Abstract
Background
Patients who survived hospitalisation for COVID‐19 experienced varying durations of illness but the factors associated with prompt recovery are unknown. This study identifies factors differentiating hospitalised patients who recovered promptly versus survived a prolonged course of illness because of COVID‐19.
Methods
This was a retrospective study from March‐August 2020 of hospitalised adults with COVID‐19 which were grouped based on time to recovery: short (≤3 days), intermediate (4‐10 days) and prolonged (>10 days). Recovery was defined as resolution of fever, tachypnea, hypotension, extubation and return of mental status at baseline. Multivariate analysis was used to evaluate factors associated with prompt recovery.
Results
Among 508 patients hospitalised for COVID‐19, 401 (79%) survived. Of those, prompt recovery (within 3 days) was achieved in 43% (174/401), whereas 23% (92/401) recovered after a prolonged period of >10 days. Overall, median age was 64 years with 73% admitted from home and 25% from a skilled nursing facility. Predictors for prompt recovery upon admission included female sex (OR, 1.8; 95% CI, 1.1‐2.7; P = .01), no fever (OR, 1.6; 95% CI, 1.1‐2.6; P = .03), longer time from symptom onset to hospitalisation (OR, 1.1; 95% CI, 1.0‐1.1; P = .001), no supplemental oxygen (OR, 1.9; 95% CI, 1.2‐3.0; P = .004), no direct ICU admission (OR, 41.7; 95% CI, 2.4‐740.4; P = .01) and absence of bacterial co‐infections (OR, 2.5; 95% CI, 1.5‐4.0, P = .0003).
Conclusions
Our study provides relevant data that could help clinicians triage competing resources in health systems that are challenged by the ebb and flow of COVID‐19 cases by identifying clinical features of COVID‐19 patients who may require less intensive management including avoidance of unnecessary antibacterial therapy.
What’s known
Patients who survived hospitalisation for COVID‐19 experienced varying durations of illness but the factors associated with prompt recovery are unknown.
What’s new
Our study provides relevant data that could help clinicians triage competing resources in health systems that are challenged by the ebb and flow of COVID‐19 cases by identifying clinical features of COVID‐19 patients who may require less intensive management including avoidance of unnecessary antibacterial therapy.
1. INTRODUCTION
The novel coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2), caused an initial outbreak in Wuhan, China in December 2019 and has spread resulting in a global pandemic with significant mortality and strain on healthcare operations. 1 , 2 , 3 To date, most of the published work focuses on the epidemiological, clinical, laboratory and radiological characteristics of patients infected with SARS‐CoV‐2 along with risk factors associated with mortality. Less is known about the varying durations of illness among those who survived hospitalisation for COVID‐19. With many health systems challenged by the demands from the ebb and flow of COVID‐19 cases nationwide and globally, our study aims to identify features associated with prompt recovery to help clinicians triage management of patients hospitalised for COVID‐19.
2. METHODS
This retrospective cohort study was conducted on adult patients (age ≥18 years) who survived hospitalisation for symptomatic COVID‐19 confirmed by positive polymerase chain reaction (PCR) result for SARS‐CoV‐2 from March to August 2020 at a 625‐bed community teaching hospital in Los Angeles. The study was approved by the hospital's Institutional Review Board (Date: June 24, 2020; Number: 00000971); informed consent was waived.
Medical charts were reviewed to obtain demographic, laboratory and clinical information, and details of antiviral and antibacterial therapy. All data were recorded on a structured data collection form and managed using REDCap electronic data capture tools hosted at the University of Southern California. 4 Sepsis‐3 criteria were used to define sepsis and septic shock. 5 Outcome measures included development of acute kidney injury (AKI) and acute respiratory distress syndrome (ARDS) and hospital length of stay (LOS).
2.1. Data analysis
Patients were grouped based on short (≤3 days), intermediate (4‐10 days), prolonged (>10 days) time to recovery defined by achievement of clinical stability (TTS): resolution of fever <38°C, respiratory rate <22 breaths/min, systolic blood pressure >100 mmHg, extubated and return of mental status to baseline. Study groups were compared on demographic and clinical characteristics, pharmacologic management and outcomes. Descriptive analyses were performed using Kruskal–Wallis and chi‐square test where appropriate. The model outcome, prompt recovery defined by short TTS was compared with combined intermediate and prolonged TTS to determine the predictability of each factor for the prompt recovery of COVID‐19 patients. Univariate logistic regression analyses were performed to determine the candidates of the predictors for prompt recovery with P value <.20. Only the factors found to be potentially significant by univariate analysis were included in the multivariate logistic regression analysis using forward model selection method. A P value <.05 was considered significant. Statistical analyses were performed using GraphPad Prism v6.07 (San Diego, CA, USA) or SAS version 9.4 (SAS Institute, Cary, NC).
3. RESULTS
Survival rate was 79% (401/508) among patients hospitalised for COVID‐19 during the study period. Overall, median age of COVID‐19 survivors was 64 years; majority (73%) were admitted from home with 25% from a skilled nursing facility (SNF). The overall median TTS was 4 days (IQR: 2‐10) for survivors, with 43% (n = 174) within 3 days, 34% (n = 135) between 4‐10 days and 23% (n = 92) after a prolonged period exceeding 10 days. More females achieved prompt versus prolonged time to recovery (51% vs 37%; Table 1).
TABLE 1.
Characteristics and outcomes of patients surviving hospitalisation with COVID‐19 a
| Characteristics |
All patients N = 401 |
Short TTS (≤3 d) N = 174 |
Intermediate TTS (4‐10 d) N = 135 |
Prolonged TTS (>10 d) N = 92 |
P‐value |
|---|---|---|---|---|---|
| Demographics | |||||
| Female | 180 (45) | 89 (51) | 57 (42) | 34 (37) | .0644 |
| Age, median (IQR), y | 64 (50‐77) | 64 (49‐78) | 66 (51‐77) | 64 (50‐73) | .5982 |
| Residence prior to admission | .0300 | ||||
| Home | 291 (73) | 135 (78) | 93 (69) | 63 (68) | |
| Skilled nursing facility | 102 (25) | 37 (21) | 41 (30) | 24 (26) | |
| Outside hospital/other | 8 (2) | 2 (1) | 1 (0.7) | 5 (5) | |
| Current/former smoker | 52 (13) | 21 (12) | 21 (16) | 10 (11) | .5261 |
| Top 5 comorbidities | |||||
| Hypertension | 213 (53) | 86 (49) | 74 (55) | 53 (58) | .3958 |
| Diabetes | 116 (29) | 46 (26) | 42 (31) | 28 (30) | .6251 |
| Hyperlipidaemia/dyslipidaemia | 107 (27) | 46 (26) | 39 (29) | 22 (24) | .7040 |
| Dementia | 71 (18) | 26 (15) | 30 (22) | 15 (16) | .2316 |
| Chronic kidney disease | 44 (11) | 15 (9) | 16 (12) | 13 (14) | .3622 |
| None | 76 (19) | 39 (22) | 23 (17) | 14 (15) | .2843 |
| Time from symptom onset to hospitalisation, median (IQR), d | 4 (1‐7) | 5 (2‐10) | 3 (1‐7) | 5 (2‐7) | .0114 |
| Top 3 COVID‐19 symptoms | |||||
| Fever | 217 (54) | 81 (47) | 78 (58) | 58 (63) | .0214 |
| Dyspnoea | 222 (55) | 87 (50) | 78 (58) | 57 (62) | .1379 |
| Non‐productive cough | 205 (51) | 88 (51) | 64 (47) | 53 (58) | .3142 |
| Disease severity upon admission | |||||
| Septic shock | 20 (5) | 0 (0) | 5 (4) | 15 (16) | <.0001 |
| Direct ICU admission | 33 (8) | 0 (0) | 10 (7) | 23 (25) | <.0001 |
| Required supplemental O2 | 265 (66) | 95 (55) | 99 (73) | 71 (77) | <.0001 |
| Mechanical ventilation | 18 (4) | 0 (0) | 3 (2) | 15 (16) | <.0001 |
| Presumed or culture‐positive co‐bacterial infection b | 138 (34) | 36 (21) | 50 (37) | 44 (47) | <.0001 |
| Receipt of antibacterial therapy upon admission | 370 (92) | 149 (86) | 129 (96) | 92 (100) | <.0001 |
| Duration of antibacterial therapy | 7 (4‐11) | 4 (2‐6) | 7 (6‐10) | 15 (11‐20) | <0.0001 |
| Outcomes | |||||
| Developed AKI | 72 (18) | 18 (10) | 23 (17) | 31 (34) | <.0001 |
| Developed ARDS | 18 (4) | 0 (0) | 2 (1) | 16 (17) | <.0001 |
| Hospital LOS, median (IQR), d | 9 (5‐15) | 5 (3‐7) | 9 (8‐13) | 22 (16‐29) | <.0001 |
Abbreviations: AKI, acute kidney injury; ARDS, acute respiratory distress syndrome; LOS, length of stay.
Data are presented as number (percentage) of patients unless otherwise indicated.
Presumed bacterial co‐infection was defined as a procalcitonin level ≥0.25 ng/mL upon admission.
At time of admission, significantly less patients in the short TTS group experienced fevers (47% vs 58% vs 63%, P = .0214), required supplemental oxygen (55% vs 73% vs 77%, P < .0001), while none had septic shock (vs 4% vs 16%, P < .0001), required direct ICU admission (vs 7% vs 25%, P < .0001) or mechanical ventilation (vs 2% vs 16%, P < .0001) compared with the intermediate and prolonged groups, respectively (Table 1). Therapy directed against COVID‐19 was less frequently prescribed to patients with short compared with intermediate and prolonged TTS: hydroxychloroquine ± azithromycin (13% vs 32% vs 37%, P < .0001), remdesivir (14% vs29% vs 48%, P < .0001), tocilizumab (0.6% vs 0.7% vs 12%, P < .0001), corticosteroids (34% vs 31% vs 58%, P < .0001) and convalescent plasma (10% vs 24% vs 40%, P < .0001).
Patients with prompt recovery were less likely to have co‐bacterial infections at time of admission (presumed 13% vs 24% vs 39%, P = .0001; culture‐positive 11% vs vs 21% vs 28%, P = .0013), especially in the respiratory site (0% [0/19] vs 10% [3/29] vs 46% [12/26], P = .0002) compared withintermediate and prolonged TTS groups, respectively. Nonetheless, 86% of the patients in the short TTS group who did not have co‐bacterial infections were prescribed broad‐spectrum antibacterial therapy for a median duration of 4 days. A notable proportion of the prolonged TTS cohort were co‐infected with multidrug‐resistant pathogens such as Pseudomonas aeruginosa (15%) and carbapenem‐resistant organisms (4%) compared with none in the short TTS group; patients who resided in a SNF prior to admission accounted for 56% (5/9) of those with Pseudomonas aeruginosa and all three cases involving carbapenem‐resistant organisms. Interestingly, co‐infection with ESBL‐producing isolates (mostly urine) occurred in 4% of patients overall, irrespective of TTS. Compared withthe intermediate and prolonged TTS cohorts, less patients with a short TTS developed AKI (10% vs 17% vs 34%, P < .0001) and none developed ARDS (vs 1% vs 17%, P < .0001) (Table 1). Ultimately, those who recovered fastest experienced a shorter hospital stay (median 5 vs 9 vs 22 days, P < .0001; Table 1).
By multivariable logistic regression analysis, independent predictors for prompt recovery from COVID‐19 were female sex, absence of fever, longer time from symptom onset to hospitalisation, no direct ICU admission, not requiring supplemental oxygen upon presentation and absence of presumed or documented co/secondary bacterial infections (Table 2).
TABLE 2.
Predictors of prompt recovery by multivariable logistic regression a
| Variables | OR (95% CI) | P‐value |
|---|---|---|
| Female | 1.76 (1.13‐2.74) | .01 |
| Time from symptom onset to hospitalisation | 1.08 (1.03‐1.13) | .001 |
| No Fevers upon admission | 1.64 (1.06‐2.57) | .03 |
| No supplemental O2 upon admission | 1.95 (1.23‐3.07) | .004 |
| No direct ICU admission | 41.73 (2.35‐740.35) | .01 |
| No presumed or culture‐positive co‐bacterial infection | 2.46 (1.51‐4.01) | .0003 |
The model dependent variable is prompt recovery (short TTS vs combined intermediate and prolonged TTS cohorts).
4. DISCUSSION
Patients who survived hospitalisation with COVID‐19 experienced varying time to achieve clinical stability before hospital discharge. While much of the focus has been on the severity of COVID‐19 among those who required hospitalisation, less is known regarding the characteristics of patients who recovered promptly. Clinical features associated with a favourable hospital course could help clinicians triage competing resources in the care of patients especially at institutions challenged by the demands from the ebb and flow in COVID‐19 cases.
Our findings indicate that female sex, longer time from symptom onset to hospitalisation, less severe presentation upon admission (fevers, septic shock, ICU, supplemental oxygen) and absence of bacterial co‐infection are factors favouring prompt recovery. We observed a similar prevalence of COVID‐19 requiring hospitalisation between male and female sex in our study. Although literature has shown a correlation between men with COVID‐19 and worse outcomes including death, our data is the first to show an association between female sex and shortened recovery time. 6 , 7 Our data is also the first to show that an extended time from symptom onset to hospital admission is a predictor for prompt recovery.
We found that significantly less patients with prompt recovery had bacterial co‐infections especially concurrent pneumonia compared with those in the prolonged TTS group. Nonetheless, 86% of patients in the short TTS group were prescribed antibacterial therapy for a median duration of 4 days despite the absence of bacterial co‐infection. This finding underscores the need for continued antimicrobial stewardship to minimise indiscriminant antibacterial use in hospitalised COVID‐19 patients. As expected, those with a short TTS had a less complicated course with less than one week length of hospital stay in this cohort. While we identified distinguishing clinical features between groups with varying time to achieve clinical stability, our study is limited by the relatively small number of patients at a single centre and the lack of viral load or host immune response measurements to provide a biological basis to our clinical observations.
5. CONCLUSION
Health system resources may be triaged accordingly based on clinical features associated with prompt recovery in patients hospitalised for COVID‐19. Concurrent antibacterial therapy needs to be judiciously prescribed to minimise the selection of antimicrobial resistance and untoward adverse effects (eg C difficile diarrhea).
DISCLOSURES
The authors declared no conflict of interest.
ACKNOWLEDGEMENTS
This work was supported by grants UL1TR001855 and UL1TR000130 from the National Center for Advancing Translational Science (NCATS) of the US National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Ny P, Kelsom C, Chron A, et al. Factors associated with prompt recovery among hospitalised patients with coronavirus disease 2019. Int J Clin Pract. 2021;75:e14818. 10.1111/ijcp.14818
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1. Phelan AL, Katz R, Gostin LO. The novel coronavirus originating in Wuhan, China: challenges for global health governance. JAMA. 2020;323:709‐710. [DOI] [PubMed] [Google Scholar]
- 2. Gorbalenya AE, Baker SC, Baric RS, et al. Severe acute respiratory syndrome‐related coronavirus: the species and its viruses—a statement of the Coronavirus Study Group. bioRxiv. 2020;. published online Feb 11. doi: 10.1101/2020.02.07.937862 [DOI] [Google Scholar]
- 3. Johns Hopkins University and Medicine Coronavirus Resource Center . Coronavirus COVID‐19 Global Cases by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University (JHU). https://coronavirus.jhu.edu/map.html. Accessed November 30, 2020.
- 4. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde Jose G. Research electronic data capture (REDCap): a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377‐381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions of sepsis and septic shock (Sepsis‐3). JAMA. 2016;315:801‐810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Jin J, Bai P, He W, et al. Gender differences in patients with COVID‐19: focus on severity and mortality. Front Public Health. 2020;8:152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Gebhard C, Regitz‐Zagrosek V, Neuhauser HK, gan R, Klein SL. Impact of sex and gender on COVID‐19 outcomes in Europe. Biol Sex Differ. 2020;11:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.