Research Epidemiological characteristics and outcomes of COVID-19 cases: mortality inequalities by socio- economic status, Barcelona, Spain, 24 February to 4 May 2020 Julieta Politi1,2, Mario Martín-Sánchez1,2, Lilas Mercuriali¹, Blanca Borras-Bermejo³, Joaquín Lopez-Contreras⁴, Anna Vilella⁵, Judit Villar⁶, COVID-19 Surveillance Working Group of Barcelona⁷, Angels Orcau1,8,9, Patricia Garcia de Olalla1,8,9,10, Cristina Rius1,8,9,11 1. Epidemiology Service, Public Health Agency of Barcelona (PHAB), Barcelona, Spain 2. Preventive Medicine and Public Health Training Unit, PSMar-UPF-PHAB (Parc de Salut Mar - Pompeu Fabra University - Public Health Agency of Barcelona), Barcelona, Spain 3. Department of Preventive Medicine and Epidemiology, Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain 4. Infectious Diseases-Internal Medicine, Hospital de Sant Pau-Universitat Autònoma de Barcelona, Barcelona, Spain 5. Preventive Medicine and Epidemiology Department, Hospital Clínic, Barcelona, Spain 6. Service of Infectious Diseases, Hospital del Mar, Barcelona, Spain 7. Members are listed under Investigators 8. CIBER Epidemiología y Salud Pública (CIBERESP), Barcelona, Spain 9. Institut d’Investigació Biomèdica Sant Pau (IIB Sant Pau), Barcelona, Spain 10. Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Barcelona, Spain 11. Department of Pediatrics, Obstetrics, Gynecology, Preventive Medicine and Public Health, Universitat Autònoma de Barcelona, Barcelona, Spain Correspondence: Patricia Garcia de Olalla (polalla@aspb.cat) Investigators: The investigators are listed at the end of the article Citation style for this article: Politi Julieta, Martín-Sánchez Mario, Mercuriali Lilas, Borras-Bermejo Blanca, Lopez-Contreras Joaquín, Vilella Anna, Villar Judit, COVID-19 Surveillance Working Group of Barcelona, Orcau Angels, de Olalla Patricia Garcia, Rius Cristina. Epidemiological characteristics and outcomes of COVID-19 cases: mortality inequalities by socio-economic status, Barcelona, Spain, 24 February to 4 May 2020 . Euro Surveill. 2021;26(20):pii=2001138. https://doi.org/10.2807/1560-7917. ES.2021.26.20.2001138 Article submitted on 07 Jun 2020 / accepted on 09 Feb 2021 / published on 20 May 2021 Background: Population-based studies characteris- death (≥ 3, adjusted odds ratio: 2.3). Age-standardised ing outcomes of COVID-19 in European settings are rates (incidence, hospitalisation, mortality) were high- limited, and effects of socio-economic status (SES) est in the most deprived SES quartile (incidence: 1,011 on outcomes have not been widely investigated. Aim: (95% confidence interval (CI): 975–1,047); hospitali- We describe the epidemiological characteristics of sation: 619 (95% CI: 591–648); mortality: 150 (95% COVID-19 cases, highlighting incidence and mortal- CI: 136–165)) and lowest in the most affluent (inci- ity rate differences across SES during the first wave dence: 784 (95% CI: 759–809); hospitalisation: 400 in Barcelona, Catalonia, Spain. Methods: This popula- (95% CI: 382–418); mortality: 121 (95% CI: 112–131)). tion-based study reports individual-level data of lab- Conclusions: COVID-19 outcomes varied markedly oratory-confirmed COVID-19 cases diagnosed from 24 across SES, underscoring the need to implement effec- February to 4 May 2020, notified to the Public Health tive preventive strategies for vulnerable populations. Agency of Barcelona and followed until 15 June 2020. We analysed end-of-study vital status and the effects Introduction of chronic conditions on mortality using logistic Coronavirus disease (COVID-19) surveillance in Spain regression. Geocoded addresses were linked to basic was implemented on 20 January 2020 to identify early health area SES data, estimated using the composed cases and minimise onward transmission. The first con- socio-economic index. We estimated age-standard- firmed case in Spain was reported in La Gomera (Canary ised incidence, hospitalisation, and mortality rates Islands) on 31 January 2020 [1]. Recent estimates sug- by SES. Results: Of 15,554 COVID-19-confirmed cases, gest that community transmission was ongoing before the majority were women (n = 9,028; 58%), median March throughout Spain [2,3], although this was not age was 63 years (interquartile range: 46–83), 8,046 evident until mid-March. In response, on 14 March, the (54%) required hospitalisation, and 2,287 (15%) cases Spanish government declared a strict lockdown, which died. Prevalence of chronic conditions varied across included social distancing measures, home confine- SES, and multiple chronic conditions increased risk of ment, banning of social gatherings, school closures, www.eurosurveillance.org 1 Figure Epidemic curve of laboratory-confirmed COVID-19 cases notified to the Public Health Agency of Barcelona, Spain, 24 February–4 May 2020 (n = 15,554) Count 600 400 600 200 0 Daily confirmed cases 400 200 0 20 Feb 27 Feb 05 Mar 12 Mar 19 Mar 26 Mar 02 Apr 09 Apr 16 Apr 23 Apr 30 Apr 07 May Date of diagnosis COVID-19: coronavirus disease. and restriction on civilian mobility, among other a unique opportunity to characterise disease epidemi- measures [4]. On 4 May, the government eased these ology thoroughly. In this regard, monitoring whether restrictions following a sustained decrease in COVID- the COVID-19 pandemic exacerbates pre-existing 19 cases throughout Spain [5]. From 5 February 2020, health inequalities is a relevant perspective within any numerous suspected COVID-19 cases were notified in context to inform rapid policy responses that ensure Barcelona. The first confirmed COVID-19 case in the health equity. However, data on the influence of socio- city of Barcelona occurred on 24 February 2020. In the economic status (SES) and other social determinants following 10 weeks, 15,554 cases were confirmed. on COVID-19 susceptibility and severity are scarce, limiting the capacity of identifying disadvantaged Europe, which is characterised by an aging popula- populations [8]. Geographically aggregated data offers tion, was strongly affected by the COVID-19 pandemic, valuable information for studying health inequalities despite well-established health systems with high by accounting for the contextual effect of the area of coverage [6]. During the start of the epidemic, Spain residence [9]. This approach is especially relevant in was one of the most affected countries worldwide in urban areas where health inequalities tend to be more terms of COVID-19 incidence and mortality rates [7]. marked [10]. Characterising and contextualising these figures is relevant to modify the course of the epidemic. In addi- The purpose of this study was to describe the epidemi- tion to describing the magnitude and severity of the ological characteristics and outcomes of the first con- crisis, regional population-wide surveillance provides secutive 15,554 laboratory-confirmed COVID-19 cases, 2 www.eurosurveillance.org as well as the cumulative incidence and mortality rate different sequences of severe acute respiratory syn- differences across SES during the first wave of COVID- drome coronavirus 2 (SARS-CoV-2) [14]. Individuals 19 in Barcelona, Catalonia, Spain from 24 February to were considered a confirmed case if both test results 4 May 2020. were positive. Since 2 April, following the Catalonia health department protocol, one positive SARS-CoV-2 Methods RT-PCR result was sufficient for confirmation [15]. Design Data sources This population-based study reports RT-PCR-confirmed Data for this study were extracted from the Mandatory COVID-19 cases notified to the Agència de Salut Pública Registry for Infectious Diseases of the city of Barcelona. de Barcelona (Public Health Agency of Barcelona, Following a case notification, the PHAB contacted each PHAB) and diagnosed in the city of Barcelona from 24 case, and trained public health professionals admin- February to 4 May 2020. istered an epidemiological questionnaire that was entered into the Registry. Case records included demo- We included all cases diagnosed up to 4 May 2020, graphic and clinical information, such as date of birth, coinciding with the easing of lockdown and restriction sex, occupation, address, date of symptom onset, date measures in Barcelona and the sustained decrease in of laboratory diagnosis, presenting symptoms, chronic COVID-19 cases within the city [5]. conditions, travel history, contact with a confirmed case, hospitalisation requirements, intensive care unit Surveillance strategy (ICU) stays, and death. We defined healthcare work- Epidemiological surveillance and control of notifiable ers (HCW) as individuals who work in a health facility. infectious diseases is a well-established PHAB-based The PHAB monitored cases’ contacts, and appropriate system in Barcelona, which is responsible for monitor- health recommendations were provided for both cases ing and controlling communicable disease among ca and contacts. Public health nurses followed cases until 1.6 million residents. Surveillance relies on active case their resolution. finding at hospitals, primary healthcare centres, nurs- ing homes, laboratories, and private practices. Prompt Of the symptoms and clinical presentation registered, notification of suspected and confirmed cases to PHAB severe forms of illness, defined as pneumonia (clinical is required. The PHAB is part of the Catalonian sur- or radiological), dyspnoea, acute respiratory distress veillance network, which is operated by the Catalonia syndrome (ARDS), and acute kidney injury (AKI) are health department. presented here. Other forms, categorised as mild, are described in the Supplement (fever, cough, diarrhoea, In response to the public health emergency of inter- chills, sore throat, headache, vomiting, weakness, national concern declared by the World Health myalgia, anosmia, and dysgeusia). Chronic conditions Organisation (WHO) on 30 January 2020 and in accord- were reported for diabetes, cardiovascular disease ance with the guidelines of the Catalonia health (including hypertension), chronic liver disease, chronic department, the PHAB strengthened surveillance of respiratory disease, chronic kidney disease, neurologic respiratory diseases [11]. The COVID-19 case definition disease, and cancer. was adapted from that published by the ECDC; a sus- pected case was defined as any individual presenting Outcomes included hospitalisation, ICU stay during with an acute respiratory disease with at least one of hospitalisation, and end-of-study vital status (alive/ the following symptoms: cough, fever or shortness of deceased). Deceased status was informed through breath, and a recent travel history to an affected area or hospitals, the city’s funeral homes, and the central being a close contact of a confirmed or probable case registry of insured persons (RCA: a registry recording [12]. Initially, affected areas included China, Japan, all individuals insured through the universal public South Korea, Singapore, and Iran [13]. The affected healthcare system). Cases’ vital status and information northern regions of Italy (Lombardy, Emilia Romagna, was updated until 15 June 2020, to allow a minimum Veneto, and Piedmont) were added in late February follow-up of 6 weeks. 2020. By early March 2020, the suspected case defi- nition evolved to include severe pneumonia cases in Individual SES was estimated using the composed hospitalised patients in whom no other aetiology could socio-economic index, an index used for allocating be determined (a minimal screening including at least resources to the Catalonian primary healthcare sys- influenza viruses was required to be negative). Case tem since 2017 [16]. The index estimates deprivation definitions evolved rapidly over the following days by basic health area (BHA), by considering the follow- driven by test shortages, prioritising laboratory confir- ing parameters: exemption from pharmaceutical co- mation for hospitalised cases and healthcare workers payments, income below EUR 18,000, income above while supplies were scarce, until early April. EUR 100,000, manual employment, insufficient school- ing (population > 16–74 years unable to read or write, or Laboratory analysis has only primary education), premature mortality (< 75 As recommended by the WHO, suspected cases were years), and potentially avoidable hospitalisations. The tested with two real-time RT-PCR assays targeting index, scaled to the Catalonian population, defines www.eurosurveillance.org 3 Table 1 Characteristics by end-of-study status in laboratory-confirmed COVID-19 cases notified to the Public Health Agency of Barcelona, Spain, 24 February–4 May 2020 (n = 15,554) Overall Alive Deceased End-of-study status n = 15,554 n = 13,267 n = 2,287 p value n % n % n % Age (median (IQR)) 63 (46–83) 59 (43–78) 85 (77–90) < 0.001 Age (years)a 0–14 44 0.3 44 0.3 0 0.0 15–44 3,372 21.7 3,361 25.3 11 0.5 45–64 4,443 28.6 4,334 32.7 109 4.8 < 0.001 65–74 1,861 12.0 1,571 11.9 290 12.7 75–84 2,421 15.6 1,723 13.0 698 30.5 ≥ 85 3,404 21.9 2,225 16.8 1,179 51.6 Sex < 0.001 Female 9,028 58.0 7,863 59.3 1,165 50.9 Healthcare workers 2,558 16.4 2,540 19.1 18 0.8 < 0.001 Hospitalised 8,406 54.0 6,610 49.8 1,796 78.5 < 0.001 ICU 859 5.5 623 4.7 236 10.3 < 0.001 Symptomsb Pneumonia 5,482 48.9 4,453 46.8 1,029 60.7 < 0.001 Dyspnoea 5,243 46.8 4,141 43.5 1,102 65.0 < 0.001 ARDS 596 5.3 379 4.0 217 12.8 < 0.001 AKI 368 3.3 209 2.2 159 9.4 < 0.001 Chronic conditions b Cardiovascular 4,010 35.8 2,855 30.0 1,155 68.1 < 0.001 Diabetes 1,707 15.2 1,222 12.8 485 28.6 < 0.001 Respiratory 1,473 13.1 1,103 11.6 370 21.8 < 0.001 Neurologic 1,297 11.6 863 9.1 434 25.6 < 0.001 Kidney disease 905 8.1 547 5.8 358 21.1 < 0.001 Cancer 870 7.8 605 6.4 265 15.6 < 0.001 Liver disease 316 2.8 235 2.5 81 4.8 < 0.001 SESc Low 3,029 19.9 2,611 20.1 418 18.3 Medium-low 3,973 26.0 3,393 26.1 580 25.4 0.033 Medium-high 4,478 29.3 3,809 29.4 669 29.3 High 3,778 24.8 3,163 24.4 615 27.0 Days from symptom onset to diagnosis (median (IQR)) 7 (3–11) 7 (3–11) 5 (2–9) < 0.001 AKI: acute kidney injury; ARDS: acute respiratory distress syndrome; COVID-19: coronavirus disease; ICU: intensive care unit upon admission; IQR: interquartile range; SES: socio-economic status. a Age was unavailable for nine cases. b Multiple symptoms and/or chronic conditions can exist for each case. c SES was unavailable for 296 cases. Values are n (%) unless otherwise specified. Percentages for symptoms and chronic conditions are calculated based on the number of surveyed cases (n = 11,210; 9,514 alive and 1,696 deceased). 4 www.eurosurveillance.org a BHA deprivation range between 0 (lowest) and 100 Spanish Law 03/2018 on Data Protection. This study (highest). We geocoded the address for each case and was approved by the ethics committee of the Parc de assigned the corresponding socio-economic index of Salut Mar (CEIC-Parc de Salut MAR, registration num- the given BHA for this study. Addresses were unavail- ber: 2020/9356), Barcelona. able for 294 cases. We used quartile cut-points based on the index’s distribution within Barcelona to define Results four SES categories: low (most deprived), medium-low, medium-high, high (most affluent). Epidemiological characteristics Overall, 25,381 COVID-19 notifications were compiled. Analysis A total of 15,554 cases were laboratory confirmed Descriptive analyses were conducted to summarise key and 7,939 cases (31.3%) were discarded. The remain- variables. Epidemiological characteristics were com- ing 1,888 cases were non-laboratory confirmed, and pared by case status at the end of the study period classified as epidemiologically-confirmed (compatible (alive/deceased) using the chi-squared, Wilcoxon symptoms and known contact with a case, but no con- rank-sum, or Kruskal–Wallis test, as appropriate. A p firmatory testing) in 873 (3.4%), chest X-ray compatible value < 0.05 was used to determine statistical signifi- with viral pneumonia (without laboratory confirmation) cance in all analyses. A subset analysis was performed in 373 (1.5%) cases, and possible cases (consistent for nursing home residents. The date of laboratory con- symptoms, without known exposures and no confirma- firmation was used to plot the epidemic curve. When tory testing) in 642 (2.5%). the date of laboratory confirmation was unknown, the notification date was used to plot the epidemic curve. The epidemic curve, based on the date of laboratory confirmation, is shown in the Figure. The first con- We calculated crude and standardised cumulative inci- firmed case in Barcelona was diagnosed on 24 February dence and mortality rates per 100,000 inhabitants, 2020. Most cases occurred between 15 March and 24 stratified by SES and sex. For individuals aged 18 years April, with one peak on 27 March and a second peak and older, we investigated the relationship between on 17 April. The median time between symptom onset chronic conditions and death using logistic regres- and date of diagnosis varied before and after the strict sion models (crude and adjusted). Individuals younger lockdown decree on 14 March: the initial delay was 3 than 18 years of age were excluded from this analysis, days, which increased over time reaching a median given their low risk of death and low prevalence of of 7 days [4]. A week after the initial case (3 March chronic conditions. We obtained adjusted odds ratios 2020), all confirmed cases (n = 16) reported either (aOR) for the risk of death by adjusting each chronic recent travel to an area with widespread transmission condition for age and sex. Crude (age-specific) rates (eight cases reported travel to Italy) or contact with a were obtained by dividing the number of cases in each confirmed imported case. Following this period, 4,495 stratum by the corresponding stratum population size. (28.9%) cases reported close contact with a case, We used a direct standardised method to calculate while the exposure was unknown for the remaining standardised cumulative incidence and mortality rates, cases. Recent travel history was reported by 71 cases, using Barcelona’s total population as the reference. with Italy being the most frequent destination (31.0%). Population estimates were based on the RCA (as of 1 Jan 2020). We calculated the case fatality rate (CFR) Clinical presentation and outcomes as the number of deaths within a category divided by Of the 15,554 confirmed cases, 9,028 (58.0%) were each category’s number of cases. All statistical analy- women, and the median age was 63 years (inter- ses were conducted using Stata (version 15, StataCorp, quartile range (IQR): 46–83) (Table 1). Almost half of College Station, Texas, United States (US)) and R soft- the cases (49.4%) were older than 65 years. Only 44 ware version 3.5.2 (R Foundation, Vienna, Austria). (0.3%) cases occurred in children 14 years or younger, of whom 12 were under 1 year of age. A total of 8,406 Ethical statement (54.0%) cases required hospitalisation, and 859 cases Patients were not directly involved in this study. Only (5.5%) required ICU admission. Death occurred in data extracted from notifiable disease surveillance 2,287 cases (14.7%). Information on symptoms and systems were used. All identifiable personal informa- past medical history were available in 11,210 cases. At tion was removed for privacy protection, and therefore presentation, 5,482 (48.9%) cases had pneumonia and no informed consent was required. Data confidentiality 5,243 (46.8%) reported dyspnoea. These proportions and other ethical considerations were handled accord- were significantly higher among cases with a fatal out- ing to the international recommendations about epi- come than those surviving at the end of the study (p demiological studies mentioned in the International value < 0.001). The prevalence of ARDS and AKI was less Guidelines for Ethical Review of Epidemiological frequent, although more than 30% of cases were fatal. Studies (Council for the International 12 Organizations HCW cases totalled 2,558 (16.4%). of Medical Sciences –CIOMS, Geneve, 1991), the Helsinki Declaration revised by the World Medical The median time between symptom onset and diagno- Organisation in Fortaleza in 2013, the General Data sis was 7 days (IQR: 3–11). This interval was shorter for Protection Regulation (GPDR) EU 2016/679, and the individuals who died by the end of the study (5 days; www.eurosurveillance.org 5 Table 2 Characteristics by end-of-study status in laboratory-confirmed COVID-19 cases in nursing home residents in the city of Barcelona, 24 February–4 May 2020 (n = 3,137) Overall Alive Deceased End-of-study status n = 3,137 n = 2,334 n = 803 p value n % n % n % Age (median (IQR)) 88 (83–92) 88 (82–92) 88 (84–93) < 0.001 Sex < 0.001 Female 2,323 74.1 1,801 77.2 522 65.0 Hospitalised 1,064 33.9 561 24.0 503 62.6 < 0.001 ICU 33 1.1 16 0.7 17 2.1 < 0.001 Symptomsa Pneumonia 581 46.0 300 40.2 281 54.4 < 0.001 Dyspnoea 686 54.3 343 46.0 343 66.3 < 0.001 ARDS 116 9.2 45 6.0 71 13.7 < 0.001 AKI 79 6.3 39 5.2 40 7.7 < 0.001 Chronic conditionsa Cardiovascular 718 56.8 397 53.2 321 62.1 < 0.001 Diabetes 297 23.5 168 22.5 129 25.0 < 0.001 Respiratory 207 16.4 124 16.6 83 16.1 < 0.001 Neurological 493 39.0 266 35.7 227 43.9 < 0.001 Kidney disease 236 18.7 128 17.2 108 20.9 < 0.001 Cancer 113 8.9 58 7.8 55 10.6 < 0.001 Liver disease 39 3.1 22 2.9 17 3.3 < 0.001 SES Low 395 12.6 295 12.6 100 12.5 Medium-low 928 29.6 712 30.5 216 26.9 0.035 Medium-high 941 30.0 695 29.8 246 30.6 High 873 27.8 632 27.1 241 30.0 Days from symptom onset to diagnosis (median (IQR)) 4 (1–9) 5 (2–11) 3 (1–8) < 0.001 AKI: acute kidney injury; ARDS: acute respiratory distress syndrome; COVID-19: coronavirus disease; ICU: intensive care unit upon admission; IQR: interquartile range; SES: socio-economic status; NA: not applicable. a Multiple symptoms and/or chronic conditions can exist for each case. Values are n (%) unless otherwise specified. Percentages for symptoms and chronic conditions are calculated based on the number of surveyed cases (n = 1,263; 746 alive and 517 deceased). IQR: 2–9) (Table 1). Among HCW, the median time from onset and diagnosis was 4 days (IQR: 1–9) and 3 days symptom onset to diagnosis was 5 days (IQR: 2–9). The (IQR: 1–8) for those who died. median time between symptom onset to diagnosis was significantly lower for the most favourable SES group, Chronic conditions which took on average a day less than the lower three For individuals aged 18 years and above, chronic condi- quartiles (6 days; IQR: 2–10; p = 0.0001). tions were absent in 5,143 (46.1%) cases (Table 3). The risk of death increased as the number of pre-existing Nursing home residents conditions increased, reaching an aOR of 2.3 (95% A total of 3,137 (20.2%) cases were nursing home res- confidence interval (CI): 1.9–2.8) when three or more idents (Table 2). The median age was 88 years (IQR: chronic conditions were present. Chronic conditions 83–92), and 74.1% were women. Hospitalisation was were more prevalent among fatal cases than those required in 33.9% of residents, and only 33 (1.1%) were who remained alive, of which cardiovascular disease admitted to the ICU. The CFR among nursing home (including hypertension) was the most frequent, fol- residents was 25.6%. The most frequent presenta- lowed by diabetes. The prevalence of chronic condi- tions among residents were dyspnoea and pneumonia tions and their associated risk of death varied across (54.3% and 46.0%, respectively). Chronic conditions SES, with a clear gradient for cardiovascular disease. were highly prevalent, with cardiovascular disease present in 56.8%, followed by neurological disease in Case fatality rate 39.0% of cases. The median time between symptom A total of 2,287 cases died by the end of the study period, with an overall CFR of 14.7%. The CFR was 6 www.eurosurveillance.org Table 3 Prevalence of chronic conditions and associated risk of death for COVID-19 cases aged 18 years and above, stratified by socio-economic status, Barcelona, Spain, 24 February–4 May 2020 (n = 11,165) Total Prevalence (%) 95% CI Deceased aOR 95% CI p value Overall Cardiovascular 3,707 33.2 32.3–34.1 1,102 1.4 1.2–1.6 < 0.001 Diabetes 1,583 14.2 13.5–14.8 460 1.3 1.1–1.5 0.001 Respiratory 1,378 12.3 11.7–13.0 358 1.3 1.1–1.5 0.001 Neurological 1,087 9.7 9.2–10.3 403 1.5 1.3–1.8 < 0.001 Kidney disease 822 7.4 6.9–7.9 336 1.7 1.4–2.0 < 0.001 Cancer 781 7.0 6.5–7.5 243 1.5 1.3–1.8 < 0.001 Liver disease 296 2.7 2.4–3.0 78 1.5 1.1–2.0 0.008 Chronic conditions (number) 0 5,143 46.1 45.1–47.0 240 Ref. 1 2,537 22.7 22.0–23.5 389 1.4 1.1–1.7 0.001 2 1,968 17.6 16.9–18.3 524 1.9 1.6–2.3 < 0.001 ≥3 1,517 13.6 13.0–14.2 543 2.3 1.9–2.8 < 0.001 SES low (n = 2,300) Cardiovascular 847 36.8 34.9–38.8 233 1.7 1.2–2.3 0.001 Diabetes 417 18.1 16.6–19.7 111 1.4 1.0–1.8 0.036 Respiratory 333 14.5 13.0–15.9 88 1.6 1.2–2.2 0.003 Neurological 175 7.6 6.5–8.7 49 1.2 0.8–1.7 0.454 Kidney disease 176 7.7 6.6–8.7 58 1.3 0.9–1.9 0.151 Cancer 127 5.5 4.6–6.5 44 2.0 1.3–3.1 0.002 Liver disease 90 3.9 3.1–4.7 27 2.2 1.3–3.7 0.005 SES medium-low (n = 2,763) Cardiovascular 927 33.6 31.8–35.3 284 1.5 1.2–1.9 0.002 Diabetes 418 15.1 13.8–16.5 121 1.2 0.9–1.5 0.303 Respiratory 346 12.5 11.3–13.8 97 1.5 1.1–2.0 0.015 Neurological 259 9.4 8.3–10.5 95 1.4 1.1- 1.9 0.024 Kidney disease 198 7.2 6.2–8.1 82 1.7 1.2–2.3 0.003 Cancer 201 7.3 6.3–8.2 56 1.3 0.9–1.8 0.223 Liver disease 62 2.2 1.7–2.8 14 1.2 0.6–2.4 0.591 SES medium-high (n = 3,225) Cardiovascular 1,034 32.1 30.5–33.7 304 1.4 1.1–1.7 0.008 Diabetes 386 12.0 10.9–13.1 110 1.2 0.9–1.6 0.193 Respiratory 419 13.0 11.8–14.2 101 1.1 0.8–1.4 0.551 Neurological 335 10.4 9.3–11.4 133 1.9 1.4–2.5 < 0.001 Kidney disease 228 7.1 6.2–8.0 101 2.0 1.5–2.8 < 0.001 Cancer 250 7.8 6.8–8.7 84 1.8 1.3–2.5 < 0.001 Liver disease 82 2.5 2.0–3.1 23 1.7 1.0–2.9 0.073 SES high (n = 2,720) Cardiovascular 866 31.8 30.1–33.6 280 1.3 1.0–1.6 0.046 Diabetes 350 12.9 11.6–14.1 118 1.4 1.1–1.9 0.009 Respiratory 269 9.9 8.8–11.0 72 1.2 0.8–1.6 0.358 Neurological 317 11.7 10.5–12.9 126 1.5 1.1–2.0 0.006 Kidney disease 216 7.9 6.9–9.0 94 1.6 1.1–2.1 0.006 Cancer 197 7.2 6.3–8.2 58 1.2 0.9–1.8 0.272 Liver disease 62 2.3 1.7–2.8 14 0.9 0.5–1.8 0.804 aOR: adjusted odds ratio by age and sex; CI: confidence intervals; COVID-19: coronavirus disease; Ref.: reference; SES: socio-economic status. This analysis included 11,165 cases (15,496 cases were > 18 years, of whom 4,331 were excluded since they were not surveyed for chronic conditions). Independent logistic regression models were fitted for each chronic condition (adjusted by age and sex). For each model, the reference category was the absence of the condition being analysed. SES was missing in 157 cases (n = 11,008). Multiple chronic conditions can exist for each case. www.eurosurveillance.org 7 highest in individuals 65 years and older (28.2%). study period, we consider that our results best describe Among the deceased, 1,796 (78.5%) required hospitali- moderate and severe COVID-19 cases. Therefore, CFR sation and 236 (10.3%) ICU admission. The CFR among should be interpreted with caution, as with any ongoing hospitalised cases was 21.4% and 27.5% among cases epidemic, because the denominator remains unknown admitted to ICU (Table 1). Differences in CFR across SES (given limited testing capacity and an unknown count were not statistically significant: the CFR was 13.8% of mild or asymptomatic cases). However, if we con- (95% CI: 12.4–15.2) for low SES, 14.6% (95% CI: 13.3– sider the results from the first round of the national 15.9) for medium-low SES, 14.9% (95% CI: 13.7–16.2) seroprevalence study, which was carried out at the end for medium-high SES, and 16.3% (95% CI: 14.9–17.7) of our study period, the estimated seroprevalence for for high SES. Barcelona was at 7.1% [20]. Considering Barcelona’s population (1.6 million), the total number of infections Incidence and mortality estimates would total 113,600. Following these assumptions and Cumulative incidence, hospitalisation, and mortal- expecting an accurate number of deaths, a rough pop- ity rates per 100,000 habitants are shown in Table 4. ulation-wide CFR estimate could approximate 2.01%, Overall cumulative incidence rate was 946.8 (95% CI: similar to those reported for Wuhan (2.2%) or China 932.0–961.5). The overall cumulative mortality rate (2.3%) [21,22]. Our surveillance system may offer more was 139.3 (95% CI: 133.6–145.0). While cumulative inci- reliable estimates of hospital CFR than previous reports dence rate was higher in female cases, the cumulative due to greater efficiency in capturing deaths occurring mortality rate was significantly higher in male cases. outside hospital settings (such as nursing homes or Age-specific incidence and mortality rates increased personal residencies) [18,23]. We report a CFR for hos- with age. Age-standardised cumulative incidence rates pitalised patients of 21.4%, which is lower than values across SES showed a clear gradient, with the highest reported for China, while similar to those reported in observed incidence rates in the group with lowest SES the US [18,23]. While these findings are surprising, (1,010.9; 95% CI: 975.1–1,046.7). The age-standardised given the differences in population and healthcare cumulative mortality rate across SES also revealed a characteristics between Spain and these two coun- gradient, with the highest mortality rate observed in tries, the nature of our data (population-wide vs hospi- the lowest SES quartile, which was significantly higher tal case series) may explain why our hospital CFR does when compared with the mortality rate observed of the not differ from those in the US. A relative preparedness most affluent strata, 150.2 (95% CI: 135.9–164.6) and gained from China’s experience could explain the hos- 121.2 (95% CI: 111.6–130.7), respectively. Cumulative pital-based CFR were lower in our setting. age-specific incidence and mortality rates by sex are shown in the Supplement, which display similar gradi- SES has long been recognised as a determinant in ents by SES in male and female cases. the incidence of infectious diseases. Previous stud- ies suggest that pandemic outcomes are influenced Discussion by SES [24-26]. Despite this, a recent review of inter- We report results from a large population-wide cohort national pandemic preparedness plans highlighted the of confirmed COVID-19 cases in Spain during the lack of consideration given to social inequalities [27]. first wave of the pandemic. The start of the epidemic We report a clear incidence and mortality rate gradi- (February to May 2020) in the city of Barcelona exem- ent between different SES within our city. For instance, plifies a rapidly evolving situation, consistent with the mortality rate spanned from 121.2 per 100,000 inhabit- ongoing COVID-19 pandemic [17]. The epidemic curve ants (95% CI: 112–131) in the most affluent SES quar- fits with a propagated epidemic pattern, with suc- tile to 150.2 (95% CI: 136–165) in the lowest and most cessively larger peaks with each incubation period, deprived SES quartile. We hypothesise that several and a peak at 4 weeks after the first confirmed case. mechanisms could explain our findings. Firstly, quar- Case distribution by age, presenting symptoms, and antine measures may increase health inequalities, chronic conditions was similar to previous reports from especially among the disadvantaged. In this sense, China and the US [18,19]. In our study, hospitalisation increased exposure between crowded household was required frequently (54.0%), and the CFR among members is highly possible and probably coupled with hospitalised cases was 21.4%. Cumulative incidence other factors including low home quality, and lack of rates were higher in women and with increasing age. ventilation or proper disinfection measures. As with Conversely, as previously reported in other studies, the influenza, household crowding, along with SES and cumulative mortality rate was higher in males and indi- poverty, have been positively associated with hospital- viduals 65 years and older [18]. Cumulative incidence isation rates [25,26,28,29]. Furthermore, most cluster and mortality rates varied across SES, with a clear transmission reported in China occurred within family gradient demonstrating higher incidence and mortality settings, which is likely to reflect our setting of large rates in those most deprived. extended families residing together in crowded homes [30]. Secondly, individuals in more disadvantaged SES Our study was based on COVID-19 laboratory-con- may be more dependent on public transportation, or firmed cases, and mortality rates reflect deaths occur- employed in essential jobs for which working from ring in this group. Since testing remained available home is not possible. Thirdly, as we and others have for cases seeking hospital care throughout the entire shown, the COVID-19 mortality rate is higher among 8 www.eurosurveillance.org Table 4 Cumulative COVID-19 incidence and mortality rates by sex and socio-economic status in the city of Barcelona, Spain, 24 February–4 May 2020 (n = 15,545) Standardised Standardised Standardised cumulative cumulative Confirmed cases Hospitalisations hospitalisation Deaths incidence/100,000 mortality/100,000 rate/100,000 (95% CI) (95% CI) (95% CI) Barcelonaa 15,545 946.8 (932.0–961.5) 8,404 511.8 (501.0–522.7) 2,287 139.3 (133.6–145.0) Sex Female 9,023 968.9 (948.9–988.9) 4,091 434.2 (420.9–447.58) 1,165 110.4 (104.1–116.8) Male 6,522 927.9 (905.4–950.5) 4,313 621.2 (602.5–639.8) 1,122 185.4 (174.0–196.0) SESb Low 3,029 1,010.9 (975.1–1,046.7) 1,837 619.4 (591.1–647.6) 418 150.2 (135.9–164.6) Medium-low 3,973 993.0 (962.5–1,023.6) 2,069 517.7 (495.5–539.9) 580 145.6 (133.8–157.4) Medium-high 4,478 968.0 (940.0–996.1) 2,417 522.3 (501.6–543) 669 145.3 (134.4–156.3) High 3,778 783.9 (759.1–808.9) 1,955 400.0 (382.3–417.7) 615 121.2 (111.6–130.7) CI: confidence intervals; COVID-19: coronavirus disease; SES: socio-economic status. a Measures for Barcelona are unadjusted, given that these represent population totals. b SES was unavailable for 287 cases, of which 126 were hospitalisations and five were deaths. Of the 15,554 cases in the study period, nine were excluded due to missing age. Population data obtained from the central registry of insured persons (as of 1 Jan 2020). Standardised mortality and hospitalisation rates by sex were calculated using Barcelona’s population as reference. those with underlying chronic conditions [18,31]. mortality rates may be under-reported. Excess mortal- Disadvantaged SES has consistently been associated ity methods may be more appropriate to evaluate fur- with a higher prevalence of chronic conditions such ther differences [36-38]. The CFR for this population as heart disease, obesity, and diabetes [32,33]. We was considerable, at 26%, although this is in range expected a higher CFR with lower SES, given the higher with other reports [39]. Our estimates underscore the prevalence of chronic conditions among people of need to implement effective surveillance in these set- lower SES. However, this was not reflected in our data, tings, as well as preventive strategies for vulnerable where no significant differences were observed across populations, especially for elderly people and other SES. While many factors could explain this finding, deprived groups [40]. Further studies comparing dif- such as testing restrictions for people without a severe ferent nursing home settings, such as size, ownership, presentation (overestimating the real CFR) or barri- nurse to patient ratios, patient safety, and quality of ers in access to care, an alternative could be that CFR life are relevant to implement appropriate prevention among severe cases is similar across SES strata in a and control measures. well-developed universal healthcare system. However, our data fail to completely explain the underlying Many factors may have influenced the widespread relationship between SES, chronic conditions, and transmission dynamics. On average, the PHAB receives COVID-19 incidence and mortality rates. Higher disease around 14,000 mandatory notifications of infectious incidence and mortality rates could be explained both diseases yearly. Roughly 10 weeks after the first con- by higher exposure and higher susceptibility to dis- firmed COVID-19 case, notifications increased rapidly, ease. More sophisticated analyses, beyond the scope reaching a total of 25,381 notifications and 15,554 of this study, are needed to better understand these confirmed cases. This rapid rise in notifications made relationships. Despite this, our results emphasise the contact tracing difficult because of limited resources, need to act on relevant determinants of health [34]. which delayed contact identification and implemen- tation of quarantine measures. The strict lockdown We observed a mortality rate gradient across low and decree may have at least partially remedied this defi- medium SES in those older than 65 years but differ- ciency, by considerably reducing the number of contacts ences were not significant. This is relevant because for each case and thereby transmission. A shortage of most nursing homes in Barcelona lie within a medium personal protective equipment (PPE) was prevalent SES BHA, which could have blurred a steeper gradient throughout, and many settings lacked proper PPE train- in mortality rates between these quartiles. To what ing. An additional source of diagnostic lag time was the extent a nursing home’s location represents an indi- requirement of public health agents’ authorisation for vidual’s SES is a valid question. Interestingly, studies all SARS-CoV-2 RT-PCR testing. Until 15 March 2020, a focusing on nursing home inequalities have associ- public health professional reviewed whether each case ated lower quality of care with both nursing home resi- fulfilled epidemiological criteria before test authorisa- dents’ SES and the SES of nursing home’s location tion. Following this period, and driven by test short- [35]. Widespread testing was not available for nurs- age, testing was reserved for HCW and hospitalised ing homes until early April 2020. Therefore, cases and cases, and hospitals were autonomous to prioritize www.eurosurveillance.org 9 tests, expediting the process. Initial case definitions Investigators were highly specific, which may have caused initial COVID-19 Surveillance Working Group of Barcelona: Anna de cases to be missed because the testing criteria were Andres, Dolores Alamo-Junquera, Carmen Gallego, Daniel G Abiétar, Montse Guillaumes, Joan P Millet, Emilia Molinero, not fulfilled. Furthermore, upon a negative RT-PCR test, Daniela Pérez León, Raquel Rodríguez, Miriam Ros, Andrés cases were not required to isolate themselves. False Antón, Xavier Martínez-Gómez, Tomás Pumarola, Magda negatives, along with asymptomatic transmission, Campins, Virginia Pomar, Ferran Navarro, Teresa Puig, Marta were likely additional contributing factors in the spread Blazquez, Inmaculada Soriano, Lourdes Barón, Clara Marín, of the epidemic, in a setting with significant testing Laura de la Torre, Xavier Castells, Margarita Posso, Juan P Horcajada, Maria A Vàrez, Laia Sentís, Miquel Gómez, Leonor constraints. Invernon, Eduardo Padilla, Mara Karaim, Noel Bordon, Francesc Fatjò, Cristina Berbel, Isabel González-Nieto, José This study had some limitations. Firstly, data used for L González, Iván Pelegrín, Eva Bargalló, Antoni Salas, Maria this study most probably represent laboratory-con- C Planes, Gloria García, Diego de Mendoza, Sònia Tortajada, firmed COVID-19 cases who sought care. While testing Natàlia Juan, Jordi Casabon may have varied across centres, standard guidelines probably maintained uniformity across health facili- ties. Secondly, a deceased status only reflects deaths Acknowledgements occurring in individuals with laboratory-confirmed We are indebted to all the ASPB staff for their invaluable COVID-19 and the direct cause of death may be of dedication: Alba Artigas, Amaia Garrido, Andrea Bastida, Andreu Queralt, Anna Fernández, Anna Hernández, Assun another reason. Thirdly, while SES was assessed Suárez, Carmen Serrano, Carolina Requeijo, Celia Sánchez, based on BHA, this may not reflect individual SES sta- Cristina Rey, Daniel Moreno, Elisabeth Carreras, Esperanza tus. While smallest available area data (census tract Peracho, Eva Masdeu, Félix Fernández, Gabriela Barbaglia, area) are recommended for ecological SES estimates Gemma Paez, Glòria Pérez, Hernán Vargas, Hugo Sánchez, to maintain a higher homogeneity within the area, we Ingrid Avellanés, Iolanda Robles, Isabel Marcos, Isabel Sala, Kilian Gozález, Laia Cases, Laura Ricart, Laura Samsó, Lluisa favoured the composed socio-economic index because Cortada, Lluisa Estruga, Lluisa Forns, Lucia Petrova, Oleguer it relied on recent area estimates for each BHA [16,41]. Parès, Mar Oriol, Marc Martí, Marc Picas, María Salvador, Furthermore, the index we used is reported to be con- Marta Arcas, Marta Caballé, Merce Gotsens, Mercè Royo, sistent with Medea index estimates [16]. Fourthly, Montse Vergara, Nabila Sanchez, Natalia Sagarra, Neus under-reporting may have been an issue, especially for Ferrer, Nuria Calzada, Nuria López, Olga Sancho, Patricia González, Pere Simon, Pilar Gorrindo, Rafael Manzanera, ICU stays, since this information was collected primar- Raquel Gallardo, Raquel Prieto, Raul Salobral, Rosa ily at admission. Finally, CFR may be especially under- Puigpinos, Roser Clos, Sandra Gabarro, Sergi Romero, Silvia estimated among cases admitted to ICU, since the Brugueras, Silvia Martín, Sonia Gil, Clara Torreblanca, Elies survival of cases that required ICU care may be more Sánchez, Montse Trullols, Ivan González, David Samblas, appropriately estimated at 90 days [42]. Antoni Guillem, Laia Nogueiro, Elena Parra, Oscar Rama, Lidia Corraliza, Rosa Garuz, Conxita Malgrat, Francesc Sabater, Lluis Rivas, Rocio Valdillo, Irene García, Xavier During the following waves of COVID-19 across Europe, Bonfill, Joan R Villalbí and many more. a higher percentage of cases were younger and asymp- tomatic in Barcelona. However, incidence rates remain higher in the most deprived neighbourhoods [43]. At Conflict of interest the time of revising this report, we could not compare None declared. the first to subsequent waves because of substantial changes regarding epidemiological procedures, the use of an abbreviated epidemiological survey, and the local Authors’ contributions registry’s transition to a centralised regional registry. PGdO and JP designed the study. JP and MM-S conducted the data analysis and performed the case review. JP wrote the Conclusion first draft of the manuscript. PGdO, AO and CR supervised During the first wave of COVID-19 in Barcelona, out- data analysis and data collection. LM, BB-B, JL-C, AV, JV, comes varied by sex, age, chronic conditions, and SES. and the COVID-19 Surveillance Working Group of Barcelona Our results underscore the need to implement effective contributed to data acquisition. All authors were involved in preventive strategies for vulnerable populations, espe- data interpretation, revised the manuscript critically for im- portant intellectual content, approved the final version, and cially for elderly and underprivileged people. Long- agreed to be accountable for all aspects of the manuscript. term social inequities should be reduced in order to lessen health inequalities. These findings may inform public health policy in large cities of Spain, as well References as other countries and regions. Rapid case diagnosis 1. Spiteri G, Fielding J, Diercke M, Campese C, Enouf V, Gaymard with prompt contact tracing should remain a priority to A, et al. First cases of coronavirus disease 2019 (COVID-19) improve system efficiency. Considerable technical and in the WHO European Region, 24 January to 21 February 2020. Euro Surveill. 2020;25(9):2000178. https://doi. human resources should be allocated to public health, org/10.2807/1560-7917.ES.2020.25.9.2000178 PMID: 32156327 without ignoring the health system and social support 2. Saez M, Tobias A, Varga D, Barceló MA. Effectiveness of the measures to flatten the epidemic curve of COVID-19. The case measures that may be required. of Spain. Sci Total Environ. 2020;727:138761. https://doi. org/10.1016/j.scitotenv.2020.138761 PMID: 32330703 3. Coma E, Mora N, Prats-Uribe A, Fina F, Prieto-Alhambra D, Medina-Peralta M. Excess cases of influenza suggest 10 www.eurosurveillance.org an earlier start to the coronavirus epidemic in Spain the New York City area. JAMA. 2020;323(20):2052-9. https:// than official figures tell us: an analysis of primary care doi.org/10.1001/jama.2020.6775 PMID: 32320003 electronic medical records from over 6 million people from 19. Yu C, Lei Q, Li W, Wang X, Liu W, Fan X, et al. Clinical Catalonia. medRxiv. 2020.04.09.20056259. https://doi. characteristics, associated factors, and predicting COVID-19 org/10.1101/2020.04.09.20056259 mortality risk: a retrospective study in Wuhan, China. Am 4. Ministerio de la Presidencia, Relaciones con las Cortes y J Prev Med. 2020;59(2):168-75. https://doi.org/10.1016/j. Memoria Democrática. Legislación consolidada: Real Decreto amepre.2020.05.002 PMID: 32564974 463/2020, de 14 de marzo, por el que se declara el estado 20. Pollán M, Pérez-Gómez B, Pastor-Barriuso R, Oteo J, Hernán de alarma para la gestión de la situación de crisis sanitaria MA, Pérez-Olmeda M, et al. Prevalence of SARS-CoV-2 ocasionada por el COVID-19. [Consolidated legislation: Royal in Spain (ENE-COVID): a nationwide, population-based Decree 463/2020, of March 14, which declares the state of seroepidemiological study. Lancet. 2020;396(10250):535- alarm for the management of the health crisis situation caused 44. https://doi.org/10.1016/S0140-6736(20)31483-5 PMID: by COVID-19]. Boletín Oficial del Estado 2020;67. Madrid: 32645347 Spanish Government. Spanish. Available from: https://www. boe.es/eli/es/rd/2020/03/14/463/con 21. Wu Z, McGoogan JM. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) 5. Ministerio De Sanidad. Orden SND/388/2020, de 3 de mayo, outbreak in China: summary of a report of 72314 cases from por la que se establecen las condiciones para la apertura al the Chinese Center for Disease Control and Prevention. público de determinados comercios y servicios, y la apertura JAMA. 2020;323(13):1239-42. https://doi.org/10.1001/ de archivos, así como para la práctica del deporte profesional jama.2020.2648 PMID: 32091533 y federado. [Order SND/388/2020, of May 3, which establishes the conditions for the opening to the public of certain shops 22. Wu JT, Leung K, Bushman M, Kishore N, Niehus R, de and services, and the opening of archives, as well as for the Salazar PM, et al. Estimating clinical severity of COVID-19 practice of professional and federated sports]. Disposición from the transmission dynamics in Wuhan, China. Nat Med. 4793 del Boletín Oficial del Estado 2020;123. Madrid: Spanish 2020;26(4):506-10. https://doi.org/10.1038/s41591-020-0822- Government. Spanish. Available from: https://www.boe.es/ 7 PMID: 32284616 boe/dias/2020/05/03/pdfs/BOE-A-2020-4793.pdf 23. Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z, et al. Clinical course and 6. Baeten R, Spasova, S, Vanhercke B and Coster S. Inequalities risk factors for mortality of adult inpatients with COVID-19 in access to healthcare. A study of national policies, European in Wuhan, China: a retrospective cohort study. Lancet. Social Policy Network (ESPN). Brussels: European Commission; 2020;395(10229):1054-62. https://doi.org/10.1016/S0140- 2018. Available from: https://ec.europa.eu/social/main.jsp?cat 6736(20)30566-3 PMID: 32171076 Id=738&langId=en&pubId=8152&furtherPubs=yes 24. Rutter PD, Mytton OT, Mak M, Donaldson LJ. Socio-economic 7. Johns Hopkins University (JHU). COVID-19 dashboard by the disparities in mortality due to pandemic influenza in England. Center for Systems Science and Engineering at Johns Hopkins Int J Public Health. 2012;57(4):745-50. https://doi.org/10.1007/ University. Baltimore: JHU. [Accessed: 11 May 2020]. Available s00038-012-0337-1 PMID: 22297400 from: https://coronavirus.jhu.edu/map.html 25. Chandrasekhar R, Sloan C, Mitchel E, Ndi D, Alden N, Thomas 8. Krieger N, Gonsalves G, Bassett MT, Hanage W, Krumholz A, et al. Social determinants of influenza hospitalization in the HM. The fierce urgency of now: closing glaring gaps in US United States. Influenza Other Respir Viruses. 2017;11(6):479- surveillance data on COVID-19. Bethesda: Health Affairs; 2020. 88. https://doi.org/10.1111/irv.12483 PMID: 28872776 Available from: https://www.healthaffairs.org/do/10.1377/ 26. Tam K, Yousey-Hindes K, Hadler JL. Influenza-related hblog20200414.238084/full hospitalization of adults associated with low census tract 9. Borrell C, Ru M, Pasarn MI, Benach J, Kunst AE. (The socioeconomic status and female sex in New Haven County, measurement of inequalities in health). La medición de las Connecticut, 2007-2011. Influenza Other Respir Viruses. desigualdades en salud. Gac Sanit. 2000;14(Supl. 3):20-33. 2014;8(3):274-81. https://doi.org/10.1111/irv.12231 PMID: Spanish. 24382111 10. Borrell C, Pons-Vigués M, Morrison J, Díez E. Factors and 27. Mamelund SE. Social inequality – a forgotten factor in processes influencing health inequalities in urban areas. J pandemic influenza preparedness. Tidsskr Nor Laegeforen. Epidemiol Community Health. 2013;67(5):389-91. https://doi. 2017;137(12-13):911-3. https://doi.org/10.4045/tidsskr.17.0273 org/10.1136/jech-2012-202014 PMID: 23413097 PMID: 28655260 11. World Health Organization (WHO). Statement on the second 28. Sloan C, Chandrasekhar R, Mitchel E, Schaffner W, Lindegren meeting of the International Health IfRegulations (2005) ML. Socioeconomic disparities and influenza hospitalizations, Emergency Committee regarding the outbreak of novel Tennessee, USA. Emerg Infect Dis. 2015;21(9):1602-10. https:// coronavirus (2019-nCoV). Geneva: WHO; 2020. Available from: doi.org/10.3201/eid2109.141861 PMID: 26292106 https://www.who.int/news/item/30-01-2020-statement-on- 29. Yousey-Hindes KM, Hadler JL. Neighborhood socioeconomic the-second-meeting-of-the-international-health-regulations- status and influenza hospitalizations among children: New (2005)-emergency-committee-regarding-the-outbreak-of-nove- Haven County, Connecticut, 2003-2010. Am J Public Health. l-coronavirus-(2019-ncov) 2011;101(9):1785-9. https://doi.org/10.2105/AJPH.2011.300224 12. European Centre for Disease Control and Prevention (ECDC). PMID: 21778498 Case definition for coronavirus disease 2019 (COVID-19), as of 30. World Health Organization (WHO). Report of the WHO-China 29 May 2020. Stockholm: ECDC; 2020. Available from: https:// joint mission on coronavirus disease 2019 (COVID-19). Geneva: www.ecdc.europa.eu/en/covid-19/surveillance/case-definition WHO; 2020. Available from: https://www.who.int/docs/default- 13. World Health Organization (WHO). Coronavirus disease source/coronaviruse/who-china-joint-mission-on-covid-19- 2019 (‎COVID-19)‎: situation report 34. Geneva: final-report.pdf WHO; 2020. Available from: https://apps.who.int/iris/ 31. Chow N, Fleming-Dutra K, Gierke R, Hall A, Hughes M, Pilishvili handle/10665/331220 T, et al. Preliminary estimates of the prevalence of selected 14. World Health Organization (WHO). Laboratory testing for underlying health conditions among patients with coronavirus 2019 novel coronavirus (2019-nCoV) in suspected human disease 2019 - United States, February 12-March 28, 2020. cases. Geneva: WHO; 2020. Available from: https://www.who. MMWR Morb Mortal Wkly Rep. 2020;69(13):382-6. https://doi. int/publications-detail/laboratory-testing-for-2019-novel- org/10.15585/mmwr.mm6913e2 PMID: 32240123 coronavirus-in-suspected-human-cases-20200117 32. Kivimäki M, Batty GD, Pentti J, Shipley MJ, Sipilä PN, 15. Generalitat de Catalunya. Procediment d’actuació enfront Nyberg ST, et al. Association between socioeconomic de casos d’infecció pel nou coronavirus SARS-CoV-2. [Action status and the development of mental and physical health procedure against cases of infection by new SARS-CoV-2 conditions in adulthood: a multi-cohort study. Lancet Public coronavirus]. Barcelona: Generalitat de Catalunya; 2020. Health. 2020;5(3):e140-9. https://doi.org/10.1016/S2468- Catalan. Available from: https://canalsalut.gencat.cat/web/. 2667(19)30248-8 PMID: 32007134 content/_A-Z/C/coronavirus-2019-ncov/material-divulgatiu/ 33. Pathirana TI, Jackson CA. Socioeconomic status and procediment-actuacio-coronavirus.pdf multimorbidity: a systematic review and meta-analysis. 16. Colls C, Mias M, García-Altés A. [A deprivation index to reform Aust N Z J Public Health. 2018;42(2):186-94. https://doi. the financing model of primary care in Catalonia (Spain)]. Gac org/10.1111/1753-6405.12762 PMID: 29442409 Sanit. 2020;34(1):44-50. Catalan. https://doi.org/10.1016/j. 34. Banks J, Marmot M, Oldfield Z, Smith JP. Disease and gaceta.2018.07.015 PMID: 30595339 disadvantage in the United States and in England. 17. World Health Organization (WHO). Coronavirus disease 2019 JAMA. 2006;295(17):2037-45. https://doi.org/10.1001/ (‎‎COVID-19)‎‎: situation report 51. Geneva: WHO; 2020. jama.295.17.2037 PMID: 16670412 https://apps.who.int/iris/handle/10665/331475 35. Park YJ, Martin EG. Geographic disparities in access to nursing 18. Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn home services: assessing fiscal stress and quality of care. T, Davidson KW, et al. , the Northwell COVID-19 Research Health Serv Res. 2018;53 Suppl 1(Suppl Suppl 1);2932-51. Consortium. Presenting characteristics, comorbidities, and https://doi.org/10.1111/1475-6773.12801 PMID: 29131339 outcomes among 5700 patients hospitalized with COVID-19 in www.eurosurveillance.org 11 36. Chen JT, Waterman PD, Krieger N. COVID-19 and the unequal surge in mortality rates in Massachusetts, by city/town and ZIP Code measures of poverty, household crowding, race/ ethnicity,and racialized economic segregation. Harvard Center for Population and Development Studies Working Paper Series. 2020;19(2). Available from: https://cdn1.sph.harvard. edu/wp-content/uploads/sites/1266/2020/05/20_jtc_pdw_ nk_COVID19_MA-excess-mortality_text_tables_figures_ final_0509_with-cover-1.pdf 37. Olson DR, Huynh M, Fine A, Baumgartner J, Castro A, Chan HT, et al. Preliminary estimate of excess mortality during the COVID-19 Outbreak - New York City, March 11-May 2, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(19):603-5. https://doi. org/10.15585/mmwr.mm6919e5 PMID: 32407306 38. Michelozzi P, de’Donato F, Scortichini M, De Sario M, Noccioli F, Rossi P, et al. Mortality impacts of the coronavirus disease (COVID-19) outbreak by sex and age: rapid mortality surveillance system, Italy, 1 February to 18 April 2020. Euro Surveill. 2020;25(19):2000620. https://doi.org/10.2807/1560- 7917.ES.2020.25.19.2000620 PMID: 32431289 39. McMichael TM, Currie DW, Clark S, Pogosjans S, Kay M, Schwartz NG, et al. Epidemiology of Covid-19 in a long- term care facility in King County, Washington. N Engl J Med. 2020;382(21):2005-11. https://doi.org/10.1056/ NEJMoa2005412 PMID: 32220208 40. Danis K, Fonteneau L, Georges S, Daniau C, Bernard-Stoecklin S, Domegan L, et al. High impact of COVID-19 in long-term care facilities, suggestion for monitoring in the EU/EEA, May 2020. Euro Surveill. 2020;25(22):2000956. PMID: 32524949 41. Carstairs V. Deprivation indices: their interpretation and use in relation to health. J Epidemiol Community Health. 1995;49(Suppl 2):S3-8. https://doi.org/10.1136/jech.49. Suppl_2.S3 PMID: 8594130 42. Jammer I, Wickboldt N, Sander M, Smith A, Schultz MJ, Pelosi P, et al. Standards for definitions and use of outcome measures for clinical effectiveness research in perioperative medicine: European Perioperative Clinical Outcome (EPCO) definitions: a statement from the ESA-ESICM joint taskforce on perioperative outcome measures. Eur J Anaesthesiol. 2015;32(2):88-105. https://doi.org/10.1097/EJA.0000000000000118 PMID: 25058504 43. Agència de Salut Pública de Barcelona. Dades diàries de la infecció per coronavirus SARS-CoV-2 (COVID-19) a Barcelona. [Daily data on SARS-CoV-2 coronavirus infection (COVID-19) in Barcelona] Barcelona: Agència de Salut Pública de Barcelona. [Accessed: 2 Jan 2021]. Catalan. Available from: https://aspb.shinyapps.io/ COVID19_BCN/#Distribució_per_sexe_i_edat_dels_casos License, supplementary material and copyright This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0) Licence. You may share and adapt the material, but must give appropriate credit to the source, provide a link to the licence and indicate if changes were made. Any supplementary material referenced in the article can be found in the online version. This article is copyright of the authors or their affiliated in- stitutions, 2021. 12 www.eurosurveillance.org