The investigators are investigating if body composition measures - in particular muscle mass, body fat, and something called the phase angle - are able to reliably predict functional outcomes for patients who are in an acute hospital for inpatient rehabilitation. The investigators also want to see if age and frailty affect these relationships.

Patients undergoing inpatient rehabilitation in our hospital will be monitored through a variety of frailty-specific tools, including strength and speed. The investigators will use a device called a body composition monitor (BCM) to record their body composition measurements. These will then be compared against their usual rehabilitation outcomes.

The investigators believe that there is a correlation between these measures and a patient's functional outcomes.

STUDY DESIGN This study is conceived as a prospective longitudinal follow-up cohort study tracking patients from admission (to the inpatient rehabilitation unit) till discharge. Ethics approval has been obtained. Trial registration will be performed. Reporting will conform to the STROBE guidelines.

STUDY POPULATION Consecutive purposive sample of all patients with stroke, deconditioning, lung disease, or amputation admitted to the CGH Rehabilitation Medicine unit.

CONDUCT OF STUDY The study will take place in the main inpatient ward of the CGH Rehabilitation Medicine unit. All patients transferred in for inpatient rehabilitation will undergo case note screening and physical examination as part of their routine post-transfer clerking. Patients fulfilling the inclusion and exclusion criteria will be approached for recruitment and enrolment within 48 work hours of transfer in to the unit. Baseline measurements will be collected, and they will undergo usual rehabilitation with the hospital's multidisciplinary team. Repeat measurements will also be collected upon discharge.

DATA COLLECTION

Patient data will be collected upon consent, including:

1. Demographic data

   1. Age, ethnicity, and gender
   2. Handedness
   3. Occupation and employment status
   4. Highest education level
   5. Living arrangements and housing type

2. Anthropometric data

   1. Height and body mass index (BMI)
   2. BIA measures of appendicular skeletal mass (ASM), phase angle (PhA), and percentage body fat (PBF) - measured twice and at least 15mins apart

3. Functional data

   1. Premorbid mobility
   2. Activities of daily living by the Katz Index
   3. Functional Independence Measure (FIM)

4. Clinical data

   1. Diagnosis group
   2. Charlson Comorbidity Index
   3. Intrinsic capacity measures per the Integrated Care for Older People screening guidance (ICOPE)
   4. Sarcopenia measured by the SARC-CalF tool
   5. Frailty measured by CFS and the FRAIL questionnaire
   6. Cognition by the Chinese version (28 marks) of the Mini-Mental State Examination (MMSE)
   7. Mood by the Hospital Anxiety and Depression Scale (HADS)
   8. Nutrition by the Mini Nutritional Assessment Short-Form (MNA-SF)
   9. Quality of Life by the 5-item EuroQOL 5-dimension questionnaire (EQ-5D-5L)
   10. Handgrip strength - measured twice
   11. Average medial gastrocnemius thickness by ultrasonography
   12. Routine blood test results (if done prior) of white cell count, C-reactive protein, albumin, iron saturation, and Vitamin D

5. Physical performance data

   1. 10-metre walk test
   2. 5-times sit-to-stand test

On discharge, the following will be collected:

1. Anthropometric data

   1. Updated body mass index (BMI)
   2. BIA measures of ASM, PhA, and PBF

2. Clinical data

   1. Mood by the Hospital Anxiety and Depression Scale (HADS)
   2. Quality of Life by the 5-item EuroQOL 5-dimension questionnaire (EQ-5D-5L)
   3. Handgrip strength

3. Physical performance data

   1. 10-metre walk test
   2. 5-times sit-to-stand test

4. Administrative data a. Length of stay (by total duration and segmental stays in the acute and rehabilitation settings respectively) b. Confounders that could have affected rehabilitation adherence or participation (occurrence during stay of: i. Acute transient or irreversible cognitive dysfunction; ii. Mood dysfunction; iii. Moderate/severe pain; and iv. Low patient self-efficacy). All tools and questionnaires used are validated and included in many clinical guidelines both local and international.

SUBGROUP ANALYSIS

The investigators will perform subgroup analysis along 3 phenotypes:

1. Admitting diagnosis group: "strokes", "deconditioning", "pulmonary", and "lower limb amputee"
2. Frailty status: "frail", "at-risk", and "non-frail" patients as defined by CFS scores of ≥5, 4, and ≤3 respectively
3. Age: "<60yrs" and "≥60yrs"

STATISTICAL AND ANALYTICAL PLANS Descriptive statistics of patient demographic and clinical characteristics will be reported as number and percent for categorical data, mean ±SD for normally distributed data, and median and interquartile range (IQR) for non-normally distributed data. Intergroup comparisons will be made using the χ²/Fisher's exact test for categorical outcomes, unpaired t-test for normally distributed outcomes, and Mann-Whitney U-test for non-normally distributed outcomes. Multiple linear regression will be used to identify independent predictors of M-FIM at discharge, M-FIMeffectiveness, and handgrip strength. The assumption of linearity between the outcome and continuous predictors will be assessed using the Box-Tidwell test. Where there is evidence of non-linearity, the investigators will use a restricted cubic spline function with three knots at the 10th, 50th, and 90th percentiles, to model the relationship between continuous predictor and the outcome. Assumptions of homoscedasticity and normality of errors will be assessed using residual and Q-Q plots respectively. The investigators will check for multicollinearity using the variance inflation factor (VIF). Beta coefficients with corresponding 95% confidence intervals (CIs) will be reported. Statistical tests will be two-sided with a 0.05 significance level. All statistical analyses will be conducted using IBM SPSS Statistics version 23.0 (IBM Corp, Armonk, NY, USA). No interim analysis is necessary as the study is not conceptualised as a randomised controlled trial.

WITHDRAWAL FROM THE STUDY Subjects may withdraw voluntarily from participation in the study at any time, including after discharge, with no prejudice to their clinical care or outpatient follow-up. Their data will be removed from the electronic records, and the hardcopy data collection form will be physically destroyed. Their withdrawal will be reported in the STROBE flowchart subsequently.

DISCONTINUATION OF THE STUDY The study may be discontinued for any of the following reasons.

1. Adverse event or serious adverse event related to data collection or study conduct, which implicates safety of subsequent recruits
2. Breakdown of the BIA machine with no cost-effective repair or replacement option available
3. Other unforeseen circumstances In the event of study discontinuation, the investigators will inform the CIRB accordingly and the study team will evaluate the incomplete data to determine if there is sufficient scholarly merit remaining to justify write-up and publication.

SAFETY MONITORING PLAN Patients will be under the care of a trained study team member during clinical measurements, who will be expected to safely operate the different measuring devices (dynamometer and BIA machine). They will be monitored for pain and discomfort during use of the devices. Other clinical deteriorations will be managed routinely. Data will be reviewed on a monthly basis by an appointed co-investigator for integrity as well as completion. This includes screening and recruitment figures, as well as demographic, anthropometric, and clinical measurements, as well as functional outcomes. Data safety will be through lock-and-key storage of physical documents as well as access-restricted, password-protection for electronic data, which can only be viewed on the hospital's intranet-facing devices.

DATA ENTRY AND STORAGE Demographic data will be collected from patients' electronic medical records. Anthropometric and clinical measurements will be collected physically. All records will be recorded in a hardcopy data collection form to be filed in the Investigator Site File (ISF) that is stored under lock-and-key in the Research Office. All data will be transcribed electronically into REDCap which is hosted on the hospital's intranet and is access-restricted as well as password-protected. Only study team members (including the clinical research co-ordinator) will have access to the data. Data will be de-identified prior to analysis by the statistician. All data will be kept for 7 years post-completion or post-publication, whichever is later. Following that, electronic data will be deleted and all paper records will be physically destroyed.

DATA QUALITY ASSURANCE All data will be collected by a small pre-defined group of trained medical doctors and clinical research co-ordinators with medical backgrounds. All study team members will be pre-trained and briefed on the use of the clinical measurement devices. Anthropometric measurements of height, weight, and BMI will be collected by the inpatient ward nurses during patients' hospitalisation. FIM is documented on a weekly basis by the inpatient clinical teams after discussion with the multidisciplinary team. All study team members will have previous training and experience in accessing the patients' electronic medical records

ETHICAL CONSIDERATIONS This study will be conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki and that are consistent with the Good Clinical Practice and the applicable regulatory requirements. The principal investigator is responsible for informing the CIRB of any amendments to the protocol or other study-related documents, as per local requirement.

INFORMED CONSENT Patients will be referred by the primary team. Eligible patients would then be approached for consent while they are inpatient, within 48 work hours of transfer in to the unit. All study team members would be trained and hence eligible for consent-taking. Consent will be taken physically and in writing. Risks and benefits of participation will be clearly explained. Patients will be given ample time to make a decision and/or decline with no impact on their standard of clinical care. If a family member's presence is requested or deemed necessary but they cannot be present at the bedside, the investigators will consider remote consent-taking per the cluster guidance document on "Guidance on Remote Consent Process and Documentation for Research. A witness will not be required for all consent-taking as the study does not involve invasive procedures. Upon enrolment into the study, a note will be entered into the patient's electronic medical record. For non-English speakers, the investigators will use appropriate translation (in Mandarin, Malay, or Tamil) with the assistance of a translator where necessary. A witness will be present. Patients with cognitive impairment may be recruited as this may be a sequela of their acute stroke, or they may have underlying impairments. The study team member assessing for cognitive impairment would have been satisfactorily trained prior. They will be treated as vulnerable population if unable to participate in the consent-taking and research process, as per the Mental Capacity Act (2008). The investigators do not foresee the recruitment of other special populations such as children, prisoners, or other vulnerable individuals. In obtaining and documenting informed consent, the investigator should comply with the GCP guidelines and to the ethical principles that have their origin in the Declaration of Helsinki.

Inclusion Criteria:

1. Transferred under the Rehabilitation Medicine service for post-acute rehabilitation

2. Rehabilitation diagnosis of:

   1. Stroke with motor weakness (modified Rankin scale 1-5), or
   2. Deconditioning (from sepsis, post-ICU admission, heart failure, falls, musculoskeletal conditions with or without pain, or other conditions), or
   3. Pulmonary rehabilitation (with or with long-term oxygen therapy), or
   4. Lower limb amputees (transtibial or transfemoral)

3. Medically stable

4. Able to understand English consent form or suitably-translated document, or have a family member/legally-authorised representative who is able to do so and willing to provide consent

Exclusion Criteria:

1. Pacemaker or implantable defibrillator in-situ
2. Medical devices or surgical implants that will interfere with BIA measurement per the manufacturer's guidelines
3. Abnormal fluid status (overloaded or depleted) at the point of screening
4. Cognitive or physical inability to tolerate sitting/lying still for 2mins (for BIA measurement)

All patients over the age of 21 years admitted to the Changi General Hospital Rehabilitation Medicine unit in Singapore, up to a target sample size of 345 patients. There are no race-based subject restrictions. There will not be any gender-based or minority-based exclusion.

Our patients come from other departments such as the Acute Stroke Unit and General Surgery, and have various medical backgrounds as well as medical conditions.

Data is protected under the SingHealth Data Protection Policy which is compliant to the Singapore Personal Data Protection Act 2012