06e2.33, p < 0.024) ( Table 4), indicating that shortening the interventional delay is essential for preventing hospitalization since the diagnostic delay were similar in all study groups. Notably, 76 out of 83 of patients with pneumonia (92%) recovered after antiviral treatment in the outpatient setting, suggesting that immediate antiviral administration after diagnosis at the clinics may prevent disease progression and lessen the economic and resource burden of hospitalization. Although no significant relationship was observed between disease severity and frequency of COVID-19 vaccination (Table 1), a meta-analysis conducted by Zou et al. showed that full-dose SARS-CoV-2 vaccination effectively reduces the Omicron variant infection [20]. Nevertheless, vaccine effectiveness has decreased since the Omicron variant has become the predominant strain [21e23] and break through infections have been reported [24]. Vaccination decreases the need for hospitalization due to severe COVID-19 [25], although our data did not show a significant relationship between the frequency of vaccination and hospitalization (Table 3), possibly because of early antiviral administration, even in patients from the  health center and quarantine facilities. The symptomtreatment interval was less than 3 days in 125 patients (60.7%) and 4e5 days in other 55 patients (26.7%); these intervals were shorter than those reported in earlier studies conducted at outpatient clinics [7e9]. Thus, the lack of a relationship between the frequency of vaccination and hospitalization may be attributed to recovery after early antiviral treatment in unvaccinated patients who would have otherwise developed severe disease. Early treatment of high-risk patients may decrease the incidence of post-acute or long COVID, which is defined as the persistence of acute-phase symptoms for more than 4 weeks [26]. A previous study reported a wide range of symptoms involving the respiratory, nervous, neurocognitive, metabolic, cardiovascular, gastrointestinal, and musculoskeletal systems, as well as fatigue and mental health problems [27]. The incidence of long COVID is related to disease severity [27], but seems to be lower in patients with Omicron infection (4.5% of patients) than in those with Delta infection. Nonetheless, a large number of patients with the associated sequelae continue to pose a tremendous longterm healthcare, social, and financial burden [28]. Therefore, early treatment even for referrals from health center or quarantine facilities may prevent long COVID and post-COVID-19 healthcare costs.
A limitation of our study was the lack of a proper control group, as we treated all the patients who visited Toshiwakai clinic and could not obtain information about the proportion of treated versus untreated patients in health center and quarantine facilities to investigate the adequacy of immediate treatment of high-risk patients after diagnosis.

Conclusion
This study showed that both diagnostic and interventional delays may be related to COVID-19 progression as viral replication is most active during the symptomatic period [18,19]. Alteration of current systematic approach for highrisk patients with COVID-19 is necessary to minimize these delays. Treatment should be executed in the shortest possible time from the onset of COVID-19 symptoms and diagnosis to prevent disease progression and the need for hospitalization. Systematic research is warranted to determine a feasible time frame for antiviral treatment from the onset of symptoms. Fig. 4 e Sub-analysis of time from symptom-onset to treatment (nonhospitalized vs hospitalized). There was no significant difference in the symptom-diagnosis interval between hospitalized and non-hospitalized patients. Both diagnosis-treatment and symptom-treatment intervals were significantly longer in the hospitalized than in non-hospitalized patients. There was no multicollinearity as the variation inflation factor for each predictor was less than 1.5. OR, odds ratio; CI, confident interval; SpO 2 , percutaneous oxygen saturation of hemoglobin.
r e s p i r a t o r y i n v e s t i g a t i o n 6 1 ( 2 0 2 3 ) 2 2 0 e2 2 9

Novelty of the study
This study is unique because we demonstrated that COVID-19 can progress rapidly after symptom onset. A diagnostic delay of even 1 day predicted moderate disease, suggesting the need for easier access to medical facilities. This is the first analysis to show that interventional delay is attributable to the current Japanese clinical management system of patient registration of each individual case before treatment.

Funding
This work was supported by a JSPS-in Aid for Scientific Research (C), Japan [grant number 21K08216].

Data availability
The dataset collected over the course of this study is available from the corresponding author on reasonable request.

Author contributions
All authors meet the ICMJE authorship criteria and have made substantial contributions to the following: (1) the conception and design of the study, or acquisition of data, or analysis and interpretation of data; (2)

Conflict of Interest
Kazuyorhi Imaizumi received financial support from Taiho Pharmaceutical Co., Ltd., Chugai PharmaceuticalCo., Ltd. and Eli Lilly Japan K.K. Masamichi Hayashi received financial support from Shionogi Pharmaceutical Co. Ltd, Japan. The other authors have no conflicts of interest.