Long COVID Frailty: A Comparative Analysis in a Veteran Population

Abstract

Long COVID is characterized by persistent symptoms affecting one or more organ systems for at least 3 months following a SARS-CoV-2 infection. Our study aimed to examine the characteristics of frailty seen in patients with Long COVID compared to the frailty seen in aging patients with multimorbidity. This is a retrospective cohort study conducted in the Miami Veterans Affairs Medical Center (VAMC). The data used to calculate the Fried phenotype through the Johns Hopkins frailty calculator was collected from two separate clinics, a Long COVID clinic and a geriatric frailty clinic. We obtained the VA Frailty Index from VA CDW (Corporate Data Warehouse). We included 106 patients from the Long COVID clinic and 97 from the frailty clinic. Patients from the Long COVID clinic were significantly younger than those from the frailty clinic (60 ± 12.6 vs. 79.8 ± 5.8, p < 0.01). Patients with frailty in the Long COVID group experienced exhaustion (96.4% vs. 53.3%) and low activity (78.6% vs. 63.3%) at a higher rate than those in the geriatric frailty clinic. Long COVID may predispose patients to develop frailty that presents with a higher frequency of exhaustion and low activity.

1. Introduction

Long COVID is a complex disease process with poorly understood pathophysiological mechanisms [1]. Per the 2024 NASEM Long COVID Definition, Long COVID is an infection-associated chronic condition that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems [2]. Although symptoms are wide-ranging, a meta-analysis found that fatigue, shortness of breath, olfactory dysfunction, and myalgia are among the most common symptoms reported [3]. Beyond these manifestations, in some cases, Long COVID fatigue can cause reduced physical and mental activity, limiting a patient’s ability to work and function in society compared to their pre-infectious baseline [4]. Unfortunately, symptoms may last for years, with active research currently investigating the long-term sequelae of this disease process [1].

Frailty has traditionally been defined as a decline in the physiological reserve of patients, leading to a vulnerability to external stressors linked to poor outcomes [5]. Similarly to Long COVID, frailty can impact function and work productivity [6]. Two major models are commonly used to determine frailty in patients. The first is the Fried phenotype, based on five physiological factors: weakness measured by grip strength, slowness measured by gait speed, weight loss, and self-reported low physical activity and exhaustion [7]. These characteristics can help providers identify appropriate interventions while monitoring the response to therapy in repeat assessments. The second model is based on deficit accumulation as a predictor of frailty and its outcomes [8]. Under this model, the accumulation of health deficits, such as chronic medical conditions, increases the risk of frailty and death [9]. One of the advantages of using the deficit accumulation model, commonly represented as a frailty index, is that it does not require a physical assessment of the patient and has been well-validated to predict mortality in multiple studies [10]. Rather than seeing these as competing frailty models, experts have suggested that both are complementary to understanding the complexity of frailty using different approaches, with each yielding valuable clinical insights [11].

Although the frailty phenotype and frailty indices have only been validated in patients over 65 years old [7,12], there have been multiple attempts to use these calculators with younger patients, and we recognize that younger people may also develop frailty. A large review identified many papers that included populations of people over and under 60 years that demonstrated predictive validity for mortality and/or hospital admissions of both the frailty phenotype and frailty indices [13]. Our study aimed to examine the development of frailty in patients with Long COVID compared to the frailty seen in aging patients with multimorbidity. Our goal was to evaluate frailty in each of these cohorts using both the Fried phenotype through the validated Johns Hopkins frailty calculator and the deficit accumulation model through the Veterans Affairs Frailty Index (VA-FI) [14]. We hypothesize that patients with Long COVID, characterized by fatigue, would experience higher rates of low activity and exhaustion than geriatric frailty patients.

2. Materials and Methods

2.1. Study Design

This was a retrospective study conducted at the Miami VAMC in the geriatric frailty clinic and the Long COVID clinic from 9 January 2024 to 9 January 2025. The frailty clinic cohort comprised patients seen in the Geriatric Research, Education, and Clinical Center (GRECC) Frailty Clinic. The GRECC Frailty Clinic evaluates patients over 65 years old who are having difficulty with daily activities, mobility, and/or cognition, among other problems. Clinic enrollment is based on referral from their PCP or other providers. Patients in the clinic are evaluated using the frailty phenotype, assessed by the Johns Hopkins (JH) frailty assessment calculator. We excluded seven patients who had a Long COVID diagnosis.

The Long COVID cohort comprised patients seen in the Miami VA Long COVID clinic. The Miami VAMC Long COVID clinic evaluates patients struggling with new or worsened symptoms lasting at least 12 weeks after a COVID-19 infection. Clinic enrollment occurs by referral and through a digital screening program. Starting in January 2024, as part of a clinical innovation project, all patients seen in person at the Miami VAMC clinic were screened for frailty by completing the JH frailty phenotype assessment as described above.

The Miami Veterans Affairs Healthcare System Institutional Review Board approved this study. The IRB granted a waiver for informed consent. The data was fully anonymized according to VHA standards prior to analysis. The data was accessed on 9 January 2024 to 17 January 2025.

2.2. Data Source for Physical Frailty Phenotype (Johns Hopkins Frailty Calculator)

We obtained the physical frailty phenotype using a combination of self-report questionnaires, physical performance tests, and, sometimes, clinical measurements, according to the validated Fried Frailty Phenotype. A health technician in both the Geriatric Specialty clinic and the Long COVID clinic evaluated the presence of unintentional weight loss, exhaustion, low physical activity, slow walking speed, and weak grip strength. Data was entered into the Johns Hopkins Frailty calculator to determine frailty status. The use of this calculator was standardized across both clinics. Weight loss was defined as loss of 10 pounds or more unintentionally in the past year or weight loss of more than 5% of body weight over the year. Exhaustion was defined as a self-reported feeling of exhaustion or low energy on at least three days a week. Low physical activity was defined as less than 383 kilocalories per week for men and less than 270 kilocalories per week for women, using a standardized physical activity questionnaire, such as the Minnesota Leisure Time Activity Questionnaire, and calculating the kilocalories expended per week. Slow walking speed was defined for men or women as less than 0.8 m/s. for men, we defined weak grip strength as <29 kg for BMI ≤ 24, <30 kg for BMI 24.1–26, and <32 kg for BMI > 26, while for women, we used the following definitions: <17 kg for BMI ≤ 23, <17.3 kg for BMI 23.1–26, and <18 kg for BMI > 26.

2.3. Data Source for VA Frailty Index (VA-FI)

We calculated the VA Frailty Index (VA-FI) for each patient using data from VA CDW (Corporate Data Warehouse) at the time they presented to the respective clinic. We also obtained the VA-FI status in December 2019—before the COVID-19 pandemic—for the two groups. The 31-item VA Frailty Index (VA-FI)15 was developed based on the deficit accumulation conceptual framework (

Table A1. VA Frailty Index variables.

Morbidity		17	Fall or Fall-Related Diagnoses
1	Anemia	18	Fatigue
2	Atrial Fibrillation	19	Gait Abnormality
3	Cancer (except basal cell skin cancer)	20	Parkinson’s Disease or Tremor Disorders
4	Cerebrovascular Disease	21	Peripheral Vascular Disease or Intermittent Claudication
5	Coronary Artery Disease	22	Muscular Wasting
6	Diabetes	Sensory Loss:
7	Heart Failure	23	Hearing Impairment/Aid
8	Hypertension	24	Peripheral Neuropathy
9	Kidney Disease	25	Vision Impairment
10	Liver Disease or Cirrhosis	Cognition and Mood:
11	Lung Disease	26	Dementias
12	Thyroid Disease	27	Anxiety
13	Osteoporosis or Pathological Fracture	28	Mood Disorders (Depression, Bipolar Disorder)
14	Incontinence	Other:
Function:		29	Chronic Pain
15	Arthritis	30	Failure to Thrive
16	Use of Durable Medical Equipment	31	Weight Loss in the Past Year

). We categorized patients based on the score as severely frail (0.40 or above, frail (>0.21), pre-frail (0.11–0.20), or robust (<0.10). Demographic information was collected via review from the VA EHR dataset and COVID-19 Shared Data Resource [15].

2.4. Inclusion and Exclusion Criteria

The study included all patients evaluated in the geriatric frailty clinic and Long COVID clinic from 9 January 2024 to 17 January 2025. Patients who were excluded from this study were those in the geriatric frailty clinic who had a diagnosis of Long COVID. The geriatric frailty clinic evaluated only patients older than 65 by referral from primary care. There were no age restrictions in the Long COVID clinic. No other restrictions for comorbidities, age, race, or gender were applied.

2.5. Statistical Analysis

Continuous variables were presented as mean ± standard deviation; categorical variables were presented as frequency and percentage. We compared numerical variables using t-tests and categorical variables using the Chi-squared test. A p-value of <0.05 was considered significant. All the statistical analyses were performed with R (the R project for statistical computing, version 4.0.5).

3. Results

We included 106 patients from the Long COVID clinic and 97 from the frailty clinic, and the patients’ demographic characteristics are presented in

Table 1. Cohort demographic information.

	Long COVID (n = 106)	Frailty Clinic (n = 90)	p-Values
Male Gender	82 (78.9%)	88 (97.8%)	<0.01
Age (years)	60 ± 12.6	77.0 ± 5.7	<0.01
Age Groups, n (%) <55	31 (29.8%)	0	<0.01
55–64	33 (31.7%)	0
65–74	25 (24.0%)	10 (11.1%)
75–84	13 (12.5%)	60 (66.7%)
≥85	2 (1.9%)	20 (22.2%)
Race, n (%)
White	51 (49%)	52 (57.8%)	0.24
Black	43 (41.3%)	32 (35.6%)
Asian	2 (1.9%)	0
Native Hawaiian or Other Pacific Islander	3 (2.9%)	0
Unknown	4 (3.8%)	4 (4.4%)
Ethnicity, n (%)
Hispanic	28 (26.9%)	8 (8.9%)
Not Hispanic	74 (71.2%)	81 (90.0%)	0.01
Unknown	1 (1.0%)	2 (2.2%)
VA-FI (baseline)
Frail	36 (34.0%)	50 (55.6%)
Pre-frail	29 (27.4%)	23 (25.6%)	<0.01
Robust	41 (38.7%)	17 (18.9%)

. Patients from the Long COVID clinic were significantly younger than those from the frailty clinic (60 ± 12.6 vs. 77.0 ± 5.7, p < 0.01) and more likely to be female, though both groups were predominantly male. There were more Hispanic patients in the Long COVID clinic compared to the frailty clinic, 28 (26.9%) vs. 8 (8.9%) (p < 0.01). Using the VA-FI deficit accumulation model, the frailty clinic also had more patients that were frail at pre-COVID baseline in 2019 compared to the Long COVID clinic (55.6% vs. 34.0%, p < 0.01).

There was a significant difference in the identification of frailty between groups using the VA-FI model but not with the Fried model. These two models demonstrated different assessments of frailty, as 66.3% of the Long COVID model was considered frail by VA-FI, while only 26.4% was considered frail by the Fried model, and among the patients in the frailty clinic, 93.3% were frail per VA-FI and only 34.4% were frail according to Fried (

Table 2. Frailty percentage by group and cohort type using the Veteran Affairs Frailty Index (VA-FI) and Fried phenotype.

Frailty Model	Frailty Group	Long COVID Clinic	Frailty Clinic	p-Values
VA-FI	Frail	69 (66.3%)	84 (93.3%)	<0.01
	Severe Frail	23 (21.7%)	39 (43.3%)
	Pre-frail	25 (24.0%)	5 (5.6%)
	Robust	10 (9.6%)	1 (1.1%)
Fried Phenotype	Frail	28 (26.4%)	31 (34.4%)	0.17
	Pre-frail	63 (59.4%)	43 (47.8%)
	Robust	15 (14.2%)	7 (7.8%)

).

Among frail patients (

Table 3. Rate of presentation by Fried phenotype category in frail and pre-frail patients from both cohorts.

	Long COVID Clinic	Frailty Clinic	p-Values
	Frail (n = 28)	Frail (n = 31)	0.06
Weakness	21 (75%)	29 (96.7%)
Slowness	13 (46.4%)	27 (90.0%)
Weight loss	9 (32.1%)	9 (30.0%)
Exhaustion	27 (96.4%)	16 (53.3%)
Low Activity	22 (78.6%)	19 (63.3%)
	Pre-frail (n = 63)	Pre-frail (n = 43)
Weakness	17 (27.0%)	22(57.9%)	<0.01
Slowness	6 (9.5%)	21(53.8%)
Weight loss	3 (4.8%)	9(23.7%)
Exhaustion	43(68.3%)	5(13.2%)
Low Activity	18(28.6%)	4(10.5%)

), those in the Long COVID group had higher rates of exhaustion (96.4%) and low activity (78.6%) compared to the frailty clinic group (exhaustion 53.3%; low activity 63.3%). Conversely, the frailty clinic patients had higher rates of weakness (96.7%) and slowness (90.0%) compared to the Long COVID group (weakness 75% and slowness 46.4%). However, the overall differences between these group components approached but did not reach significance (p = 0.06). Among the pre-frail patients, a similar pattern was seen, with Long COVID patients having higher rates of exhaustion (68.3% vs. 13.2%) and low activity (28.6 vs. 10.5%) as compared to the frailty clinic’s patients. In addition, frailty clinic patients had higher rates of weakness (57.9% vs. 27%) and slowness (53.8% vs. 9.5%). Within the pre-frail group, there was a significant difference between the Long COVID and frailty clinic phenotype components (p < 0.01).

4. Discussion

4.1. Differences in Fried Phenotype Between Group

In the patients evaluated in the GRECC frailty clinic (standard frailty groups), weakness and slowness were the predominant features in both the frail and pre-frail groups, with increasing exhaustion and lower activity in the frail group. Frailty in older adults is thought to be driven by weakness and reduced muscle strength, with sarcopenia often developing as a precursor to frailty [16]. Studies have shown that decreased muscle strength is highly predictive of frailty status over several years [17].

Patients with frailty in the Long COVID group experienced exhaustion and low activity at a higher rate than those in the geriatric frailty clinic. This relationship was even more pronounced in the pre-frail category. These findings were anticipated, given the pathophysiological drivers of Long COVID within each frailty category. Fatigue is a dominant complaint of individuals with Long COVID, which can last for months to years at a time [18]. Additionally, studies have shown evidence of lower physical activity and impaired function, which significantly affect the quality of life of patients with Long COVID [19]. As a result, patients with post-viral conditions, such as Long COVID, have a higher probability of exhaustion and a lower physical activity phenotype, as seen in our results. These differences in phenotype presentation may be partly driven by the underlying pathophysiological mechanisms, response to exertion, and treatment outcomes.

4.2. Limitations of Group Comparison in Fried Phenotype

The frailty tools used in this study have been validated for patients over 65. Our study attempted to look at how frailty presented using the same frailty markers in both a young and older cohort. One of the reasons for this is that, through our study, we wanted to understand if patients with Long COVID were meeting the clinical frailty criteria of older adults after contracting the virus. Our results show a pattern of low activity and exhaustion that was higher than expected for a younger group. However, caution should be taken in the interpretation of these results because we did not have an age-matched comparison group, as the Long COVID group tended to be younger. Therefore, the observed difference may be due to Long COVID, the age difference, or a combination of both. Regarding frailty determination, the actual rate of frailty may have been underestimated in the younger cohort as the thresholds for the physical frailty phenotype were lower than expected for age-matched peers. Further studies should adjust the frailty phenotype to account for age-specific frailty markers concerning weight loss, grip strength, and gait speed.

4.3. Changes in Frailty Index

In our study, we were able to determine the baseline VA-FI scores from 2019, before the emergence of COVID-19. As expected, the Long COVID group had higher baseline rates of robust patients (38.7% vs. 18.9%) and lower rates of frailty (34% vs. 55.6%) compared to the frailty clinic cohort. After developing Long COVID, the number of frail patients by deficit accumulation almost doubled (from 34% to 66.3%), while the number of robust patients dropped considerably (38.7% to 9.6%). The same effect was observed in the frailty clinic patients, with the majority becoming frail (55% to 93.3%), while few remained in the pre-frail or robust categories.

These findings in the Long COVID group were unexpected given the rapid change in frailty status. Patients in the frailty clinic, with more than half meeting the criteria for frailty at baseline and subsequent referral for evaluation in the frailty clinic, were expected to show further progression to frailty. However, in the Long COVID cohort, comprising a younger and healthier patient population, there was also a major shift in VA-FI frailty characteristics. These results should be interpreted with caution. Other factors may be driving these changes, such as general aging, progression, changes in underlying comorbidities, or changes in healthcare utilization patterns after developing these conditions. Further research is needed to determine the rate of change and the factors that may be driving this shift in the VA-FI for patients with Long COVID, including a comparison of expected deficit accumulation over 5 years in a similar (age and comorbidity population) non-Long COVID cohort.

4.4. Strengths and Limitations

One of the major strengths of this study is the use of both the deficit accumulation model and the Fried phenotype to understand the frailty characteristics within both groups. Currently, there are limited data regarding this physical frailty phenotype in Long COVID patients, as many studies have been conducted using only the deficit accumulation model. Another strength of this study includes the longitudinal use of the deficit accumulation model, demonstrating a comparison of frailty development over a 5-year period between groups. Given the design of the study, we were able to analyze frailty characteristics across multiple patient groups within the veteran community.

The limitations of this study include the lack of validation of frailty phenotypic measures in younger populations and the lack of a non-Long COVID control group with similar age and comorbidity distribution for comparison of the development of frailty and frailty characteristics using these methods. Another limitation may be generalizability, as clinic patients were referred by their primary care providers to either the frailty clinic or the Long COVID clinic within the VA health system. Patients who were subjected to this referral method may have had more disabling symptoms or concerns, requiring further specialty evaluation. As a result of this referral process, our sample population may be skewed towards more symptomatic individuals or individuals more motivated in seeking care. Another limitation was the lack of longitudinal assessments for physical frailty. Although this is beneficial for comparing the phenotypic presentation of frailty within both groups, further research is required to understand the changes in frailty characteristics over time as a function of specific interventions.

5. Conclusions

Patients with Long COVID had a physical frailty phenotype, with higher rates of exhaustion and low physical activity being the predominant drivers. Using the deficit accumulation model among a cohort of patients evaluated in a Long COVID clinic, there was nearly double the number of patients with frailty as compared to the pre-pandemic baseline. While these results should be interpreted with caution, our findings highlight a risk for frailty development in Long COVID patients, warranting clinical evaluation and treatment. Additional research is required to understand intervention strategies that may be beneficial for these patients.