Incidence of New-Onset and Worsening Heart Failure Before and After the COVID-19 Epidemic Lockdown in Denmark
What Is New?
-
Anecdotally, most clinicians have noted a decline in the number of patients seeking medical care for non-coronavirus disease 2019 (COVID-19)–related causes, which has raised concerns for significant collateral damage. We observed that the admission rates for worsening heart failure and the incidence rates of new-onset heart failure declined by 30% in Denmark after the country locked down.
-
As yet, the lockdown has not impacted mortality in patients with heart failure at a population-based level.
What are the Clinical Implications?
Introduction
On March 11, 2020, the World Health Organization designated coronavirus disease 2019 (COVID-19) a global epidemic (pandemic). The Danish government response and recommendation was a public lockdown and a rapid shift in the focus of the care in the nationwide and public healthcare system. Inpatient and outpatient medical and surgical treatment was restricted to urgent cases, and all emergency departments and intensive care units were ordered to be prepared to receive patients infected with COVID-19. It was also advocated that patients communicated with their general practitioners by telephone to reduce the infection risk. At the same time, several facilities including schools, department stores, malls, and restaurants were closed, and people were strongly encouraged to stay home. Major focus was on preventing infection of high-risk patients with chronic disease and the elderly. Consequently, patients with serious chronic diseases such as heart failure (HF) may have changed their behavior if symptoms occurred to avoid hospitalizations. How and whether collateral damage of the public management of the COVID-19 pandemic has affected the treatment of HF, exemplified by outpatient HF visits and inpatient treatment for worsening HF, is unknown. HF is a common comorbidity that is associated with high mortality, and prompt initiation and titration of guideline-directed therapy has shown to improve short- and long-term clinical outcome, so focus on optimal care is necessary during and after the COVID-19 pandemic.1,2 Anecdotally, most clinicians have noted a decline in the number of patients seeking medical care for non–COVID-19–related causes, which has raised concerns for significant collateral damage in many cardiology patients and in particular, patients with HF, who are tenuous at baseline. For patients with ST-segment–elevation myocardial infarction, a recent study noted a ≈40% decline of the number of ST-segment–elevation myocardial infarction activations during the COVID-19 period in the United States, but data on the consequences for HF patients are unknown.3 The objectives of the present study were, therefore, to determine the incidence of new-onset and worsening HF before and during the COVID-19 pandemic, as well as the mortality in patients with established HF, in a nationwide and public healthcare system.
Methods
Registries, Population, and Outcomes
Due to Danish laws and regulations, data will not be publically available. The Danish healthcare system is owned by the government and funded by public taxes. All citizens are eligible to use it, and utilization is without copayment. For appropriate healthcare planning and economic reasons, the government keeps databases on most healthcare-related metrics. For the present study, we accessed the Danish Nationwide Patient Registry, where all in-hospital and outpatient clinic visits are registered on the date of occurrence with one main diagnosis and any secondary diagnoses of relevance to the specific contact. This registry is continuously updated, and Danish research institutions can apply for access for research purposes. The registry was linked with the Danish population registry, where all births and deaths are registered on the date of occurrence. Before data are delivered to the individual research institutions, they are encrypted, disabling personal identification. The quality of the Danish healthcare-related registries for cardiovascular outcomes research is generally high.4–6 The present study was approved by the Danish Data Protection Agency.
The present study included all adult Danes (ie, people aged >18 years residing in Denmark) between January 1, 2019, and March 31, 2020. We studied incidence rates of new-onset HF and of hospitalization for worsening HF in the non-COVID versus COVID era (January 1 to March 11 versus March 12 to March 31, respectively). In addition, we compared the differences in patient characteristics among those with new-onset and worsening HF in the COVID-19 era (March 12, 2020, to March 31, 2020) with the corresponding period in 2019. We considered a comorbidity as present if the date of onset occurred before the event (new-onset HF or worsening HF). We identified all diagnoses of HF (International Classification of Diseases, Eighth Revision codes 42709, 42710, 42711, 42719, 42899, and 42849 and Tenth Revision codes I50, I110, I130, and I132, respectively; main or contributing diagnoses) that had been registered at any time since initiation of the registries (1978). We defined a first-time diagnosis (based on in-hospital and outpatient visits) of HF as new-onset HF. Worsening HF was defined as an admission for HF as the primary diagnosis that required a hospital stay duration of at least 1 night.
Statistics
Tests for differences in patient characteristics were performed by the χ2 test and t test for discrete and continuous variables, respectively. For each day between January 1 and March 31 in 2019 and 2020, we calculated incidence rates of new-onset HF and hospitalizations for worsening HF based on the whole Danish population aged >18 years and mortality rates in HF patients. We used age- and sex-adjusted Poisson regression models to calculate the incidence rate ratios of new-onset HF and hospitalizations for worsening HF comparing after the lockdown versus before the lockdown. The Kaplan-Meier method was used to estimate mortality in HF patients with diagnosed COVID. All analyses were performed in SAS, version 9.4.
Results
Daily incidence rates of new-onset HF and hospitalizations for worsening HF are shown in Figure 1A through 1D. Rates of new-onset HF between January 1 and March 11 were comparable for 2020 and 2019 (1.83 versus 1.78 per 10 000 person-years; P=0.19), while hospitalizations for worsening HF were slightly higher in 2020 versus 2019 (1.04 versus 0.93 per 1000 person-years; P=0.02). In the lockdown period, rates of new-onset HF diagnoses (1.26 versus 2.25 per 1000 person-years) and of hospitalizations for worsening HF (0.63 versus 0.99 per 1000 person-years) were significantly lower in 2020 versus 2019 (P for both, <0.0001). Age- and sex-adjusted incidence rate ratios for new-onset HF and hospitalization for worsening HF in the lockdown period (versus the other periods) were 0.69 (95% CI, 0.63–0.77) and 0.70 (95% CI, 0.61–0.80), respectively. Tables 1 and 2 present the baseline characteristics of patients diagnosed with new-onset HF and hospitalized with HF in the 2 periods in 2020 versus 2019, respectively. Overall, the characteristics of patients with new-onset and worsening HF in 2020 and 2019 were comparable, including a similar age distribution and comorbidity burden in 2019 versus 2019 at the date of diagnosis, both for the early period and in the COVID-19 era. The prevalence of patients with chronic obstructive pulmonary disease was, however, observed to be lower among those with new-onset HF in the COVID-19 period versus the non–COVID-19 period.
| Early Period (January 1 to March 11) | Late Period (March 12 to March 31) | |||||
|---|---|---|---|---|---|---|
| 2020 | 2019 | P for Difference | 2020 | 2019 | P Values | |
| n | 2197 | 2099 | 398 | 720 | ||
| Age, y | 74.8 (12.7) | 74.3 (13.0) | 0.21 | 73.3 (12.9) | 73.4 (13.0) | 0.91 |
| Sex, male | 1333 (61%) | 1307 (62%) | 0.28 | 247 (62%) | 445 (62%) | 0.93 |
| Diabetes mellitus | 443 (20%) | 433 (21%) | 0.71 | 82 (21%) | 167 (23%) | 0.32 |
| Vascular | 525 (24%) | 522 (25%) | 0.46 | 114 (29%) | 179 (25%) | 0.17 |
| MI | 443 (20%) | 440 (21%) | 0.52 | 96 (24%) | 150 (21%) | 0.20 |
| Stroke | 323 (15%) | 317 (15%) | 0.71 | 59 (15%) | 101 (14%) | 0.72 |
| AF | 741 (34%) | 701 (33%) | 0.82 | 106 (27%) | 227 (32%) | 0.09 |
| CKD | 188 (9%) | 202 (10%) | 0.22 | 41 (10%) | 64 (9%) | 0.44 |
| Cancer | 516 (23%) | 464 (22%) | 0.28 | 103 (26%) | 157 (22%) | 0.12 |
| COPD | 315 (14%) | 320 (15%) | 0.40 | 48 (12%) | 119 (17%) | 0.04 |
| Asthma | 120 (5%) | 125 (6%) | 0.48 | 26 (7%) | 43 (6%) | 0.71 |
| Immigrants | 133 (6%) | 110 (5%) | 0.47 | 18 (5%) | 48 (7%) | 0.32 |
| Hospitalization for Worsening HF | Early Period (January 1 to March 11) | Late Period (March 12 to March 31) | ||||
|---|---|---|---|---|---|---|
| 2020 | 2019 | P for Difference | 2020 | 2019 | P for Difference | |
| n | 1149 | 1066 | 215 | 353 | ||
| Age, y | 75.2 (12.7) | 75.2 (12.6) | 0.99 | 74.0 (13.9) | 75.3 (13.2) | 0.26 |
| Sex, male | 791 (69%) | 708 (66%) | 0.22 | 155 (72%) | 236 (67%) | 0.19 |
| Diabetes mellitus | 271 (24%) | 274 (26%) | 0.25 | 45 (21%) | 76 (22%) | 0.87 |
| Vascular | 316 (28%) | 314 (29%) | 0.31 | 57 (27%) | 109 (31%) | 0.27 |
| MI | 265 (23%) | 279 (26%) | 0.09 | 47 (22%) | 89 (25%) | 0.36 |
| Stroke | 154 (13%) | 157 (15%) | 0.37 | 29 (13%) | 47 (13%) | 0.95 |
| AF | 333 (29%) | 309 (29%) | 0.99 | 54 (25%) | 99 (28%) | 0.45 |
| CKD | 90 (8%) | 98 (9%) | 0.25 | 17 (8%) | 27 (8%) | 0.91 |
| Cancer | 207 (18%) | 166 (16%) | 0.12 | 33 (15%) | 61 (17%) | 0.55 |
| COPD | 131 (11%) | 127 (12%) | 0.71 | 30 (14%) | 53 (15%) | 0.73 |
| Asthma | 57 (5%) | 54 (5%) | 0.91 | 8 (4%) | 16 (5%) | 0.64 |
| Immigrant | 81 (7%) | 63 (6%) | 0.32 | 13 (6%) | 21 (6%) | 0.94 |
Figure 1. Daily incidence rates (per 1000 person-years) of new-onset diagnoses of heart failure (HF) and hospitalizations for worsening HF in the period January 1 to March 31, 2019 and 2020, respectively. Horizontal lines represent the average incidence rates in the prelockdown period. Vertical lines represent the day of lockdown.
Mortality
Daily mortality rates among patients with known HF are shown in Figure 2. Compared with the non–COVID-19 period, we observed no statistically significant increase in the mortality rate for the period of March 12 to March 30 versus the other periods for patients with HF (142 versus 132 per 1000 person-years; age- and sex-adjusted mortality rate ratio, 1.05 [95% CI, 0.93–1.18]; P=0.45).
Figure 2. Daily mortality rates in patients with heart failure in the period January 1 to March 31, 2020 vs 2019. Horizontal lines represent the average mortality rates in the prelockdown period. Vertical lines represent the day of lockdown.
HF Patients Admitted With COVID-19
A total of 90 patients with prevalent HF (0.08% of all patients with diagnosed HF in Denmark) were admitted with COVID-19 between February 22 (when the first COVID-19–positive patient was diagnosed in Denmark) and March 31, 2020. Of these, 25 died before April 2, 2020 (end of the available data), corresponding to a mortality of 37% (95% CI, 23%–50%) within 15 days after the COVID-19 diagnosis (Figure 3).
Figure 3. Kaplan-Meier curve of mortality in heart failure patients diagnosed with coronavirus disease 2019 (COVID-19).
Discussion
The present study investigated the collateral consequences for HF patients of the COVID-19–related lockdown of the entire Danish society occurring on March 12, 2020. Overall, we observed a 30% reduction in rates of admissions for worsening of HF and a similar reduction in rates of patients who were diagnosed with new-onset HF in the period, compared with earlier in 2020 and compared with a similar period of 2019. The negative effect of the lockdown on the number of HF-related healthcare contacts appeared to occur across all ages, and we observed no difference in comorbidity burden among those who were diagnosed after lockdown versus before, although fewer patients with new-onset HF had chronic obstructive pulmonary disease among those diagnosed after lockdown, possibly because they to a larger extent than others are self-quarantined to limit their exposure to the virus or hope that progression of symptoms is explained by an already known chronic disease.
New-Onset HF
The combination of fewer visits at the general practitioners, fewer hospitalizations overall (for cardiac and noncardiac causes), and fewer myocardial infarctions (missed ST-segment–elevation myocardial infarctions may increase the rate of new-onset HF later in the epidemic) may explain the decrease in the number of new-onset HF diagnosed during the COVID-19 epidemic. It should be kept in mind that HF is a syndrome with preexisting comorbidities.7 Patients do, therefore, often have symptoms before their HF diagnosis, and the reduced propensity of being diagnosed with HF could potentially also indicate that the level of care in general is reduced in the healthcare system during the COVID-19 epidemic.
Worsening HF
The observed reduction in patients hospitalized for worsening HF may be caused by multiple factors. Patients may be afraid of being admitted during the epidemic and have improved self-care, adherence to guideline-directed medical therapy, and changed dosages of diuretics according to their disease management education or telemedicine advices. In that way, hospitalizations for universal overhydration might be avoided, and only patients with congestion due to shifts in the fluid compartments will be admitted. Prior data suggest that ≈50% of patients with acute decompensated HF are fluid overloaded and the other half have a fluid shift.8 Another explanation could be physical isolation and lack of exposures like an infection that trigger a HF hospitalization. A longer follow-up will, however, be needed to understand whether some patients are also slowly deteriorating at home, and it is possible that we will observe a surge in HF hospitalizations in the weeks to months to come.
Mortality
The number of patients who have tested positive for COVID-19 in Denmark is low compared with many other countries. This was supported by our study showing that 0.08% of HF patients were diagnosed with COVID-19 infection during our observational time. Thus, the COVID-19 epidemic has not rooted in Denmark yet. HF has been reported as a risk factor for adverse outcomes on COVID-19 infections and patients with HF infected with COVID-19 have a high mortality.9,10 This was confirmed in our limited sample, where the cumulative mortality was 37% (and may be higher than so, given the short follow-up available). Therefore, minimizing exposure to the virus in this vulnerable group is justified. However, a prolonged time to diagnosis and delayed treatment for HF may lead to multiorgan injury and increased mortality in the long term.7,11 Although we did not observe any differential mortality in the COVID-19 period, our data raise concern for undertreating HF in the current era, which may translate into worse outcomes. More research is needed to understand whether the patients not presenting to the healthcare system may have improved their compliance to diet and medications in the COVID-19 era or whether they are slowly deteriorating at home.
Methodological Considerations
The present study is unique due to the complete and nationwide data. Selection bias and patients lost to follow-up is, therefore, avoided. We have been able to compare 2020 with 2019 and have in that way implemented a control group. However, the study is observational in its design, and it has, therefore, inborn limitations. First, no causal relationship between the public lockdown and the reduced number of patients with new-onset and worsening HF can firmly be concluded. Second, the present study is based on administrative registries and important clinical variables like left ventricular ejection fraction, functional class, NT-proBNP (N-terminal pro-B-type natriuretic peptide), and renal function are lacking, and we could not differentiate between HF with reduced and preserved ejection fraction. However, we do not think that a change over time in these unmeasured confounders explains the association between public lockdown and the reduction in numbers of new-onset and worsening HF. Third, Denmark is a country with 4 seasons, and HF hospitalization rates and mortality may differ with season and also with, for example, influenza season.12,13 To take this into account, we have compared 2020 with 2019, but to what extent influenza may have influenced our data is unknown. Fourth, people with and in risk of HF are elderly and sometimes frail. It may be suggested that the observed decrease in HF hospitalizations is due to a competing risk of a COVID-19 hospitalization. However, the registered number of HF patients hospitalized with COVID-19 does not equivalent the reduction in worsening HF hospitalizations. Fifth, although based on a nationwide sample with complete follow-up, the study sample was of limited size, and we applied multiple testing without correcting for the number of tests performed, which may have introduced statistical type 1 and type 2 errors. The generalizability of our results may also be questioned. Denmark is a country with 6 million habitants with a nationwide and public healthcare system and a low Gini coefficient (a measure of economic inequality, where a low coefficient indicates a homogeneous population).5 Most people speak and understand Danish and the recommendations from the authorities. The effects of collateral damage of a public lockdown on HF care in relation to an epidemic in countries with less uniform healthcare systems and higher Gini coefficients may be different.
Clinical Perspectives
Our results underscore a need for an improved communication from cardiology societies to populations, patients, and general practitioners to focus on the diagnosis and treatment of HF during the COVID-19 pandemic that may continue for a long period, at least at an endemic level. HF clinics may expect that more patients are referred for evaluation when the pandemic has ended. Finally, if mortality rates continue to be neutral for HF patients despite fewer observed hospitalizations, it may be speculated that improved clinical evaluation of symptomatic HF should result in more frequent decongestion of patients in the outpatient setting and fewer hospitalizations for worsening HF to avoid the toxic effects of being admitted.14,15
Conclusions
The incidence rates of patients hospitalized with worsening HF or receiving a diagnosis of new-onset HF were reduced by ≈30% after lockdown but has, as yet, not impacted mortality in HF patients at a population-based level. However, these data raise concerns for a potential undertreatment of HF in the current time that may impact prognosis in the longer term. As the mortality associated with COVID-19 infection in HF patients was high, means to reduce the exposure to COVID-19 without affecting HF treatment overall is needed. The effectiveness of telemedicine in this setting remains to be determined. Whether COVID-19 lockdown will affect long-term mortality risk warrants follow-up data once the pandemic has settled.
| COVID-19 |
coronavirus disease 2019 |
| HF |
heart failure |
| NT-proBNP |
N-terminal pro-B-type natriuretic peptide |
Footnotes
References
- 1.
Lee DS, Stukel TA, Austin PC, Alter DA, Schull MJ, You JJ, Chong A, Henry D, Tu JV . Improved outcomes with early collaborative care of ambulatory heart failure patients discharged from the emergency department.Circulation. 2010; 122:1806–1814. doi: 10.1161/CIRCULATIONAHA.110.940262LinkGoogle Scholar - 2.
Gattis WA, Hasselblad V, Whellan DJ, O’Connor CM . Reduction in heart failure events by the addition of a clinical pharmacist to the heart failure management team: results of the Pharmacist in Heart Failure Assessment Recommendation and Monitoring (PHARM) Study.Arch Intern Med. 1999; 159:1939–1945. doi: 10.1001/archinte.159.16.1939CrossrefMedlineGoogle Scholar - 3.
Garcia S, Albaghdadi MS, Meraj PM, Schmidt C, Garberich R, Jaffer FA, Dixon S, Rade JJ, Tannenbaum M, Chambers J, . Reduction in ST-Segment Elevation cardiac catheterization laboratory activations in the United States during COVID-19 pandemic.J Am Coll Cardiol. 2020. doi: 10.1016/j.jacc.2020.04.011CrossrefGoogle Scholar - 4.
Sundbøll J, Adelborg K, Munch T, Frøslev T, Sørensen HT, Bøtker HE, Schmidt M . Positive predictive value of cardiovascular diagnoses in the Danish National Patient Registry: a validation study.BMJ Open. 2016; 6:e012832. doi: 10.1136/bmjopen-2016-012832CrossrefMedlineGoogle Scholar - 5.
Schmidt M, Schmidt SAJ, Adelborg K, Sundbøll J, Laugesen K, Ehrenstein V, Sørensen HT . The Danish health care system and epidemiological research: from health care contacts to database records.Clin Epidemiol. 2019; 11:563–591. doi: 10.2147/CLEP.S179083CrossrefMedlineGoogle Scholar - 6.
Adelborg K, Sundbøll J, Munch T, Frøslev T, Sørensen HT, Bøtker HE, Schmidt M . Positive predictive value of cardiac examination, procedure and surgery codes in the Danish National Patient Registry: a population-based validation study.BMJ Open. 2016; 6:e012817. doi: 10.1136/bmjopen-2016-012817CrossrefMedlineGoogle Scholar - 7.
Christiansen MN, Kober L, Torp-Pedersen C, Gislason GH, Schou M, Smith JG, Vasan RS, Andersson C . Preheart failure comorbidities and impact on prognosis in heart failure patients: a nationwide study.J Intern Med. 2020. doi: 10.1111/joim.13033CrossrefMedlineGoogle Scholar - 8.
Fudim M, Hernandez AF, Felker GM . Role of volume redistribution in the congestion of heart failure.J Am Heart Assoc. 2017; 6:e006817. doi: 10.1161/JAHA.117.006817LinkGoogle Scholar - 9.
Guo T, Fan Y, Chen M, Wu X, Zhang L, He T, Wang H, Wan J, Wang X, Lu Z . Cardiovascular implications of fatal outcomes of patients with coronavirus disease 2019 (COVID-19).JAMA Cardiol. 2020. doi: 10.1001/jamacardio.2020.1017CrossrefGoogle Scholar - 10.
Ruan Q, Yang K, Wang W, Jiang L, Song J . Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China.Intensive Care Med. 2020; 46:846–848. doi: 10.1007/s00134-020-05991-xCrossrefMedlineGoogle Scholar - 11.
Madelaire C, Gustafsson F, Stevenson LW, Kristensen SL, Køber L, Andersen J, D’Souza M, Torp-Pedersen C, Gislason G, Schou M . Favorable five-year outcomes for heart failure diagnosed in younger patients without severe comorbidity.Int J Cardiol. 2020; 305:106–112. doi: 10.1016/j.ijcard.2020.01.055CrossrefMedlineGoogle Scholar - 12.
Kytömaa S, Hegde S, Claggett B, Udell JA, Rosamond W, Temte J, Nichol K, Wright JD, Solomon SD, Vardeny O . Association of influenza-like illness activity with hospitalizations for heart failure: the atherosclerosis risk in communities study.JAMA Cardiol. 2019; 4:363–369. doi: 10.1001/jamacardio.2019.0549CrossrefMedlineGoogle Scholar - 13.
Panhwar MS, Kalra A, Gupta T, Kolte D, Khera S, Bhatt DL, Ginwalla M . Effect of influenza on outcomes in patients with heart failure.JACC Heart Fail. 2019; 7:112–117. doi: 10.1016/j.jchf.2018.10.011CrossrefMedlineGoogle Scholar - 14.
Krumholz HM . Post-hospital syndrome–an acquired, transient condition of generalized risk.N Engl J Med. 2013; 368:100–102. doi: 10.1056/NEJMp1212324CrossrefMedlineGoogle Scholar - 15.
Okumura N, Jhund PS, Gong J, Lefkowitz MP, Rizkala AR, Rouleau JL, Shi VC, Swedberg K, Zile MR, Solomon SD, ; PARADIGM-HF Investigators and Committees*. Importance of clinical worsening of heart failure treated in the outpatient setting: evidence from the prospective comparison of ARNI With ACEI to determine impact on global mortality and morbidity in heart failure trial (PARADIGM-HF).Circulation. 2016; 133:2254–2262. doi: 10.1161/CIRCULATIONAHA.115.020729LinkGoogle Scholar