Impact of COVID-19 on health-related quality of life in patients with cardiovascular disease: a multi-ethnic Asian study

Coronavirus disease 2019 (COVID-19) is a global pandemic with unprecedented medical, economic and social consequences. Much attention has been on the potentially severe cardiovascular complications associated with COVID-19 [1, 2]; patients with pre-existing cardiovascular disease (CVD) also suffer worse outcomes with COVID-19 infection [3]. Beyond these direct consequences, COVID-19 has reshaped the delivery of cardiovascular care—non-critical and elective procedures are postponed, and distancing imperatives have led to rapid scaling of telemedicine and cancellation of cardiac rehabilitation. The repeated public health messages that those with chronic conditions should practice social and physical distancing may adversely affect the health-seeking behavior of patients with CVD. Indeed, data from affected countries show reductions in hospital admission for myocardial infarctions and strokes [4], but those who presented were sicker and had worse outcomes, suggesting delayed presentation [5, 6].

Studies evaluating the impact of COVID-19 on non-infected persons have also looked at the psychological impact of the pandemic [7, 8]; both small and large studies have been conducted in general population, healthcare workers, and those with pre-existing psychiatric diseases. Little is known about the impact of COVID-19 on physical and mental health of patients with CVD, the biggest killer globally [9].

Singapore is a multi-ethnic city state in the Asia–Pacific; majority of its 5.7 million population are Chinese, followed by Malays and Indians. It reported its first case of COVID-19 on 23rd January 2020, and imposed a partial national lockdown, termed as the Circuit Breaker, from 3rd April 2020 [10]. There was discontinuation of non-critical medical services, mask advisory and closure of public places. There was gradual easing of the restrictions from 2nd June 2020 (Phase 1 Safe Re-opening) [11], followed by Phase 2 Re-opening from 18th June 2020, where almost all activities were allowed except for large-scale events and venues, as well as entertainment venues [12].

Our study aimed to, using the EQ-5D, evaluate the impact of COVID-19 on health-related quality of life (HRQoL) in a multi-ethnic Asian cohort with CVD in Singapore, hypothesizing that the physical capacity and mental health of these patients would be adversely affected by COVID-19.

A total of 85 patients with pre-pandemic HRQoL assessments were approached between 29th April 2020 to 19th June 2020 for this study, and 84 (99%) agreed to participate. Data from 3 (4%) participants were excluded from the final analysis, as the baseline HRQoL survey was performed more than 2 years ago. The final number of participants whose data were analyzed was 81.

The mean age of the participants was 60 years (SD 11), the majority were men (93%), and the ethnic distribution reflected that of the general population. The majority of the participants had coronary artery disease (69%), followed by chronic heart failure (28%). Along with their CVD, the participants also reported having other comorbidities such as hypertension (62%), dyslipidemia (61%), and diabetes mellitus (49%). There was little functional limitation at baseline with 94% reporting NYHA functional classes I or II. There was no change in NYHA functional status during the pre- and during-pandemic visits for any of the participants with 97% reporting NYHA functional classes I or II during pandemic. The baseline characteristics of the patients are summarized in Table 1.

Among 81 patients who participated in the study, all completed EQ-5D-3L at both visits. A subset of 66 patients completed EQ-5D-5L in addition to EQ-5D-3L at both visits. There were no notable differences in baseline characteristics between patients who completed both questionnaires versus who completed only EQ-5D-3L, except for NYHA status and CVD diagnoses (Additional file 1: Table S1).

Using EQ-5D-3L at pre-pandemic, 24 of 81 participants (29.6%) reported problems in at least one dimension; the most commonly affected dimension was pain/discomfort (23.5%) followed by anxiety/depression (12.4%). During the pandemic, 31 participants (38.3%) reported problems in at least one dimension with the severity of the problems in each dimension similar to or worse than pre-pandemic (Fig. 1). Among all the dimensions, the highest increase in problems was reported for anxiety/depression with mean domain-specific score increased by 0.12 (SES = 0.373; p = 0.012) (Table 2). There was no meaningful change in EQ-VAS (mean change = − 0.01; 95% CI − 2.87 to 2.85) as well as EQ-5D-3L utility score (mean change = − 0.015; 95% CI − 0.063 to 0.034).

A higher proportion of patients reported problems using the EQ-5D-5L compared to the EQ-5D-3L both at pre-pandemic and during pandemic visits. A total 32 of 66 participants (48.5%) reported problems in at least one dimension at pre-pandemic with the majority reported problems in pain/discomfort (39.4%), followed by anxiety/depression (25.8%). During the pandemic, 37 participants (56.1%) reported problems in at least one dimension with the severity of the problems in each dimension similar to or worse than pre-pandemic. Similar to EQ-5D-3L data, the highest increase in problems was reported for anxiety/depression with mean domain-specific score increased by 0.26 (SES = 0.516; p = 0.007) (Additional file 1: Table S2). There was no meaningful change in the EQ-5D-5L utility score (mean change = − 0.036; 95% CI − 0.084 to 0.012).

Our study involving patients with chronic CVD reported worse HRQoL in several aspects of health during the COVID-19 outbreak, despite stability in their CVD. Specifically, there was increased anxiety and depression on serial EQ-5D-3L assessments. Results were more pronounced using EQ-5D-5L. The increase in anxiety and depression during the outbreak was clinically (small-to-moderate effect size) as well as statistically significant. To our knowledge, this is one of the first studies evaluating the impact of COVID-19 pandemic on the HRQoL of non-infected persons with CVDs.

Infectious diseases outbreaks are known to have health, social and economic implications, for infected and non-infected persons alike. Prior studies have shown the adverse psychological impact of COVID-19 and its containment measures in COVID-19 patients, frontline workers, patients with pre-existing psychiatric diseases and the general population [7, 8]. Our findings are consistent with prior literature, and further extended those reports by providing the perspectives of those with CVD.

The increase in anxiety and depression amongst our study participants, is unsurprising and may be explained by the following: (1) the swathe of information, from official organizations to social media platforms, re-iterating the increased risks of COVID-19 with CVDs, (2) postponements of non-critical CVD follow-up appointments, and the emphasis on social and physical distancing may have altered their perception of healthcare accessibility, (3) separation from family and friends during the Circuit Breaker, leading to social isolation, (4) mask-wearing requirement in public areas, resulting in dyspnea for some people, and (5) economic concerns, given the closures of businesses. Lingering memories of the devastating impact of the Severe Acute Respiratory Syndrome (SARS) outbreak in 2003, with 238 cases and 33 deaths in Singapore, may have also contributed to anxiety and depression [20]. An understanding of the relative contribution of these factors would have been helpful in shaping future policies and interventions; unfortunately, such data were not collected in this brief study.

The magnitude of change in HRQoL was small. In particular, there was little change in the physical dimensions of HRQoL, which partly reflected the stability of the participants’ underlying CVDs between the surveys done at baseline and during the COVID-19 outbreak. The subgroups bearing the brunt of the pandemic and its containment measures are usually the elderly [21], females [22‐24], less educated [24, 25], and those from the low socioeconomic status [26, 27]. Our study participants of mostly older, middle-class males may not have been the vulnerable group. Furthermore, support from the government in response to the pandemic—rapid ramping up of resources and capabilities in terms of COVID-19 testing and provision of care, assurance that non-COVID care would not be compromised, shift towards telemedicine, and massive stimulus packages to cushion the economic impact, might have somewhat mitigated the impact of the pandemic.

There are several strengths of our study. First, all the HRQoL assessments were completed by participants themselves rather than proxies. Second, all participants had baseline assessments prior to the outbreak, thus acting as their own controls. Third, the impact of the pandemic on HRQoL was evaluated by two validated HRQoL measures (EQ-5D-3L and EQ-5D-5L) and both gave consistent results, leaving less room for random noise in the results. Fourth, the follow-up HRQoL assessment was carried out during the partial national lockdown which enabled us to adequately capture the impact of outbreak and containment measures on HRQoL.

Our study has a few limitations. First, the administration of HRQoL assessments was different pre-pandemic and during pandemic for some participants. All pre-pandemic HRQoL assessments were self-administered; majority of the follow-up assessments were administered by the interviewer over the phone. The follow-up assessments were conducted when physical outpatient visits were restricted. Although our preference was for self-administered assessments via email or post, some participants preferred interviewer-administered assessments over the phone. Therefore, we would not rule out some impact of this administration on their responses (response bias). Nevertheless, previous research has shown that interviewer-administration is likely to overestimate HRQoL, rather than underestimate [28]. That is, some participants could have under-reported their health problems. Therefore, our estimates of the impact of the pandemic on HRQoL should be considered as conservative estimates. Second, lack of data on external factors (social, environmental and economic factors) and limited sample size precluded us from studying the potential impact of patient and external factors on the change in HRQoL with certainty. Furthermore, our study performed several hypotheses, including evaluating the pandemic impact on each dimension of EQ-5D separately, without multiplicity correction. There could be chances of inflated type-I error (false positive), and results should be considered as exploratory. Third, as in all survey-based studies, sampling bias is an inherent limitation. Although consecutive recruitment without restrictions on type and stage of CVD allowed for a representative sample of adult patients with CVD, we included only those who were literate (English and Mandarin) and who had completed HRQoL questionnaires prior to the pandemic, thereby selecting the subgroups who may not have been most affected by the pandemic. It is conceivable that the mental health impact could be greater on the subgroups not represented in this study. Finally, our study was conducted in Asian patients with CVD. The findings may not be generalizable to other ethnicities or patient populations. We encourage further research in other countries and other disease groups to understand the impact of the pandemic.

The findings of adverse psychological impact on patients with CVD have important implications. It is plausible that the impact would be greater if our findings are extrapolated to more vulnerable populations—elderly, females, less educated, lower socioeconomic classes and poorly controlled CVD. Psychological ramifications can be long-lasting even after the pandemic has ended, and should not be ignored. At time of writing, there are no authoritative organizations that plan and coordinate psychological interventions in Singapore during the COVID-19 outbreak. In the meantime, cardiologists can do their part by proactively screening for psychological issues in patients who come for consultations, be it in-person or via telemedicine. Where necessary, psychosocial interventions can be implemented in collaboration with psychiatrists and psychologists.

The COVID-19 pandemic and its containment measures is associated with a significant decline in the psychological health components of the HRQoL of patients with pre-existing CVD. Our data suggest a need for improved screening of psychological issues and timely implementation of psychosocial interventions in these patients during the pandemic.