Pan-European Study on Geriatric Rehabilitation After COVID-19 Disease

INTRODUCTION AND RATIONALE The COVID-19 pandemic has hit the world hard, and older and vulnerable persons are more likely to get very ill when infected, and have the highest morbidity and mortality. The majority of patients that are admitted to the hospital are older (>70 years), and some of them have been, or are being, admitted to Intensive Care Units. Many patients stay on the ICU/hospital for weeks. To get back to the highest state of functional independence, these patients will need specialised care and treatment, therefore geriatric rehabilitation will be very likely required.

Geriatric Rehabilitation (GR) is defined as a multidimensional approach of diagnostic and therapeutic interventions, the purpose of which is to optimise functional capacity, promote activity and preserve functional reserve and social participation "in older people with disabling impairments." The goal of geriatric rehabilitation has been defined as "to assist [older people] to manage personal activities of daily living without the assistance of another person. If this is not possible, the goal is to minimize the need for external assistance through the use of adaptive techniques and equipment." Thereby focussing on the amelioration of dependency in activities of daily living. However, we are still in the dark about the amount of functional recovery of geriatric post-COVID-19 patients. We do know from the first lessons from Italy, that the rehabilitation process is difficult, and the course is capricious.

Before being able to assure the best outcomes for older post-COVID geriatric rehabilitation patients, we first have to be informed about what the clinical characteristics are, what treatment is provided, and what outcomes geriatric post-COVID-19 patients have in geriatric rehabilitation. Therefore, members of the Special Interest Group Geriatric Rehabilitation of the European Geriatric Medical Society (EuGMS) decided to conduct this multi-center observational cohort study into usual care data on the course of functional and clinical outcomes in post-COVID-19 geriatric rehabilitation patients in several European countries.

OBJECTIVES The primary objective of this study to get insight into the course of functional and medical recovery in (geriatric) rehabilitation patients affected by COVID-19 in Europe. The secondary objective is to get insight into the treatment modalities employed and the organisation of geriatric rehabilitation that post-COVID patients in Europe receive.

The study was conducted according to the principles of the Declaration of Helsinki (2013 version) and in accordance with the General Data Protection Regulation (GDPR) and in full conformity to any applicable state or local regulations in the participating countries.

RECRUITMENT In principle, patients admitted for geriatric rehabilitation due to COVID-19 morbidity will be informed by letter and asked if they have objections that their regular care data will be anonymously used for research purposes to improve practice in geriatric rehabilitation. If they do, the data will not be used (opt-out procedure). In some countries this will be conducted under the guidelines for service evaluation and audit, as it uses routine data, held entirely by the clinical care team. Where this is not possible because of local legislation or guidelines, informed consent will be obtained (opt-in procedure). During one-year (October 2020 - November 2021), patients were enrolled in the study.

Participants will receive standard rehabilitation treatment according to the discretion of clinical teams, which will likely be adapted to the specific needs of post-COVID-19 patients. This treatment includes physical therapy, occupational therapy, and medical treatment by an advanced nursing practitioner, geriatrician or medical specialist. This study will not influence any therapy provided or decisions about the medical treatment or treatment programs already used in the participating geriatric rehabilitation setting.

This study is observational and it is not anticipated that the observations will influence therapy or treatment, particularly since such indices, questionnaires and measurements are a regular part of the rehabilitation program.

STUDY MANAGEMENT For each participating geriatric rehabilitation facility in Europe one instructed local care professional will complete an online CASTOR database in which anonymous data from the clinical records of the included patients will be collected at admission, discharge and 6 weeks and 6 months follow-up. For countries with multiple participating sites, a country coordinator was appointed. Most country coordinators were the members of the Special Interest Group Geriatric Rehabilitation of the EuGMS.

DATA COLLECTION Variables collected

• Demographic factors: In all participants, selected demographic factors such as year of birth/age, gender, and residency premorbid will be documented.
• Information from hospital / general practitioner: hospital length of stay (days), ICU length of stay (days), complications during hospital stay or at home (Thromboembolism, Delirium, Pressure Ulcer), and COVID-19 diagnosis confirmed (PCR, Serology, clinically)
• At admission in geriatric rehabilitation: In all participants, only already routinely collected data will be documented. Such as ADL (Barthel index), quality of life (EQ-5D-5L), frailty (Clinical Frailty Scale), body mass index (BMI), nutritional status (MNA/NRS/MUST/SNAQ65/..), comorbidity (wFCI (0-36 pts)), cognition (MMSE/Demtect/ MOCA)) , delirium (DOS/SQiD), mood/anxiety (/GDS/HADS), pressure ulcer (yes/no), fatigue (VAS-scale), dyspnoea (VAS-scale), pain (NRS-P), mobility/balance (Timed Up and Go), muscle strength (MRC biceps /quadriceps), and speech/swallowing problems (yes/no)
• At discharge from geriatric rehabilitation: At discharge information about: hospital readmissions, length of stay in GR, discharge destination, ADL, frailty, BMI, nutrition, cognition, mood/anxiety, fatigue, dyspnoea, pain, mobility/balance, muscle strength, quality of life and complications during GR stay (Thromboembolism, Delirium, Pressure Ulcer) will be documented. At discharge we will additionally collect information about which professionals have been involved during geriatric rehabilitation an which treatment interventions have been applied.
• Follow-up after discharge: At 6 weeks and 6 months follow-up, and if desirable at flexible time points following regular care in the participating countries, we will document: ADL, Quality of life, post-traumatic stress syndrome.
• Qualitative data about countries' GR care services and triage systems: country coordinators in the study filled in a questionnaire about in what facility types GR for post-COVID-19 patients is provided (e.g. nursing homes, skilled nursing facility, acute care hospital ward, or on outpatient basis), which person refers the post-COVID-19 patients to GR (e.g. physician in the hospital, or GP), at which moment the referral takes place, which patient characteristics the referral is based on (e.g. standardized age cut-offs, frailty, multimorbidity, functional status, or psychosocial needs), which formal assessment is used for GR referral decisions, additional selection criteria by health insurance/policies, and criteria that need to be met by post-COVID patients for discharge from GR.

Data checks Data checks were performed during data collection to prepare for the steering board meetings that were held every two to three months with the country coordinators. In order to stimulate the achievement of the sample size, numbers of patients included were discussed during these meeting. Also structural missing or other remarkable things in the data were discussed during these meetings. Local care professionals were made aware of other missings or outliers in the data via email and asked to check and if possible complete or correct these data.

Sample Size calculation The primary outcomes are ADL functioning (Barthel index or derived from the USER or FIM) and quality of life (EQ-5D-5L). We performed a power calculation for a paired sample t-test on the primary outcome ADL-functioning (Barthel index). We assume a minimal clinically important difference of 2 points on the Barthel index as relevant, this is the mean difference between Barthel score at admission and Barthel score at discharge. The study would require a sample size of 52 (number of pairs) to achieve a power of 80% and a level of significance of 5% (two sided), for detecting a mean of the differences of 2 between pairs, assuming the standard deviation of the differences to be 5. We wanted to include a minimum of 52 patients per country which was agreed by the special interest group SIG EuGMS as being appropriate to generate the data we require. We aimed for a minimum of 250 participants in total.

STATSISTICAL ANALYSIS Different techniques will be used to analyze the data: Descriptive statistics will be used to give an overview of characteristics of the participants in the participating countries for all primary and secondary outcomes.

• For cross-country comparisons, we will perform a paired sample student t-test will be used to test the difference between ADL functioning at admission and follow-up per participating country, multilevel models will be built to model the recovery (in various outcome measures) of patients over time per country, supplemented with qualitative data about the countries' GR care services and triage systems.
• The course of recovery of post-COVID-19 patients over time (in various outcome measures) will be modelled with multilevel unconditional growth models. These multilevel models take into account that measurements are nested in patients and that patients are nested in countries. In doing so, the model with the best fitting shape (e.g. linear or quadratic) to our data will reveal the amount and speed of recovery.
• The impact of frailty on this course of recovery will be modelled, by adding the variable for frailty measured by the clinical frailty scale (CFS) to the models for the course of recovery.
• The impact of experiencing delirium during or prior to GR on this course of recovery will be modelled, by adding the variable delirium (yes/no) to the models for the course of recovery.
• (Potential) confounders of the impact of frailty on course of recovery and of the impact of delirium on the course of recovery such as age, sex, frailty status, delirium, and hospital stay, will be added to these
• By using descriptive statistics it will be described which post-COVID-19 GR patients receive which therapies/interventions during GR, and/or which professionals are involved in their GR trajectory.

No imputation will be used for missing data; complete case analysis will be performed, under the assumption that the data is missing at random.

The level of significance will be set at p \< 0.05.