Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was first identified in December 2019 in Wuhan, Hubei, China.1 2 According to World Health Organization Coronavirus (COVID-19) data, as of 7 May 2022, 188 countries and territories had reported more than 510.2 million cumulative confirmed cases and more than 6.23 million deaths3; in Hong Kong, there were 330 670 confirmed cases and 9308 (2.81%) deaths.4 Common symptoms of COVID-19 include fever, sore throat, cough, malaise, dyspnoea, and anosmia or aguesia.1 Although most people have mild symptoms, some develop acute respiratory distress syndrome, which may lead to cytokine storm, multiorgan failure, septic shock, and even death.5

There is evidence that rash is an early symptom or the only symptom in patients who are ‘asymptomatic’ or paucisymptomatic.6 7 8 9 Early detection of this ‘silent’ sign and corresponding diagnosis are important for epidemiologic management because asymptomatic or paucisymptomatic cases may function as sources of community spread. Various dermatologic manifestations of COVID-19 have been reported including maculopapular eruption, urticarial eruption, livedo reticularis, pernio/chilblain, vasculitis, vesicular eruption, and papulo-necrotic eruption.10 11 12 13 14 15 The incidences of cutaneous manifestations in patients with COVID-19 have varied among case series (from 0.2% to 20.4%10 11 12 13), possibly because of the under-recognition of asymptomatic or paucisymptomatic cases.

The spread of SARS-CoV-2 mainly involves droplets; it can also occur via direct contact and is speculated to occur through faecal excretion.2 The primary target of SARS-CoV-2 is the upper respiratory mucosa, where angiotensin-converting enzyme 2 (ACE2) serves as a functional receptor for viral spikes and eventual viral entry into host cells. Gene expression of the SARS-CoV-2 cellular receptor ACE2 has been demonstrated in multiple human tissues, including skin and adipose tissue.16 17 18 Therefore, the proposed mechanisms by which SARS-CoV-2 affect cutaneous tissues include direct attacks on epidermal basal cells and vascular endothelial cells (possibly targeting ACE2 expressed on skin keratinocytes) and indirect impacts through the antiviral inflammatory response.16 17 18

There is speculation that patients with rash occurrence may have a better prognosis because they display better antiviral immunity.19 Early in the COVID-19 pandemic, little was known about relationships among cutaneous manifestations, viral load, co-morbidities, and clinical outcomes. A recent systematic review showed inconclusive results about the relationship between COVID-19 severity and viral load; however, it suggested that older age and higher SARS-CoV-2 viral load were directly related.20 Likewise, some rashes such as maculopapular rash and chilblain-like lesions were found to be strongly associated with paucisymptomatic disease course and lower severity of COVID-19 while skin changes such as acro-ischaemia, livedo reticularis and purpura may be useful indicators of higher severity of COVID-19.21 22 In 2020, according to the Public Health Ordinance of Hong Kong, all patients with SARS-CoV-2–positive test results were hospitalised for quarantine, regardless of symptoms.4 Here, we explored the incidences and patterns of clinical and cutaneous manifestations among hospitalised patients with confirmed COVID-19, then investigated associations with viral load, co-morbidities, and prognosis.

This retrospective cohort study was conducted from 1 July to 30 September 2020 in an acute tertiary hospital, Queen Mary Hospital (ie, a major public hospital within one of seven hospital clusters) serving one-fifth of the population of 7.5 million in Hong Kong. Electronic hospital records were used to identify adult patients aged ≥18 years who were admitted during the study period for suspected COVID-19.

The flow of patient recruitment is illustrated in Figure 1. Patients included in this study were adults with laboratory confirmation of COVID-19 by real-time reverse transcription polymerase chain reaction (rRT-PCR) assay from a nasopharyngeal swab. Clinical information was collected from electronic clinical photographs of patients who had provided informed consent to receive treatment. A physical examination was performed by a dermatologist within 48 hours of rash onset to confirm clinical signs; follow-up was conducted monthly until 3 months after discharge. Rashes were considered COVID-19–related if they were new, could not be explained by the patient’s previous or pre-existing skin conditions or an alternative diagnosis (eg, drug eruption or other viral exanthem of varicella, parvovirus, enterovirus, influenza, parainfluenza, adenovirus, or respiratory syncytial virus detected in nasopharyngeal swab [performed as clinically indicated and excluded]), occurred along with the SARS-CoV-2–positive rRT-PCR test results, and resolved when other symptoms improved.

Clinical and laboratory data, including patient demographics, initial COVID-19 viral load according to cycle threshold (Ct) value, treatment received, co-morbidities (diabetes mellitus, hypertension, and chronic kidney disease [CKD]), and pre-existing skin diseases, were retrieved from electronic medical records for analysis. For the detection of viral nucleic acids, rRT-PCR is considered a gold standard diagnostic assay. The Ct value refers to the number of rRT-PCR cycles needed to amplify viral RNA to a detectable level; it is inversely related to viral load.23 Thus, the Ct value can indicate the relative quantity of viral RNA in a specimen (lower Ct values reflect greater quantities of viral RNA). In this study, Ct values of <26, 26-30, and ≥31 were regarded as high, intermediate, and low viral load, respectively.24 25

Continuous variables were expressed as medians (interquartile ranges) or means (± standard deviations), as appropriate. The Mann-Whitney U test and Kruskal-Wallis test were used to compare median values between two groups and among ≥3 groups, respectively. Categorical variables, expressed as proportions, were compared using the Chi squared test or Fisher’s exact test, as appropriate.

To identify factors independently associated with outcomes, variables with P values <0.1 in univariate analyses were subsequently entered into binary logistic regression multivariate analyses; odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. All statistical analyses were performed using SPSS (Windows version 26.0; IBM Corp, Armonk [NY], United States). Two-tailed P values <0.05 were considered statistically significant.

From 1 July to 30 September 2020, 414 patients with suspected COVID-19 were admitted to our hospital. This study included 219 patients who had SARS-CoV-2–positive rRT-PCR results in analyses of nasopharyngeal swab samples (from 213 recovered patients and six patients who had died). One hundred and ninety-five patients were excluded because of non–COVID-19 diagnosis, unconfirmed status, non-Asian ethnicity, or refusal to consent (Fig 1).

The mean patient age was 54.7 ± 17.5 years (range, 18-99), the male-to-female ratio was approximately 1:1, and 90.4% of the patients were Chinese (Table 1). The mean duration of hospitalisation was 9.87 ± 6.99 days and the overall mortality rate was 2.7%. The mean SARS-CoV-2 rRT-PCR Ct values for nasopharyngeal swab on admission was 24.2 ± 7.1. The median time to the first post-discharge visit was 38 days (range, 28-42) and the median duration of follow-up was 14 weeks (range, 13.1-15.5).

The three most frequent symptoms were upper respiratory symptoms: cough (51.5%), fever (42.5%), and sputum production (27.8%). Among the 219 patients with positive SARS-CoV-2 test results, 58 (26.5%) were asymptomatic and had undergone compulsory SARS-CoV-2 testing in accordance with the Public Health Ordinance. Of the 58 patients, 75.9% reported contact with identifiable index cases, such as household members, domestic helpers, or work colleagues.

Twenty patients presented with new rash. The incidence of new rash was 9.1% in this 3-month study period (95% CI=6.25%-14.4%). At the time of this study, there were no biomarkers or diagnostic tests for COVID-19–related cutaneous manifestations. Any new cutaneous manifestation not attributable to a previous/pre-existing skin disease or alternative diagnosis was considered COVID-19–related. Upon review by a dermatologist, six patients were diagnosed with localised urticarial eruptions after interferon injection treatment; 6.4% of patients (14/219) displayed various forms of COVID-19–related rash (Fig 2 and Table 2).

The most common manifestations were maculopapular exanthem (n=6, 42.9%, median Ct value: 24.8), followed by livedo reticularis (n=4, 28.6%, median Ct value: 21.3), varicella-like lesions (n=2, 14.3%, median Ct value: 19.3), urticaria (n=1, 7.1%, median Ct value: 14.4), and acral chilblain and petechiae (n=1, 7.1%, median Ct value: 33.1) [Fig 2]. The median Ct values for patients with and without rash were 22.9 and 24.1, respectively (P=0.58). The timing of symptom onset ranged from day 1 to day 9 (median, 4; mean, 4.28 ± 2.26). Skin symptoms were the sole symptoms in two patients with COVID-19 (0.91%), highlighting the importance of carefully evaluating patients who only display initial cutaneous symptoms or signs.

Compared with patients without rash, patients with rash were more likely to exhibit fever (OR=5.73; P=0.008) and display pulmonary infiltrates on chest X-ray (OR=5.06; P=0.013). Among patients with pulmonary infiltrates (n=19), four of them had rash. The episodes of desaturation requiring supplemental oxygen were less common in patients with rash (25%, 1/4) than in those without (93.3%, 14/15; OR=0.02, 95% CI=0.001-0.49; P=0.02). Furthermore, among these 19 patients with pulmonary infiltrates, systemic corticosteroids were less frequently required by patients with rash (25%, 1/4) than by those without (73.3%, 11/15; OR=0.12, 95% CI=0.01-1.53; P=0.10), but it was not statistically significant. There were no significant differences in age, sex, co-morbidities or Ct values between patients with and without rash. The estimated glomerular filtration rate (eGFR) was slightly lower in older patients without rash (P=0.024); 10.2% of these patients had CKD stage ≥3. In terms of outcomes, patients with and without rash had mortalities of 0.0% and 2.9%, respectively (P=0.97). The length of hospitalisation was similar in both groups (Table 3).

Patients aged ≥70 years had a significantly higher viral load (as reflected by a lower Ct value), compared with those aged <70 years (mean Ct value: 21.97 vs 24.65, P=0.03). Regardless of age, patients with hypertension and CKD stage ≥3 had a significantly higher viral load and lower initial Ct value on admission (OR=2.65, 95% CI=1.08-6.45 and OR=3.65, 95% CI=1.18-11.3, respectively; both P<0.05).

All six patients who died were men; their mean age was 87.0 ± 7.3 years. The rates of hypertension, diabetes mellitus, a glycated haemoglobin level of ≥6.5%, CKD stage ≥3, and higher viral load (ie, lower Ct value on admission) were significantly greater among patients who died than among those who survived (Table 4). Older age, hypertension, and low eGFR were associated with a higher risk of mortality (all P<0.05) [Table 5].

Treatment varied in this cohort because there was no standard of care in the early days of the COVID-19 pandemic. Symptomatic treatment was administered to 53 patients (24.2%); 166 patients (75.8%) received early treatment within the first week of symptom onset, including interferon beta-1b and ribavirin, which were administered based on the results of a triple therapy clinical study.21 Emollients and topical corticosteroids of mild to moderate potency (1% hydrocortisone cream and 0.1% mometasone furoate cream) were prescribed for symptomatic relief.

During follow-up, we observed that urticarial eruption after interferon injection resolved within 10 to 14 days upon completion of treatment. With respect to COVID-19–related skin eruptions, most lesions (maculopapular exanthem, livedo reticularis, and urticaria) were self-limiting and spontaneously resolved without specific treatment; there were no severe sequelae. Two patients with varicella-like lesions had mild post-inflammatory hyperpigmentation without scarring.

Cutaneous manifestations of the COVID-19 pandemic have been gaining increasing attention because they may be useful in the early diagnosis of COVID-19, triage of patients with SARS-CoV-2–positive test results, and risk stratification. There is speculation that the mechanism involves the direct action of SARS-CoV-2 on tissues, the complement/interferon-driven immune response, and the coagulation system; alternatively, it involves nonspecific skin symptoms of systemic viral infection.16 17 22 26 27 28 Although more investigations are needed, it is possible that some symptoms are clinical signs of milder COVID-19, whereas others are indicators of more severe clinical illness.

Our study showed that patients with confirmed COVID-19 could display various cutaneous manifestations. The most common manifestation attributed to COVID-19 was maculopapular exanthem, followed by livedo reticularis. Because most skin lesions were transient and self-limiting, skin biopsy was only performed in one patient. In that 40-year-old female patient, skin biopsy of the left trunk revealed low to moderate numbers of perivascular lymphocytes and histiocytes, as well as sparse eosinophils, in the superficial dermis; focal parakeratosis was present in the epidermis. There was no evidence of vasculitis or interfacial changes. These findings were compatible with maculopapular exanthem.

In previous reports, erythema multiforme–like lesions, chilblain-like acral eruptions, and livedo erythema were identified in children and young adult patients with asymptomatic or mild disease.26 28 29 In contrast, maculopapular rash and acro-ischaemic lesions were often observed among adult patients with more severe disease. Among our patients who presented with rash, there were no instances of mortality; the duration of hospitalisation was similar regardless of rash status. The results of a previous study has suggested that the cutaneous manifestation is the only manifestation of COVID-19 in some patients30; thus, careful documentation of any cutaneous symptoms during the COVID-19 pandemic may be necessary for early recognition and diagnosis.30 Additionally, urticaria with fever has diagnostic implications because this combination may be an early symptom of subsequently confirmed SARS-CoV-2 infection.19 In our cohort, patients with cutaneous manifestations were more likely to present with fever. Although most of our patients had symptoms other than rash alone, two patients (0.9%) presented with rash only (one with urticaria and one with maculopapular exanthem); the clinical significance of these symptoms should not be ignored. Informal extrapolation of these results to the general population in Hong Kong suggested that 2477 cases (2/219; ie, 0.91% × 272 235 confirmed cases)4 solely involve rash presentation; these patients would remain undiagnosed if they did not undergo SARS-CoV-2 testing. This lack of diagnosis is a potential health threat and could facilitate viral spread.

In our cohort, the incidence of new rash was 13.6%. In the study by Guan et al31 in China, the prevalence of rash was much lower in patients with COVID-19 (0.2%; 2/1099). In that study, patients with rash may have been underdiagnosed because patients with suspected COVID-19 were managed by general practitioners or hospitalists who had less familiarity with cutaneous manifestations.31 In contrast, our patients underwent prompt assessment by in-hospital dermatologists to detect cutaneous manifestations. In an Italian study, the prevalence of rash presentation was much higher (20.4%),13 presumably because asymptomatic patients were excluded through a lack of testing. However, if we exclude the 58 asymptomatic patients in our cohort (all of whom underwent compulsory testing in accordance with the Public Health Ordinance), the incidence of new rash in our study was 16.7% (95% CI=14.5-18.8), which remains lower than the incidence in the Italian study. We speculate that this difference is related to the early initiation of combined treatment (ribavirin and interferon beta-1b) in our cohort, which may modify or halt the SARS-CoV-2–induced inflammatory process.21 Importantly, the genomic characteristics of SARS-CoV-2 spread are under investigation worldwide; this approach helps identify transmission routes in various regions. In a case series in the United States, SARS-CoV-2 genomes in one region were predominantly associated with isolates that originated in Europe (>80%), similar to the distributions of viral strains in other regions in the United States32; a smaller subgroup of SARS-CoV-2 genomes displayed similarity to strains that originated in Asia (15%), indicating multiple sources of viral spread within the community.32 Differences in the prevalences of cutaneous manifestations may represent variations in SARS-CoV-2 genomic characteristics among regions; in Hong Kong, a cosmopolitan city with many travellers from mainland China and other countries, the prevalences of cutaneous manifestations may be the result of viral strains from all provinces of China as well as Europe and other regions. Further studies are needed concerning genomic variations and clinical manifestations.

In terms of viral load and prognosis, higher viral load on admission was significantly associated with greater mortality in patients with older age, history of hypertension, and CKD stage ≥3. Univariate analysis showed that the risk of mortality was the greatest among patients with older age, hypertension, higher glycated haemoglobin level, and renal impairment. Multivariate Cox regression analysis confirmed that older age, hypertension, and low eGFR were significantly associated with greater mortality risk.

Conversely, patients with renal impairment were less likely to present with rash, suggesting that the immune response is weaker in patients with renal impairment. However, the length of hospitalisation was similar regardless of cutaneous manifestations; the presence of cutaneous manifestations was not associated with other co-morbidities. There was no clear association between Ct values and rash occurrence. Additional studies with larger sample sizes may be necessary to explore the relationship between rash subtype and viral load.

The results of a previous study suggested that cutaneous manifestations of COVID-19 were related to immunological responses rather than the direct results of viral invasion17; cutaneous manifestations may be an early sign of immunological responses elsewhere in the body, similar to pulmonary infiltrates secondary to cytokine storm. The present study showed that the incidence of pulmonary infiltrates was considerably higher among patients with rash (28.6%) than among those without (7.3%) [Table 3]; conversely, patients with rash were less likely to display further deterioration, such as oxygen desaturation and a requirement for oxygen supplementation (P=0.016). Only one patient with rash (25%) received dexamethasone, whereas multiple patients without rash required such treatment (73.3%) [P=0.11]. Another explanation is that, overall, patients with rash tended to seek medical attention earlier than those without, which would increase the likelihood of prompt treatment. A previous study has indicated that patients with cutaneous manifestations may have a better prognosis because those patients develop a more robust immune response.17

In patients with new pulmonary infiltrates as well as evidence of respiratory decompensation/failure (eg, desaturation and/or tachypnoea), systemic corticosteroids have been used to prevent tissue destruction from cytokine storm after other causes had been ruled out. In this context, patients receiving systemic corticosteroids had more severe disease that involved evidence or features of respiratory decompensation and carried a greater risk of mortality.

In the present study, after the exclusion of patients with nosocomial/secondary bacterial pneumonia, heart failure, or pulmonary changes related to prior disease, 19 patients (8.7%) had new pulmonary infiltrates on admission. All 19 patients received interferon beta-1b and ribavirin treatment; 12 patients received dexamethasone (daily dosage range, 6-8 mg; mean duration, 8.63 ± 2.53 days) [Table 4]. Among the 12 patients receiving dexamethasone, five patients (41.7%) died despite the use of systemic corticosteroids, together with empirical antibiotics, interferon beta-1b, and ribavirin; in contrast, only one death (14.3%) occurred among seven patients receiving interferon beta-1b and ribavirin without corticosteroids (OR=4.28, 95% CI=0.38-47.6; P=0.23). The mean interval from symptom onset to systemic corticosteroid initiation was shorter among patients who recovered than among those who died (5.14 ± 2.14 days vs 8.61 ± 2.30 days; P=0.0026). These results suggest that the early use of systemic corticosteroids may lead to a better survival outcome.

Although no deaths occurred among patients with cutaneous manifestations, the mortality rate did not significantly differ from the rate of 2.9% among patients without rash. Most patients received treatment within the first week after diagnosis of COVID-19 (according to detection of SARS-CoV-2–specific immunoglobulin G within 14 days after symptom onset; mean, 7.71 ± 3.05 days; range, 4-13), which may have improved disease outcomes and shortened hospitalisation. These findings highlighted the importance of early treatment beginning at symptom onset (ie, in the first week) and supported the use of interferon therapy described in a previous report.21

First, this study had a small number of patients. Second, there was potential selection bias because only hospitalised patients with SARS-CoV-2–positive test results were included in the analysis; patients with COVID-19 who did not undergo screening or seek medical consultation were not diagnosed, and thus they were excluded from the study. Third, Ct value analysis was not conducted according to rash subtype and severity because of the limited number of patients. Fourth, some viral laboratory tests (eg, test for human herpesvirus 6) were not routinely available in our hospital, which may have hindered the interpretation of possible causes of rash or the identification of coexisting infections. Nevertheless, most other possible viral infections were excluded from this study. Additional studies with larger sample sizes and comparisons with treatment outcomes are needed.

This study did not demonstrate direct relationships among rash, viral load, and mortality. Furthermore, cutaneous manifestations may be early signs of immunological responses (similar to pulmonary infiltrates). Patients with older age, hypertension, and renal impairment have greater mortality risk and higher viral load. These high-risk groups should be prioritised in early screening and vaccination efforts to avoid poor clinical outcomes.

Concept or design: CSM Wong, IFN Hung.
Acquisition of data: CSM Wong, MMH Chung, MWM Chan, AKC Cheng, YM Lau.
Analysis or interpretation of data: CSM Wong, IFN Hung. Drafting of the manuscript: CSM Wong, IFN Hung, CK Yeung, HHL Chan.
Critical revision of the manuscript for important intellectual content: CSM Wong, IFN Hung, MYW Kwan, CK Yeung, HHL Chan, CS Lau.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors declare no conflicts of interest.

The authors thank all patients for their participation.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

This research was approved by the Institutional Review Board of The University of Hong Kong/Hospital Authority Hong Kong West Cluster (Ref No.: UW20-725) and was conducted in full compliance with the ICH E6 guideline for Good Clinical Practice and the principles of the Declaration of Helsinki. Appropriate patient consent was obtained for clinical information and images to be publicly reported. All participants’ clinical data and reports were deidentified to maintain anonymity.

1. Chen N, Zhou M, Dong X, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet 2020;395:507-13. Crossref

2. Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med 2020;382:727-33. Crossref

3. World Health Organization. WHO coronavirus (COVID-19) dashboard. Situation by region, country, territory & area. Available from: https://covid19.who.int/table. Accessed 7 May 2022.

4. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Latest situation of coronavirus disease (COVID-19) dashboard in Hong Kong. Available from: https://chp-dashboard.geodata.gov.hk/covid-19/en.html. Accessed 7 May 2022.

5. Ye Q, Wang B, Mao J. The pathogenesis and treatment of the ‘cytokine storm’ in COVID-19. J Infect 2020;80:607-13. Crossref

6. Joob B, Wiwanitkit V. COVID-19 can present with a rash and be mistaken for dengue. J Am Acad Dermatol 2020;82:e177. Crossref

7. Hassan K. Urticaria and angioedema as a prodromal cutaneous manifestation of SARS-CoV-2 (COVID-19) infection. BMJ Case Rep 2020;13:e236981. Crossref

8. Andina D, Noguera-Morel L, Bascuas-Arribas M, et al. Chilblains in children in the setting of COVID-19 pandemic. Pediatr Dermatol 2020;37:406-11. Crossref

9. Marzano AV, Genovese G, Fabbrocini G, et al. Varicella-like exanthem as a specific COVID-19-associated skin manifestation: multicenter case series of 22 patients. J Am Acad Dermatol 2020;83:280-5. Crossref

10. De Giorgi V, Recalcati S, Jia Z, et al. Cutaneous manifestations related to coronavirus disease 2019 (COVID-19): a prospective study from China and Italy. J Am Acad Dermatol 2020;83:674-5. Crossref

11. Tammaro A, Adebanjo GA, Parisella FR, Pezzuto A, Rello J. Cutaneous manifestations in COVID-19: the experiences of Barcelona and Rome. J Eur Acad Dermatol Venereol 2020;34:e306-7. Crossref

12. Marraha F, Al Faker I, Gallouj S. A review of the dermatological manifestations of coronavirus disease 2019 (COVID-19). Dermatol Res Pract 2020;2020:9360476. Crossref

13. Recalcati S. Cutaneous manifestations in COVID-19: a first perspective. J Eur Acad Dermatol Venereol 2020;34:e212-3. Crossref

14. Galván Casas C, Català A, Carretero Hernández G, et al. Classification of the cutaneous manifestations of COVID-19: a rapid prospective nationwide consensus study in Spain with 375 cases. Br J Dermatol 2020;183:71-7. Crossref

15. Zhao Q, Fang X, Pang Z, Zhang B, Liu H, Zhang F. COVID-19 and cutaneous manifestations: a systematic review. J Eur Acad Dermatol Venereol 2020;34:2505-10. Crossref

16. Hamming I, Timens W, Bulthuis ML, Lely AT, Navis G, van Goor H. Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis. J Pathol 2004;203:631-7. Crossref

17. Novak N, Peng W, Naegeli MC, et al. SARS-CoV-2, COVID-19, skin and immunology—what do we know so far? Allergy 2021;76:698-713. Crossref

18. Li MY, Li L, Zhang Y, Wang XS. Expression of the SARS-CoV-2 cell receptor gene ACE2 in a wide variety of human tissues. Infect Dis Poverty 2020;9:45. Crossref

19. Rahimi H, Tehranchinia Z. A comprehensive review of cutaneous manifestations associated with COVID-19. Biomed Res Int 2020;2020:1236520. Crossref

20. Dadras O, Afsahi AM, Pashaei Z, et al. The relationship between COVID-19 viral load and disease severity: a systematic review. Immun Inflamm Dis 2022;10:e580. Crossref

21. Hung IF, Lung KC, Tso EY, et al. Triple combination of interferon beta-1b, lopinavir-ritonavir, and ribavirin in the treatment of patients admitted to hospital with COVID-19: an open-label, randomised, phase 2 trial. Lancet 2020;395:1695-704. Crossref

22. Wollina U, Karadağ AS, Rowland-Payne C, Chiriac A, Lotti T. Cutaneous signs in COVID-19 patients: a review. Dermatol Ther 2020;33:e13549. Crossref

23. Infectious Diseases Society of America; Association for Molecular Pathology. IDSA and AMP joint statement on the use of SARS-CoV-2 PCR cycle threshold (Ct) values for clinical decision-making. Available from: https://www.idsociety.org/globalassets/idsa/public-health/covid-19/idsa-amp-statement.pdf. Accessed 12 Apr 2022.

24. Aranha C, Patel V, Bhor V, Gogoi D. Cycle threshold values in RT-PCR to determine dynamics of SARS-CoV-2 viral load: an approach to reduce the isolation period for COVID-19 patients. J Med Virol 2021;93:6794-7. Crossref

25. Cevik M, Tate M, Lloyd O, Maraolo AE, Schafers J, Ho A. SARS-CoV-2, SARS-CoV, and MERS-CoV viral load dynamics, duration of viral shedding, and infectiousness: a systematic review and meta-analysis. Lancet Microbe 2021;2:e13-22. Crossref

26. Genovese G, Moltrasio C, Berti E, Marzano AV. Skin manifestations associated with COVID-19: current knowledge and future perspectives. Dermatology 2021;237:1-12. Crossref

27. Bastard P, Rosen LB, Zhang Q, et al. Autoantibodies against type I IFNs in patients with life-threatening COVID-19. Science 2020;370:eabd4585. Crossref

28. Chua GT, Wong JS, Lam I, et al. Clinical characteristics and transmission of COVID-19 in children and youths during 3 waves of outbreaks in Hong Kong. JAMA Netw Open 2021;4:e218824. Crossref

29. Chua GT, Wong JS, Chung J, et al. Paediatric multisystem inflammatory syndrome temporally associated with SARS-CoV-2: a case report. Hong Kong Med J 2022;28:76-8. Crossref

30. Leung TY, Chan AY, Chan EW, et al. Short- and potential long-term adverse health outcomes of COVID-19: a rapid review. Emerg Microbes Infect 2020;9:2190-9. Crossref

31. Guan WJ, Ni ZY, Hu Y, et al. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med 2020;382:1708-20. Crossref

32. Zhang W, Govindavari JP, Davis BD, et al. Analysis of genomic characteristics and transmission routes of patients with confirmed SARS-CoV-2 in southern California during the early stage of the US COVID-19 pandemic. JAMA Netw Open 2020;3:e2024191. Crossref

In the early days of the coronavirus disease 2019 (COVID-19) pandemic, telemedicine was recommended as a solution to provide safe access to healthcare.1 In 2020, the World Health Organization reported that most global health authorities regarded telemedicine as a potential method to provide services for patients with non-communicable diseases.2 For example, in a national survey of healthcare providers in Germany, approximately 60% of participants reported routine or partial use of telemedicine during the COVID-19 pandemic.3 In the US in 2020, telemedicine was used for COVID-19 screening, monitoring of patients with positive COVID-19 test results, management of chronic diseases, and virtual monitoring and follow-up.4 Telemedicine reduces patient travel costs and improves access, while reducing the use of personal protective equipment and the risk of COVID-19 transmission. In China, ‘internet hospitals’ offered essential medical support to the public during the early days of the COVID-19 pandemic.5

However, from a patient perspective, there are limitations and barriers to the use of telemedicine. A qualitative study conducted in Hong Kong in 20166 revealed that patient concerns included technical and logistical issues (eg, difficulty in accessing and using computers), limited personal interactions (eg, lack of in-person physical examinations and risk of incorrect diagnosis related to poor communication), concerns regarding cybersecurity and safety, and problems with prescriptions (eg, distrust of local community pharmacies). In 2020, clinical guidelines were published concerning the performance of remote primary care assessments and treatments for patients with suspected COVID-19 in the United Kingdom.7 During the COVID-19 pandemic, a digital health ecosystem may benefit the healthcare system, as well as the broader population (eg, through tracking and communication strategies) and research and health technology sectors (eg, online activity monitoring and digital support for isolation and quarantine situations).8

Telemedicine consists of remote healthcare service delivery by healthcare professionals for diagnosis, treatment, and prevention efforts, as well as research and continuing education.9 The fundamental goal of telemedicine is to increase access to care, thereby serving populations that otherwise would not receive timely medical evaluation and treatment. Research concerning telemedicine effectiveness has been controversial.10 11 12 13 14 A previous systematic review found that many studies showed no difference between telemedicine and usual care, and there remains limited evidence concerning the cost-effectiveness of telemedicine.10 With respect to chronic disease management via telemedicine, another review found publication bias and a tendency to report only short-term outcomes.11 Moreover, a Cochrane review revealed improved control of blood glucose among patients with diabetes using telemedicine, compared with patients receiving in-person care; conversely, the care modality did not affect health outcomes among patients with heart failure.12 Regarding the impacts and costs of telemedicine services, economic analyses have been problematic because of complex and unpredictable collaborative achievement processes13; generalisability has been limited by poor quality and low reporting standards.14 There is a need to focus on patient perspectives and telemedicine innovations.

In 2020, the health system in Hong Kong was considerably impacted by COVID-19 transmission risk and public health responses.15 During the early days of the COVID-19 pandemic, the number of in-person medical consultations decreased worldwide.16 17 In the Netherlands, the use of telemedicine offset this decrease.17 Before the COVID-19 pandemic, telemedicine was not widely used in Hong Kong6 18; user perspectives concerning the role of telemedicine during the COVID-19 pandemic and future pandemics require investigation. This study of telemedicine users in Hong Kong explored their perceptions of telemedicine use during the COVID-19 pandemic, their perceptions of disease risk, and their COVID-19 preparedness measures. We hypothesised that telemedicine users have distinct perceptions of risk, compared with the general public; we sought to determine the effects of such differences on the delivery of health services.

We conducted a cross-sectional online survey from 6 April to 11 May 2020. Participants in this study were users of COVID-19 testing services provided by The Chinese University of Hong Kong (CUHK) Medical Clinic,19 a multispecialty clinic offering specialist consultations, health screenings, vaccinations, and COVID-19 testing services.

All service users completed a telemedicine consultation, followed by COVID-19 testing. The consultation, delivered using a standard telephone, focused on assessment of the user’s COVID-19 risk prior to the time of testing. Users were offered a deep throat saliva test for COVID-19 detection. After the telemedicine consultation and COVID-19 test, all users were invited to complete an anonymous online survey within 1 month for service evaluation and data collection.

The online survey consisted of 24 questions regarding telemedicine services, COVID-19 risk perception, and preparedness measures in Hong Kong. The survey specifically focused on reasons for using CUHK telemedicine and testing services, user perceptions and attitudes regarding telemedicine consultations, user perceptions and attitudes regarding the COVID-19 pandemic and prevention, and user characteristics (eg, age and sex). Questions about user perceptions and attitudes regarding the current COVID-19 pandemic and prevention were phrased in a manner identical to a previous study,20 allowing direct comparison with survey results from a study focused on the Hong Kong general public. The survey was pilot tested and subsequently modified to ensure content validity. Most questions included ‘yes’ or ‘no’ answers and a 5-point Likert scale. Written consent to take part in the study was obtained from all participants before they began the survey.

Descriptive statistics were reported for participant characteristics, perceptions, and attitudes regarding telemedicine consultations. Perceptions and attitudes regarding the COVID-19 pandemic and prevention were recorded in a manner that allowed comparison with the results of a prior COVID-19–focused telephone survey of the Hong Kong general public.20 The prior telephone survey—a cross-sectional, population-based landline telephone survey using a computerised random-digit dialling method—included 765 adult Hong Kong residents during the period from 22 March to 1 April 2020. The participants in that study were representative of Hong Kong Census population data with respect to age, sex, marital status, and residential district, although they had higher levels of education and household income.20 Chi squared tests were used for comparisons between the prior survey and the present survey. The threshold of statistical significance was set at 0.05. All statistical analyses were conducted using SPSS (Windows version 21.0; IBM Corp, Armonk [NY], US).

Among 187 telemedicine users of COVID-19 testing services at the CUHK Medical Clinic, we identified 150 responses to the online survey during the period from 6 April to 11 May 2020. In total, 141 telemedicine users (response rate of 75.4%) were willing to participate in this study. Table 1 shows the participants’ characteristics. Notably, 56.1% were men, over half (50.7%) were aged 18 to 24 years (all participants were aged <65 years), most (59.4%) lived in private housing, and most (65.4%) resided in the New Territories. The most common reason for COVID-19 testing was a work-related requirement (56.7%, 80/141), followed by recent international travel (39.7%, 56/141); 14.9% (21/141) of the participants sought testing because of concerns about the spread of COVID-19. Overall, 14.2% (20/141) of the participants had either been in close contact with a confirmed case or suspected that they had symptoms of COVID-19.

In total, 95.1% (116/122) participants believed that COVID-19 telemedicine consultations were useful. Most participants believed that telemedicine consultations could perform the functions of ‘health protection, promotion and disease prevention’ (73.6%) and ‘diagnosis’ (64.0%) [Fig 1]. Concerning the choice of telemedicine provider, almost all participants (99.2%) would accept medical doctors; more than half of the participants (54.1%) would accept trained nurses, but only 13.1% would accept non-clinical staff who had been trained to provide telemedicine services.

Nearly half of the participants (49.0%) agreed or strongly agreed that telemedicine consultations were appropriate during the COVID-19 pandemic (Fig 2). More than half (61.0%) of the participants reported satisfaction with telemedicine consultations (‘agree’ or ‘strongly agree’) and services provided by clinic staff (73.8% responded ‘agree’ or ‘strongly agree’). However, only 36.2% agreed or strongly agreed that service quality was identical between telemedicine consultations and in-person consultations.

Household capacities for potential COVID-19—related quarantine were compared between telemedicine users in this study and respondents in the prior Hong Kong general public telephone survey20 (Table 2). Telemedicine users reported having less space at home, fewer masks, fewer gloves, and less detergent for potential quarantine situations, compared with respondents in the Hong Kong general public telephone survey (all P<0.05).

In response to the question ‘How many designated caregivers are appropriate for each isolated/quarantined person?’, telemedicine users were less likely to answer correctly (limit to one main carer, 8.1%), compared with respondents in the Hong Kong general public telephone survey (51.3%)20 [P<0.001, Chi squared with continuity correction].

Most telemedicine users (76.4%) believed that household prevention could prevent COVID-19; approximately 67% of them believed that they had sufficient knowledge to manage COVID-19—related health risks. These percentages were higher than the percentages in the Hong Kong general public telephone survey20 (Table 3). Additionally, >80% of telemedicine users believed that Hong Kong had achieved better control of COVID-19, compared with other major cities.

Regarding their main channel for infectious disease information, approximately 64% of participants were using the internet or mobile applications, whereas 24% were using television; these percentages differed from the Hong Kong general public telephone survey, in which 56.5% of respondents used the internet or mobile applications and 36.2% used television20 (P<0.001). Overall, telemedicine users preferred their main channel for infectious disease information to be the internet or mobile applications and personal sources (eg, family, friends, or healthcare professionals).

This study demonstrated high satisfaction with telemedicine consultations among users and revealed that users considered telemedicine to be appropriate during the COVID-19 pandemic. Despite the perception that telemedicine users have sufficient knowledge to manage health risks from COVID-19, when responses were compared between telemedicine users and the Hong Kong general public, we found that household preventative measures were inadequate among telemedicine users.

Most of our telemedicine users did not agree that quality was identical between telemedicine consultations and in-person consultations. This perspective possibly resulted from the provision of telemedicine consultations via telephone without a video component; moreover, the users might not have been familiar with the concept of telemedicine consultation. Indeed, a 2014 nationwide survey in the US revealed that only 15% of family physicians reported using telemedicine in the previous year; barriers included a lack of training, a lack of reimbursement, the cost of equipment, and potential liability issues.21 Furthermore, Schwamm22 has described telemedicine as a disruptive technology that may threaten traditional healthcare delivery. Obstacles to the expansion of telemedicine in the US include state-level statutes that require the clinician to be located in the same state as the patient and to have previously completed an in-person consultation with that patient.23 Similar regulatory requirements exist in Hong Kong.24

Various studies in the US have detected increasing uptake of telemedicine, particularly in primary care.25 26 27 28 29 A study of telemedicine users in a large commercially insured population in the US from 2005 to 2017 showed that the mean age was 38.3 years; on average, users of primary care telemedicine were younger than users of telemental healthcare and more likely to reside in urban areas.25 A study regarding Teladoc, one of the largest telemedicine providers in the US, revealed that Teladoc users were younger and less likely to have used healthcare before the introduction of Teladoc; they were also less likely to have a follow-up visit in any setting, compared with patients who visited a clinic for a similar condition.26 These findings are consistent with our observations that most telemedicine users were young; they also suggest that the use of telemedicine services can provide opportunities for healthcare access that are not otherwise available.

Intriguingly, although more telemedicine users agreed that household preparation could prevent COVID-19, they were less likely to believe that their household preventative measures were adequate, compared with respondents in the Hong Kong general public telephone survey.20 This disparity may be attributed to differences in participant characteristics: telemedicine users in this study were younger (51% aged 18-24 years, vs 9% in the Hong Kong general public20), were male (56% vs 47%20), and were living in the New Territories (65% vs 51%20). Telemedicine users were also more likely to use the internet (64% vs 57%20) as the main channel for infectious disease information and to prefer using the internet for such information (62% vs 54%20). These findings have important implications for the use of telemedicine to fill gaps in health promotion and disease prevention. Wu et al30 found that secondary cases from household transmission of COVID-19 were common in Zhuhai, China, and one-third of these secondary cases were asymptomatic. Sufficient household preparation measures are needed to limit the spread of COVID-19.31

Telemedicine in Hong Kong had a ‘late start’ (in 1994); in 1998, Hjelm32 predicted that the telemedicine would be rapidly implemented in Hong Kong. However, in the December 2019 version of the Ethical Guidelines on Practice of Telemedicine by the Medical Council of Hong Kong, it was noted that telemedicine in Hong Kong has not fully developed.24 Despite the obvious convenience benefit and reduced risk of COVID-19 transmission, the guidelines remind practitioners that they remain fully responsible for legal and ethical obligations during telemedicine consultations.24 Furthermore, the guidelines mention that standards of care to protect patients are applicable during in-person and telemedicine consultations; practitioners should familiarise themselves with the World Medical Association Statement on the Ethics of Telemedicine.33

Training and patient assessment guidelines for healthcare practitioners are urgently needed, considering the unique circumstances surrounding the use of telemedicine, such as technology (eg, technical limitations of virtual consultations, including assessments), patient education and informed consent, cybersecurity, and other concerns.6 These guidelines would ensure that the same standards of telemedicine consultations can be implemented, as described by the Medical Council of Hong Kong.

The present study focused on 141 telemedicine users of a single private clinic providing deep throat saliva polymerase chain reaction tests for detection of severe acute respiratory syndrome coronavirus 2; thus, the study participants may not be representative of all telemedicine users in Hong Kong. Responder bias may have affected the results (the response rate was 75.4%), but it was not possible to compare participants with individuals who refused to participate. Furthermore, the recent experience of a telemedicine consultation may have biased participants’ responses in favour of telemedicine.

Concerning the comparison of COVID-19 risk perception and preparedness measures, methodological discrepancies between the online survey of telemedicine users and the telephone survey of the Hong Kong general public may have led to differences in responses, particularly with respect to including younger and more computer-literate individuals in the online survey. Although the present study was conducted from 6 April to 11 May 2020 (after a surge of COVID-19 cases in Hong Kong), the Hong Kong general public telephone survey used for comparison was conducted from 22 March to 1 April 2020 (during a surge in COVID-19 cases).20 The difference in data collection periods may have contributed to different perceptions of COVID-19 preparedness. Various physical distancing measures were implemented during the COVID-19 surge (from late March to early April), which may also have contributed to differences between the telephone survey and online survey cohorts.

Finally, because of sample size limitations and problems with representativeness, the findings of the study may be restricted to understanding views regarding telemedicine consultations among participants in the present study. However, factors such as young age, residence in private housing, and residence in the New Territories may have contributed to response bias; because no information was collected concerning occupation, education level, income, or ethnicity, we could not control for bias related to these factors.

In this study, telemedicine users in Hong Kong agreed that telemedicine consultations were appropriate during the COVID-19 pandemic. Participants agreed that telemedicine consultations could perform the functions of health protection, promotion, disease prevention, and diagnosis. The use of telemedicine for screening and patient education can be encouraged during the COVID-19 pandemic in Hong Kong.

Concept or design: KKC Hung, EYY Chan, JCY Wu.
Acquisition of data: KKC Hung, JCY Wu, CA Graham.
Analysis or interpretation of data: KKC Hung, EYY Chan, ESK Lo, Z Huang.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: KKC Hung.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

This research was approved by the Survey and Behavioural Research Ethics Committee of The Chinese University of Hong Kong (Ref No.: SBRE-19-730). Patients were treated in accordance with the tenets of the Declaration of Helsinki and provided written informed consent for all treatments and procedures, as well as publication of their anonymised data.

1. Hollander JE, Carr BG. Virtually perfect? Telemedicine for COVID-19. N Engl J Med 2020;382:1679-81. Crossref

2. World Health Organization. Rapid assessment of service delivery for NCDs during the COVID-19 pandemic. Available from: https://www.who.int/publications/m/item/rapid-assessment-of-service-delivery-for-ncds-during-the-covid-19-pandemic. Accessed 10 Aug 2020.

3. Peine A, Paffenholz P, Martin L, Dohmen S, Marx G, Loosen SH. Telemedicine in Germany during the COVID-19 pandemic: multi-professional national survey. J Med Internet Res 2020;22:e19745. Crossref

4. Ford D, Harvey JB, McElligott J, et al. Leveraging health system telehealth and informatics infrastructure to create a continuum of services for COVID-19 screening, testing, and treatment. J Am Med Inform Assoc 2020;27:1871-7. Crossref

5. Gong K, Xu Z, Cai Z, Chen Y, Wang Z. Internet hospitals help prevent and control the epidemic of COVID-19 in China: multicenter user profiling study. J Med Internet Res 2020;22:e18908. Crossref

6. Kung K, Wong HF, Chen JY. An exploratory qualitative study on patients’ views of medical e-consultation in a public primary care setting. Hong Kong Pract 2016;38:120-7.

7. Greenhalgh T, Koh GC, Car J. COVID-19: a remote assessment in primary care. BMJ 2020;368:m1182. Crossref

8. Fagherazzi G, Goetzinger C, Rashid MA, Aguayo GA, Huiart L. Digital health strategies to fight COVID-19 worldwide: challenges, recommendations, and a call for papers. J Med Internet Res 2020;22:e19284. Crossref

9. World Health Organization. A health telematics policy in support of WHO’s Health-for-all strategy for global health development: report of the WHO Group Consultation on Health Telematics, 11-16 December, Geneva, 1997. 1998. Available from: https://apps.who.int/iris/handle/10665/63857. Accessed 7 Sep 2023.

10. McLean S, Sheikh A, Cresswell K, et al. The impact of telehealthcare on the quality and safety of care: a systematic overview. PLoS One 2013;8:e71238. Crossref

11. Wootton R. Twenty years of telemedicine in chronic disease management—an evidence synthesis. J Telemed Telecare 2012;18:211-20. Crossref

12. Flodgren G, Rachas A, Farmer AJ, Inzitari M, Shepperd S. Interactive telemedicine: effects on professional practice and health care outcomes. Cochrane Database Syst Rev 2015;2015:CD002098. Crossref

13. Ekeland AG, Bowes A, Flottorp S. Effectiveness of telemedicine: a systematic review of reviews. Int J Med Inform 2010;79:736-71. Crossref

14. Eze ND, Mateus C, Cravo Oliveira Hashiguchi T. Telemedicine in the OECD: an umbrella review of clinical and cost-effectiveness, patient experience and implementation. PLoS One 2020;15:e0237585. Crossref

15. Hung KK, Walline JH, Graham CA. COVID-19: emergency medicine perspectives from Hong Kong. Eur J Emerg Med 2020;27:163-4. Crossref

16. Hung KK, Walline JH, Chan EY, et al. Health service utilization in Hong Kong during the COVID-19 pandemic—a cross-sectional public survey. Int J Health Policy Manag 2022;11:508-13. Crossref

17. Auener S, Kroon D, Wackers E, Dulmen SV, Jeurissen P. COVID-19: a window of opportunity for positive healthcare reforms. Int J Health Policy Manag 2020;9:419-22. Crossref

18. Sheng OR, Hu PJ, Chau PY, et al. A survey of physicians’ acceptance of telemedicine. J Telemed Telecare 1998;4 Suppl 1:100-2. Crossref

19. CUHK Medical Clinic. COVID-19 Testing Service (Company/Groups). Available from: https://cuclinic.hk/programmes/covid-19-testing-service/covid-19-testing-testing-service-company-group. Accessed 30 Jul 2020.

20. Chan EY, Huang Z, Lo ES, Hung KK, Wong EL, Wong SY. Sociodemographic predictors of health risk perception, attitude and behavior practices associated with Health-Emergency Disaster Risk Management for biological hazards: the case of COVID-19 pandemic in Hong Kong, SAR China. Int J Environ Res Public Health 2020;17:3869. Crossref

21. Moore MA, Coffman M, Jetty A, Petterson S, Bazemore A. Only 15% of FPs report using telehealth; training and lack of reimbursement are top barriers. Am Fam Physician 2016;93:101.

22. Schwamm LH. Telehealth: seven strategies to successfully implement disruptive technology and transform health care. Health Aff (Millwood) 2014;33:200-6. Crossref

23. Pearl R. Kaiser Permanente Northern California: current experiences with internet, mobile, and video technologies. Health Aff (Millwood) 2014;33:251-7. Crossref

24. The Medical Council of Hong Kong. Guidelines for all registered medical practitioners. Available from: https://www.mchk.org.hk/files/newsletter-26th.pdf. Accessed 30 Jul 2020.

25. Barnett ML, Ray KN, Souza J, Mehrotra A. Trends in telemedicine use in a large commercially insured population, 2005-2017. JAMA 2018;320:2147-9. Crossref

26. Uscher-Pines L, Mehrotra A. Analysis of Teladoc use seems to indicate expanded access to care for patients without prior connection to a provider. Health Aff (Millwood) 2014;33:258-64. Crossref

27. Neufeld JD, Doarn CR. Telemedicine spending by Medicare: a snapshot from 2012. Telemed J E Health 2015;21:686-93. Crossref

28. Mehrotra, A. The convenience revolution for treatment of low-acuity conditions. JAMA 2013;310:35-6. Crossref

29. Courneya PT, Palattao KJ, Gallagher JM. HealthPartners’ online clinic for simple conditions delivers savings of $88 per episode and high patient approval. Health Aff (Millwood) 2013;32:385-92. Crossref

30. Wu J, Huang Y, Tu C, et al. Household transmission of SARS-CoV-2, Zhuhai, China, 2020. Clin Infect Dis 2020;71:2099-108. Crossref

31. Chan EY, Gobat N, Kim JH, et al. Informal home care providers: the forgotten health-care workers during the COVID-19 pandemic. Lancet 2020;395:1957-9. Crossref

32. Hjelm NM. Telemedicine: academic and professional aspects. Hong Kong Med J 1998;4:289-92.

33. World Medical Association. WMA Statement on the Ethics of Telemedicine. Available from: https://www.wma.net/policies-post/wma-statement-on-the-ethics-of-telemedicine/. Accessed 15 Apr 2021.

Because of its close geographical proximity to mainland China, Hong Kong was one of the first regions outside of the Mainland to be affected by the severe acute respiratory syndrome (SARS) and coronavirus disease 2019 (COVID-19) epidemics.1 Since 2020, Hong Kong experienced an initial epidemic wave of imported COVID-19 cases and spillover effects, as well as subsequent epidemic waves of imported COVID-19 cases and associated local transmission.

The COVID-19 epidemic has resulted in millions of deaths worldwide. By late 2020, there had been multiple country-level analyses in other regions; however, there were limited data concerning outcomes among patients with severe or critical COVID-19 in Hong Kong.2 This study analysed 28-day mortality in these patients and explored risk factors for mortality among them during the first several months of the COVID-19 pandemic.

This retrospective multi-centre cohort study included all adult patients aged ≥18 years with severe or critical COVID-19 who were admitted to the medical wards or intensive care units (ICUs) of three public acute hospitals in Hong Kong from 22 January to 30 September 2020. The three hospitals were Pamela Youde Nethersole Eastern Hospital, Princess Margaret Hospital, and Ruttonjee and Tang Shiu Kin Hospitals.

Patients were diagnosed with COVID-19 if their respiratory samples contained severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA, according to reverse transcription polymerase chain reaction (RT-PCR) analysis. Respiratory samples included nasopharyngeal aspirate, nasopharyngeal swab, nasopharyngeal aspirate or swab paired with throat swab, or deep throat saliva. Coronavirus disease 2019 grade was considered mild, severe, or critical, in accordance with the guidelines of the Chinese Center for Disease Control and Prevention.3 Severe COVID-19 was characterised by dyspnoea, respiratory rate of ≥30 breaths/minute, blood oxygen saturation <94%, ratio of arterial oxygen partial pressure to fractional inspired oxygen (P/F ratio) <300, and/or a 50% increase in lung infiltrates within 2 days.3 Critical COVID-19 was characterised by respiratory failure, septic shock, and/or multiple organ dysfunction or failure.3

Hong Kong experienced three epidemic waves during the study period.4 The first epidemic wave, from mid-January to early March 2020, occurred after travellers from mainland China arrived in Hong Kong during the Chinese New Year. The second epidemic wave occurred when Hong Kong residents returned from overseas during the Easter Holiday, from mid-March to May 2020. The third epidemic wave extended from early July until late September 2020; disease transmission may have originated among commercial airline crews.4 Cases were classified as imported or local by the Centre for Health Protection within the Department of Health. Medical comorbidities were defined according to the International Classification of Diseases, Tenth Revision. Comorbidities were selected based on the STOP-COVID (Study of the Treatment and Outcomes in Critically Ill Patients With COVID-19) in the US, which analysed critical COVID-19 cases at 65 ICUs.5

All medical records and data from the Clinical Management System of Hospital Authority and Clinical Information System (IntelliVue Clinical Information Portfolio; Philips Medical, Amsterdam, Netherlands) used by the ICUs were retrospectively reviewed. Upper respiratory tract infection (URI)–related symptoms included cough, rhinorrhoea, and sore throat. For patients admitted to the ICU, disease grade on admission was determined using the APACHE IV (Acute Physiology and Chronic Health Evaluation IV) score and the SOFA (Sequential Organ Failure Assessment) score. For patients requiring mechanical ventilatory support, disease grade was determined by the P/F ratio on the day of intubation. Blood test parameters included minimum lymphocyte count, maximum C-reactive protein level, maximum lactate dehydrogenase (LDH) level, and maximum alanine aminotransferase level. Clinical outcome data included the use of oxygen supplementation, mechanical ventilation, vasopressor or inotrope, renal replacement therapy (RRT), extracorporeal membrane oxygenation, and cardiopulmonary resuscitation, as well as mortality and length of stay in the ICU and hospital. Patients were followed up until death or 31 March 2021, whichever occurred earlier.

The primary outcome was 28-day mortality. Secondary outcomes included length of stay and mortality in the ICU and hospital, duration of oxygen supplementation and mechanical ventilatory support, and time to viral clearance or time to development of antibodies against SARS-CoV-2. Viral clearance was defined as the collection of two consecutive respiratory samples at least 24 hours apart that were both SARS-CoV-2–negative, according to RT-PCR analysis. Cycle threshold (Ct) value indicated the number of RT-PCR cycles required to amplify the viral RNA to a detectable level; this value was inversely related to viral load.6 Minimum Ct values were recorded to determine maximum viral load. Each patient’s blood was collected and qualitatively tested for antibodies against SARS-CoV-2 (ie, antibodies to SARS-CoV-2 nucleoprotein) using the Abbott SARS-CoV-2 Immunoglobulin G assay. Patients were generally released from isolation when they met the requirement for viral clearance or when they displayed serum antibodies, in accordance with recommendations from the Scientific Committee on Emerging and Zoonotic Diseases under the Centre for Health Protection within the Department of Health.7

The clinical characteristics and outcomes of patients with severe or critical COVID-19 were compared between 28-day survivors and non-survivors. Subgroup analysis was performed among patients who received ICU care between 28-day survivors and non-survivors; it was also performed among patients whose laboratory reports provided information regarding viral load (ie, Ct values) in respiratory samples. The frequencies of these characteristics and outcomes were expressed as medians ± interquartile ranges (IQRs) or as numbers of patients and corresponding percentages.

Based on the population of patients with COVID-19 (n=5080) in Hong Kong on 29 September 2020, where 351 patients had ever displayed serious or critical disease, the prevalence of outcome factors in this patient population was 6.9%. Using a confidence limit of 5% and a design effect of 2, the sample sizes required to achieve statistical powers of 80% and 90% were 84 patients and 138 patients, respectively.

For univariate analyses, the Pearson Chi squared test or Fisher’s exact test was used to compare categorical variables; the Mann-Whitney U test was used to compare continuous variables. Variables with a P value of <0.1 in univariate analyses were included in subsequent multivariable analyses. Independent predictors of 28-day mortality were assessed by logistic regression analysis using a forward stepwise approach. Considering the potential for unstable or extreme estimates because of the small sample sizes in subgroup analyses, logistic regression was not performed. All statistical analyses were performed using SPSS (Windows version 23.0; IBM Corp, Armonk [NY], US).

From 22 January to 30 September 2020, 1312 adult patients with COVID-19 were admitted to Princess Margaret Hospital, Pamela Youde Nethersole Eastern Hospital, and Ruttonjee and Tang Shiu Kin Hospitals. In total, 125 patients had severe or critical COVID-19.

The median age was 64 years (IQR=57-75); 69.6% of the patients were men and 93.6% were Chinese (Table 1). In total, 68.8% of the patients were admitted in the third epidemic wave and 86.4% acquired COVID-19 in Hong Kong. Almost half of the patients had hypertension or obesity (48.0% and 54.4%, respectively); 27.2% of patients had diabetes mellitus, and 21.6% had ever smoked. The median duration of symptoms before respiratory samples were confirmed SARS-CoV-2–positive was 3 days. The most common presenting symptom was fever (80.0%), followed by URI-related symptoms (64.8%). Overall, 5.6% of the patients were asymptomatic before their positive test result.

As shown in Table 1, 60% to 80% of patients received lopinavir-ritonavir, ribavirin, interferon beta-1b, or corticosteroids. Nearly half of the patients (48.8%) received anticoagulation treatment; <20% received remdesivir, tocilizumab, or convalescent plasma. More than 80% of the patients received antibiotics during hospitalisation. In total, 45.6% of the patients received ICU care. One patient received non-invasive ventilation, 41 patients (32.8%) received mechanical ventilation, and 11 patients (8.8%) received prone ventilation. Only one patient required extracorporeal membrane oxygenation. Approximately one-fourth of the patients required vasopressor support; 14 patients (11.2%) received acute RRT.

Fifteen patients (12%) died within 28 days (Table 1). Four additional patients died during hospitalisation; thus, the overall hospital mortality rate among patients with severe or critical COVID-19 was 15.2%. The median hospital length of stay was 21.8 days (IQR=15-31.8) and the median duration of oxygen supplementation was 7 days (IQR=4-12.5).

Twenty-eight–day non-survivors were older than 28-day survivors (84 years [IQR=71-86] vs 62 years [IQR=56-71.3]; P<0.001) and more frequently had a history of ischaemic heart disease (26.7% vs 5.5%; P=0.019) or stroke (26.7% vs 4.5%; P=0.012). Moreover, non-survivors had a shorter duration of symptoms before RT-PCR confirmation of SARS-CoV-2 positivity in respiratory samples (1 day [IQR=1-2] vs 3 days [IQR=1-6]; P=0.014); fewer non-survivors displayed URI-related symptoms (33.3% vs 69.1%; P=0.010). Non-survivors had a higher maximum C-reactive protein level (209 mg/L [IQR=82-248] vs 103 mg/L [IQR=63.6-153.5]; P=0.031); they more frequently received mechanical ventilation (66.7% vs 28.2%; P=0.006), acute RRT (40.0% vs 7.3%; P=0.002), and vasopressor support (66.7% vs 19.1%; P<0.001). Finally, non-survivors received a shorter duration of lopinavir-ritonavir treatment (1 day [IQR=0-2] vs 6 days [IQR=0.75-11]; P=0.007) [Table 1].

Logistic regression analysis revealed that age (odds ratio [OR] per 1-year increase in age=1.12, 95% confidence interval [CI]=1.04-1.21; P=0.002), history of stroke (OR=15.96, 95% CI=1.65-154.66; P=0.017), use of acute RRT (OR=15.32, 95% CI=2.67-87.83; P=0.002), and lopinavir-ritonavir duration (OR per 1-day increase=0.82, 95% CI=0.68-0.98; P=0.034) were independent predictors of 28-day mortality (Table 2).

Among the 57 patients admitted to the ICU, nine died; the ICU mortality rate was 15.8% (Table 3). The median ICU length of stay was 9.6 days (IQR=4.6-14.9). Univariate analysis demonstrated that 28-day non-survivors were older; more frequently had a history of ischaemic heart disease or stroke; had a shorter duration of symptoms; less frequently presented with URI-related symptoms; more frequently received mechanical ventilation, acute RRT, and vasopressor support; and received a shorter course of lopinavir-ritonavir treatment. Other significant differences between non-survivors and survivors were the minimum lymphocyte count (0.32 × 10⁹/L [IQR=0.17-0.4] vs 0.4 × 10⁹/L [IQR=0.3-0.6]; P=0.042), maximum LDH level (706 U/L [IQR=492.2-5255.5] vs 474.5 U/L [IQR=406.3-616]; P=0.043), anticoagulation duration (8 days [IQR=3- 10] vs 12.5 days [IQR=7.5-26.8]; P=0.045), APACHE IV score upon ICU admission (83 [IQR=64.5-98] vs 54.5 [IQR=46-61.8]; P<0.001), and SOFA score upon ICU admission (10 [IQR=5.5-13] vs 4 [IQR=3-5.8]; P=0.001).

The median P/F ratio on the day of intubation was 94.1 (IQR=81.9-117.7) for patients with COVID-19 requiring mechanical ventilation, and the median duration of ventilation for all ICU patients with COVID-19 was 8.5 days (IQR=5-18); these values were similar between survivors and non-survivors (Table 3).

As shown in Table 4, 28-day non-survivors were older and more frequently had a history of ischaemic heart disease or stroke; fewer of them had URI-related symptoms (36.4% vs 68.8%; P=0.010). Non-survivors more frequently received mechanical ventilation (81.8% vs 29.9%; P=0.001), acute RRT (45.5% vs 7.8%; P=0.004), and vasopressor support (81.8% vs 23.4%; P<0.001); they received a shorter course of lopinavir-ritonavir treatment (1 day [IQR=0-2] vs 6 days [IQR=0.75-11]; P=0.010). Additionally, non-survivors had a lower minimum lymphocyte count (0.37 × 10⁹/L [IQR=0.16-0.4] vs 0.49 × 10⁹/L [IQR=0.3-0.7], P=0.041) and lower minimum Ct value (15.1 [IQR=12.6-18.5] vs 19 [IQR=16.4-22.9]; P=0.004).

Among the 125 patients with severe or critical COVID-19, 38 underwent regular monitoring of viral load via RT-PCR analysis of respiratory samples for SARS-CoV-2 RNA; viral clearance was achieved within a median of 26 days (IQR=19-35). Among the 79 patients with access to antibody testing, a median of 14 days (IQR=10-18) was elapsed between the first positive RT-PCR result and the emergence of antibodies against SARS-CoV-2.

In this cohort of Hong Kong patients with severe or critical COVID-19, the 28-day mortality rate was 12.0%; it was 15.8% among such patients who were admitted to the ICU. In 2020, higher mortality rates were observed among cohorts in the US (35.4% in the STOP-COVID cohort5), Italy (ICU and hospital mortality rates of 48.8% and 53.4%, respectively8), and China (28-day mortality of 38.7% among severely and critically ill patients9). Patient baseline characteristics were similar across the cohorts—the median age was 60 to 70 years and the most common comorbidity was hypertension (40%-50%).5 8 9 The proportion of patients requiring mechanical ventilation varied across cohorts, ranging from 67% to 87% in the US5 and Italian8 cohorts, whereas it was 30% in the Chinese cohort.9 Overall, 70.2% of patients in our ICU subgroup received mechanical ventilation, which is comparable with the percentages in the US5 and Italy.8 The P/F ratio in our cohort was similar to the ratio in the STOP-COVID (median, 94.1 vs 124),5 reflecting moderate to severe hypoxaemia. Extracorporeal membrane oxygenation was required by <3% of patients in our cohort and the three comparison cohorts. In addition to patient factors, non-patient factors (eg, ICU bed availability and patient-to–hospital staff ratio) may affect quality of care and patient outcomes. Among hospitals of the Department of Veterans Affairs in the US, COVID-19 mortality was significantly greater when ICU demand exceeded 75%.10 Hospitals in the US with fewer ICU beds and nurses per COVID-19 case also had greater mortality.11 The relatively low prevalence, limited local transmission, and better ICU availability in Hong Kong may have contributed to the lower mortality rate observed in this study.

In our cohort, older age, history of stroke, use of acute RRT, and shorter duration of lopinavir-ritonavir treatment were independent predictors of 28-day mortality. Across studies in 2020, older age was commonly identified as a risk factor for COVID-19 mortality.5 8 9 A history of stroke was also identified as a significant risk factor for COVID-19 mortality in a large cohort study in China; higher neutrophil and interleukin-6 levels were observed in patients with a history of stroke, possibly because of a stronger inflammatory response to COVID-19.12 In an American cohort, the hospital mortality rate was 50% among patients who developed acute kidney injury, and the highest mortality risk was present in patients requiring dialysis.13 In 2020, the mechanism of acute kidney injury was speculated to involve direct viral invasion of renal tissue, considering post-mortem findings in China.14 Such injury was closely related to respiratory failure; in another American cohort, the median time from mechanical ventilation to the initiation of acute RRT was 0.3 hour.15

A shorter duration of lopinavir-ritonavir was identified as a risk factor in our cohort. Lopinavir-ritonavir was originally a protease inhibitor for the treatment of human immunodeficiency virus infection.16 In 2020, this repurposed drug was included in treatment recommendations in Hong Kong, where it effectively reduced viral load when used in combination with ribavirin and/or interferon beta-1b16; however, there was no difference in mortality between treatment and control groups as no patient died during the study.16 Two Chinese cohorts did not show significant outcome differences when using lopinavir-ritonavir,17 consistent with interim results from the World Health Organization Solidarity trial.18 Lopinavir-ritonavir treatment has been associated with gastrointestinal upset and liver dysfunction. Our findings may differ from the results of other cohort studies because, in the present study, lopinavir-ritonavir treatment was not administered to patients with severe liver dysfunction or to those who were presumably unable to tolerate side-effects because of their illness. In contrast, up to 80% of the ICU patients in an American cohort received hydroxychloroquine and/or azithromycin,5 despite doubtful efficacy and significant arrhythmia risk with QT prolongation.19 The selection of antiviral treatment may partly explain the differences in mortality among cohorts.

Regarding viral load, univariate analysis in our cohort showed that lower minimum Ct values were associated with greater 28-day mortality. In a study at Cornell in the US,6 the SARS-CoV-2 viral load on admission independently predicted the risks of intubation and mortality. In a Brazilian cohort, Ct values of <25 were associated with greater mortality.21 The implications of Ct values may become clearer if patients with mild disease are analysed together.

Lymphopenia on admission has been associated with worse outcomes in terms of ICU care requirement and mortality,22 presumably because of the cytokine storm phenomenon and the infection of T cells by SARS-CoV-2,19 which infection of T cells by SARS-CoV-2 was confirmed both in vitro and by flow cytometry and immunofluorescence studies.20 In the present study, the minimum lymphocyte count was significantly lower among 28-day non-survivors in both subgroups. Multi-centre COVID-19 studies have shown that an elevated LDH level is associated with a six-fold increase in the likelihood of severe disease and a 16-fold increase in the likelihood of mortality.24 In the present study, although univariate analysis showed that the maximum LDH level was associated with greater 28-day mortality in the overall cohort, it was not an independent predictor in logistic regression; this finding may have been influenced by the sample size.

To our knowledge, this large study was the first investigation in Hong Kong concerning the clinical characteristics and outcomes of patients with severe or critical COVID-19; it included three public acute hospitals and covered three waves of the COVID-19 pandemic in Hong Kong. This study also explored the impacts of viral parameters and treatment modalities on 28-day mortality.

First, this retrospective study in Hong Kong may have been subject to confounding factors and selection bias. The results may not be generalisable to patients with COVID-19 worldwide. Second, this study lacked information concerning viral parameters (eg, specific SARS-CoV-2 strains or mutations). Third, although some studies showed that inborn errors in type 1 interferon immunity and autoantibodies to type 1 interferons were associated with critical COVID-19,25 26 this study did not explore such factors because patient immunity data were unavailable. Fourth, the use of remdesivir was limited; most treatment courses were solely available to patients enrolled in studies by the pharmaceutical company concerned. Finally, other novel treatment options, including antivirals such as nirmatrelvir/ritonavir and molnupiravir, anti-inflammatory agents such as baricitinib and tocilizumab, and neutralising monoclonal antibodies against SARS-CoV-2, were not available or introduced during the study period. Fifth, the sample size of this study did not reach the statistical powers of 90% and may then not be of high enough power.

In this Hong Kong cohort, the 28-day mortality among patients with severe or critical COVID-19 was 12.0%. Age, history of stroke, use of RRT, and shorter course of lopinavir-ritonavir treatment were associated with greater 28-day mortality. In the future, larger studies with a focus on viral and host factors (eg, mutations in SARS-CoV-2 spike genes and interferon-1 immunity status) could improve prognosis prediction.

All authors contributed to the concept or design of the study, acquisition of data, analysis or interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

The authors thank the guidance and support of the following seniors and colleagues: Dr KK Chan from the Department of Medicine of Pamela Youde Nethersole Eastern Hospital, Dr Jenny YY Leung and Dr Alwin WT Yeung from the Department of Medicine and Geriatrics of Ruttonjee and Tang Shiu Kin Hospitals, and Dr Dominic HK So from Intensive Care of Princess Margaret Hospital.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The study protocol was approved by the Hong Kong East Cluster Research Ethics Committee (Ref No.: HKECREC—2020-115) and the Kowloon West Cluster Research Ethics Committee (Ref No.: KW/EX-21-005 [155-05]). The requirement for written informed patient consent was waived by both Committees due to the retrospective nature of the research.

1. Wong SY, Kwok KO, Chan FK. What can countries learn from Hong Kong’s response to the COVID-19 pandemic? CMAJ 2020;192:E511-5. Crossref

2. Ling L, So C, Shum HP, et al. Critically ill patients with COVID-19 in Hong Kong: a multicentre retrospective observational cohort study. Crit Care Resusc 2020;22:119-25. Crossref

3. Wu Z, McGoogan JM. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72 314 cases from the Chinese Center for Disease Control and Prevention. JAMA 2020;323:1239-42. Crossref

4. To KK, Chan WM, Ip JD, et al. Unique clusters of severe acute respiratory syndrome coronavirus 2 causing a large coronavirus disease 2019 outbreak in Hong Kong. Clin Infect Dis 2021;73:137-42. Crossref

5. Gupta S, Hayek SS, Wang W, et al. Factors associated with death in critically ill patients with coronavirus disease 2019 in the US. JAMA Intern Med 2020;180:1436-47. Crossref

6. Magleby R, Westblade LF, Trzebucki A, et al. Impact of severe acute respiratory syndrome coronavirus 2 viral load on risk of intubation and mortality among hospitalized patients with coronavirus disease 2019. Clin Infect Dis 2021;73:e4197-205. Crossref

7. Scientific Committee on Emerging and Zoonotic Diseases, Centre for Health Protection, Hong Kong SAR Government. Updated consensus recommendations on criteria for releasing confirmed COVID-19 patients from isolation (July 29, 2020). Available from: https://www.chp.gov.hk/files/pdf/updated_consensus_recommendations_on_criteria_for_releasing_confirmed_covid19_patients_from_isolation29july2020.pdf. Accessed 1 Oct 2020.

8. Grasselli G, Greco M, Zanella A, et al. Risk factors associated with mortality among patients with COVID-19 in intensive care units in Lombardy, Italy. JAMA Intern Med 2020;180:1345-55. Crossref

9. Yang X, Yu Y, Xu J, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study. Lancet Respir Med 2020;8:475-81. Crossref

10. Bravata DM, Perkins AJ, Myers LJ, et al. Association of intensive care unit patient load and demand with mortality rates in US Department of Veterans Affairs hospitals during the COVID-19 pandemic. JAMA Netw Open 2021;4:e2034266. Crossref

11. Janke AT, Mei H, Rothenberg C, Becher RD, Lin Z, Venkatesh AK. Analysis of hospital resource availability and COVID-19 mortality across the United States. J Hosp Med 2021;16:211-4. Crossref

12. Qin C, Zhou L, Hu Z, et al. Clinical characteristics and outcomes of COVID-19 patients with a history of stroke in Wuhan, China. Stroke 2020;51:2219-23. Crossref

13. Chan L, Chaudhary K, Saha A, et al. AKI in hospitalized patients with COVID-19. J Am Soc Nephrol 2021;32:151-60. Crossref

14. Su H, Yang M, Wan C, et al. Renal histopathological analysis of 26 postmortem findings of patients with COVID-19 in China. Kidney Int 2020;98:219-27. Crossref

15. Hirsch JS, Ng JH, Ross DW, et al. Acute kidney injury in patients hospitalized with COVID-19. Kidney Int 2020;98:209-18. Crossref

16. Hung IF, Lung KC, Tso EY, et al. Triple combination of interferon beta-1b, lopinavir-ritonavir, and ribavirin in the treatment of patients admitted to hospital with COVID-19: an open-label, randomised, phase 2 trial. Lancet 2020;395:1695-704. Crossref

17. Wong CK, Wan EY, Luo S, et al. Clinical outcomes of different therapeutic options for COVID-19 in two Chinese case cohorts: a propensity-score analysis. EClinicalMedicine 2021;32:100743. Crossref

18. WHO Solidarity Trial Consortium; Pan H, Peto R, et al. Repurposed antiviral drugs for COVID-19—interim WHO solidarity trial results. N Engl J Med 2021;384:497-511. Crossref

19. Sanders JM, Monogue ML, Jodlowski TZ, Cutrell JB. Pharmacologic treatments for coronavirus disease 2019 (COVID-19): a review. JAMA 2020;323:1824-36. Crossref

20. Pontelli MC, Castro ÍA, Martins RB, et al. SARS-CoV-2 productively infects primary human immune system cells in vitro and in COVID-19 patients. J Mol Cell Biol 2022;14:mjac021. Crossref

21. Faíco-Filho KS, Passarelli VC, Bellei N. Is higher viral load in SARS-CoV-2 associated with death? Am J Trop Med Hyg 2020;103:2019-21. Crossref

22. Huang I, Pranata R. Lymphopenia in severe coronavirus disease-2019 (COVID-19): systematic review and meta-analysis. J Intensive Care 2020;8:36. Crossref

23. Tavakolpour S, Rakhshandehroo T, Wei EX, Rashidian M. Lymphopenia during the COVID-19 infection: what it shows and what can be learned. Immunol Lett 2020;225:31-2. Crossref

24. Henry BM, Aggarwal G, Wong J, et al. Lactate dehydrogenase levels predict coronavirus disease 2019 (COVID-19) severity and mortality: a pooled analysis. Am J Emerg Med 2020;38:1722-6. Crossref

25. Zhang Q, Bastard P, Liu Z, et al. Inborn errors of type I IFN immunity in patients with life-threatening COVID-19. Science 2020;370:eabd4570. Crossref

26. Bastard P, Rosen LB, Zhang Q, et al. Autoantibodies against type I IFNs in patients with life-threatening COVID-19. Science 2020;370:eabd4585. Crossref

Prostate cancer (PCa) is the second most common cancer in men worldwide, and its incidence is rapidly increasing in Hong Kong.1 Despite increased awareness of PCa, many patients in Hong Kong are diagnosed with advanced disease involving metastasis, which does not respond to curative treatment. Previously, there was considerable interest in using serum prostate-specific antigen (PSA) for screening and early detection of PCa, with the hope that this approach would improve treatment outcomes. However, a Cochrane review showed that screening was associated with more frequent detection of localised disease, but there was no cancer-specific survival benefit; patients may even be harmed by complications associated with diagnostic and therapeutic procedures.2 Thus, the use of PSA-based screening has declined in the United States in the past decade.3 Notably, a concomitant epidemiologic shift from early to advanced disease (ie, reverse stage migration) has occurred, leading to serious concerns within the urological community.4

The slow progression and protracted clinical course of PCa are well-known.5 Patients with metastatic disease can survive for several years before death, which may also be caused by other medical conditions.6 However, during the course of disease, patients may experience complications related to PCa, including skeletal-related events and urinary tract complications. These complications can threaten a patient’s quality of life and lead to increased medical expenses. To our knowledge, minimal information is available regarding the course of disease (particularly complications) in patients with advanced PCa.

Here, we assessed the incidences of complications and unplanned hospital admissions among patients with metastatic hormone-sensitive PCa (mHSPC). This information may provide useful insights regarding the disease course and treatment needs of patients with mHSPC. It may also provide a more comprehensive understanding of the potential benefits of PSA-based screening in the management of patients with PCa.

The Asian Prostate Cancer (A-CaP) Study, a prospective multi-nation study designed to investigate real-world clinical management of PCa in Asia, began in January 2016. Patients with a diagnosis of PCa were recruited into the study.7 8 Clinical information was prospectively collected, including baseline patient and disease parameters, treatment received, and clinical progress. Thus far, >30 000 patients from 14 Asian countries have been recruited (based on an unpublished annual meeting report of A-CaP meeting held on 24 November 2020).

In Hong Kong, five hospitals (Alice Ho Miu Ling Nethersole Hospital, North District Hospital, Pok Oi Hospital, Prince of Wales Hospital, and Tuen Mun Hospital) formed the Hong Kong Prostate Cancer (HK-CaP) study group as part of the A-CaP project. Since 2016, all patients who presented to these hospitals (as an outpatient or inpatient) and received a diagnosis of PCa were recruited into the Hong Kong cohort, which currently includes >1000 cases of PCa (across all stages). In this study, we identified the first 100 consecutive patients with mHSPC (with no additional inclusion or exclusion criteria), then extracted their data from the HK-CaP database. Analyses were conducted using the extracted data, along with clinical information that had been retrospectively retrieved from electronic medical records.

Continuous variables are presented as medians with interquartile ranges, while categorical variables are presented as frequencies and percentages. All statistical analyses were performed using R version 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria). Two-sided P values <0.05 were considered statistically significant.

From January 2016 to August 2017, 100 consecutive patients with mHSPC were included in this study. The median age was 74 years (range, 50-93) [Table]. Overall, 17% of patients had no history of chronic disease. In contrast, 55%, 25%, 10%, and 63% of patients had pre-existing hypertension, diabetes, cerebrovascular disease, and history of smoking, respectively. The median serum PSA level at diagnosis was 202.5 ng/mL (range, 0.9-6260), and 83% of patients had abnormal findings on digital rectal examination at initial presentation. Thirteen patients (13%) had a family history of PCa. Most patients presented with symptoms, including lower urinary tract symptoms (62%), haematuria (9%), and constitutional symptoms (22%). Only 7% of patients had incidental findings of an abnormal serum PSA level during wellness screening.

Most patients (95%) had histologically proven PCa; the remaining 5% of patients had a clinical diagnosis of PCa based on a high serum PSA level (219-2779 ng/mL), with or without abnormal findings on digital rectal examination. Among patients with a histological diagnosis (n=94), the proportions with International Society of Urological Pathology grades 1 to 5 were 3.2%, 7.5%, 2.2%, 19.4%, and 67.7%, respectively. Most patients had bone metastases (96%); among them, eight patients also had visceral metastases. The remaining four patients were diagnosed with non-pelvic lymph node metastases (M1a) at initial presentation (Table).

With the exception of one patient who selected watchful waiting, all patients received androgen deprivation therapy. Initial androgen deprivation therapies included luteinising hormone-releasing hormone antagonists (46 patients, 46%), luteinising hormone-releasing hormone agonists (with short-term antiandrogen treatment during flares) [28 patients, 28%], and bilateral orchidectomy (25 patients, 25%). Seventeen patients also received upfront chemotherapy for advanced disease; no patient received upfront abiraterone.

The mean follow-up period was 32.9 months (range, 0.3-54.2). During follow-up, 59 patients developed metastatic castration-resistant prostate cancer (mCRPC). Among these patients, older-generation antiandrogens, docetaxel, abiraterone, enzalutamide, and prednisolone alone were used by 26 (44.1%), nine (15.3%), 17 (28.8%), 18 (30.5%), and five (8.5%) patients, respectively. Thirty-two patients (32%) also received palliative radiotherapy for symptom control. Seven (11.9%) patients with mCRPC used denosumab for bone protection. The median overall survival time was 3.7 years; 33 (67.3%) patients died of PCa and 16 (32.7%) patients died of other causes, and none of these causes were cardiovascular events (Fig 1).

Only 30 patients (30%) had no disease-related complications. Among the observed complications, skeletal-related events were most common: 41 patients reported bone pain during follow-up. Of these 41 patients, 35 (85%) required regular analgesics, and 12 (29%) required opioid analgesics for pain control. Approximately half of the 41 patients (ie, 20 patients, 49%) required palliative radiotherapy for bony metastases. Moreover, 21 patients developed pathologic fractures, and eight patients had cord compression.

The second most common complication was retention of urine secondary to prostatic obstruction (28 patients, 28%). Among these 28 patients, only seven (25%) were able to discontinue urethral catheter use. Of the remaining patients, 10 (36%) required endoscopic prostatic surgery and 11 (39%) required long-term urethral catheter use.

Among 15 patients who developed ureteral obstruction, four (27%) required ureteral stenting and four (27%) required long-term nephrostomy drainage. The remaining seven (47%) patients received conservative management. During follow-up, 17 patients developed gross haematuria.

Forty-one patients (41%) developed anaemia (haemoglobin level <10 g/dL); 22 of these patients (53.7%) required transfusion. Furthermore, only four of the 41 patients received chemotherapy to manage mCRPC. Therefore, most cases of anaemia were presumably the direct result of PCa. Other complications included deep vein thrombosis (4%), psychiatric problems (adjustment disorder or depression) [4%], and suicidal ideation (1%).

Fifty-nine (59%) patients experienced ≥1 unplanned hospital admissions during the course of disease. The proportions of patients with 1-5, 6-10, and >10 unplanned hospital admission episodes were 43%, 10%, and 6%, respectively. The indications for these admissions included skeletal-related events (bone pain, fracture, and fall, 19%), urinary complications (haematuria, ureteric obstruction, and bladder outcome obstruction, 16%), and sepsis (urosepsis, pneumonia, and infection of other origin, 28%) [Fig 2]. At least half of the admissions were presumably the direct result of PCa, such as bone pain and urinary complications.

In this prospective observational study, over a mean follow-up period of approximately 32 months, PCa-related complications occurred in 70% of 100 patients with newly diagnosed mHSPC; around 60% of these patients had unplanned hospital admissions during the course of disease. Slightly less than half of the patients died during this study period. More than two-thirds of the patients died of PCa. Also, many of the patients experienced PCa-related complications had received various treatments for their disease and complications. These real-world data provide insights concerning the natural course of disease in patients with advanced PCa; they also suggest a need to reconsider management approaches for such patients. Additionally, these data may help to evaluate the potential benefits of PSA-based screening for PCa.

The primary purpose of disease screening involves identifying a disease in its early or asymptomatic stages, which can allow more effective treatment and support better outcomes. Consequently, disease-related mortality and complications can be minimised, while improving patient quality of life.9 Early intervention may also help to reduce medical expenses through disease treatment at an earlier stage, rather than a later and more complex stage. Prostate cancer fulfils some of the criteria for disease screening: it is a common cancer, displays a latent disease stage, and has acceptable diagnostic tests and effective treatments.10

However, PSA-based PCa screening is among the most controversial topics in urology. In a review based on data from five randomised trials of PCa screening, no cancer-specific survival benefit was identified; moderate harm was caused by diagnostic procedures.2 Moreover, overdiagnosis and overtreatment were common; they could cause treatment-related harm. Therefore, Chou et al11 recommended against PSA-based screening. This recommendation led to a decline in the use of PSA testing during the past decade.3 As expected, the overall incidence of PCa, particularly low-risk disease, has decreased in recent years.12 Unfortunately, there has been a concomitant increase in diagnoses of advanced and higher-grade disease (ie, reverse stage migration).4 13 Furthermore, the European Randomized Study of Screening for Prostate Cancer revealed a 30% reduction in the relative risk of metastatic disease in the screened population, compared with the non-screened population.14 Therefore, PSA-based screening may at least reduce the number of patients who present with metastatic disease.

Discussions of PSA-based screening have mainly focused on survival benefits (ie, decreases in overall and cancer-specific or all-cause mortality), as well as the harms associated with screening procedures and overtreatment of low-risk disease.2 However, there has been minimal consideration of the potential advantages of screening in terms of preventing disease-related complications, as well as the negative effects of advanced disease on quality of life. Moreover, there has been little discussion regarding the potential financial implications of managing advanced PCa and its complications.

Rather than investigating the value of PSA-based screening, the present study was conducted to fill the gap in knowledge regarding the course of disease in patients who present with mHSPC. In our cohort, 70% of patients developed PCa-related complications (eg, bone, urinary tract, and anaemia) during the course of disease. We found that nearly 60% of patients had unplanned hospital admissions for various complications; this proportion was much greater than the observed readmission rate of 6.5% (6 of 93 patients) for localised PCa over a period of >5 years in Hong Kong.15 Therefore, early diagnosis may be the only way to minimise the incidence of PCa-related complications in patients with mHSPC.

In addition to PCa-related complications, cancer treatment can cause adverse effects in patients with mHSPC. Typical androgen deprivation therapy is notorious for causing cardiovascular and metabolic complications.16 17 18 Treatments specifically for mCRPC, such as chemotherapy or next-generation androgen receptor targeting agents, are also associated with adverse effects such as febrile neutropenia, gastric distress, hypertension, and cardiac events.19 Additionally, the direct and indirect costs of treatment place additional financial burdens on patients and their families (ie, ‘financial toxicity’), impose a psychological burden, and adversely affect the quality of life of patients.20 Therefore, earlier diagnosis of PCa may help patients to avoid progression to advanced or metastatic disease, thereby reducing suffering associated with advanced disease and its treatment-related complications.

This study had some limitations. First, bone-protecting agents, which are relatively expensive, are not commonly used in Hong Kong. The absence of such agents may have led to a higher rate of skeletal-related events in our patients. Second, only 17 patients received upfront chemotherapy and no patients received upfront next-generation androgen receptor–targeting agents. Thus, we could not assess whether the use of these newer treatment approaches could minimise disease-related complications. Third, we did not collect data regarding the quality of life of patients, which would help to clarify the effects of disease-related complications on patients and their families.

In recent years, there have been many advances in PCa diagnosis and treatment. The use of new markers for PCa, such as the Prostate Health Index,21 urinary exosomes,22 and multiparametric magnetic resonance imaging,23 has considerably improved diagnostic accuracy and reduced the need for prostate biopsy (ie, to rule out false-positives based on elevated PSA levels). The use of the transperineal route for prostate biopsy has also minimised prostate biopsy–related complications.24 In addition, active surveillance for low-risk PCa has mitigated possible harms associated with overtreatment.25 In combination, these new advances and better knowledge of the disease course will improve support for PCa screening, thereby minimising disease-related suffering.

In this observational study, 70% of patients with metastatic PCa at initial presentation had various PCa-related complications and many unplanned hospital admissions during the course of disease. Although there remains controversy concerning whether PSA-based PCa screening is beneficial for cancer-specific survival, the recent observation of reverse stage migration in PCa related to decreased PSA testing is problematic for PCa management. Advanced PCa may be associated with significant disease-related complications; it can also place an increased burden on the healthcare system by contributing to more unplanned hospital admissions. Thus, there is a need for more comprehensive assessment of the value of PSA-based PCa screening in terms of preventing disease-related morbidity and mortality. Advances in diagnostic tools and the use of active surveillance may help reduce the harms associated with diagnostic procedures and overtreatment of early disease.

Concept or design: CF Ng.
Acquisition of data: CWH Mak, SYS Chan, ML Li, CH Leung.
Analysis or interpretation of data: CWH Mak, CH Leung.
Drafting of the manuscript: CF Ng, CWH Mak.
Critical revision of the manuscript for important intellectual content: JYC Teoh, PKF Chiu, PSK Chu.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

As editors of the journal, CF Ng and JYC Teoh were not involved in the peer review process. Other authors have disclosed no conflicts of interest.

This research has been presented as oral presentation in the 25th Annual Scientific Meeting of Hong Kong Urological Association held on 25 October 2020 in Hong Kong SAR, China.

The research received support from J-CaP (Japan Study Group of Prostate Cancer) and Takeda Pharmaceutical Company Limited. The funder had no role in study design, data collection/analysis/interpretation, or manuscript preparation.

The study protocol was approved by the Joint Chinese University of Hong Kong–New Territories East Cluster Clinical Research Ethics Committee (Ref No.: 2014.251) and registered in ClinicalTrials.gov (Identifier: NCT03344835). Informed patient consent has been waived by the Committee because of the observational nature of the study.

1. Teoh JY, Hirai HW, Ho JM, Chan FC, Tsoi KK, Ng CF. Global incidence of prostate cancer in developing and developed countries with changing age structures. PLoS One 2019;14:e0221775. Crossref

2. Ilic D, Neuberger MM, Djulbegovic M, Dahm P. Screening for prostate cancer. Cochrane Database Syst Rev 2013;(1):CD004720. Crossref

3. Fleshner K, Carlsson SV, Roobol MJ. The effect of the USPSTF PSA screening recommendation on prostate cancer incidence patterns in the USA. Nat Rev Urol 2017;14:26-37. Crossref

4. Reese AC, Wessel SR, Fisher SG, Mydlo JH. Evidence of prostate cancer “reverse stage migration” toward more advanced disease at diagnosis: data from the Pennsylvania Cancer Registry. Urol Oncol 2016;34:335.e21-8. Crossref

5. Popiolek M, Rider JR, Andrén O, et al. Natural history of early, localized prostate cancer: a final report from three decades of follow-up. Eur Urol 2013;63:428-35. Crossref

6. Tangen CM, Faulkner JR, Crawford ED, et al. Ten-year survival in patients with metastatic prostate cancer. Clin Prostate Cancer 2003;2:41-5. Crossref

7. Akaza H, Hirao Y, Kim CS, et al. Asia prostate cancer study (A-CaP Study) launch symposium. Prostate Int 2016;4:88-96. Crossref

8. Lojanapiwat B, Lee JY, Gang Z, et al. Report of the third Asian Prostate Cancer study meeting. Prostate Int 2019;7:60-7. Crossref

9. The Society of Radiographers. NHS screening programmes–purpose of the programmes. Available from: https://www.sor.org/getmedia/160c519f-7d43-4138-95f4-84e955bc2857/nhs_population_screening_2.pdf. Accessed 10 Jun 2023.

10. Wilson JM, Jungner G, World Health Organization. Public Health Papers. Principles and practice of screening for disease. World Health Organization, 1968. Available from: https://apps.who.int/iris/handle/10665/37650. Accessed 10 Jun 2023.

11. Chou R, Croswell JM, Dana T, et al. Screening for prostate cancer: a review of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2011;155:762-71. Crossref

12. Jemal A, Culp MB, Ma J, Islami F, Fedewa SA. Prostate cancer incidence 5 years after US Preventive Services Task Force Recommendations against screening. J Natl Cancer Inst 2021;113:64-71. Crossref

13. Hu JC, Nguyen P, Mao J, et al. Increase in prostate cancer distant metastases at diagnosis in the United States. JAMA Oncol 2017;3:705-7. Crossref

14. Schröder FH, Hugosson J, Carlsson S, et al. Screening for prostate cancer decreases the risk of developing metastatic disease: findings from the European Randomized Study of Screening for Prostate Cancer (ERSPC). Eur Urol 2012;62:745-52. Crossref

15. Ng CF, Kong KY, Li CY, et al. Patient-reported outcomes after surgery or radiotherapy for localised prostate cancer: a retrospective study. Hong Kong Med J 2020;26:95-101. Crossref

16. Hu JR, Duncan MS, Morgans AK, et al. Cardiovascular effects of androgen deprivation therapy in prostate cancer: contemporary meta-analyses. Arterioscler Thromb Vasc Biol 2020;40:e55-64. Crossref

17. Ng CF, Chiu PK, Yee CH, Lau BS, Leung SC, Teoh JY. Effect of androgen deprivation therapy on cardiovascular function in Chinese patients with advanced prostate cancer: a prospective cohort study. Sci Rep 2020;10:18060. Crossref

18. Shore ND, Saad F, Cookson MS, et al. Oral relugolix for androgen-deprivation therapy in advanced prostate cancer. N Engl J Med 2020;382:2187-96. Crossref

19. Tonyali S, Haberal HB, Sogutdelen E. Toxicity, adverse events, and quality of life associated with the treatment of metastatic castration-resistant prostate cancer. Curr Urol 2017;10:169-73. Crossref

20. Rotter J, Spencer JC, Wheeler SB. Financial toxicity in advanced and metastatic cancer: overburdened and underprepared. J Oncol Pract 2019;15:e300-7. Crossref

21. Chiu PK, Ng CF, Semjonow A, et al. A multicentre evaluation of the role of the Prostate Health Index (PHI) in regions with differing prevalence of prostate cancer: adjustment of PHI reference ranges is needed for European and Asian settings. Eur Urol 2019;75:558-61. Crossref

22. Wang WW, Sorokin I, Aleksic I, et al. Expression of small noncoding RNAs in urinary exosomes classifies prostate cancer into indolent and aggressive disease. J Urol 2020;204:466-75. Crossref

23. Kasivisvanathan V, Rannikko AS, Borghi M, et al. MRI-targeted or standard biopsy for prostate-cancer diagnosis. N Engl J Med 2018;378:1767-77. Crossref

24. Chiu PK, Lo KL, Teoh JY, et al. Sectoral cancer detection and tolerability of freehand transperineal prostate biopsy under local anaesthesia. Prostate Cancer Prostatic Dis 2021;24:431-8. Crossref

25. Kinsella N, Helleman J, Bruinsma S, et al. Active surveillance for prostate cancer: a systematic review of contemporary worldwide practices. Transl Androl Urol 2018;7:83-97. Crossref

Healthcare spending in Hong Kong continues to grow, as demonstrated by an 8.6% increase (to HK$69.7 billion [US$8.99 billion]) from fiscal year 2018 to fiscal year 2019.1 According to internal statistics drawn from the Clinical Data Analysis and Reporting System of the Hospital Authority (HA), the annual number of emergency department (ED) visits in all public EDs in the New Territories West Cluster increased by 11.5% (to 410 707) between 2018-19 and 2019-20. Similar trends have been observed in many industrialised countries.2 Moreover, the proportion of older adults (age ≥65 years) is projected to increase from 17.2% in 2019 to 25.1% in 2029.3 The increasing demand for ED services may result in diminished quality of care and negative outcomes for vulnerable patients.

Frequent ED use is often regarded as a major component of ED utilisation burden. Substantial gaps remain in research concerning individuals who frequently use EDs in Hong Kong. Efforts to understand this population and the impacts of frequent use on ED utilisation burden are needed to identify interventions that can improve healthcare delivery, both in and out of the EDs.

This region-based study analysed individuals who frequently used public EDs and explored specific interventions that may reduce ED visits by these patients.

This multi-centre retrospective cohort study analysed ED visits from all acute care hospitals located within the New Territories West Cluster (serving a population of 1.14 million in 20183). During the study period, only three EDs (in Tuen Mun Hospital, Pok Oi Hospital, and Tin Shui Wai Hospital4) provided acute emergency care in this region. All three EDs are public and managed by the HA, the statutory body that manages all public hospitals in Hong Kong. In the study region, individuals who called ambulances were typically taken to one of these three EDs.

The study population included all patients, regardless of age, who visited any of these three EDs during the fiscal year from 1 April 2018 to 31 March 2019. Patients without acceptable identification documents were excluded.

Usage classes were defined according to the total number of ED visits by an individual patient to any of the 18 public EDs in Hong Kong5 during the study period, using cut-off values commonly reported in the literature to facilitate comparison6: normal users (NUs) had ≤3 visits,6 frequent users (FUs) had 4 to 9 visits,6 and high-intensity users (HIUs) had ≥10 visits.7

Two additional groups were defined: non-ED users had no ED visits over a 12-month period, whereas superusers had very high numbers of ED visits (≥35 over a 12-month period, almost 1 visit every 10 days).

Data were collected from the Clinical Data Analysis and Reporting System of the HA, a comprehensive electronic patient database that includes patient demographics, diagnoses, surgical records, ED visits, hospitalisation episodes, and radiological investigations from all public hospitals and most public clinics in Hong Kong. It allows episode-based and patient-based analyses according to user-defined criteria. For this study, multiple visits were linked using a unique record linkage number. Population demographic data were obtained from the 2016 Population By-census8 and other official government statistics.3 9

Two dimensions of analysis were performed on the three classes NUs, FUs, and HIUs: patient-based and episode-based, in which patients and individual visits were the respective units of analysis.

Demographic variables were assigned based on values reported at the index visit; they included age, ethnicity (Chinese/non-Chinese), sex (male/female), institutionalisation (residency in an old-age home or not), and ED payment status (exempt or not). In Hong Kong, identity card holders and residents aged <11 years are charged HK$180 (US$23.2) per ED visit. Civil servants and their family members, individuals receiving social security assistance from the government, and individuals in vulnerable groups (low income, chronically ill, and older adults with minimal income/assets) are exempt from the requirement to pay for ED visits. To evaluate ED utilisation trends, data were retrieved regarding utilisation in the 12-month period before the study.

In Hong Kong, all hospital admissions and ED visits (excluding patients who leave before consultation) are coded using an appropriate International Classification of Diseases, Ninth Revision (ICD-9) diagnosis recommended by the World Health Organization.9 A list of diagnoses that represent chronic conditions with significant morbidity was compiled and classified into nine categories with reference to ICD-9 (online supplementary Appendix 1).9 All in-patient procedures, except minor bedside and clinical procedures, are also appropriately coded and classified as minor, intermediate, major, or ultra-major10; they are then divided into elective and emergency categories. These coding processes, and the processes described below, are routine procedures that undergo strict internal auditing for completeness and accuracy. This study analysed ICD-9 diagnosis codes, as well as major and ultra-major procedure records, from hospital and clinic encounters in the 5 years prior to the study period. The full list of major and ultra-major operations included in the current study is mostly based on the HA’s List of Private Services.10

Deaths occurring in Hong Kong were retrieved from the Hong Kong Death Register. Two-year mortality was defined as death from any cause between 1 April 2018 and 31 March 2020.

Variables in the episode-based analysis included age, ambulance utilisation, triage category (in descending order of urgency: critical, emergency, urgent, semi-urgent, and non-urgent), imaging performed in ED (plain radiography and computed tomography), and visit details (to be discussed below). Admission rates and median length of stay for admitted patients were calculated.

For each episode, the attending specialty, trauma status, and ICD-9 diagnosis or chief complaint were recorded by the attending emergency physician. The attending specialty was regarded as the specialty to which a patient’s chief complaint belongs. Four main specialties (medicine and geriatrics, general surgery, orthopaedics, and paediatrics) were included in the data analysis. The ICD-9 diagnosis codes were grouped into musculoskeletal pain (eg, joint pain, cervicalgia, and lumbago) and minor infections (eg, acute upper respiratory tract infection). A full list of diagnoses within each of these two categories is included in online supplementary Appendix 2.9

Descriptive statistics were compiled for the demographic characteristics of non-ED users, NUs, FUs, and HIUs, with reference to data from the 2016 Population By-census.8 Age-specific distributions of ED visits and numbers of ED visits per capita were analysed. In the patient-based analysis, demographic and clinical characteristics of NUs, FUs, and HIUs were compared by univariate analysis with appropriate statistical tests. The episode-based analysis compared age distribution, ambulance use, triage category, investigations performed, principal diagnosis during each ED visit, and admission statistics among NUs, FUs, and HIUs. The clinical and demographic characteristics of superusers were also analysed.

Two models were used to identify independent predictors of FU and HIU statuses, compared with NU status and specific numbers of ED visits. Multinomial logistic regression was conducted, using the patient as the unit of analysis, to compare the FU and HIU groups with the reference group. Zero-truncated Poisson regression was performed to predict the independent association of each predictor with the number of ED visits. All bivariate predictors associated with the outcome (P<0.1) were entered into the model and used as categorical variables. Age-stratified subgroup analysis (≤17 years, 18-64 years, and ≥65 years) was performed to identify independent predictors of FU and HIU statuses in each age-group. P<0.05 was regarded as the threshold for statistical significance in all analyses. Zero-truncated Poisson regression was performed with R11 with vector generalised linear and additive model.12 All other statistical analyses were conducted using SPSS (Windows version 25.0; IBM Corp, Armonk [NY], United States).

As shown in the Figure, the number of per capita ED visits was the highest in the youngest and oldest age-groups (0.66 for age 0-4 years and 1.02 for age ≥85 years); patients aged 10 to 54 years had a stable and low number of per capita ED visits (<0.3).

During the study period, 215 862 patients with valid identity documents accessed EDs in the study region; there were 371 915 visits in total (Table 1). Frequent users and HIUs constituted small percentages of the total number of patients (8.2% and 0.8%, respectively), but they represented larger percentages of visits (21.3% and 5.9%, respectively). The total number of ED visits by a single HIU ranged from 10 to 263.

Compared with NUs (median age, 47 years), FUs and HIUs were older (55 and 52 years, respectively). Frequent users and HIUs more often had underlying chronic illnesses, such as cardiac, respiratory, neurological, and gastrointestinal/hepatobiliary diseases. Mental health disorders and substance abuse were also much more prevalent among FUs and HIUs. Moreover, 23.5% and 59.0% of FUs and HIUs were FUs or HIUs in the previous year, implying habitual attendance behaviour. Frequent users and HIUs also had higher all-cause mortality rates, compared with NUs.

Reasons for ED visits differed among NUs, FUs, and HIUs (Table 2). Normal users were much more likely to visit the ED for an injury, whereas primary diagnoses related to musculoskeletal pain were more common in HIUs. Levels of urgency on presentation also differed among NUs, FUs, and HIUs. More visits by FUs were triaged as critical, emergency, or urgent (38.1%), compared with visits by NUs (31.8%). However, a lower percentage of visits by HIUs were triaged as urgent or above (28.6%), compared with visits by NUs; more HIUs attended EDs by calling an ambulance (HIUs: 21.2% vs NUs: 17.9%).

Multinomial logistic regression and zero-truncated Poisson regression showed similar independent predictors of FU and HIU statuses (Table 3). Payment exemption and number of visits in the prior year were the strongest predictors. Residency in an old-age home was a risk factor for FU status (adjusted odds ratio [OR]=1.131) but a protective factor for HIU status (adjusted OR=0.48). Non-Chinese ethnicity, pre-existing systemic diseases including cardiac, respiratory, gastrointestinal/hepatobiliary and renal, mental health disorders, and substance abuse were risk factors for frequent ED use. Age-stratified analysis (Table 4) showed similar predictors among the three groups. Respiratory diseases and gastrointestinal/hepatobiliary diseases were prevalent in paediatric FUs; neurological diseases and mental health disorders were prevalent in both paediatric FUs and paediatric HIUs. Non-Chinese ethnicity was a risk factor for FU and HIU statuses among patients aged <18 years (adjusted ORs=1.949 and 2.107, respectively), but it was not associated with ED use in older patients (age >64 years).

Superusers had particularly high prevalences of mental health disorders (26.9%) and gastrointestinal/hepatobiliary diseases (36.5%) [Table 5]. Many of their visits were less serious or complex, compared with visits by NUs, FUs, and HIUs. Only 9.4% of visits by superusers required the use of ambulance services. Most visits by superusers were triaged as semi-urgent or non-urgent (90.9%). Musculoskeletal pain and minor infections represented 54.3% of the principal diagnoses for superusers.

The characteristics of individuals who frequently use EDs in Hong Kong have not been extensively investigated. Frequent users and HIUs constituted 9% of all ED patients and represented 27.2% of all visits. Region-based studies in developed countries have revealed similar findings.6 7 13 14 15 16 Although frequent ED use is often assumed to indicate inappropriate use, our data do not support this assumption. The health statuses and presentation conditions of FUs strongly suggest that these individuals have greater health needs than the rest of the population. Frequent users were more likely to have underlying illnesses, require ambulance services, have visits triaged as urgent or above, require hospital admission, and have a longer length of hospitalisation. In contrast, HIUs had visits triaged as less urgent, required less investigations, often attended for conditions (eg, musculoskeletal pain and minor infections) that could potentially be managed in primary care clinics, and had lower admission rates compared with both NUs and FUs. These trends were much more pronounced among superusers.

High percentages of visits involved musculoskeletal pain and minor infections, conditions that could potentially be managed in primary care clinics. Such visits could be related to a misunderstanding of ED function, a misperception of condition severity, or use of the ED as a substitute for unavailable forms of care. Policies to decrease ED utilisation burden should focus on improving healthcare delivery services and primary care; such policies are likely to benefit NUs with similar needs. Recent advances in information technology and the widespread use of smartphones offer many opportunities to improve information dissemination and increase accessibility to alternative sources of care.

Our findings suggest disproportionate use of ED services by ethnic minorities. The number of Hong Kong residents of non-Chinese ethnicity rose by 70% over a 10-year period; they constituted 8% of the total population in 2016.8 The Chinese literacy rate was <20% in some age-groups.8 More frequent ED use may result from language barriers to access of care, limited knowledge of local healthcare alternatives, or cultural differences in health-seeking behaviour. Effective policy measures targeting these populations should promote an understanding of the local healthcare system through informative advertisements about alternative services offered; measures should also reduce language barriers (eg, through translator services or the provision of online booking instructions in other languages).

Frequent users constitute a sicker population requiring medical care that cannot be easily provided in primary care clinics. They need support to facilitate integration and maintenance in the community, which would reduce the need for emergency medical care at EDs. The burden placed on EDs may be reduced by developing new community-based support systems, such as enhanced coverage of community nursing services and extended service hours (eg, weekends and public holidays); dedicated multidisciplinary teams to provide rehabilitation services and caregiver training in the community; and centres for the provision of coordinated community services. Eligible geriatric patients could benefit from a holistic community geriatric care strategy triggered by an ED visit, with protocol-driven assessments in the ED to identify potentially reversible risk factors for subsequent deterioration and repeated ED visits. This strategy should include comprehensive geriatric assessment prior to ED discharge, focused on factors such as fall risk, delirium screening, and frailty assessment. In addition to physical assessments, psychosocial assessments and support may improve patient outcomes and minimise repeated ED visits. Pre-discharge care planning that empowers patients and family members to seek help from non-ED sources may also prevent repeated ED visits. Thus far, there remains uncertainty about the effectiveness of community-based interventions for people with multimorbidity because of the relatively small number of randomised controlled trials focused on this area of healthcare.17

Reformation of the chronic disease service model may improve quality of care and reduce repeated ED visits. Patient-based care, rather than disease-based care, may be beneficial. Patients with multimorbidity (especially older adults) receive medical treatment through multiple specialty or subspecialty clinics, which can result in fragmented and duplicative care. It is not uncommon for patients with chronic diseases to attend the ED for minor problems and questions about their chronic diseases because they cannot find an alternative source of medical advice. Case management helps improve outcomes in some chronic diseases.18 Efforts to strengthen the abilities of primary care clinics to function as ‘case managers’ for patients with chronic diseases may improve quality of care and reduce the number of ED visits.

In this study, mental health disorders and substance abuse were significant predictors of frequent ED use. A previous work indicated that one in seven Hong Kong residents aged 16 to 75 years has anxiety, depression, or another common mood disorder.19 The HA is the main specialist service provider for patients with mental health disorders. It provides in-patient facilities, day hospitals, specialist out-patient clinics, and community outreach services. The HA is experiencing increased demand for specialist mental health services,20 which may be causing patients to use EDs instead. Emergency department visits and readmissions for psychiatric problems may be reduced by reforming the current service model to expand community psychiatric services, with a focus on personalised care for psychiatric patients and their caregivers through a case management approach that facilitates community re-integration and strengthens recovery. Enhanced screening to identify early features of mental health disorders may allow earlier detection and treatment, thereby reducing ED utilisation. Patients and caregivers should receive education about health-seeking behaviours during instances of acute deterioration (eg, using a 24-hour psychiatric advisory hotline or undergoing urgent assessment at a psychiatric specialist out-patient clinic), rather than simply using the ED as a safety net.

This study was limited to the three EDs in the New Territories West Cluster. Its findings may not be generalisable to other regions in Hong Kong with different demographics, health-seeking behaviours, and socio-economic statuses. Also, this study only investigated ED visits within a specific time period and did not consider past or future periods. Thus, it may have underestimated ED visits for patients who were born or died during the study period.

By reviewing diagnosis codes, we were able to include many visits and patients in the analysis; however, we could not analyse individual charts. Although coding is a routine component of hospital procedures, codes are only required for the current condition or presenting problem. Generally, coding is not mandatory for appointments at out-patient clinics. This difference in coding information may have led to underestimation of patient comorbidities. To mitigate this possibility, we examined all diagnosis codes from the past 5 years to acquire a more complete representation of underlying comorbidities.

Because this study excluded patients without valid identification documents, homeless persons may have been underrepresented. These individuals potentially have a heavier disease burden and a disproportionate share of frequent visits. The study may also have excluded visitors to Hong Kong and individuals who do not have residency status.

Frequent users and HIUs are a small but diverse population that represents a substantial proportion of annual ED visits. Demographic factors, economic considerations, and medical conditions all contribute to increased numbers of ED visits. Our data suggest that there are many opportunities for improvement via streamlining and enhancement of healthcare delivery to reduce ED utilisation.

Concept or design: PYT Ng, CT Lui.
Acquisition of data: PYT Ng, CT Lui.
Analysis or interpretation of data: PYT Ng, CT Lui.
Drafting of the manuscript: PYT Ng, CT Lui.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

We thank Mr Chun-ho Lam, Statistical Officer at Research Assist Team of the New Territories West Cluster of Hospital Authority, Hong Kong, for his advice on the zero-truncated Poisson regression model.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

This research was approved by the New Territories West Cluster Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: NTWC/REC/19081). Informed patient consent was waived by the Committee due to the retrospective nature of the study.

1. Hospital Authority, Hong Kong SAR Government. Hospital Authority annual report 2018-2019. Available from: https://www.ha.org.hk/ho/corpcomm/AR201819/PDF/HA_Annual_Report_2018-2019.pdf. Accessed 1 May 2020.

2. Pines JM, Hilton JA, Weber EJ, et al. International perspectives on emergency department crowding. Acad Emerg Med 2011;18:1358-70. Crossref

3. Data.gov.hk, Hong Kong SAR Government. Projections of population distribution 2021-2029. Available from: https://data.gov.hk/en-data/dataset/hk-pland-pland1-projections-of-population-distribution-2021-to-2029. Accessed 17 Jul 2023.

4. Hospital Authority, Hong Kong SAR Government. List of all accident & emergency departments. 2023. Available from: https://www.ha.org.hk/visitor/ha_visitor_index.asp?Content_ID=200246&lang=ENG. Accessed 19 Jul 2023.

5. Hospital Authority, Hong Kong SAR Government. Introduction of clusters: New Territories West Cluster. 2023. Available from: https://www.ha.org.hk/visitor/ha_visitor_index.asp?Content_ID=10181&Lang=ENG&Dimension=100&Parent_ID=10084. Accessed 19 Jul 2023.

6. Krieg C, Hudon C, Chouinard MC, Dufour I. Individual predictors of frequent emergency department use: a scoping review. BMC Health Serv Res 2016;16:594. Crossref

7. Dr Foster UK. High intensity users: reducing the burden on accident & emergency departments. Available from: https://www.telstrahealth.com/content/dam/telstrahealth/pdf-downloads/Dr-Foster_High-Intensity-Users-Report.pdf. Accessed 10 Aug 2023.

8. 2016 Population By-census, Hong Kong SAR Government. District profiles. Available from: https://www.bycensus2016.gov.hk/en/bc-dp.html. Accessed 1 May 2020.

9. World Health Organization & International Conference for the Ninth Revision of the International Classification of Diseases. Manual of the international statistical classification of diseases, injuries, and causes of death: based on the recommendations of the ninth revision conference, 1975, and adopted by the Twenty-ninth World Health Assembly, 1975 revision. 2021. Available from: https://apps.who.int/iris/handle/10665/40492. Accessed 19 Jul 2023.

10. Hospital Authority, Hong Kong SAR Government. Operations. List of private services. Available from: https://www3.ha.org.hk/fnc/Operations.aspx?lang=ENG. Accessed 19 Jul 2023.

11. R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. Available from: https://www.R-project.org/. Accessed 17 Jul 2023.

12. Yee TW. Vector generalized linear and additive models: with an implementation in R. Springer. Available from: https://link.springer.com/book/10.1007/978-1-4939-2818-7. Accessed 17 Jul 2023. Crossref

13. Hunt KA, Weber EJ, Showstack JA, Colby DC, Callaham ML. Characteristics of frequent users of emergency departments. Ann Emerg Med 2006;48:1-8. Crossref

14. Lee WL, Chen WT, Hsiao FH, Huang CH, Huang LY. Characteristics and resource utilization associated with frequent users of emergency departments. Emerg Med Int. 2022;2022:8064011. Crossref

15. Leporatti L, Ameri M, Trinchero C, Orcamo P, Montefiori M. Targeting frequent users of emergency departments: prominent risk factors and policy implications. Health Policy 2016;120:462-70. Crossref

16. Fuda KK, Immekus R. Frequent users of Massachusetts emergency departments: a statewide analysis. Ann Emerg Med 2006;48:9-16. Crossref

17. Smith SM, Wallace E, O’Dowd T, Fortin M. Interventions for improving outcomes in patients with multimorbidity in primary care and community settings. Cochrane Database Syst Rev 2021;1:CD006560. Crossref

18. Reilly S, Miranda-Castillo C, Malouf R, et al. Case management approaches to home support for people with dementia. Cochrane Database Syst Rev 2015;1:CD008345. Crossref

19. Lam LC, Wong CS, Wang MJ, et al. Prevalence, psychosocial correlates and service utilization of depressive and anxiety disorders in Hong Kong: the Hong Kong Mental Morbidity Survey (HKMMS). Soc Psychiatry Psychiatr Epidemiol 2015;50:1379-88. Crossref

20. Food and Health Bureau, Hong Kong SAR Government. Mental health review report. Available from: https://www.fhb.gov.hk/download/press_and_publications/otherinfo/180500_mhr/e_mhr_full_report.pdf. Accessed 1 May 2020.