The COVID-19 pandemic has had an immense impact on the elderly, who are among the most vulnerable and affected groups. This population experiences a higher proportion of severe cases and complications during infection 1 and exhibits intrinsic and extrinsic factors that contribute to increased clinical fragility and susceptibility to infectious processes, such as weakened immune systems or immunosenescence,2 heightened comorbidity,3 malnutrition,4 and a higher rate of institutionalization.

Since the onset of the COVID-19 pandemic, advanced age has been identified as one of the strongest risk factors for poor outcomes, complications, and mortality.5–7 Age is an easily measurable prognostic marker; however, its prognostic utility by itself is limited.8 In this regard, emerging evidence suggests that frailty may be a significant predictor of poor outcomes in older people hospitalized due to COVID-19.9,10 Frailty has also been used for clinical decision-making during this pandemic,11 but further clinical research is still needed to determine the usefulness of frailty screening in predicting adverse disease.

Frailty is defined as "a medical syndrome with multiple causes and contributors characterized by diminished strength, endurance, and reduced physiological function, which increases an individual's vulnerability to developing increased dependency and/or death".12 The likelihood of frailty increases with age, estimated to affect around 40% of older patients.13 Fried's frailty criteria are widely used for diagnosing frailty.14 According to these criteria, a diagnosis of frailty is established if the patient meets three of the following criteria: unintentional weight loss, exhaustion, muscle weakness, motor slowness, and low activity. Over the years, multiple instruments have been developed to assess frailty, including rapid detection scales that are more feasible in clinical practice and require only a few minutes of application, such as the Clinical Frailty Scale (CFS)15 and the FRAIL scale.16

The relationship between the degree of pre-infection clinical frailty and the progression of COVID-19 has been the subject of several studies to date. However, most studies assessing the prognostic capacity of clinical frailty have either been conducted in the general population or, if they do evaluate prognostic factors in older patients, have not included the degree of frailty. These studies have examined the relationship between frailty and mortality, hospital infection rates, intensive care admission rates, and disease phenotypes.10,17,18

Other factors, such as advanced age, male sex, severe functional dependence, and comorbidities like hypertension, diabetes mellitus, and obesity, as well as analytical parameters (C-reactive protein, lymphopenia, neutrophilia, etc.), clinical parameters at admission (hypoxia, high SOFA score, temperature, etc.), and the presence of radiological abnormalities, have also been identified as the main risk factors for poor outcomes in older people with COVID-19 infection.19–21

Systematic frailty assessment in older patients with COVID-19 infection allows for the early identification of frail elderly patients. This enables better care for those at higher risk of severe disease and facilitates improved resource allocation. Therefore, the objectives of this study in older patients hospitalized due to COVID-19 infection are as follows: a) to determine the prognostic value of frailty on intrahospital patient survival compared to other previously identified predictors of poor prognosis, and b) to emphasize the importance of early detection of frailty in this population.

Qualitative variables were presented as absolute and relative frequencies and compared using the chi-square test or Fisher's exact test, as appropriate. Quantitative variables were expressed as mean and standard deviation and compared using Student's t-test for independent groups.

To assess the prognostic value of frailty on patient survival, a multivariate Cox proportional hazards analysis was conducted. This analysis included variables that showed a significant association in the univariate analysis, as well as other variables of recognized prognostic value and potentially confounding factors reported in the literature, particularly those found in a previous article based on this COVID-19 patient registry.19 Kaplan–Meier curves were generated to visually represent patient survival according to frailty categories. A significance level of 0.05 (95% confidence level) was assumed. The data were stored and analyzed using the SPSS statistical package, version 25, for Windows.

Informed consent was obtained from all patients. In cases where biosafety concerns or patient discharge had occurred, verbal informed consent was requested and documented in the medical records. Data confidentiality and patient anonymity were strictly maintained in accordance with Spanish regulations governing observational studies. Patient identifiable information was removed before analyzing the database, ensuring that individual patients cannot be identified either in this article or in the database.

In the SEMI-COVID-19 Registry, a total of 1,920 patients aged ≥70 years who had been hospitalized due to COVID-19 infection between March 1 and December 31, 2020 were identified, and their degree of frailty was assessed. Forty-two participants were excluded due to incomplete registration of minimum clinical characteristics. Ultimately, the study included 1,878 participants of both sexes, with 52% men and 48% women. Among the included patients, 1,351 (71.9%) were discharged alive from the hospital, while 527 (28.1%) died during their hospital stay. There were no significant differences in the survival rates between males and females (Table 1).

The clinical and demographic characteristics of the population are shown in Table 1. The mean age of the population was 81.7±6.9 years, being higher in the group of non-survivors (83.5 vs. 81 years, p<0.001). The most prevalent comorbidities were hypertension (75.2%), dyslipidaemia (51.9%), diabetes (32%), atrial fibrillation (21%), and dementia (20.8%). The proportion of comorbidities such as dyslipidaemia, diabetes mellitus with target organ involvement, atrial fibrillation, dementia, degenerative neurological disease, and heart failure was significantly higher in the non-survivor group (p<0.001). However, although the proportion of cases with hypertension and COPD was also higher in the non-survivor group, no significant differences were found compared to survivors. Charlson index was high in all population, but more in the non-survivor group (6.3 vs. 5.3 points, p<0.001). Most patients (62,5%) were independent or with a middle level of dependence; nonetheless, the proportion of patients with a moderate or severe level of dependence was higher among non-survivors (34.2% vs 25.8% patients with moderate frailty, and 34.5% vs. 14.7% patients with severe frailty, respectively, p<0.001).

Regarding the degree of frailty, 174 patients (9.3%) were classified as very fit/fit (CFS 1−2); 795 patients (42.3%) as vulnerable (CFS 3−4); 529 patients (28.9%) were classified as moderately fragile (CFS 5−6) and 380 patients (20.2%) as severely fragile or terminal (CFS 7−9). The proportion of patients with moderate or severe frailty is significantly higher in non-survivors (p<0.001)

A higher proportion of patients in the non-survivor group had significantly higher levels of leukocytes, neutrophils, CRP, creatin, LDH, ferritin, and procalcitonin (Table 2) than survivors (p<0.001). Non-survivors also had higher lymphopenia and higher D-Dimer values, although this did not reach significance. Hypoxemia (Oxygen saturation <90%) was more frequent in non-survivor group (p<0.001). Radiological findings (condensations, bilateral infiltrates and pleural effusion) were significantly more frequent in the non-survivor group (p<0.001).

Frailty was associated with higher all-cause mortality after adjustment for age, sex, and comorbidities, showing worsening of clinical outcome with increased frailty (Table 3 and Fig. 1). The crude HR for time from hospital admission to mortality was 1.634 (95% CI 1.023−2.611; p0.04) for CFS 3−4; 2,887 (1,816−4,591; p<0.001) for CFS 5−6 and 5,557 (3,493−8,841; p<0.001) for CFS 7−9, all compared to CFS 1−2. The adjusted HR was 1.458 (95% CI 0.903−2.355; p0.001) for CFS 3−4; 2,344 (1,437−3,823; p<0.001) for CFS 5−6 and 3,694 (2,155−6,330; p<0.001) for CFS 7−9. Elevated CRP, lymphopenia, neutrophilia, hypoxemia, and bilateral chest X-ray infiltrates were also significantly associated with mortality (p<0.001). Severe dependence was also significantly associated with mortality in the univariate analysis with a crude HR of 2.409 (1.984−2.926; p<0.001); however, in the multivariate analysis this association was not significant adjusted HR 1.184 (0.879−1.595; p0.266).

Figure 1.

Overall survival by CFS category.

(0.08MB).

Our study determines the importance of early detection of frailty in older patients hospitalised due to COVID-19, considering it as main risk factor associated with adverse outcomes compared to other risk factors already identified in previous studies. The main predictors of poor prognosis during the acute phase of infection have been described in previous studies. However, most of them have been performed in the general population and only some have targeted the older patient population or have evaluated the degree of frailty.17

Older age is associated with increased mortality but is not sufficient on its own for risk stratification in patients with COVID-19 and is subject to ethical controversy.11 Pre-admission functional status plays an important role in the evolution of this patient profile, especially those with a moderate-severe degree of frailty who are at higher risk of adverse outcomes. This relationship between frailty and increased mortality has been extensively studied in other diseases. In COVID-19 an increasing number of studies have identified frailty as one of the main prognostic factors of the disease, but more evidence is still needed.

Previously, the SEMI-COVID registry19 had identified the main predictors of poor prognosis in very older patients (>80 years) hospitalised due to COVID-19 infection and this study was the first one to identify the prognostic significance of pre-admission clinical status on the outcome of geriatric patients, finding that a severe degree of dependency (defined as a Barthel Index ≤60) was an independent predictor of mortality. Nonetheless, one of the main limitations of this study was that it did not evaluate frailty. Advanced age, severe level of dependency, male sex, certain laboratory, and chest X-ray abnormalities were identified as the main predictors of poor clinical outcome in this population. In contrast, comorbidities were not associated with increased mortality. Our data reflect concordance with the results of this study except for some aspects. No association was found between intrahospital survival and sex or level of severe dependency. However, a high degree of comorbidity (defined by Charlson index ≥4) was associated with lower intrahospital survival.

Studies assessing frailty in COVID-19 such as the COPE study,17 and others27 demonstrated that frailty was associated with mortality and longer hospital stay, showing a worsening of clinical outcome with increasing frailty. Our results do not only confirm these assertions, but also establish the presence of moderate-severe frailty as the main prognostic factor independently associated with all-cause hospital survival compared to the other factors identified in previous studies. In addition, there is a direct relationship between a higher degree of frailty and lower intrahospital survival.

The assessment of the degree of frailty was carried out using the quantitative method CFS, which is the most widely used in other studies.28,29 The CFS is a reliable and potentially useful screening tool to identify frailty. It is also easily applicable even in a situation of limited human resources and increasing demand for medical services,30 as it was the case in the COVID-19 pandemic. Other frailty measures are available for the assessment of frailty in hospitalised patients but are either more time consuming to apply or rely on routinely collected data to score frailty.31 The National Institute for Health and Care Excellence (NICE) published in 2020 the COVID-19 rapid guidelines for adult critical care, recommending the use of the CFS in patients aged 65 and over to aid clinical decision making and avoid age discrimination.32

Therefore, the early assessment of frailty represents a valuable opportunity to provide higher quality care to older adults with COVID-19. Early detection and careful monitoring of frailty can alert us to the possibility of adverse outcomes and help us to provide appropriate clinical management in this patient profile. The initial approach to these patients should incorporate an appropriate functional assessment including the evaluation of the degree of frailty. In addition, we would like to emphasise the importance of the use of the CFS scale as a predictor of unfavourable events in this population.

This study has some limitations. First, this is a retrospective series focusing on hospitalised patients. Since these patients had more severe disease and a higher mortality rate, our data may overestimate the overall mortality in the totality of adults over 70 years of age with COVID-19. Second, as a retrospective cohort study, the data were collected by a large number of investigators, which could have led to heterogeneity in data entry and validation.

In older patients hospitalised due to COVID-19 infection, the degree of frailty is the main predictor of intrahospital survival, showing that it increases the risk of all-cause mortality after adjustment for age, comorbidities, and other prognostic factors related to the severity of the infection. These findings highlight the need for early detection of frailty using clinical scales, which is of vital importance in establishing a prognosis in this population.

The authors declare that they have not received funding for conducting this study.

The authors declare that they do not have any conflicts of interest.