Assessment of activities of daily living in patients post COVID-19: a systematic review

Introduction

In December 2019 a SARS-COV-2 virus emerged in Wuhan, China, which has rapidly spread around the world (Phelan, Katz & Gostin, 2020). In March 2020, the World Health Organization (WHO) declared a global pandemic due to the disease caused by this novel coronavirus (COVID-19) (Adhikari et al., 2020).

The magnitude of the sequelae produced by the disease at different functioning levels are still under research. The consequences, such as fatigue and breathlessness, have been reported to persist after hospital discharge and the impact at different functioning levels over time remains unclear (Carfì, Bernabei & Landi, 2020; Halpin et al., 2020). These findings are in line with the presence of symptoms such as myalgia that has been reported in about 15% of patients 4 months post-COVID-19 infection (Garrigues et al., 2020).

The immediate and short-term assessment of physical capacity in patients with COVID-19 has been a challenge, with a variable number of subjects being able to perform the tasks included on the most commonly used physical capacity assessment tests (Belli et al., 2020; Curci et al., 2020; Sakai et al., 2020; Liu et al., 2020; Bousquet et al., 2020; Zerah et al., 2020). In this way, it seems necessary and reasonable to complement the clinical assessment of physical capacity with a further analysis of functional ability related to perform activities of daily living (ADL).

Impairments in body functions and structure, such as dyspnea, weakness, myalgia and other pain sensations limit the ability to perform both basic ADL (BADL) and instrumental ADL (IADL). BADL are related to personal care and mobility (such as dressing, eating, ambulating, toileting, hygiene) whereas IADL is associated with the person’s ability to interact with his/her environment. Limitations in the ability to perform ADL increase the susceptibility to care dependency, reducing the quality of life of both the affected person and the caregiver (Dunsky, 2019). Dependence tends to be accentuated in the elderly population and through all age ranges after an extended ICU stay (Herridge et al., 2016); a reduced level of ADL independence has been related to worse prognosis after COVID-19 (Bousquet et al., 2020; Zerah et al., 2020).

The WHO consider two main aspects when defining functional ability. First, the intrinsic capacity related to all the physical and mental capacities of an individual. Second, the environment related to all the factors in the extrinsic world that form the context of an individual’s life (World Health Organization, 2015). In this way, functional ability during ADL is determined by the impairment affecting the intrinsic capacity as consequence of COVID-19, and the environment is determined by the context of each individual’s life, including variables such as the time course of the disease or the previous functional status. Measuring ADL is relevant since it provides an essential framework of the individual’s current functional state, and thus the requirements for rehabilitation and other resources (Rivera-Lillo et al., 2020).

Scales commonly used to assess functional status regarding ADL in inpatients and survivors of critical illness are: Barthel Index (BI), Functional Independence Measure (FIM), Modified Rankin Scale (mRS), EQ-5D, among others (Elliott et al., 2011). The mRS has been widely used as the primary outcome measure for acute stroke (Broderick, Adeoye & Elm, 2017), whereas FIM was identified as a less biased assessment with greater reliability and validity compared to other instruments (BI or Katz) (Cohen & Marino, 2000). Most of the scales allow quantification of the magnitude of functional independence, based on the assistance requirements across different items related to activities or participation during ADL.

The combination of variables leading to different scenarios related to intrinsic capacity and environment places different limitations on ADL. It is therefore necessary to detect the best way to know and understand functional ability during ADL before and after COVID-19. To fill this knowledge gap, in this work we analyze the most common test reported to date to evaluate the relationship between ADL functional ability and COVID-19. We also describe the factors related to context and intrinsic capacity to provide an overview of the necessity to assess the impact of COVID-19 in ADL limitations. Appropriate assessment scales will help to objectively evaluate the effect of COVID-19 on the functional ability of patients over time, leading to optimal care and rehabilitation strategies for affected subjects. At the same time, ADL functional ability status previous to COVID-19 provides a fine-grained understanding of the importance of individual context in the ADL functional status post-COVID-19. The aim of this report is to understand the impact of COVID-19 on ADL performance of adult patients and to describe the common scales used to assess ADL functional ability status on patients before and after COVID-19.

Materials and Methods

We conducted a systematic review of the literature according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al., 2009). The review was registered in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42020208449).

Results

The initial search found a total of 1,228 studies, 1,005 studies remained after removing duplicates. After abstract and title screening, 983 studies were excluded. Twenty-one full texts remained, which were assessed for eligibility leading to the exclusion of 12 studies due to wrong publication type (n = 5), research protocol (n = 3), wrong study design (n = 2), incomplete data (n = 1), and preprint (n = 1). The process was embodied in a flowchart (Fig. 1). Nine studies (Halpin et al., 2020; Belli et al., 2020; Curci et al., 2020; Sakai et al., 2020; Liu et al., 2020; Bousquet et al., 2020; Zerah et al., 2020; Mcloughlin et al., 2020; Agarwal et al., 2020) met all the inclusion criteria. Four retrospective (Belli et al., 2020; Sakai et al., 2020; Zerah et al., 2020; Agarwal et al., 2020), 2 prospective (Bousquet et al., 2020; Mcloughlin et al., 2020), 2 cross-sectional (Halpin et al., 2020; Curci et al., 2020) and one RCT (Liu et al., 2020) studies were included. A total of 1,465 patients were studied in the included articles. The median age was 68.98 years (± 8.29) of which 716 (48.9%) were women. In total 1,190 (81.2%) were inpatients (Table 1).

The included studies used different instruments such as BI (assesses BADL) (Belli et al., 2020; Curci et al., 2020; Sakai et al., 2020), ADL score (assesses BADL and IADL) (Bousquet et al., 2020; Zerah et al., 2020), FIM (assesses BADL in motor and cognitive domain) (Liu et al., 2020), Composite Functional Score (CFS) (assesses BADL and IADL) (Mcloughlin et al., 2020), mRS (assesses disability in BADL and IADL) (Agarwal et al., 2020) and EQ-5D-5L (assesses BADL) (Halpin et al., 2020). For each study, design, sample size, gender, age, body mass index, setting or intervention and main patient characteristics were extracted (Table 1).

A quality assessment was performed using the National Heart, Lung and Blood Institute’s study assessment tools. Six (66.6%) of the nine studies included obtained a “Good” quality ranking with a total score of over 75% (Curci et al., 2020; Liu et al., 2020; Bousquet et al., 2020; Zerah et al., 2020; Mcloughlin et al., 2020; Agarwal et al., 2020) while three (33.3%) of them pointed to a “Fair” quality ranking with a total score of between 50% and 75% (Halpin et al., 2020; Belli et al., 2020; Sakai et al., 2020). No “Poor” quality ranked studies were included (Supplemental File 2).

Discussion

This review identified nine studies that assessed performance of ADL on COVID-19 patients using eight different scales: BI (Belli et al., 2020; Curci et al., 2020; Sakai et al., 2020), ADL score (Bousquet et al., 2020; Zerah et al., 2020), FIM (Liu et al., 2020), Composite Functional Score (CFS) (Mcloughlin et al., 2020), mRS (Agarwal et al., 2020) and EQ-5D-5L. In these studies, the performance of ADL on COVID-19 patients was mainly used to explore the association between baseline functional ability as a mortality predictor (Bousquet et al., 2020; Zerah et al., 2020) and to identify changes in functional ability outcomes after the acute period of infection (Mcloughlin et al., 2020; Agarwal et al., 2020) (Table 2).

In this way, ADL assessment can be used to identify functional limitations, to understand the patient’s prognosis and to evaluate the intervention’s effect on the subject. The majority of studies included in this review evaluated ADL: (1) On admission to a rehabilitation unit, mostly as a prognostic factor, and (2) At discharge and during the follow-up period, to evaluate the impact of the disease but also to compare the progress obtained as result of an intervention or rehabilitation process.

In all of the studies included, the findings revealed a decline in ADL performance after COVID-19 infection regardless of the scale applied. Those who had the worse results were older patients and/or patients who had complications during their hospital stay, such as being admitted to ICU (Halpin et al., 2020), mechanical ventilation (Belli et al., 2020), delirium (Mcloughlin et al., 2020), cerebral microbleeds, leukoencephalopathy (Agarwal et al., 2020) or greater oxygen requirements (Curci et al., 2020). Participation in a rehabilitation program was also a factor likely to modify the outcomes in functional capacity, however, not all cases were able to improve the performance of ADL (Liu et al., 2020). Several factors regarding a patient’s context are likely to affect the success, or lack of it, in improving the functional capacity and ADL performance of the subjects.

Our results show the different tools used to evaluate functional abilities during ADL. However, not all evaluate the same dimensions. Among the tools used are the BI and the FIM (Belli et al., 2020; Curci et al., 2020; Liu et al., 2020; Sakai et al., 2020). The BI assesses the performance in BADL, such as feeding, personal toileting, bathing, or dressing, with a score between 0 and 100 points (Bouwstra et al., 2019). A higher number reflects greater ability to function independently and has the advantage that it can be applied to self-reporting or direct administration (Bouwstra et al., 2019). The FIM assesses the functional status with scores which range from 18 (lowest) to 126 (highest), but there is a difference to BI, which in addition to BADL incorporates psychological and social functions as communication or special cognition (Maritz et al., 2019). Both instruments are widely used, particularly in patients with neurological conditions (Bouwstra et al., 2019; Maritz et al., 2019).

The EQ-5D-5L is a variation of the EQ-5D, a standardized instrument for measuring generic health status (Janssen et al., 2013). This instrument assesses the mobility, self-care, usual activities, pain/discomfort and anxiety/depression, allowing the evaluation of BADL (Janssen et al., 2013). The ADL score assesses washing, getting dressed, moving about indoors, going to the toilet, eating, and continence with scores ranging from 0 to 6. A total score ≤ 5/6 indicates ADL-dependency (Burton & Potter, 2017). It is likely that the simplest test is the mRS, which measures the degree of disability or dependance in the ADL of people who have suffered a stroke or other causes of neurological disability (Broderick, Adeoye & Elm, 2017). However, it does not allow precise discrimination of the dimensions affected when planning an intervention. On the other hand, if a tool is not capable of meeting the clinician’s requirements, composite tools can be used, as was the case in one of the selected studies.

ADL performance acts as an indicator of a patient’s prognosis (Bousquet et al., 2020; Zerah et al., 2020; Level et al., 2018). Complementing rehabilitation efforts on ADL training may lead to a major impact on the patient’s life, rather than just considering physical function parameters (Elliott et al., 2011). The importance of ADL as a prognostic factor can be seen in the present association between dependance or low functional capacity with short- and medium-term mortality (Guidet et al., 2020).

Another aspect related to independence level is cognitive performance, which can be included to complement the assessment of ADL (Torres-Castro et al., 2021). The cognitive sphere is particularly relevant in IADL due to higher complexity invested on the involved tasks at all age groups and particularly on elderly people. Therefore, tools such as the Mini-Mental State Examination (MMSE) or Montreal Cognitive Assessment (MOCA) could complement ADL evaluation. These tests are highly cost-effective since they are easy to implement by healthcare professionals in all clinical settings. Other relevant variables to keep in mind for a more complete patient assessment when COVID-19 survivors are comprehensively evaluated is respiratory function, quality of life and physical capacity (Torres-Castro et al., 2021). In particular, we must have a special focus on respiratory function, due to COVID-19 being mainly a respiratory disease and which has reported sequelae in lung function after infection (Torres-Castro et al., 2020). We consider that these should be evaluated more exhaustively in further long-term studies.

Patients are out of their normal context during hospitalization, thus some ADL evaluations, especially IADL, must be adapted or omitted according to the nature of the assessment. Therefore, ADL evaluations should be completed once the patient is discharged and returns to his context so that the impact of COVID-19 after the acute phase can be more accurately estimated. In addition to the context, it becomes cardinal to select the adequate assessment tool to obtain accurate results (Bai et al., 2020).

In the literature we find a specific tool to assess functional status, the Post-COVID-19 Functional Status (PCFS) Scale, which could be used for patient follow-up (Klok et al., 2020). This scale assesses the patient’s functional capacity upon discharge from the hospital and allows functional sequelae to be assessed. Although it does not strictly correspond to a test that evaluates ADL, its information can be useful to complement the functional evaluation of these patients, especially since studies have already appeared that have validated it in post-COVID-19 patients (Lorca et al., 2020).

In this work, we identified six tests used to explore the independence level during ADL to estimate the impact in functional ability and the relationship with prognostic factors and mortality associated with COVID-19. A key aspect of integrating information from different studies requires knowing the features of the sample related to intrinsic capacity and context described in each study. Thus, in the studies, including subjects with severe intrinsic capacity impairment, most of the tests are suitable for exploring the independence level during ADL, independently of the context. A different scenario could be observed in studies conducted in a rehabilitation program context (leading to selection bias). Liu et al. (2020) reported a higher level of independent functional ability post COVID-19 in comparison with samples coming from studies conducted in other different contexts not related to rehabilitation. In this way, the combination of intrinsic capacity and context gives us valuable information which can be used to select the appropriate assessment tool.

Until specific instruments that assess ADL appear and are validated in post-COVID-19 patients, we recommend the use of tools designed for this purpose, such as the BI, the FIM or the ADL Score. Although the FIM did not show any discrimination in the effect of an intervention (Liu et al., 2020), we recommend that future studies use the FIM in a population with greater functional compromise. Further studies, including a deeper analysis of ADLs, both BADL and IADL, are required to explore the impact of COVID-19 on functional performance among adults, especially on elderly people.

Limitations and strengths

The limitations of this review were mostly related to the heterogeneity of the studies, including patient characteristics, study design, methods, and assessment timeframe; in most of the studies there is no detailed description of demographic and clinical features of the sample. Also, we reviewed a small number of studies which only show short-term outcomes (long-term are unknown), and ADL/functional performance were not the main aim of the studies. Another important limitation is that our results are based on articles where the mean age of people was over 60 years old, so the findings cannot be extrapolated to people under that age. The main strength is its focus on the assessment of ADL as an essential outcome to optimize future rehabilitation protocols and resource allocation. Furthermore, it provides an overview of various assessment tools that have been used to explore ADL among COVID-19 patients.

Conclusions

All included studies found a reduction of ADL despite the test or scale used, revealing a vital worsening of physical function, deterioration of ADL performance and consequently loss of independence of COVID-19 patients after the acute phase of infection. Besides, the functional ability status previous to COVID-19 is vital for predicting the severity of the disease and mortality. Considering the results shown in our work, the choice of the best test to assess ADL is closely related to the intrinsic capacity and the subjects’ context. Thus, Barthel Index and ADL score were the most used assessments across subjects with different levels of intrinsic capacity and contexts.

Assessment of activities of daily living provide relevant information regarding the functional impact on COVID-19 patients, which contributes towards identifying the rehabilitation needs of this population and could inform the allocation of resources to support their recovery.

Well-designed studies should research short and long-term functional impact on post-COVID-19 patients with the objective of optimizing intervention strategies and supporting decision-making in clinical practice. It is also essential to further test the psychometric properties of these tests on COVID-19 patients. The data collected to date in this systematic review is a useful starting point for further studies.

Supplemental Information