Het monitoren van de fysieke fitheid is belangrijk. Op deze wijze kan voor- of achteruitgang van de fysieke fitheid in de gaten worden gehouden. Ook gedurende gepersonaliseerde trainingsprogramma’s vóór, tijdens, of na de oncologische behandeling is het belangrijk en kan er beoordeeld worden of de beoogde doelen (kunnen) worden behaald. Het is echter niet duidelijk welke meetinstrumenten geschikt zijn om fysieke fitheid van patiënten met kanker te monitoren in de praktijk.

One Repetition Maximum

Handgrip dynamometry

Thirty-Second Chair Stand Test

Five-Time Sit to Stand Test

Stair Climbing Test

Timed Up and Go test

Six-Minute Walking Test

Incremental Shuttle Walking Test

Steep Ramp Test

Physician-based Assessment and Counceling for Exercise questionnaire

Fitmáx questionnaire

Veterans-Specific Activity Questionnaire

Duke Activity Status Index

Description of studies

Aabo (2021) examined the reliability and measurement error of the 30-second chair-stand test in a convenience sample of male outpatients with prostate cancer (n=60). Participants were excluded when they had opioid-demanding skeletal pain. All participants were on androgen deprivation therapy (median time since start: 1 year and 23 days, range: 3 months and 1 week to 8 years and 11 months) and were already familiar with the 30-second chair-stand test. The mean age was 70.8 years (SD: 6.6, range: 51-86). Comorbidities in the sample were cardiovascular disease (n=16), diabetes (n=5), hypertension (n=29), dyslipidemia (n=16), and osteoporosis (n=3). Two or more comorbidities were present in 25 men. Additional therapies were provided in the form of chemotherapy or novel endocrine therapy to ten men with castration-resistant prostate cancer. Metastatic disease was prevalent in 38 men. The original manual was followed for conducting the 30-second chair-stand test, using a chair with 44cm seat height. Participants were instructed to sit in the middle of the chair with their backs straight and feet placed approximately at shoulder-with a little posterior to the knees. One foot was placed slightly in front of the other in order to maintain balance. Arms were crossed over the chest and the number of sit-to-stand repetitions within 30 seconds were recorded. If a participant was not able to stand up with crossed arms, he was allowed to use the armrests. The rater demonstrated one session of the test. The participant was given a practice trial before the actual test trial was performed. Two raters were used for the test procedures and calibrated test procedures prior to the test trials. A manual was used to standardize the same information and instructions to participants. Encouragements or talking was not allowed during the test trials. Test trials were performed twice under identical conditions on the same day by the same rater. The two test trials were separated by one hour of rest, where the participants were instructed not to be physically active. Intra-rater reliability and measurement error were reported.

Blackwood (2020) examined the reliability and measurement error of the Timed Up and Go test in 50 older breast (n=26), prostate (n=17), lung (n=5), and colorectal (n=1) cancer survivors. Patients were included when 65 years or older, were English speaking, had a medically confirmed diagnosis of breast / lung / prostate / colorectal cancer, had at least 3 months treatment completion for the primary cancer before testing, were able to get up from a chair and walk 50 feet with or without an assistive device. The mean age was 74.2 years (SD: 6.74), with 58% of the sample being female. Mean functional comorbidity index was 2.72 (range 0-7), with 17 participants having a fall history. The mean years since last active cancer treatment was 11.71 years (SD: 9.05). Participants received chemotherapy only (n=1), radiotherapy only (n=6), surgery only (n=19), chemotherapy and radiotherapy (n=5), chemotherapy and surgery (n=5), radiotherapy and surgery (n=5), chemotherapy and radiotherapy and surgery (n=5), or chemotherapy and surgery and hormone therapy (n=1). Treatment type was not reported for three participants. Participants were instructed to stand up from the chair (46cm seat height) and walk 3 meters to a line on the floor whereafter to turn around, walk back and sit on the chair again. Time needed to complete the task was recorded with a stopwatch. Two trials were performed and the average of those two trials was recorded while a two-week interval was used between test sessions.

Blackwood (2021) examined the reliability and measurement error of the 30-second Chair-Stand Test and Five-Time Sit To Stand test in older breast (n=28), prostate (n=15), lung (n=3), or colorectal (n=1) cancer survivors. Construct validity of the 30-second Chair-Stand Test, Timed Up and Go test and Five-Time Sit To Stand test were examined by correlating these measures. Participants were included when 65 years or older, were English speaking, had a medically confirmed diagnosis of breast / lung / prostate / colorectal cancer, had at least 3 months treatment completion for the primary cancer before testing, were able to get up from a chair and walk 50 feet with or without an assistive device. Participants were excluded when they reported a cancer recurrence or metastasis, had a history of chronic neurologic condition, had more than one cancer diagnosis, had an acute illness, or had an unstable medical condition. Sixty participants were recruited, however only forty-seven participants had complete data for all measures. Reasons were lost to follow-up, missing data at one or both time points, or not able to complete a test without upper extremity support. The mean age of the resulting 47 participants was 73.7 years (SD: 6.38, range: 65-89), with 66% being female. Participants received radiotherapy only (n=6), surgery only (n=14), hormonal treatment only (n=1), chemotherapy and radiotherapy (n=4), chemotherapy and surgery (n=5), radiotherapy and surgery (n=6), chemotherapy and radiotherapy and surgery (n=7), or chemotherapy and surgery and hormone therapy (n=1). Cancer stage at diagnosis was stage 0 (n=2), stage 1 (n=22), stage 2 (n=8), stage 3 (n=3), or unknown (n=12). For the 30-second chair-stand test, participants were instructed to be seated with arms crossed over their chest. When the test started, participants stood up and sat back down repeatedly. The number of repetitions within 30 seconds were recorded. Tests were performed by one rater. Participants performed one trial per session, with a two-week interval between sessions.

Booth (2001) examined the measurement error of the Shuttle Walking Test in patients with advanced cancer receiving palliative care. Patients had a WHO performance status of 1 or 2 and were recruited from the day center and hospice. Patients who could walk without assistance from another person were eligible for the study. Patients with cognitive impairment, short-term memory loss, or walking difficulties due to pain were excluded from the study. Initially, 71 patients were recruited but 30 declined due to impracticality (n=16), felt too frail (n=8), limited by pain or giddiness (n=5), tasked of saying ‘no’ [n=1]). Thereafter, 9 patients dropped out due to deterioration and unable to walk for a second trial (n=8) and declining further involvement (n=1). From the patients who completed all three tests, 10 participants were excluded from analysis because of experiencing pain (n=4), feeling giddy (n=1), lack of mobility (n=1), complications between trials (n=3), and baseline shortness of breath was very different between trials (n=1). Median age of the resulting 22 analyzed patients was 70 years (range: 31-84), where n=7 were males. Patients had a lung carcinoma (n=3), metastatic breast carcinoma (n=5), or other types of cancer (n=14). The Shuttle Walking test was performed on three different days. The first trial was a familiarization trial. Median interval between the second and third trial was 7 days (range: 1-35). Walking distance in meters was recorded per trial.

De Backer (2007) examined the reliability and criterion validity of the steep ramp test in patients with breast cancer (n=20), ovarian cancer (n=5), colorectal cancer (n=4), testicular cancer (n=2), non-Hodgkin’s lymphoma (n=5), and Hodgkin’s lymphoma (n=1) treated curatively. Patients were excluded when patients were not able to sit or lie down, had cognitive disorders or severe emotional instability, or had other serious illnesses limiting physical performance. Thirty-seven patients (n=10 male, n=27 female) were included, having a mean age of 48 years (SD: 11, range: 24-71). Patients received chemotherapy (n=37), chemotherapy and radiotherapy (n=1), chemotherapy and surgery (n=11), or chemotherapy and radiotherapy and surgery (n=20). A VO₂max test was performed on a cycle ergometer using an oximeter connected to a computer for breath-by-breath analysis. A ramp protocol was used in where the patient was suspected to reach their maximum load within 10 minutes. The ramp protocol was applied after 4 minutes of unloaded cycling. Patients were instructed and encouraged to cycle until exhaustion and with a pedal frequency between 70 and 80rpm. After the ramp protocol, the patients cycled unloaded until VO₂ values were back to baseline. The VO₂max test was repeated after 18 weeks of training. The Steep Ramp Test started with cycling at 25W for 30 seconds. Thereafter, the load was increased every 20 second by 25W until patients were exhausted. Patients were instructed to cycle with a frequency between 70-80rpm and the test was terminated when the pedal frequency fell below 60rpm. The Steep Ramp Test was performed before training, at week 5, week 9, week 13, and week 18. The test-retest reliability of the Steep Ramp Test was performed in a subset of the sample (n=23).

Eden (2018) examined the reliability of the Six Minute Walking Test and the 30-Seconds Chair-Stand test in patients with head and neck cancer. Construct validity was assessed by correlating the six-minute walking test to the 30-second chair stand test. Patients were included when they had surgery in the past three months for head and neck cancer or were currently undergoing radiotherapy, chemotherapy, or chemoradiotherapy for their head and neck cancer. Patients furthermore had to be between 18-85 years old, be community-dwelling during the study period, and have adequate proficiency in English. Patients were excluded when patients needed an interpreter, were unable or unwilling to return for the second trial within one week, or had comorbidities limiting the safe completion of trials. Sixty-six eligible patients consented, of which 1 patient was lost to follow-up and 23 patients were not included due to a diagnosis of skin or thyroid cancer. The mean age of the remaining 42 participants (81% male) was 63.1 years (SD: 9.3, range: 32.5-76.8) and had a mean BMI of 26.9 (SD: 5.4, range: 32.5-76.8). Tumor diagnosis was oral cavity (n=34), oropharynx or hypopharynx (n=3), nasal cavity or paranasal sinus (n=3), or parotid (n=2) cancer. Tumor stage was either T1 (n=14), T2 (n=10), T3 (n=9), or T4 (n=9). Therapy received was surgery for 40 patients (n=8 with radiation, n=6 with radiation and chemotherapy), radiotherapy only for one patient, and radiotherapy and chemotherapy only for one patient. Trials used for the reliability testing were performed on separate days (mean time between days: 4.3 days, range: 1-12). On the first day, two trials of the Six Minute Walking Test were performed with one hour of rest in between in a 22 meters long hallway. The first trial was a familiarization trial, where the second trial was used for reliability testing. The 30-Seconds Chair-Stand test was performed once on day one. Patients kept their arms crossed over their chest. On the second testing day, one trial for both the Six Minute Walking Test and 30-Seconds Chair-Stand test was performed.

England (2012) examined the construct validity of the Incremental Shuttle Walking Test by correlating the test to measures of inspiratory and peripheral muscle strength. Patients (n=41, of which 19 females) with incurable thoracic cancer and an ECOG-status of 0-2 were included. Patients were excluded when a palliative intervention to improve exercise capacity was appropriate, when they had ischaemic heart disease, had pain affecting their ability to walk, or if they had received chemotherapy or radiotherapy within the last four weeks. The dose of drugs potentially affecting the ability to exercises needed to be stable for at least a week. The mean age was 64 years (SD: 8) with diagnosis of non-small cell cancer (n=26), mesothelioma (n=11), or small cell cancer (n=4). ECOG-status was either 0 (n=16), 1 (n=21), or 2 (n=4). Cancer was either local (n=21) or advanced (n=20), where patients had metastases in the lymph nodes (n=12), lung (n=6), bones (n=4), liver (n=2), and adrenal metastases (n=1). Thirty-seven patients received palliative treatments previously in the form of chemotherapy (n=26), radiotherapy (n=10), or radical surgery (n=1). Smoking status was ‘never’ (n=7), ‘current’ (n=7), or ‘ex-smoker’ (n=27). Comorbidities observed in the sample were COPD (n=5), osteoarthritis (n=3), hypertension (n=4), diabetes (n=4), and a recent pulmonary embolism (n=1). Patients performed a sniff nasal inspiratory pressure test for measuring the inspiratory muscle strength and the leg extensor power test for peripheral muscle strength. Five minutes of rest were taken between assessments. Patients listened to pre-recorded instructions for the Incremental Shuttle Walking Test prior to performing a familiarization trial. After the trial, patients were seated until their heart rate and breathlessness returned to normal and rested for an hour thereafter. A second trial was then performed.

Granger (2015) assessed the criterion validity of the Six-Minute Walking Test and Incremental Shuttle Walking Test. Construct validity was assessed by correlating the Six-Minute Walking Test to the Incremental Shuttle Walking Test. Participants were included when they had non-small cell lung cancer (histologically confirmed), received or were schedules to receive treatment in the past six months, and had an ECOG-status of 0-2. Patients were excluded when they were unable to give consent, had a cognitive disorder, had extensive siceral or skeletal metastases, had stage IV disease after treatment, had co-morbidities that prevented exercise, Insufficient proficiency of English, or had contraindications for performing a PET (as recommended by the American Thoracic Society). Included patients (n=20) had a mean age of 66.1 years (SD: 6.5), of which 8 were male (40%). Mean Body Mass Index was 28.5 (SD:4.0). In the sample, 7 patients (35%) had COPD and smoking status was ‘never’ (n=2, 10.5%), ‘current’ (n=2, 10.5%), ‘ex-smoker’ (n=15, 78.9%), or unknown (n=1). The lung cancer stage was either stage I (n=12), stage II, (n=5), or stage III (n=3). Adenocarcinomas (n=15), squamous cell carcinomas (n=2), and other types (n=3) were observed in the sample. Time of performing the study tests was pre-treatment (n=3), post-surgery only (n=12), or post-surgery and chemotherapy (n=5). ECOG-status was either 0 (n=11), 1 (n=8), or unknown (n=1). A CPET was performed on a cycle ergometer using an incremental protocol. The workload increased every minute until symptomatic limitations. Breathlessness and heart rate data were used to determine whether the test was maximal.

The Six-Minute Walking Test was performed in a 30-meters long corridor according to the American Thoracic Society’s and European Respiratory Society’s recommendations. The Incremental Shuttle Walking Test was performed around a 10-meters shuttle course while the walking pace increased every minute. Walking pace started at 0.5 m/sec. and the test terminated when the participant was no longer able to maintain the pace. Two tests for both the Six-Minute Walking Test and the Incremental Shuttle Walking Test were performed to account for learning effects and the highest result was used in the analyses. Assessments were performed on each single visit. Patients returned for a second test visit within seven days. Patients were stable and did not receive medical or exercise interventions in-between visits. At least 15 minutes of rest were taken between tests on a single visit, while the test order was randomized.

Koegelenberg (2007) examined the criterion validity by correlating the Stair Climb Test with a cycle ergonometry CPET. Patients were included when 18 years or older, were scheduled for a lung resection, were under optimal medical therapy, had an FEV1 under 80% of the predicted FEV1, were physically able to perform cycle ergometry and stair climbing, and could provide written consent. Patents were excluded when a lung volume reduction surgery was planned, had serious cardiopulmonary disease, or were unable to exercise due to musculoskeletal or lower limb pathology. Patients (n=44, of which 13 females) had a mean age of 47.6 years (SD: 12.5). Indication for surgery was either bronchiectasis (n=27), non-small cell bronchial carcinoma (n=13), aspergilloma (n=2), or other (n=2). COPD was present in 24 patients. Cycle ergometry was performed on the same day as the Stair Climb Test with 2 hours of rest in-between. Cycle ergometry was performed following the American Thoracic Society guidelines. A 2-minute warm-up period at 0W was followed by the test start at 20W, using increases of 20W every minute. When a clear plateau for VO₂max was not observed, the observed VO₂peak was used. For the Stair Climb Test, patients needed to climb a stair as fast and as high as possible. The test was terminated when patients stopped for more than three seconds or reached the height of 20 meters. Use of the handrail was allowed to maintain balance, but not for elevation of themselves.

Li (2018) examined the criterion validity of the Duke Activity Status Index in 43 patients with cancer. Patients were included when scheduled for a major cancer surgery, when referred to the CPET service during pre-operative work-up, and who had a concurrent Duke Activity Status Index questionnaire administered. Patients were excluded when unable to perform the test due to pain, had neurological deficits, or had severe cognitive deficits. Cases with missing data were removed from analysis. The sample contained 25 males and had a median age of 63 years (IQR: 18). Fifteen participants had received chemotherapy within the last 6 months. Median Body Mass Index in the sample was 25.92 (IQR: 8.07). The CPET was performed following the American Thoracic Society/American College of Chest Physicians’ guidelines. Three minutes of unloaded cycling was used before a ramp protocol was initiated with 20W increases every minute at 60-70RPM until peak exercise. Thereafter, patients pedalled unloaded for 5 minutes as a recovery period. Tests were terminated at peak exercise due to fatigue, dyspnoea, chest pain, leg pain, signals of myocardial ischemia, hypotension, or arrhythmia at the patient’s or clinician’s discretion. The Duke Activity Status Index was used as a self-administered questionnaire, part of routine pre-operative assessments. The questionnaire was immediately administered before CPET testing.

Rogers (2017) examined the construct validity of handgrip dynamometry by correlating the dynamometry to a One-Repetition Maximum barbell bench press. Patients were included when they were a female breast cancer survivor between 1-15 years after diagnosis, were free from cancer, had ≥1 lymph nodes removed, had no medical conditions or medications that contraindicated an exercise program, had a BMI ≤50, had no future plans for surgery during the study period, had no bilateral lymph node removal, did not perform weight lifting in the past year, and had a stable body weight while not trying to lose weight. The sample contained 295 participants with a mean age of 55.9 years (SD: 8.8) and a mean Body Mass Index of 29.2 (SD: 6.1). A large proportion (86%) of the sample was post-menopausal. Disease stage at diagnosis was predominantly stage I (45%) or stage III (31%), while the mean time since diagnosis was 60.8 (SD: 39.3) months. About half of the sample had lymphedema (48%) and breast cancer on their dominant side (50%). Therapies received in the same were chemotherapy (n=224) and/or radiotherapy (n=229). Hand grip strength was assessed with the participants seated having their elbows at 90 degrees flexion. Three maximal contractions in each hand were performed, alternatingly, with a 1-minute rest in-between. A warm-up session for the Weight for the One-Repetition Maximum barbell bench press performed with 4 to 6 repetitions using a 2.6kg weight. Weight was progressively increased based on the participant’s self-reported rating of difficulty until the participant indicated a maximal effort, was unable to lift with proper biomechanics, was unwilling or unable to attempt to lift more, or reported a problem that required termination of the test.

Schmidt (2013) examined the criterion validity, reliability and measurement error of the Six-Minute Walking Test. Patients were included when they had a histologically confirmed cancer (or recurrence), had ongoing or recently finished chemotherapy, radiotherapy, and/or hormone therapy, were between 18-75 years old, and had an EGOC status of 0-2. Patients were excluded when they had brain or bone metastases, had a hemoglobin concentration less than 8g/dl, had chronic infection, had significant respiratory or cardiac disease, had any medical condition that limited participation, or had orthopedic or neurologic conditions that influenced successful completion of exercise tests. Fifty patients were recruited (n=36 females) with a mean age of 57.4 (SD: 10.2) years. Mean BMI was 25.3 (SD: 4.2) and median time since diagnosis was 12 months. Most common cancer site was breast (38%), colorectal (26%), or prostate (10%). Disease status was either advanced (n=14) or there was a curative option (n=36). The Six-minute Walking Test was performed according to the American Thoracic Society’s guideline using a 30-meters course in a corridor. Patients were instructed to walk back and forth and stopped when six minutes elapsed. The test was repeated by a subsample (n=30) within 2 to 7 days after the first test, at the same time of day. CPET was performed on a cycle ergometer. The load was increased with 25W every 3 minutes until exhaustion, initially starting the test at 0W. Criteria for exhaustion were symptom limitations or the inability to keep pedaling at 60RPM.

Schumacher (2018) examined the construct validity of the Six-Minute Walking test in cancer survivors. Patients were included with a wide range of cancers. Patients were excluded when there was a history of congestive heart failure, had myocardial infarction, had a chronic lung disease (including asthma), had significant ambulatory problems, or had a history of hemoptysis, fainting, or epilepsy. The sample consisted of 187 patients (115 females) having a mean age of 61 years (SD: 13). During the study, 31% of the participants were undergoing (chemo)radiation. Participants completed a treadmill protocol of twenty-one on-eminute stages. Both speed and inclination could increase with each stage. Patients could terminate the test at any time but were encouraged to perform maximally. Tests were terminated when patients indicated that they reached their maximum performance or when they grabbed onto the handrails. Gas-analyses were not used. Instead, a prediction model was used to estimate the VO₂peak from the treadmill test. Patients furthermore completed a single Six-Minute Walking Test. The treadmill test and the Six-Minute Walking Test were performed one week apart in randomized order.

Sebio-Garcia (2017) examined the reliability and measurement error of the Six-Minute Walking Test in patients with cancer awaiting surgery. Patients were included when undergoing major oncological surgery, had a high risk for post-operative complications, and had severe deconditioning. Patients were excluded when they were unable to complete tests and questionnaires in a prehabilitation program, were scheduled for surgery within 3 weeks, or had severe musculoskeletal, neurological, or cognitive impairments. The sample consisted of 170 patients (114 males) with a mean age of 71.1 (SD 14.9) years and a mean Body Mass Index of 26.5 (SD: 4.8). ASA score was either I (n=69), II (n=84), III (n=13), or IV (n=4). Planned type surgery was colorectal (n=65), upper gastrointestinal (n=36), pancreatic (n=8), urologic (n=44), gynecolgic (n=7), cytoreductive (n=3), or other (n=7). Most patients did not receive neoadjuvant treatment (n=112), while others received chemotherapy (n=37), radiotherapy (n=17), or chemoradiotherapy (n=17). Only 2.9% of the sample scored a 0 on the Charlson Comorbidity index, the others ranged from 1 to 5 with the highest proportion scoring 1 (44.9%). The Six-Minute Walking Test was performed following the American Thoracic Society/European Respiratory Society’s guideline using a 30-meters course in a corridor. Patients received 30 minutes of rest before starting the second trial, assessed by the same rater.

Stuiver (2017) assessed the criterion validity of the Steep Ramp Test in cancer survivors. The study used patient data recorded in two randomized controlled trials. Data of 283 patients were used (68 males) with patients having a mean age of 53 years (SD: 11). Types of cancer in the sample were breast cancer (n=162), colon cancer (n=49), lymphomas (n=56), ovarian cancer (n=8), cervix cancer (n=4), and testis cancer (n=4). Patients received chemotherapy (n=283), radiotherapy (n=123), surgery (n=227), stem cell transplantation (n=32), immunotherapy (n=51), and/or hormone therapy (n=114). A CPET at baseline followed by the participants first Steep Ramp Test within 30 days after CPET while not starting exercise training in this interval were used for validation (median interval: 8 days, IQR: 6-10). Both tests were performed on a cycle ergometer. The CPET used an incremental protocol adjusted to each participant, aiming for a maximal performance within 8-12 minutes. Patients cycled between 60-80RPM and received encouragement. The test was terminated when exhausted or when unable to maintain 60-80RPM. The Steep Ramp Test had a four-minute warm-up period at 10W. The starting workload was 25W followed by an increase of 25W every 10 seconds. The Steep Ramp Test was terminated when the pedal frequency dropped below 60RPM.

Trutschnigg (2008) examined the reliability and measurement error of handgrip dynamometry. Patients aged between 18 and 36 years old with recently diagnosed (up to 6 months) advanced non-small cell lung cancer or gastrointestinal cancer were included. Patients (n=70, 27 females) had a mean age of 61.5 years (SD: 13.2) and a mean Body Mass Index of 24.4 (SD: 4.9). The Jamar dynamometer and the Biodex System 3 (with handgrip attachment) were used for measurements of handgrip strength. Participants performed one or two familiarization trials on both instruments. For both instruments, three consecutive maximal contractions (lasting 3 seconds) were performed for the dominant hand with a 1-minute break between instruments. Participants were tested twice and the mean of three trials was used as the outcome. Testing position of the participants was standardized. Patients had their elbow flexed at 90 degrees and the wrist at 0 degrees for the Biodex. Patients had their elbow flexed at 90 degrees and their arm rested on the armrest. The non-dominant arm rested neutrally and both feet were firmly on the ground at shoulder’s width.

Tsuji (2022) examined the criterion validity of the Six-Meter Walking Test. Construct validity of the One-Repetition Maximum (leg press) was assessed and the construct validity of the hand grip strength predicting the One-repetition maximum (along with other predictors) was assessed as well. Patients were included when they were females aged between 20 to 59 years at diagnosis, were diagnosed with stage I-IIa invasive breast cancer, were within 2-13 months after surgery, had no chemotherapy next to hormone or radio therapy, did not exercise more than 30 minutes on moderate intensity on two weekdays. Patients were excluded when exercise was considered too riskful, had a smoking history in the past year, had a Body Mass Index of ≥30, were judged unfit for participation for other reasons, or were administered beta-adrenergic blocking agents. The sample (n=50) had a mean age of 48 years (SD: 6) and a mean Body Mass Index of 21.0 (SD: 2.1). Breast cancer stage was either stage I (n=36, 72%) or stage IIa (n=14, 28%). The tumor was estrogen receptor positive (n=49), progesterone receptor positive (n=48), and/or HER2 positive (n=1). Forty-seven patients received hormone therapy and twenty-three patients received radiotherapy. Mean time since surgery was 11 months (SD: 22). A CPET was performed on a cycle ergometer using an incremental multistage protocol. The test commenced at 29.4W and was increased every minute by 14.7W. Patients performed the test until exhaustion. That is, when the pedal frequency dropped below 55RPM for the third time. The Six-Meter Walking Test was performed according to the American Thoracic Association on a 30-meters course in a corridor. Participants were instructed to walk as far as possible in their own pace for 6 minutes. The One-Repitition Maximum was performed on a leg press. Patients performed a 10 repetition warm-up at 20-30 kilograms and rested for 2-3 minutes thereafter. Weight was increased (10-20%) in single trials, with 1-5 minutes of rest in-between, until the One-Repetition Maximum was achieved. Grip strength was measured twice for both the left and right hand, with measures alternating between hands. The test was performed while standing up straight with the arms in neutral position.

Van Hinte (2020) examined the reliability and measurement error of handgrip dynamometry, the 30-second Chair-Stand Test, the Timed Up and Go test, and the Six-Minute Walking Test in survivors of head and neck cancer. Patients were included when they were survivors of head and neck cancer, had completed medical treatment, were 18 or over, and were able to walk unaided. Patients were excluded when not able to speak or understand Dutch, when receiving palliative care, or when being at risk when performing physical measurements. Fifty patients were recruited (22 females) with a mean age of 68.6 years (SD: 9.9) and a median Body Mass Index of 25.0 (IQR: 23.5-26.7). Smoking status was ‘current’ (n=4), ‘never’ (n=7), or ‘ex-smoker’ (n=39). Patients had a median of 3.0 years (IQR: 1.0-5.25) since cancer treatment. Tumor location was at the oral cavity (n=28), nasopharynx (n=1), oropharynx (n=2), larynx (n=12), or other (n=7). Patients received surgery (n=19), surgery and radiotherapy (n=18), radiotherapy (n=4), Surgery with radiotherapy and chemotherapy (n=7), or radiotherapy and chemotherapy (n=2). Twenty-eight patients received a neck dissection, performed unilateral (n=22) or bilateral (n=6). Time interval between two test trials were at least one hour with a maximum of two hours. A single test trial for the 30-second Chair-Stand test and the Six-Minute Walking Test both contained one measurement. The handgrip strength and Timed Up and Go test were performed three times each test trial, where the best score was used. Test and re-test assessments were performed by the same rater.

Weemaes (2021) assessed the criterion validity and responsiveness of the Steep Ramp Test in cancer survivors. Patients were included when they had completed active medical treatment, were suffering physically and psychosocially (as identified by a sports physician, psychologist, and occupational therapist), had completed a CPET and Steep Ramp Test before participation in the rehabilitation programme, and gave consent to use their usual care data. Patients were excluded when they were unable to cycle until exhaustion during the tests. The sample consisted of 106 patients (n=28 males) with a mean age of 56.6 years (SD: 11.0) and a mean Body Mass index of 27.5 (SD: 4.8). Cancer type in the sample was breast cancer (n=51), colorectal cancer (n=9), lung cancer (n=7), lymphomas (n=6), prostate cancer (n=4), or other types (n=29). Metastases were prevalent in a part of the sample: lymphatic (n=17), hepatic (n=5, skeletal (n=4), other (n=3). Patients had received surgery (n=80), chemotherapy (n=62), radiotherapy (n=55), hormone therapy (n=32), immunotherapy (n=11), and/or stem cell transplantation (n=4). Participants performed a CPET and a Steep Ramp Test (2 to 7 days apart) before the start of an exercise program and after 10 weeks of training. For the CPET, patients started with an unloaded 3-minute warm-up period. Increase in workload was adjusted to the patient, aiming at reaching a maximal effort between 8 to 12 minutes. Patients needed to keep pedaling at least at 60RPM until exhaustion. The test was terminated when the patient stopped or dropped under 60RPM. The Steep Ramp Test was performed on a cycle ergometer and patients started with a 3-minute warming-up at 25W. Thereafter, every 10 seconds the load increased with 25W.The test was terminated when the pedaling frequency dropped below 60RPM or when voluntary exhaustion was achieved by clinical signs of intense effort.

Willcock (2018) examined the measurement error of the Incremental Shuttle Walking Test in patients with thoracic cancer. Patients were included when they had incurable thoracic cancers and had an ECOG-status of 0-2 (reporting a limitations to undertake daily activities). Patients were excluded when they received chemotherapy or radiotherapy within the prior four weeks or when their symptoms were related to a palliative intervention. The sample (n=41) had a mean age of 64 (SD: 8) and a median ECOG-status of 1 (range: 0-2). Cancer types observed in the sample were non-small cell lung cancer (n=26), small cell lung cancer (n=4), and mesothelioma (n=11). The incremental shuttle walking test was performed two times at the same time of day on consecutive days. A 10-meters course was used and walking speed was externally paced. Patients were instructed to walk and not to run. Each minute the walking speed would increase until patients reached their symptom-limited maximum (by breathlessness and leg fatigue). Patients were advised to wear comfortable shoes and take their usual medication. On the first day of testing, a familiarization trial and a test trial was performed, with one hour of rest in-between. The second test day consisted of two trials again.

Win (2006) assessed the criterion validity of the Incremental Shuttle Walking Test in patients with operable lung cancer. Consecutive patients with potentially resectable lung cancer were recruited. Patient were excluded when they had unstable angina, had a myocardial infarction in the prior 6 weeks, or had disorders that physically influenced exercise performance. Patients (n=125) had a mean age of 68.8 years (SD: 7.7, range: 42-85). Thirty-three percent of the sample had an FEV1 under 1.5 liters (for lobectomy) and under 2.0 liters (for pneumonectomy). For the Incremental Shuttle Walking Test the participants walked back and forth on a 10-meter course. Walking speed was externally paced by a cassette playing signals. The test was terminated when the patient could no longer keep up with the required speed or became too breathless to continue. The CPET was performed on a treadmill using the Standardized Exponential Exercise Protocol. Every minute the workload increased by 15% using an increase in speed or inclination for a maximum of 20 minutes (including baseline and recovery). The test proceeded until patients were sign or symptom limited. The Incremental Shuttle Walking Test and treadmill CPET were performed on the same day with at least four hours in-between. Patients were fully familiarized with both tests before performing the tests.

Results

Validity of instruments for muscle strength

See Table Validity results regarding the tests of interest intending to measure muscle strength for a summary of the results found for the validity of the One-Repetition Maximum, handgrip dynamometry, Thirty-Second Chair-Stand Test, Five-Times Sit to Stand test, Timed-Up and Go test, Six-Minute Walking Test, and the Incremental Shuttle Walk Test to assess muscle strength.

No information was found for the Stair Climb Test, Steep Ramp Test, Physician-based Assessment and Counseling for Exercise questionnaire, Fitmáx, Veteran-Specific Activity Questionnaire, and Duke Activity Status Index, in the population of interest.

One repetition maximum (1RM)

Tsuji 2020 reported data about the construct validity (hypotheses testing) of the 1RM on a leg press.

Handgrip Dynamometry

Information about the construct validity (hypotheses testing) of handgrip dynamometry to assess muscle strength was reported by Rogers (2017) and Tsuji (2022). Rogers (2017) correlated several different measures of handgrip strength to the performance on a 1RM barbell bench press. Tsuji (2022) used a prediction equation including hand grip strength, among others, as predictors for the 1RM on a leg press.

Thirty-second Chair-Stand Test (30sCST)

Blackwood (2021) correlated the 30sCST with the Timed Up and Go test for construct validity (hypotheses testing).

Five-Times Sit-To-Stand test (5TSTS)

Blackwood (2021) correlated the 5TSTS with the Timed Up and Go test for construct validity (hypotheses testing).

Timed Up and Go test (TUG)

Blackwood (2021) had correlated both the 30sCST and 5TSTS test with the TUG test for construct validity (hypotheses testing).

Six-Minute Walking Test (6MWT)

Eden (2018) correlated the walking distance from the 6MWT to the number of repetitions from the 30sCST for construct validity (hypotheses testing).

Incremental shuttle walking test (ISWT)

England (2014) correlated the ISWT to the inspiratory muscle strength and peripheral muscle strength (leg extension power) for the information on the construct validity (hypotheses testing).

Table Validity results regarding the tests of interest intending to measure muscle strength

Validity of instruments for cardiorespiratory fitness

See Table Validity results regarding the tests of interest intending to measure cardiorespiratory fitness for a summary of the results found for the Thirty-Second Chair-Stand Test, Stair Climb Test, Six-Minute Walking Test, Incremental Shuttle Walk Test, and the Steep Ramp Test.

No information was found for the One-Repetition Maximum, handgrip dynamometry, Five-Times Sit to Stand test, Timed-Up and Go test, Physician-based Assessment and Counseling for Exercise questionnaire, Fitmáx questionnaire, and the Veteran-Specific Activity Questionnaire in the population of interest.

Thirty-second Chair-stand test (30sCST)

Blackwood (2021) and Eden (2018) examined the construct validity of the 30sCST by correlating the test to the Timed-up and Go test and Six-Minute Walking Test, respectively.

Stair climbing test (SCT)

Koegelenberg (2020) examined the criterion validity by correlating the average speed of ascend in the SCT to the VO₂peak as measured by a CPET.

Six-minute walking test (6MWT)

Granger (2015), Schmidt (2013), and Tsuji (2022) examined the criterion validity by correlating the walking distance to the VO₂peak as assessed by CPET. Tsuji (2022) also used the 6MWT walking distance as a predictor, among other predictors, in a prediction equation to estimate the VO₂peak. Eden (2018) and Granger (2015) also correlated the walking distance to the Thirty-Second Chair Stand Test and Incremental Shuttle Walking Test for the construct validity, respectively. Schumacher (2018) correlated in several prediction models, with walking distance as a predictor, the VO2peak which was estimated from a treadmill walking test also using a prediction equation to obtain the VO₂peak.

Incremental shuttle walking test (ISWT)

Granger (2015) and Win (2006) assessed the criterion validity of the ISWT by correlating the test with the VO₂peak obtained from a CPET. Granger (2015) furthermore assessed the construct validity (hypotheses testing) by correlating the ISWT with the Six-Minute Walking Test.

Steep ramp test (SRT)

De Backer (2007) and Weemaes (2021) correlated the Wmax from the Steep Ramp Test to the VO2max or VO2peak from a CPET, respectively. Stuiver (2017) used prediction equations to estimate the VO2peak obtained from a CPET, using the workload of the last SRT stage plus 2.5W (i.e. maximal short exercise capacity) as a predictor in these equations.

Duke activity status index (DASI)

Li (2018) used several prediction models to estimate the VO₂ obtained by CPET using the DASI score or specific DASI questions as predictors.

Table Validity results regarding the tests of interest intending to measure cardiorespiratory fitness

Reliability

Table Reliability results regarding the instruments of interest summarizes the extracted results for reliability of the instruments of interest. No data regarding the reliability of the one-repetition maximum, stair climbing test, incremental shuttle walking test, Physician-based Assessment and Counseling for Exercise questionnaire, Fitmáx questionnaire, Veterans-Specific Activity Questionnaire, or Duke Activity Status Index were found.

Handgrip dynamometry

Two studies assessed the within-day test-retest reliability of handgrip dynamometers (Trutschnigg, 2008; Van Hinte, 2020). One study reported reliability results for both the left and right hand separately (Van Hinte, 2020).

Thirty-second chair-stand test

Two studies assessed the within-day intra-rater test-retest reliability (Aabo, 2021; Van Hinte, 2020) and found ICCs of 0.97 and 0.92, respectively. Two other studies assessed the between-day test-retest reliability (Blackwood, 2021; Eden, 2018), resulting in relatively similar ICCs of 0.89 and 0.948, respectively.

Five-time sit to stand test

Only one study reported reliability results. The intra-rater between-day test-retest reliability had an ICC of 0.86 (Blackwood, 2021).

Timed up and go test

Two studies reported reliability results. One study reported the between-day reliability, showing an ICC of 0.95 (Blackwood, 2020). The other study reported an ICC of 0.98 for within-day reliability (Van Hinte, 2020).

Six-minute walking test

Between-day reliability was assessed in two studies (Eden, 2018; Schmidt, 2013), reporting ICCs of 0.96 and 0.93, respectively. Two other studies investigated the within-day reliability and found ICCs of 0.98 and 0.97 (Sebio-Garcia, 2017; Van Hinte, 2020).

Steep ramp test

One study (De Backer, 2007) assessed the reliability of the steep rap test (Wmax). An ICC of 0.996 was reported.

Table Reliability results regarding the instruments of interest

Measurement error

See Table Measurement error results regarding the tests of interest for a summary of the results regarding measurement error found for handgrip dynamometry, Thirty-Second Chair-Stand Test, Five-Times Sit to Stand test, Timed-Up and Go test, Six-Minute Walking Test, and the Incremental Shuttle Walk Test.

No information was found for the measurement error of the One-Repetition Maximum, Stair Climb Test, Steep Ramp Test, Physician-based Assessment and Counseling for Exercise questionnaire, Fitmáx questionnaire, and the Veteran-Specific Activity Questionnaire in the population of interest.

Handgrip dynamometry

Trutschnigg (2008) and Van Hinte (2020) reported data about the measurement error of handgrip dynamometry. Jamar (Truschnigg, 2008; Van Hinte, 2020) and the Biodex System 3 (Trutschnigg, 2018) was used.

Thirty-Second Chair-Stand Test (30sCST)

Aboo (2021), Blackwood (2021), and Van Hinte (2020) provided information about the measurement error. Minimal Detectable Differences were calculated for repetitions (range: 2.6 to 3).

Five-Times Sit to Stand test (5TSTS)

Blackwood (2021) calculated a Minimal Detectable Difference of 3.19 seconds.

Timed-Up and Go test (TUG)

Blackwood (2020) and van Hinte (2020) calculated the measurement error of the TUG test (Minimal Detectable Differences of 2.494 and 1.54, respectively)

Six-Minute Walking Test (6MWT)

Schmidt (2013), Sebio-Garcia (2017), and Van Hinte (2020) provided information about the measurement error. Van Hinte (2020) reported a Minimal Detectable Difference of 56.67 meters.

Incremental Shuttle Walk Test (ISWT)

Booth (2001) and Willcock (2018) reported data about the measurement error of the ISWT in Bland-Altman plots.

Table Measurement error results regarding the tests of interest

Responsiveness

Only one study reported data about the responsiveness of an instrument. Weemaes (2021) assessed the responsiveness of the Steep Ramp Test. Patients received a 10-week exercise intervention. The before-after change scores of the steep ramp test (Wmax) were correlated to the change scores on the CPET (r=0.51). Using a cutoff of ≥6% increase, the steep ramp test had an area under the curve (AUC) of 0.74 (95%CI: 0.60-0.87), a sensitivity of 70.7%, a specificity of 66.7%, a positive predictive value of 82.9%, and a negative predictive value of 50.5% (Weemaes, 2021).

Level of evidence of the literature

See Table Summary of GRADE assessments for each instrument per measurement property for a summary of the GRADING for each measurement instrument per measurement property of interest.

Table Summary of GRADE assessments for each instrument per measurement property

One Repetition Maximum

The level of evidence regarding the outcome measure construct validity (muscle strength) was downgraded by 4 levels because of study limitations (3 levels for risk of bias: only one study which was judged inadequate); number of included patients (1 level for imprecision: n=50 in sample); publication bias was not assessed.

The level of evidence regarding the outcome measures validity (cardiorespiratory fitness), measurement error, reliability, and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Handgrip dynamometry

The level of evidence regarding the outcome measure construct validity (kilograms for muscle strength) was downgraded by 3 levels because of study limitations (3 levels for risk of bias: only one study judged to have inadequate quality); publication bias was not assessed.

The level of evidence regarding the outcome measure construct validity (predicted kilograms for muscle strength) was downgraded by 4 levels because of study limitations (3 levels for risk of bias:); conflicting results (inconsistency); applicability (bias due to indirectness); number of included patients (1 level for imprecision: n=50 in body of evidence); publication bias was not assessed.

The level of evidence regarding the outcome measure reliability was downgraded by 2 levels because of study limitations (1 level for risk of bias: one study judged as doubtful and one as adequate); number of included patients (1 level for imprecision: between 50 and 100 patients in the total body of evidence); publication bias.

The level of evidence regarding the outcome measure measurement error was downgraded by 2 levels because of study limitations (1 level for risk of bias: only one study was judged as adequate); conflicting results (not downgraded for inconsistency based on mean differences); number of included patients (1 level for imprecision: sample size in the body of evidence was between 50 and 150); publication bias was not assessed.

The level of evidence regarding the outcome validity (cardiorespiratory fitness) and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Thirty-Second Chair Stand Test

The level of evidence regarding the outcome measure construct validity (muscle strength) was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only 1 study which was judged as doubtful); number of included patients (2 levels for imprecision: less than 50 participants in sample); publication bias was not assessed.

The level of evidence regarding the outcome measure construct validity (cardiorespiratory fitness) was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only one study which was judged as doubtful); number of included patients (2 levels for imprecision: less than n=50 in the total sample); publication bias was not assessed.

The level of evidence regarding the outcome measure reliability was downgraded by 1 level because of study limitations (1 level for risk of bias: three studies judged as doubtful and one as adequate); publication bias was not assessed.

The level of evidence regarding the outcome measure measurement error was downgraded by 1 level because of study limitations (1 level for risk of bias: one study judged as adequate, multiple as doubtful); publication bias was not assessed.

The level of evidence regarding the outcome responsiveness was not GRADEd. None of the included studies reported data about this measurement property.

Five-Time Sit to Stand Test

The level of evidence regarding the outcome measure construct validity was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only 1 study which was judged as doubtful); number of included patients (2 levels for imprecision: less than 50 participants in sample); publication bias was not assessed.

The level of evidence regarding the outcome measure reliability was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only one study which was judged as doubtful); number of included patients (2 levels for imprecision: less than n=50 in the body of evidence); publication bias was not assessed.

The level of evidence regarding the outcome measure measurement error was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only 1 study which was judged as doubtful); number of included patients (2 levels for imprecision: less than 50 participants in sample); publication bias was not assessed.

The level of evidence regarding the outcomes validity (cardiorespiratory fitness) and responsiveness were not GRADEd. None of the included studies reported data about this measurement properties.

Stair Climbing Test

The level of evidence regarding the outcome measure construct validity could not be GRADEd, since none of the included studies reported data about its validity to measure muscle strength.

The level of evidence regarding the outcome measure criterion validity (cardiorespiratory fitness) was downgraded by 2 levels number of included patients (2 levels for imprecision: less than n=50 in the total sample); publication bias was not assessed.

The level of evidence regarding the outcome reliability, measurement error and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Timed Up and Go test

The level of evidence regarding the outcome measure construct validity was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only 1 study which was judged as doubtful); number of included patients (2 levels for imprecision: less than 50 participants in sample); publication bias was not assessed.

The level of evidence regarding the outcome measure was not GRADEd. None of the included studies reported data about this measurement property.

The level of evidence regarding the outcome measure reliability was downgraded by 2 levels because of study limitations (1 level for risk of bias: one study judged as doubtful and one as adequate); conflicting results (inconsistency); applicability (bias due to indirectness); number of included patients (1 level for imprecision: between 50 and 100 patients in the total body of evidence); publication bias was not assessed.

The level of evidence regarding the outcome measure measurement error was downgraded by 2 levels because of study limitations (1 level for risk of bias: one study judged as adequate and one as doubtful); conflicting results (not downgraded for inconsistency: the observed difference may be related to differences in measurement protocols); number of included patients (1 level for imprecision: body of evidence contained n=100); publication bias was not assessed.

The level of evidence regarding the outcomes validity (cardiorespiratory fitness) and responsiveness reliability were not GRADEd. None of the included studies reported data about these measurement properties.

Six-Minute Walking Test

The level of evidence regarding the outcome measure construct validity (muscle strength) was downgraded by 3 levels because of study limitations (1 level for risk of bias: only one study judged as adequate); number of included patients (2 levels for imprecision: less than n=50 in the sample); publication bias was not assessed.

The level of evidence regarding the outcome measure criterion validity (walking distance for cardiorespiratory fitness) was downgraded by 1 level because of conflicting results (1 level for inconsistency: heterogeneous results might not (fully) be explained by differences between studies); publication bias was not assessed.

The level of evidence regarding the outcome measure construct validity (walking distance for cardiorespiratory fitness) was downgraded by 4 levels because of study limitations (3 levels for risk of bias: only one study which judged as inadequate); number of included patients (2 levels for imprecision: less than n=50 in the total body of evidence); publication bias was not assessed.

The level of evidence regarding the outcome measure construct validity (predicted VO₂peak for cardiorespiratory fitness) was downgraded by 2 levels because of study limitations (2 levels for risk of bias: only one study which judged as doubtful); conflicting results (not downgraded for inconsistency: different models with different predictors will probably result in different results); publication bias was not assessed.

The level of evidence regarding the outcome measure criterion validity (predicted VO₂peak for cardiorespiratory fitness) was downgraded by 2 levels because of study limitations (2 levels for risk of bias: the largest (ca. 3 times) study was of doubtful quality); conflicting results (not downgraded for inconsistency: different models with different predictors will probably result in different results); publication bias was not assessed.

The level of evidence regarding the outcome measure reliability was downgraded by 1 level because of study limitations (1 level for risk of bias: multiple studies judges as doubtful); publication bias was not assessed.

The level of evidence regarding the outcome measure measurement error was downgraded by 1 level because of study limitations (1 level for risk of bias: multiple studies were judged as doubtful); conflicting results (not downgraded for inconsistency: differences in measurement protocols may explain the differences, although the limits of agreement were deemed relatively comparable); publication bias was not assessed.

The level of evidence regarding the outcome responsiveness was not GRADEd. None of the included studies reported data about this measurement property.

Incremental Shuttle Walking Test

The level of evidence regarding the outcome measure construct validity (muscle strength) was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only one study, judged as doubtful); number of included patients (2 levels for imprecision: less than n=50 in the sample); publication bias.

The level of evidence regarding the outcome measure criterion validity (cardiorespiratory fitness) was not downgraded. No evidence of study limitations (risk of bias), conflicting results (inconsistency), applicability (bias due to indirectness), or limited number of included patients (imprecision) were found. Publication bias was not assessed.

The level of evidence regarding the outcome measure construct validity (cardiorespiratory fitness) was downgraded with 4 levels because of study limitations (2 levels for risk of bias: only one study which was judged as doubtful), conflicting results (inconsistency), applicability (bias due to indirectness), or limited number of included patients (2 levels for imprecision: less than n=50 in the total body of evidence) were found. Publication bias was not assessed.

The level of evidence regarding the outcome measure measurement error was downgraded by 2 levels because of study limitations (1 level for risk of bias: multiple studies judged as doubtful); conflicting results not downgraded for (inconsistency: some heterogeneity may be explained due to different sample characteristics [e.g. tumor location]); applicability (bias due to indirectness); number of included patients (1 level for imprecision: between 50 and 100 patients in the sample); publication bias was not assessed.

The level of evidence regarding the outcome reliability and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Steep Ramp Test

The level of evidence regarding the outcome measure criterion validity (Wmax for cardiorespiratory fitness) was not downgraded. No evidence of study limitations (risk of bias), conflicting results (inconsistency), applicability (bias due to indirectness), or limited number of included patients (imprecision) were found. Publication bias was not assessed.

The level of evidence regarding the outcome measure criterion validity (predicted VO₂peak for cardiorespiratory fitness) was downgraded by 1 level because of the number of included patients (1 level for imprecision: between 50 and 100 patients were included); publication bias was not assessed.

The level of evidence regarding the outcome measure reliability was downgraded by 4 levels because of study limitations (2 levels for risk of bias: only one study which was judged as doubtful); number of included patients (2 levels for imprecision: less than n=50 in the body of evidence); publication bias was not assessed.

The level of evidence regarding the outcome measure responsiveness (cardiorespiratory fitness) was downgraded by 1 levela because the number of included patients (1 level for imprecision: between 50-100 patients in the total body of evidence); publication bias was not assessed.

The level of evidence regarding the outcomes validity (muscle strength) and measurement error were not GRADEd. None of the included studies reported data about these measurement properties.

Physician-based Assessment and Counceling for Exercise questionnaire

The level of evidence regarding the outcomes validity, reliability, measurement error, and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Fitmáx questionnaire

The level of evidence regarding the outcomes validity, reliability, measurement error, and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Veterans-Specific Activity Questionnaire

The level of evidence regarding the outcomes validity, reliability, measurement error, and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

Duke Activity Status Index

The level of evidence regarding the outcomes validity (for muscle strength), reliability, measurement error, and responsiveness were not GRADEd. None of the included studies reported data about these measurement properties.

The level of evidence regarding the outcome measure criterion validity (cardiorespiratory fitness) was downgraded by 2 levels because of the number of included patients (2 levels for imprecision: less than n=50 included in the total body of evidence); publication bias was not assessed.

A systematic review of the literature was performed to answer the following question:

What is the validity, the reliability, the measurement error, and responsiveness of measurement instruments (as defined in the PICO) to monitor the physical fitness (i.e. cardiorespiratory fitness and muscle strength) of patient diagnosed with cancer?

Relevant outcome measures

The guideline development group considered validity, reliability, measurement error, and responsiveness as a critical measurement properties for decision making.

The working group defined the measurement properties following the Consensus-based Standards for the selection of health Measurement Instruments (COSMIN) taxonomy (Mokkink, 2010).

The working group used the updated criteria for good measurement properties reported in Mokkink (2018), based on Terwee (2007) and Prinsen (2018):

For criterion validity

• Positive (+): The correlation with the gold standard was ≥0.70 or the Area Under the Curve ≥0.70
• Unclear (?): Not all information for a positive judgement (+) of the measurement property was reported
• Negative (-): The correlation with the gold standard was <0.70 or the Area Under the Curve was <0.70

For construct validity (hypotheses testing)

• Positive (+): The result is in accordance with the hypothesis
• Unclear (?): No hypothesis was defined
• Negative (-): The result is not in accordance with the hypothesis

For reliability

• Positive (+): The ICC or weighted Kappa was ≥0.70
• Unclear (?): The ICC or weighted Kappa was not reported
• Negative (-): The ICC or weighted Kappa was <0.70

For measurement error

• Positive (+): The Smallest Detectable Change or Limits of Agreement was < Minimal Important Change
• Unclear (?): The Minimal Important Change was not defined
• Negative (-): The Smallest Detectable Change or Limits of Agreement was > Minimal Important Change

For responsiveness

• Positive (+): The result in in accordance with the hypothesis or the Area Under the Curve was ≥0.70
• Unclear (?): No hypothesis was defined
• Negative (-): The result is not in accordance with the hypothesis or the Area Under the Curve was <0.70

Search and select (Methods)

The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms until the 2^nd of May 2022. The detailed search strategy is depicted under the tab Methods. The systematic literature search resulted in 688 hits. Studies were selected based on the following criteria: participants were diagnosed with any kind of cancer, one of the measurement instruments of interest were used, data about one of the measurement properties of interest were reported. For criterion validity, studies or study data were excluded when it was unclear (i.e. not reported and not deducible) which parameter of a test (e.g. VO₂peak, distance walked, seconds, etc.) was used to examine the measurement properties of interest. For construct validity (e.g. hypotheses testing), studies or study data were excluded that correlated an instrument of interest with any instrument not of interest when it was not specified which construct the concurrent instrument intended to measure.

Forty-six studies were initially selected based on title and abstract screening, of which five systematic reviews. Potentially relevant systematic reviews were screened for studies they had included which our search strategy had missed. After reading the full text, 25 studies were excluded (see the table with reasons for exclusion under the tab Methods), and 21 studies were included (of which 3 were identified through the screened systematic reviews: England, 2012; Koegelenberg, 2007; Win, 2006).

Results

Twenty-one studies were included in the analysis of the literature. Important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables and was carried out using the COSMIN-tool (Mokkink, 2018).