Decline in Mobility and Balance in Persons With Post–COVID-19 Condition

Abstract

Objectives

Post–COVID-19 condition (PCC) may impact mobility and balance and affect physical function. The objectives of the study were to estimate the prevalence of decline in balance and mobility in individuals with PCC; explore the association between comorbidities and sociodemographic characteristics with decline in balance and mobility; and evaluate correlations between decline in mobility and balance with change in performance of usual activities, personal care, and global health perception.

Methods

The design was a cross-sectional study of persons with a COVID-19 diagnosis that was confirmed at least 3 months before the study. Those with PCC, defined as those still troubled by symptoms, were evaluated for decline in mobility and balance and with associated clinical and demographic factors using bivariate analysis and multivariable logistic regression. Correlations between decline in mobility and balance were also examined with change in ability to perform usual activities, personal care, and global health perception.

Results

In 1031 persons with PCC, mobility deteriorated in 44.9%, and balance deteriorated in 37.1%. Older age, hospitalization, comorbidities, and obesity were associated with decline in mobility, while decline in balance was associated with older age and comorbidities. Reduced mobility was associated with changes in ability to carry out usual activities (rp = 0.6), conduct personal care (rp = 0.6), and global health status (rp = 0.5). Correlations between decline in balance and these same outcomes were 0.5, 0.5, and 0.45, respectively.

Conclusions

Almost half the participants with PCC had reduced mobility, and over a third reported deterioration in balance, with associated difficulties with daily functioning. Factors associated with greater decline help identify those most at risk.

Impact

Many people with PCC experience changes in mobility and balance, which can affect functional capacities and lead to physical therapist consultations. Further study should assess specific needs of these patients and determine effective physical therapist interventions to meet these needs.

Lay Summary

Many persons with post–COVID-19 condition (PCC) experience symptoms resulting in functional problems, such as difficulties with personal care and performing usual activities. This study focused on difficulty in mobility and problems with balance. Almost half of persons with PCC in the study had declines in getting around and over a third had declines in balance. These problems were associated with older age and having other comorbid health conditions and were linked with decline in ability to perform personal care, carry out usual activities, and perceived global health status.

Balance, Cross-Sectional Study, Function, Mobility, Post–COVID-19 Condition

Introduction

There have been >700 million confirmed COVID-19 cases throughout the world since the beginning of the pandemic in March 2020,¹ with >4 million identified cases in Canada.² Among the most frequent symptoms reported are fatigue, breathlessness, difficulty sleeping, muscle aches, brain fog, joint pain, and heart issues.³ Some persons continue to experience symptoms following the acute COVID-19 infection period. The World Health Organization defines post–COVID-19 condition (PCC) as the continuation or development of new symptoms 3 months after the initial infection, with symptoms lasting for ≥2 months, and with no other explanation.⁴

Patients with persistent symptoms of COVID-19 have a lower quality of life when compared to a healthy population,⁵ and they experience difficulties with activities of daily living.⁶ Furthermore, fatigue, the most common symptom of PCC, influences long-term functional ability.⁷ Indeed, it is reported that many individuals who were hospitalized for COVID-19 have persistent symptoms that affect their daily functional activities and experience substantial reductions in health status, mobility, usual activities, and personal care.⁸

Studies have looked at the prevalence of PCC and associated symptoms.⁹ However, the literature is scarce regarding the impact of PCC on mobility and balance. Evaluating the frequency of mobility and balance disability as well as the associated factors would be helpful for the planning and implementation of physical therapist services for persons with PCC, as there is evidence that physical therapist intervention can improve physical function.¹⁰

The objectives of this study were to estimate the prevalence of decline in balance and mobility in individuals with PCC; explore the influence of comorbidities and sociodemographic characteristics on decline in balance and on mobility associated with PCC; and evaluate the correlation between decline in mobility and balance with changes in the ability to perform usual activities, personal care, and global health perception in people with PCC.

Methods

The research design is a cross-sectional electronic study. The population included residents of the city of Laval, Quebec (approximately 450,000), with a positive diagnosis of COVID-19, confirmed by a polymerase chain reaction (PCR) test between March 1, 2020 and January 31, 2022. Participants were recruited at least 12 weeks after the positive diagnosis of COVID-19.

Ethics approval was granted by the Comité scientifique et d’éthique de la recherche du CISSS Laval. Subsequently, the personal information of potential participants was obtained from the Public Health Department of the City of Laval, and all eligible participants who provided their email address were contacted. The inclusion criteria were adults (≥18 years old) living at home (those living in long-term care facilities were excluded from the study). We sent an email to potential participants describing the project, and those who consented electronically proceeded to respond to the questionnaires.

The study was conducted between June 10, 2022 and October 12, 2022 and included 2 standardized questionnaires: the Newcastle Post-COVID Syndrome Follow Up Screening Questionnaire¹¹ and the COVID-19 Yorkshire Rehabilitation Screen (C19-YRS).¹² The Newcastle Post-COVID Syndrome Follow Up Screening Questionnaire is used to identify people with long-term sequelae of COVID-19. The questionnaire includes 14 questions, and the question we used to classify participants as having PCC or not was “Have you made a full recovery, or are you still troubled by symptoms?”

Participants who were identified as having PCC proceeded to answer the C19-YRS. The C19-YRS includes questions where participants were asked to rank from 0 (no impact) to 10 (severe impact or intensity of symptom) their pre-COVID state as well as their current status regarding breathlessness, airway complications, changes in voice, mobility, balance, fatigue, and participation in activities of daily living, including household roles and leisure activities, personal care, and perceived global health status. The specific questions for mobility were “On a scale of 0 to 10, how severe are any problems you have in walking about or moving about (if normally mobilizes in another way)? 0 means I have no problems, and 10 means I am completely unable to get around. NOW (0–10). PRE-COVID (0–10).” The specific questions for balance were “On a scale of 0–10, how severe are any balance problems you have? 0 means I have no problems, 10 means I am completely unable to balance myself. NOW (0–10). PRE-COVID (0–10).”

We also collected sociodemographic and clinical factors. These included sex, age, marital status, education level, current employment status, body mass index, hospitalization due to COVID-19, vaccination status against COVID-19, duration since positive PCR test, and comorbid chronic conditions.

Data Analysis

We calculated the prevalence of PCC in our sample as per our definition of still being troubled by symptoms. In those with PCC, we focused on 2 functional outcomes: mobility and balance. Analysis consisted of changes from pre-COVID to current status for each of these 2 outcomes as reported on the C19-YRS. The changes for each of the variables were ranked: no deterioration (0), slight deterioration (−1 to −3), moderate deterioration (−4 to −6), and severe deterioration (−7 to −10). The change was calculated by subtracting the current score from the pre-COVID score for each item. Next, we explored the associations between sociodemographic and clinical factors and each of our outcomes (mobility and balance changes). This was done by first using bivariate analysis (χ²) and then implementing multivariable logistic regression models. Finally, using Pearson correlation coefficients, we evaluated the association between decline in mobility and balance with change in ability to perform usual activities, personal care, and perception of global health status in people with PCC. For all analyses, we used only those with complete data and did not impute values for missing data.

Role of the Funding Source

The funders played no role in the design, conduct, or reporting of this study.

Results

Among the 41,099 residents from the city of Laval who were sent the invitation to participate by email (at least ≥12 weeks after their positive PCR test), 4592 residents opened the email, and 2764 participants completed the questionnaires.

There were 1031 (37.8%) participants with PCC (were still troubled by symptoms) and they formed our study sample. Most (77%) were female, and the mean age was 45.4 years (SD = 13.5). Almost 55% reported at least 1 chronic comorbid condition; 26.4% had 1 comorbidity, 18.8% had 2 comorbidities, and 9.5% reported ≥3 comorbid conditions. The most frequently reported comorbidities were depression and anxiety, followed by hypertension, hypercholesterolemia, and arthritis. Table 1 lists a description of clinical and sociodemographic characteristics of our sample of persons with PCC.

Table 1

Sociodemographic and COVID-19–Related Characteristics of Persons With Post–COVID-19 Condition (N = 1031)^a

Characteristic	N	%
Sex (n = 1024)
Female	791	77.2
Male	233	22.8
Age, y (n = 1017)
<45	467	45.9
45–64	475	46.7
≥65	75	7.4
Marital status (n = 1024)
Single/separated/divorced/widowed	322	31.4
Married or living together	702	68.6
Education level (n = 1022)
Bachelor or more	398	38.9
More than high school but less than bachelor	447	43.7
High school or less	177	17.3
Current employment status (n = 963)
Unemployed	183	19.0
Working	780	81.0
Obesity (n = 977)
No	581	59.5
Yes	396	40.5
Hospitalized due to COVID (n = 1021)
No	943	92.4
Yes	78	7.6
COVID vaccination (n = 1026)
No	73	7.1
Yes	953	92.9
Duration since positive PCR test, mo (n = 1007)
<12	472	46.9
≥12	535	53.1
Any chronic diseaseb (n = 1028)
No	466	45.3
Yes	562	54.7

^a

PCR = polymerase chain reaction.

^b

Chronic diseases include: congestive heart failure, chronic lung disease, diabetes, arthritis, hypertension, high cholesterol, blood circulation problems, stroke, cancer, depression, anxiety, psychological or psychiatric problems, cancer, other (specify).

Participants reported perceived level of mobility and balance (before and after infection). The Figure illustrates the proportion of participants who report no change, a small deterioration, a moderate deterioration, and a severe deterioration in mobility and balance. Approximately, 45% of participants with PCC reported a change in their mobility (small = 25.9%; moderate =15.3%; severe = 3.7%). With respect to balance, 37% reported a change from pre-COVID status (small = 24.3%; moderate = 9.9%; severe = 2.9%).

Figure

Deterioration in mobility and deterioration in balance in persons with post-COVID-19 condition (PCC) as compared to the before COVID status (n = 1031). 55.1% had no deterioration in mobility and 62.9% had no deterioration in balance. There was slight deterioration in 25.9% for mobility and 24.3% for balance. There was moderate deterioration in 15.3% for mobility, and 9.9% for balance. Severe deterioration was present in 3.7% for mobility and 2.9% for balance. Numbers in parentheses refer to difference between severity before COVID and at time of study (≥12 weeks after COVID).

Table 2 illustrates the associations (bivariate χ² analysis and multivariable logistic regression) between decline in mobility and clinical and sociodemographic factors. In the bivariate analyses, older age, being obese, having been hospitalized for COVID, having a chronic comorbid disease, and having been diagnosed with COVID >1 year earlier were associated with decline in mobility. In the multivariable model, the following were independently associated with decline in mobility: older age, obesity, hospitalization for COVID, and having at least 1 comorbid chronic condition.

Table 2

Factors Associated With a Decline in Mobility in Persons With PCC (n = 1031)^a

Characteristic	No Change (N = 492)	Decline in Mobility (N = 401)	Bivariate Analysis	Multivariable Logistic Analysis
	n (%)	n (%)	P	Adj OR (95% CI)
Sex (n = 888)			.53
Female	370 (54.49)	309 (45.51)		1.22 (0.87–1.70)
Male	119 (56.94)	90 (43.06)		Ref.
Age, y (n = 885)			<.001b
<45	256 (64.32)	142 (35.68)		Ref.
45–64	202 (48.33)	216 (51.67)		1.82 (1.34–2.46)b
≥65	31 (44.93)	38 (55.07)		1.85 (1.02–3.33)b
Marital status (n = 890)			.77
Single/separated/divorced/widowed	147 (54.44)	123 (45.56)		1.05 (0.76–1.44)
Married or living together	344 (55.48)	276 (44.52)		Ref.
Education level (n = 889)			.37
Bachelor’s degree or higher	205 (57.91)	149 (42.09)		Ref.
More than high school but less than bachelor’s degree	209 (54.15)	177 (45.85)		1.06 (0.77–1.45)
High school or less	77 (51.68)	72 (48.32)		1.16 (0.75–1.79)
Obesity (n = 856)			<.001b
No	299 (59.80)	201 (40.20)		Ref.
Yes	177 (49.72)	179 (50.28)		1.44 (1.07–1.93)b
Hospitalized due to COVID (n = 886)			<.001b
No	463 (56.95)	350 (43.05)		Ref.
Yes	25 (34.25)	48 (65.75)		2.02 (1.17–3.51)b
COVID vaccination (n = 889)			.49
No	32 (50.79)	31 (49.21)		Ref.
Yes	457 (55.33)	369 (44.67)		0.82 (0.47–1.45)
Duration since positive test, mo (n = 880)			0.01b
<12	243 (59.71)	164 (40.29)		Ref.
≥12	243 (51.37)	230 (48.63)		1.26 (0.94–1.69)
Any chronic disease (n = 890)			<.001b
No	246 (61.96)	151 (38.04)		Ref.
Yes	245 (49.70)	248 (50.30)		1.50 (1.11–2.01)b

^a

Adj OR = adjusted odds ratio; PCC = post–COVID-19 condition; Ref. = reference category.

^b

Factors significantly associated with a decline.

Similarly, we found that being 45 to 64 years old compared with being younger, having been hospitalized for COVID, having a chronic comorbid condition, and having been diagnosed with COVID >1 year earlier were associated with decline in balance in the bivariate analysis. The multivariable logistic regression analysis indicated that older age (45–64 years) and chronic comorbidity were independently associated with decline in balance (Tab. 3).

Table 3

Factors Associated With a Decline in Balance in Persons With PCC (n = 1031)^a

Characteristic	No Change (N = 561)	Decline in Balance (N = 3 31)	Bivariate Analysis	Multivariable Logistic Analysis
	n (%)	n (%)	P	Adj OR (95% CI)
Sex (n = 887)			.55
Female	424 (62.54)	254 (37.46)		1.16 (0.83–1.64)
Male	135 (64.59)	74 (35.41)		Ref.
Age, y (n = 883)			<.001b
<45	278 (69.85)	120 (30.15)		Ref.
45–64	245 (58.61)	173 (41.39)		1.64 (1.20–2.24)b
≥65	34 (50.75)	33 (48.25)		1.60 (0.89–2.89)
Marital status (n = 889)			.91
Single/separated/divorced/widowed	169 (62.59)	101 (37.41)		0.95 (0.69–1.32)
Married or living together	390 (63.00)	229 (37.00)		Ref.
Education level (n = 888)			.14
Bachelor’s degree or higher	236 (66.48)	119 (33.52)		Ref.
More than high school but less than bachelor’s degree	239 (61.76)	148 (38.24)		1.19 (0.87–1.64)
High school or less	84 (57.53)	62 (42.47)		1.16 (0.74–1.80)
Obesity (n = 855)			.244
No	325 (65.13)	174 (34.87)		Ref.
Yes	218 (61.24)	138 (38.76)		1.05 (0.78–1.42)
Hospitalized due to COVID (n = 885)			<.01b
No	523 (64.41)	289 (35.59)		Ref.
Yes	34 (46.58)	39 (53.42)		1.68 (1.00–2.85)
COVID vaccination (n = 888)			.25
No	36 (56.25)	28 (43.75)		Ref.
Yes	523 (63.47)	301 (36.53)		0.78 (0.45–1.38)
Duration since positive test, mo (n = 879)			.04b
<12	273 (67.08)	134 (32.92)		Ref.
≥12	285 (60.38)	187 (39.62)		1.21 (0.90–1.63)
Any chronic disease (n = 889)			<.01b
No	270 (67.50)	130 (32.50)		Ref.
Yes	289 (59.10)	200 (40.90)		1.37 (1.01–1.86)b

^a

Adj OR = adjusted odds ratio; PCC = post–COVID-19 condition; Ref. = reference category.

^b

Factors significantly associated with a decline.

We also calculated associations between sociodemographic and clinical factors with the outcome of moderate-to-severe decline in both mobility and balance. Females (OR = 1.60; 95% CI = 1.02–2.49), persons between 45 and 64 years old as opposed to younger persons (OR = 1.90; 95% CI = 1.28–2.83), persons who were obese (OR = 1.45; 95% CI = 1.00–2.29), and those who had been hospitalized for COVID (OR = 2.13; 95% CI = 1.21–3.75) were more likely to have moderate to severe deterioration in mobility. For balance, factors that were independently associated with moderate-to-severe decline were: 45 to 64 years old versus those who were younger (OR = 1.81; 95% CI = 1.12–2.90) and having been infected with COVID >12 months earlier (OR = 1.81; 95% CI = 1.15–2.86).

In Table 4, we report the correlations between decline in mobility and decline in balance with ability to perform usual activities, personal care, and perception of global health status. The Pearson correlation coefficients range from 0.45 to 0.63 and are all statistically significant.

Table 4

Pearson Correlations (R_p) Between Change in Mobility, Change in Balance, and Change in Ability to do Usual Activities, Personal Care, and Perception of Global Health Status

	Change in Ability to do Usual Activities Rp(95% CI)	Change In Ability to do Personal Care Rp(95% CI)	Change in Perception of Global Health Status Rp(95% CI)
Change in mobility	0.63 (0.59–0.67)	0.60 (0.56–0.64)	0.52 (0.47–0.57)
Change in balance	0.53 (0.48–0.57)	0.54 (0.50–0.59)	0.45 (0.39–0.50)

Discussion

In this study, 37.8% of participants were troubled with persisting symptoms ≥12 weeks after their positive COVID test (PCC). Our results align with other studies with similar selection criteria (ie, ≥12 weeks since the positive PCR test).^3,9 For participants with PCC, declines in mobility and balance were found in 44.9% and 37.1%, respectively. Older age, hospitalization, comorbidities, and obesity were associated with decline in mobility, and decline in balance was associated with older age and comorbidities.

In terms of mobility, Tabacof et al¹³ reported that 56% of their nonhospitalized COVID long-haulers had mobility changes, which is higher than what is reported in this current study (44.91%). This higher rate may be explained due to their selection criteria; their sample consisted of persons who were undergoing medical follow-up (attending a PCC clinic) as opposed to our community-based study. Even higher rates are reported in a Danish study of persons referred to occupational therapy at a post–COVID-19 clinic: 95% of persons had slight to severe functional limitations at 6 to 12 months after COVID-19 infection. In this study, the authors included any limitations, whereas we looked specifically at decline in mobility to determine changes related to PCC, thus people with preexisting mobility issues who perhaps did not decline further would not be included. In an Italian study of persons who had been hospitalized for COVID 4 months earlier, 53.8% had some functional impairment measured by the 2-Minute Walk Test. Furthermore, chronic obstructive pulmonary disease was associated with physical impairment,¹⁴ which supports our finding of association of decline in mobility with having a chronic comorbid condition. Again, the higher frequency of functional impairment can be explained by the fact that the participants had been all hospitalized (and were likely more severe cases as well as having been immobile for longer periods of time due to hospitalization) versus our study where most had their COVID managed in the community. Also, their study did not measure change from pre-COVID status, thereby including everyone who reported a functional impairment regardless of their pre-COVID status.

A Canadian study of community-dwelling seniors who had COVID-19 (self-reported but not necessarily PCC) were almost twice as likely to report a negative change in mobility and physical function compared to those who had not had COVID.¹⁵ Although the timing of COVID was not documented, it is possible that many had ongoing symptoms and possibly PCC, supporting our results of declining mobility following COVID-19 infection. Similar to our findings, older persons, and those having multiple chronic conditions were more likely to report worsening mobility. The authors also found that lower socioeconomic status was associated with mobility problems. While we did not measure income directly, we did measure education which served as a proxy for socioeconomic status; however, this was not found to be associated with mobility limitations.

In our community-based sample with PCC, 37% reported a decline in balance. Balance problems have been reported in some smaller studies. Yılmaz et al¹⁶ reported that 22% of 37 study participants recovering from mild COVID-19 had balance problems which impacted their activities of daily living. In a study with 50 participants also recovering from mild COVID-19, Guzik et al¹⁷ reported higher rates of postural instability when compared with a control group. Balance issues due to COVID-19 may be explained by inner ear damage caused by the infection, resulting in vestibular dysfunction. This may result in slower postural compensation and dizziness in those with PCC.¹⁸ De Sousa et al¹⁹ also reported worse balance and overall physical function in 40 persons with PCC who had not been hospitalized compared to controls who had not had COVID.

These results from the literature, as well as findings from this study, support a potential negative impact of PCC on mobility and balance. It also reinforces the importance of addressing these issues for practitioners. However, in the literature, there seems to be a gap as to what risk factors are associated with changes in mobility and balance. Thus, the current study provides information that can be useful to clinicians. For example, it may be especially important for physical therapists to evaluate mobility and balance problems in persons with PCC who are older, obese, had been hospitalized, and have other comorbidities.

PCC has important repercussions on health and functional status. These sequalae highlight the importance of assessing the multiple factors associated with this disease. Considering that more than a third of the people with a positive PCR test in this sample had persisting symptoms months or even years after the original infection, there is a need for specialized assessment and treatment of this condition. Specifically, many experience changes in their mobility and balance. These changes are correlated with the decline in functional capacities and may lead to the need for consultations with physical medicine clinicians as well as physical therapists and occupational therapists. However, many with PCC do not receive adequate treatment.²⁰ Our study underscores the need to provide effective rehabilitation interventions to persons with PCC who have mobility and balance limitations.

Limitations

There are several limitations to our study. The data collected in this study were self-reported by the participants as opposed to data collected by clinicians using standardized tests (eg, 6-Minute Walk Test for mobility, Berg Balance Test). The response rate to the survey (among all those with a positive COVID diagnosis) was low, which may affect generalizability. Furthermore, those who responded to the online study may be those who have more problems after COVID infection, which would likely lead to selection bias, inflating the prevalence figures. Despite this, the prevalence of PCC appears consistent with other studies. The questionnaires used in this study compared the state of the participants before COVID-19 infection and at least 12 weeks after infection. There may be recall bias with respect to pre-COVID level of function. On the other hand, documenting functional differences from pre– to post–COVID status to ascertain any change in function due to PCC reduces the overestimation of disability in persons who may have had difficulties in mobility and balance pre-COVID. This study was also limited to midsize Canadian city (ie, Laval) and may not be generalizable to other contexts.

We defined PCC as a response to 1 question on the Newcastle questionnaire (ie, having made a full recovery as opposed to still being troubled by symptoms). The rationale for choosing this definition is that persons who are no longer troubled by their symptoms would be less likely to seek health services. This is in line with Andersen behavioral model, whereby seeking health care is associated with predisposing factors (mainly sociocultural and health beliefs), enabling factors (availability of services and ability to access these), and need (perceived or evaluated need).²¹ Our main outcome focuses on the need aspect. Thus, our definition excludes persons who still have symptoms but consider themselves to be fully recovered (low perceived need).²²

Also, this study was observational and cross-sectional; we can only conclude that there are associations between various factors and our outcomes and cannot assume causation between PCC and decline in mobility and balance.

A strength of our study is that all cases were confirmed COVID cases by the Public Health Department. However, it is important to note that the majority of cases were pre-Omicron, as the first cases of omicron were detected in Quebec at the end of November 2021. Severity of PCC may be related to the COVID variant type.²³ Nevertheless, the type of COVID variant would not make a difference in the management of problems with mobility and balance.

Conclusion

The objectives of this study were to estimate the prevalence of decline in balance and mobility in individuals with PCC and to explore the association of comorbidities and sociodemographic characteristics on these problems as well as the relationship to functioning and perceived health status.

We found nearly half of those with PCC had a decline in mobility, and over a third reported deteriorations in balance. The substantial prevalence of PCC, its symptoms, and effects on mobility and balance not only reinforces the importance of controlling the spread of COVID-19 but also emphasizes the need for providing effective services that address the complexity of this condition. This suggests that further studies should be done to assess the specific needs of these patients and to adapt current physical therapist practice to address those needs.

Author Contributions

Debbie Ehrmann Feldman (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing—original draft), Albert Guillemette (Conceptualization, Formal analysis, Methodology, Project administration, Writing—original draft), Juliana Sanzari (Conceptualization, Formal analysis, Methodology, Writing—original draft), Stéphane Youkheang (Conceptualization, Formal analysis, Methodology, Writing—original draft), and Barbara Mazer (Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Validation, Writing—review & editing).

Ethics Approval

This study was approved by the Comité Scientifique et d’Éthique de la Recherche du CISSS Laval.

Funding

This study was funded by the Fondation Cité de la Santé and the Jewish Rehabilitation Hospital Foundation.

Disclosures and Presentations

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.