Skip to content

Patient Care36 min read

How Physical Therapists Test Functional Movement: Which Weight-Bearing Test for Which Patient, and What a Finding Means

Single-leg squat, step-down, heel raises, chair stands or TUG: which weight-bearing test fits which patient, how to run it, and what a finding suggests.

Reza

Founder, OTR/L·

There isn't one test. You pick the weight-bearing test by the person in front of you and the goal they came in with, watch the movement, and then break it into parts until you find what limits it. For a young athlete that usually means a single-leg squat or lateral step-down, single-leg heel raises, a Y-Balance-type reach test and, late in knee rehab, hop tests. For most adults it means watching them squat, hinge and lunge, because those are the movements their goals depend on. For an 85-year-old it means a chair stand test, the Timed Up and Go, a standing balance test and gait speed. Whatever you see, a movement fault is a hypothesis to test, not a diagnosis.

This came from a question in a physical therapy group: how do PTs test functional movement these days, to see where the chain is weak in a weight-bearing position? A single-leg squat, single-leg calf raises, a heel tap? And what do the findings suggest? The best replies were short. Watch them squat, hinge and lunge, tie it to their goals, and break it down if something is off. And it depends on the population: a single-leg squat isn't practical for an 85-year-old, so watch them walk and stand up from a chair. Those replies are right. This guide adds the detail behind them: how to run each test, the numbers that come with it, and how much weight each finding can carry.

A note on who is writing. I'm an occupational therapist, not a PT, and the founder of Logicly, a home health scheduling app for clinicians. Every cutoff and norm below is cited to the study or the organization that published it, with the year, because several of them are older or narrower than people assume.

The Short Answer

  • Match the test to the person and the goal. Athletes: single-leg squat, lateral step-down, single-leg heel raises, a Y-Balance-type test, hop tests. Most adults: squat, hinge and lunge. Older adults: chair stands, the Timed Up and Go, the 4-Stage Balance Test, gait speed.
  • Break a faulty movement into parts. Range of motion, strength, balance and control, pain or fear, and load. Test each before you name a cause.
  • Knee valgus is not proof of weak hips. The hip strength link is inconsistent across studies. Reduced ankle dorsiflexion has a more consistent association.
  • Screening tests predict injury poorly. That holds for the Functional Movement Screen composite score, drop-jump knee valgus and general Y-Balance cut points.
  • Every cutoff belongs to the people it was studied in. The heel raise "25" came from adults under 60. The Timed Up and Go has two common cutoffs, 12 and 13.5 seconds. In people in their 80s, average scores sit close to several fall-risk cutoffs.
  • In home health, a standardized test is good OASIS observation. It doesn't produce a GG code, but it shows you the sit to stand and the walk.

Pick the Test by the Person

Who Start with Add when the goal needs it
Athletes and active younger adults Single-leg squat or lateral step-down; single-leg heel raises A Y-Balance-type reach test; hop tests late in knee rehab
Most adults with a pain or movement problem Watch the squat, hinge and lunge, or the task they can't do Single-leg versions; tests of the parts (ankle range, strength, balance)
Older adults, including patients at home 30-Second Chair Stand or Five Times Sit to Stand; Timed Up and Go; 4-Stage Balance Test; gait speed Berg Balance Scale or Functional Gait Assessment when balance is the main problem

Each test answers a different question. A single-leg squat asks how someone controls their body over one leg through depth. A chair stand asks whether they can get up, and how fast. A balance test asks how narrow a base they can hold. Decide which question matters for this person's goal, then pick the test.

For older adults, the choice is less personal than it sounds. CDC's STEADI program (Stopping Elderly Accidents, Deaths and Injuries), which puts the American and British Geriatrics Societies' fall prevention guideline into practice, lists the Timed Up and Go, the 30-Second Chair Stand and the 4-Stage Balance Test as the common assessments of gait, strength and balance. For adults with neurologic conditions, a 2018 clinical practice guideline developed with support from APTA and its Academy of Neurologic Physical Therapy recommends a core set: the Berg Balance Scale, the Functional Gait Assessment, the 10 Meter Walk Test, the 6-Minute Walk Test, the Five Times Sit to Stand and the Activities-specific Balance Confidence Scale. That guideline is for neurologic conditions, not for every patient, but it is the closest thing PT has to an agreed set.

Athletes and Active Younger Adults

These tests load one leg at a time, which is where a side-to-side difference shows up.

Single-Leg Squat

The patient stands on one leg and squats down and back up a few times while you watch from the front. You are looking at the trunk, the pelvis, the knee's path over the foot, and whether they can stay balanced.

What the research says about reading it:

  • Raters can agree. In Crossley and colleagues' 2011 study of 34 people without symptoms (mean age 24), three clinicians matched a consensus panel's good, fair or poor ratings 73% to 87% of the time.
  • Poor performers had weaker hip abduction. Compared with the 9 good performers, the 12 poor performers had less hip abduction torque, less trunk side flexion force and later onset of gluteus medius activity. Hip external rotation torque did not differ.
  • The visual scales are weak as a diagnostic. A 2023 systematic review of three visual rating methods (the Crossley scale, the Whatman score and medial knee displacement) found most had insufficient discriminative validity. Only the Crossley scale showed sufficient validity for patient-centred outcomes, on very low certainty evidence. The authors advise using visual single-leg squat scales with caution.

So the single-leg squat is a good way to see the problem. It is not a good way to name its cause.

Lateral Step-Down (the "Heel Tap")

This is the test people often mean by a heel tap. Piva and colleagues described it in 2006 for patients with patellofemoral pain:

  1. The patient stands on one leg on a 20 cm step, hands on the waist, knee straight, foot near the edge.
  2. The other leg hangs over the floor beside the step, knee straight.
  3. They bend the standing knee until the other heel gently touches the floor, then straighten back up.
  4. Five repetitions, with you facing them.

Score one point each for an arm strategy to keep balance, a trunk lean, the pelvis rotating or hiking, and an unsteady stance (stepping down on the other side, or wavering). For the knee, add 1 point if it moves medially so the tibial tuberosity crosses a vertical line over the second toe, or 2 points if it crosses a line over the medial border of the foot. A total of 0 or 1 is good quality, 2 or 3 medium, and 4 or more poor. In 30 patients, agreement between testers was moderate (the measures in that group had intraclass correlations of 0.67 to 0.79).

One study is worth knowing because of how it was run. Rabin and Kozol (2010) taught 29 healthy women the optimal movement pattern first, then rated their step-down. The women who still moved with moderate quality had less ankle dorsiflexion, measured with the knee bent and in weight bearing, than those who moved well. A fault that survives clear instruction tells you more than one you see cold.

Single-Leg Heel Raise Test

This answers the "single-leg calf raises" part of the question. The patient rises onto the ball of one foot and lowers, as many times as they can, while you hold them to a set height, pace and posture. It is the everyday clinical test of plantar flexor strength and endurance. It uses body weight as the resistance because, as Lunsford and Perry put it, manual muscle testing can load every muscle group except the ankle plantar flexors.

The norms depend on who was tested and how:

  • The "25" is from younger adults. Lunsford and Perry (1995) tested 203 people aged 20 to 59. They averaged 27.9 heel raises (SD 11.1, range 6 to 70), with no difference between men and women. The lower 99% confidence limit was 25, so the authors proposed 25 repetitions as the standard for a grade of Normal.
  • Counts fall with age and are lower in women. Jan and colleagues (2005) tested 180 sedentary adults aged 21 to 80 and found repetitions decreased with age and in women. They argued the traditional "20 or more is normal" grade, applied regardless of age and sex, needs adjusting for both.
  • The best current norms are by decade. Hébert-Losier and colleagues (2017) tested 566 adults aged 20 to 81. Each person did single-leg heel raises on a 10-degree incline, at 60 per minute set by a metronome, with fingertips on a wall at shoulder height, until they could no longer keep the height, pace or posture. Between-day reliability was excellent (ICC 1.0), with 95% limits of agreement of about 6 repetitions either way. Medians fell by four to five repetitions per decade, and after about age 60 women did more than men.

Median single-leg heel raises, right side, from Hébert-Losier and colleagues (2017):

Age Men Women
20 37.5 30.7
30 33.0 28.0
40 28.5 25.3
50 24.0 22.6
60 19.5 19.9
70 14.9 17.2
80 10.4 14.5

The left-side medians are within 1.6 repetitions of these. The estimates are for someone with a body mass index of 24.2 and the study's typical activity level (4 on a self-rated scale of 1 to 6), and the 2.5th percentile is close to zero from age 70 on. Two things follow. A count from a flat floor without a metronome isn't directly comparable with these numbers. And a 75-year-old who does 12 is inside the normal range by these norms, though a difference between their two legs may still matter.

What a low count suggests: plantar flexor strength or endurance is limited on that side. That's as direct a link as functional testing gets. The authors also note that older adults with weaker plantar flexors are at greater risk of falls and have poorer balance, which is one reason the test belongs in an older adult's exam too.

Y-Balance and Star Excursion Reach Tests

The patient stands on one leg and reaches as far as they can with the other foot in three directions: anterior, posteromedial and posterolateral. Reach distances are expressed as a percentage of leg length and combined into a composite score, and you compare sides.

The famous cut points come from one cohort. Plisky and colleagues (2006) tested 235 high school basketball players on the Star Excursion Balance Test before a season. Players with more than 4 cm of right-left difference in anterior reach were 2.5 times more likely to have a lower extremity injury. Girls whose composite reach was under 94.0% of leg length were 6.5 times more likely.

The later evidence tempers that. The instrumented Y-Balance Test is reliable (intrarater ICCs of 0.85 to 0.91 in the 2009 development study). But a 2021 systematic review of 57 studies by the same group found composite scores vary by as much as 13% of leg length with sex, sport and level of competition. Only 3 of 13 studies linked anterior reach asymmetry to injury. The authors concluded that if general cut points are used, the test may not predict injury, that cut points need to be specific to the population, and that an asymmetry cut point should be larger than the test's standard error of measurement, 3.2 cm.

Use it to measure dynamic single-leg control and to compare sides over time. Don't use one cut point to tell an athlete they are at risk.

Hop Tests

Hop tests belong late in rehab, when the person's goals involve jumping, landing or cutting and their surgeon's or protocol's criteria allow it. A common set is four hops, the ones Reid and colleagues tested: a single hop for distance, a triple hop for distance, a crossover hop for distance and a 6-meter timed hop. You express the involved leg's result as a percentage of the other leg's, the limb symmetry index.

  • Reliable enough to track. In 42 patients about 16 weeks after ACL reconstruction, Reid and colleagues (2007) found limb symmetry index ICCs of 0.82 to 0.93 and minimal detectable changes of 7% to 13%.
  • 85% is the old normal. Barber and colleagues (1990) found that more than 90% of 93 people with normal knees scored 85% or more on the single hop for distance and the timed hop.
  • A normal score doesn't rule out a problem. Noyes and colleagues (1991) found only 50% of 67 patients with ACL-deficient knees had an abnormal score on one hop test, and 62% on two. Low sensitivity, high specificity: an abnormal result confirms a deficit; a normal one doesn't clear it.
  • Symmetry can flatter. Wellsandt and colleagues (2017) followed 70 athletes. Six months after reconstruction, 57.1% met a 90% symmetry index on quadriceps strength and all four hops, the kind of criterion often used for return to sport. Only 28.6% reached 90% of their uninjured leg's pre-surgery numbers. The pre-surgery comparison was more sensitive for second ACL injuries (0.818 against 0.273). The uninjured leg loses capacity too, so compare against its earlier numbers when you have them.

Most Adults: Squat, Hinge, Lunge, Then Break It Down

For a general adult population, the reply in the thread is the method. Watch the squat, the hinge and the lunge, because those are how people get down to and up from a chair or the floor, lift things, kneel and climb stairs. Pick the ones their goals depend on. A roofer and a retired gardener need different ones.

If a movement looks wrong, break it into parts and test each one:

Ask How to check What it would point to
Is it range of motion? Measure the joints the movement needs. For the ankle, the weight-bearing lunge (knee to wall) test A mobility limit, for example ankle dorsiflexion in a squat or step-down
Is it strength? Test the muscles the movement loads: hip abductors and extensors, quadriceps, a single-leg heel raise count A capacity limit
Is it balance or control? Does it change with one clear cue, a slower pace, or a fingertip of support? A control or habit problem if it cleans up quickly
Is it pain or fear? Ask, and watch where in the range the movement changes A protective pattern rather than a weakness
Is it load or fatigue? Does it hold for two reps and fall apart at ten, or with a weight in hand? Endurance, or a gap between capacity and the task

On the ankle check: a 2015 systematic review of 12 reliability studies found the weight-bearing lunge test reliable between clinicians (ICC 0.80 to 0.99). Its minimal detectable change was about 4.6 degrees or 1.6 cm between clinicians, and 4.7 degrees or 1.9 cm for the same clinician. A change smaller than that is within the test's error.

The control question has the least evidence behind it. I didn't find a study that tests "it improved with a cue" as a diagnostic. What Rabin and Kozol's step-down study does suggest is the reverse: when a fault persists after you have taught the pattern, look harder at range and strength.

What About the Functional Movement Screen?

The Functional Movement Screen is the best-known packaged movement test, and its composite score has been studied widely as an injury predictor. A 2017 systematic review of 24 studies found a small association between a composite score of 14 or less and injury in male military personnel (pooled risk ratio 1.47). It found moderate evidence against using it to predict injury in soccer, and limited or conflicting evidence elsewhere. The authors concluded it shouldn't be used as an injury prediction tool. Bahr's 2016 review goes further: no screening test for sports injury has shown adequate test properties, and no study has shown that screening athletes and treating the high-risk ones works better than giving everyone the same program.

That doesn't make watching movement useless. It means you use what you see to plan treatment for this person's problem, not to forecast who will get hurt.

A Finding Is a Hypothesis: Knee Valgus as the Example

Knee valgus in a single-leg squat or step-down is the finding people most often over-read. Here is what each usual explanation has behind it.

Possible cause What the evidence says How to test it
Hip abductor or external rotator weakness Mixed. Crossley 2011 found less hip abduction torque in poor single-leg squatters, but no difference in external rotation. Dix 2018 (meta-analysis) found weaker hips related to valgus in single-leg landing, but not in the single-leg squat or double-leg landing. A 2021 review of 7 studies called the evidence conflicting. A 2026 review of 22 studies found mostly no correlation and no significant pooled result. Measure hip abduction and external rotation strength, with a dynamometer if you have one
Limited ankle dorsiflexion More consistent. Lima 2017 (meta-analysis of 17 studies) found less dorsiflexion in people with dynamic valgus (standardized mean difference -0.65), whether measured in weight bearing or not. Rabin 2010 found the same in the step-down. Weight-bearing lunge test, both sides
Motor control or habit Least studied as a cause. Some exercise programs reduce valgus: a 2021 review found 7 of 10 did, mostly those working muscles that cross the knee. Does one cue change it, and does it stay changed after practice?
Future injury Weak. In 710 elite female soccer and handball players, medial knee displacement in a drop jump had an area under the curve of 0.6, and in players with no previous ACL injury none of the drop-jump measures predicted one (Krosshaug 2016). Not something a single test can tell you

One more result keeps this honest. A 2020 systematic review of hip strengthening for knee osteoarthritis found clear benefits for pain, function and hip strength, but no change in dynamic alignment or the knee adduction moment. People got better without the "fault" changing. A movement finding can guide treatment without being the thing that treatment has to fix.

The same caution applies to the other tests. A low chair stand count correlates with leg strength (in Jones and colleagues' 1999 study, r = 0.78 for men and 0.71 for women against a weight-adjusted leg press), but pain, fear, balance and a soft low couch can all lower it too. A slow Timed Up and Go mixes standing up, walking and turning. CDC's TUG form notes that slow tentative pace, loss of balance, short strides, little or no arm swing, steadying on walls, shuffling, turning en bloc and misusing the device may signal neurological problems that need further evaluation. The number tells you something is limited. Your observation and the follow-up tests tell you what.

Older Adults and Patients at Home

For the 85-year-old in the reply, the questions change. Can they get up? Can they walk? Can they turn, and stand on a narrow base, without falling? These tests answer that with a chair, a stopwatch and a tape measure.

30-Second Chair Stand

CDC's STEADI form: a straight-backed chair without arms, seat 17 inches high. The patient sits in the middle, arms crossed at the wrists against the chest, feet flat. On "Go" they stand fully and sit back down as many times as they can in 30 seconds. Count the full stands; if they are more than halfway up at 30 seconds, count it. If they have to use their arms, stop the test and score 0. Stand next to them for safety.

CDC's below-average scores:

Age Men Women
60–64 fewer than 14 fewer than 12
65–69 fewer than 12 fewer than 11
70–74 fewer than 12 fewer than 10
75–79 fewer than 11 fewer than 10
80–84 fewer than 10 fewer than 9
85–89 fewer than 8 fewer than 8
90–94 fewer than 7 fewer than 4

CDC says a below-average score indicates a risk for falls. These numbers match the bottom of the middle-50% range (the 25th percentile) in Rikli and Jones's 1999 norms for 7,183 community-dwelling adults aged 60 to 94, as the Shirley Ryan AbilityLab Rehabilitation Measures Database lists them. So "below average" means roughly the bottom quarter of that sample. In the original 1999 validation, test-retest reliability was 0.84 for men and 0.92 for women.

A patient who can't stand once without their arms scores 0, which tells you little about change. For them, use the arms, write down that you did, and retest exactly the same way.

Five Times Sit to Stand

From the Rehabilitation Measures Database: a standard-height chair (43 to 45 cm, 17 to 18 inches) with a backrest, not pushed against a wall. Arms folded across the chest. "Stand up and sit down 5 times as quickly as you can when I say Go." Time from "Go" until they sit after the fifth stand. Give one practice trial. Needing arm support or help to finish counts as failing the test; document any change you make.

Three numbers get quoted for this test, and they mean three different things:

  • Slower than average for age. Bohannon's 2006 meta-analysis of 13 papers: worse than average if it takes more than 11.4 seconds (age 60 to 69), 12.6 seconds (70 to 79) or 14.8 seconds (80 to 89).
  • Balance dysfunction. Whitney and colleagues (2005), 174 adults with and without balance disorders: the Database lists 14.2 seconds as the cutoff for those over 60. The test correctly identified 65% of people with balance problems overall (81% under age 60). The Activities-specific Balance Confidence Scale (80%) and Dynamic Gait Index (78%) did better.
  • Recurrent falls. Buatois and colleagues (2008), 2,735 healthy community-living adults aged 65 and older: 15 seconds predicted recurrent falls. This was published as a research letter.

Timed Up and Go

CDC's STEADI form: the patient wears their regular shoes and uses their walking aid if they need one. They start sitting back in a standard armchair, with a line on the floor 3 meters (10 feet) away. On "Go" they stand, walk to the line at their normal pace, turn, walk back and sit down. Time from "Go" until they are seated. Stay by them throughout. The Database adds a practice trial, a chair that stands free rather than against a wall, and the same assistive device every time you retest.

The cutoff depends on the source:

  • CDC STEADI: an older adult who takes 12 seconds or more is at risk for falling.
  • Shumway-Cook and colleagues (2000): more than 13.5 seconds, with 87% sensitivity and 87% specificity. That came from 30 people: 15 with two or more falls in six months and 15 with none.
  • Barry and colleagues (2014), meta-analysis: at 13.5 seconds or more, pooled specificity was 0.74 and sensitivity 0.31. The authors said the TUG is better at ruling falls in than out and shouldn't be used alone to identify high fall risk in community-dwelling older adults.

For context, Steffen and colleagues (2002) found mean TUG times of 8 seconds in community-dwelling adults in their 60s, 9 seconds in their 70s, and 10 (men) and 11 (women) seconds in their 80s. Podsiadlo and Richardson's original 1991 paper found people under 20 seconds were independent in basic transfers, while those over 30 seconds were dependent in transfers and needed help in the shower or tub.

4-Stage Balance Test

CDC's STEADI form. Four positions, each harder than the last: feet side by side; the instep of one foot touching the big toe of the other; tandem stance, heel touching toe; and standing on one foot. Show each one, help the patient into it while holding their arm, let go once they are steady, and time up to 10 seconds. They can move their arms or body but not their feet. If they hold a position for 10 seconds without moving their feet or needing support, go to the next; if not, stop. No cane or walker, eyes open.

CDC: an older adult who can't hold the tandem stand for at least 10 seconds is at increased risk of falling, and referral to physical therapy for gait and balance exercise, or to an evidence-based program such as Tai Chi, is worth considering.

Gait Speed

The usual clinical version is the 10 Meter Walk Test. The Database describes timing the person over a set distance after they've had room to reach their speed, two trials at comfortable pace and two fast, averaged, in meters per second. Document the device. Walk at least half a step behind so you don't set the pace.

Studenski and colleagues (2011) pooled 9 cohorts, 34,485 community-dwelling adults aged 65 and older, average gait speed 0.92 m/s:

  • Every 0.1 m/s faster was associated with better survival (hazard ratio 0.88).
  • About 0.8 m/s sat at the median life expectancy for age and sex. Speeds of 1.0 m/s or more consistently went with longer survival than age and sex alone predicted.
  • The authors suggested speeds under 0.6 m/s may identify people at higher risk of early death, and that gait speed is less informative in people who already depend on others for basic daily activities.
  • Their suggested clinic method: a 4-meter course, a stopwatch, a standing start, and an instruction to walk at their usual pace, with no further encouragement.

For change over time, Perera and colleagues (2006) put a small meaningful change at about 0.05 m/s and a substantial one at about 0.10 m/s in older adults.

One trap: the categories of household walker (under 0.4 m/s), limited community walker (0.4 to 0.8) and community walker (over 0.8) come from people after stroke. Don't apply them to every older adult.

Berg Balance Scale and Functional Gait Assessment

When balance is the main problem, the two longer scales give you more detail.

The Berg Balance Scale has 14 items scored 0 to 4, maximum 56, and takes 15 to 20 minutes. Its cutoffs come from different studies: under 45 for greater fall risk (Berg, 1992), and under 51 with a fall history or under 42 without one (Shumway-Cook, 1997, in 44 older adults). The Rehabilitation Measures Database states plainly that no common interpretation of Berg scores exists, and it notes ceiling effects in some groups, such as mid-stage Parkinson disease. In older adults scoring 45 to 56 at the start, the minimal detectable change is 3.3 points (Donoghue, 2009). The neurologic core set guideline found strong evidence for using the Berg to measure change in balance.

The Functional Gait Assessment has 10 walking items scored 0 to 3, maximum 30. It needs a marked walkway 6 meters (20 feet) long and 30 cm (12 inches) wide, a 9-inch obstacle and a set of steps with rails. It was adapted from the Dynamic Gait Index to reduce that test's ceiling effect, and adds walking with a narrow base, walking backward and walking with eyes closed. In Wrisley and Kumar's 2010 study of 35 adults aged 60 to 90, a score of 22 or less predicted falls over the next 6 months with 100% sensitivity and 72% specificity. In a later study of 135 older adults (Beninato, 2014), the same cutoff had 66% sensitivity and 84% specificity, and the minimal clinically important difference was 4 points.

What the Cutoffs Can and Can't Tell You

Look at the norms for people in their 80s next to the cutoffs:

Test Average in the 80s Common fall-risk cutoff
Timed Up and Go 10 s (men), 11 s (women) (Steffen 2002) 12 s (CDC) or more than 13.5 s
Functional Gait Assessment 20.8 of 30 (Walker 2007, 33 people) 22 or less
Berg Balance Scale 53 (men), 50 (women) (Steffen 2002) Under 51 with a fall history

An average person in their 80s is near or past several of these lines. Several cutoffs also came from small samples. So read a score as one piece of the picture. STEADI's own algorithm pairs these tests with a medication review, orthostatic blood pressure, vision, feet and footwear, vitamin D and home hazards. For home hazards, STEADI's step is a referral to an occupational therapist to evaluate home safety. Where these tests earn their keep is change: the same test, the same way, at evaluation and discharge, judged against its minimal detectable change.

Running These Tests in a Patient's Home

Home testing is mostly about keeping the setup the same each time.

  • Chair. The protocols differ: no arms and a 17-inch seat for CDC's chair stand, a standard armchair for the TUG (the Database gives 44 to 47 cm), and 43 to 45 cm with a backrest for the Five Times Sit to Stand. Home chairs vary. Note which chair you used and its seat height, and use it again.
  • Space. The TUG needs a 3-meter line. Studenski's gait speed method needs a 4-meter course plus room to start. The Functional Gait Assessment needs a 6-meter walkway. Measure a hallway once and write it down.
  • Device. Same walker or cane every retest. The 4-Stage Balance Test is done without one.
  • Safety. CDC's forms say to stand next to the patient for the chair stand and stay by them for the TUG. Clear the path first.
  • Record the protocol with the score, so the next clinician, and your own discharge retest, can match it.

When the tests point to balance, our free Balance Program Builder makes a graded home program in four levels, from seated to moving, with figures. For the patient's family, preventing falls at home during home health covers the moments falls happen, room by room, and home exercises when the therapist isn't there covers keeping up the program between visits.

How These Tests Relate to OASIS GG Items

For home health clinicians: the standardized tests and Section GG measure different things, but one can feed the other.

  • Different scales. GG0130 (self-care) and GG0170 (mobility) code the type and amount of help a helper gives, from 06 (independent) to 01 (dependent), plus codes for activities not attempted. Seconds and repetitions don't convert to a GG code. A patient can be 06 on sit to stand and still take 16 seconds for five stands.
  • The test is observation. The OASIS-E2 manual lets clinicians assess GG activities by direct observation, which it calls preferred, as well as by patient or caregiver report, similar activities, or other staff. A chair stand test is direct observation of GG0170D Sit to stand. A TUG shows a sit to stand, then a walk out and back with a turn, which shows you what GG0170I Walk 10 feet asks.
  • Order matters at start of care. At SOC and ROC, the GG code reflects the patient's ability before the benefit of the agency's services. The manual suggests having the patient attempt the activity before giving any instruction that could make them more independent. Run your tests cold, note the help needed, then teach.
  • Your presence isn't supervision. The manual says the assessing clinician's presence shouldn't automatically be coded as supervision. Code the help needed to do the activity safely. If you had to steady them, that is supervision or touching assistance (04).
  • Why the agency cares. The Discharge Function Score, used in Home Health Quality Reporting and, since the 2025 performance year, in the expanded HHVBP Model, is built from GG items including sit to stand, walk 10 feet and walk 50 feet with two turns. Skipped, dashed and "activity not attempted" items count as missing and are filled in by statistical imputation. An activity you actually observed is better than an estimate.

For scoring each item, see how to score the OASIS-E2 GG items. For what the functional measures mean for agency payment, see HHVBP 2026 and functional outcomes.

Key Takeaways

  • Pick the test by the person and their goal: single-leg tests for athletes, squat, hinge and lunge for most adults, chair stands, the TUG, a balance test and gait speed for older adults.
  • When a movement looks wrong, test the parts: range, strength, control, pain or fear, and load.
  • Knee valgus points more consistently to ankle dorsiflexion than to hip weakness, and it doesn't predict injury well. Test the hypothesis before you treat it.
  • Use norms that match the person and the protocol: the heel raise 25 is for adults under 60, and older adults' averages sit near several fall cutoffs.
  • Track change with the same test, set up the same way, against its minimal detectable change.
  • In home health, run standardized tests before teaching at start of care. They give you direct observation for GG coding.

Frequently asked questions

Is a single-leg squat a good test for an older adult?

Often not as the first test. For an 85-year-old, the questions that matter are whether they can get up from a chair, walk, turn and stand on a narrow base. Start with a chair stand test, the Timed Up and Go, the 4-Stage Balance Test and gait speed; CDC's STEADI program lists the first three for gait, strength and balance. A fit older adult whose goals need it can still do a supported step-down or single-leg heel raises.

Does knee valgus in a single-leg squat mean weak hips?

Not on its own. Studies of hip strength and knee valgus disagree: a 2018 meta-analysis found the link in single-leg landing but not in the single-leg squat, and a 2026 review of 22 studies found no significant pooled relationship. Reduced ankle dorsiflexion has a more consistent association. Treat valgus as a hypothesis: measure hip strength and ankle dorsiflexion, and see whether a clear cue changes the movement.

How many single-leg heel raises are normal?

It depends on age, sex and how you test. The often-quoted 25 comes from a 1995 study of adults aged 20 to 59. In a 2017 study of 566 adults tested on a 10-degree incline at one raise per second, the median fell from about 37 for men and 30 to 31 for women at age 20 to about 10 for men and 13 to 15 for women at age 80. Compare the two legs, and use norms that match how you tested.

What Timed Up and Go time means a fall risk?

CDC's STEADI materials flag 12 seconds or more. Other studies use more than 13.5 seconds. A 2014 meta-analysis found the TUG better at ruling falls in than ruling them out, with a pooled sensitivity of 0.31 at 13.5 seconds, and advised against using it alone. Read it with the patient's fall history, the other tests and what you saw during the walk.

Can a standardized test give me an OASIS GG code?

Not directly. GG items code how much help a helper gives, on a six-level scale, not seconds or repetitions. But running a chair stand or Timed Up and Go is direct observation of sit to stand and walking, which the OASIS-E2 manual names as the preferred way to assess. At start of care, run the test before you teach anything, and code the help the patient needed.

Sources

Older adults and fall risk:

Athletes and single-leg tests:

Knee valgus, components and screening:

OASIS and home health:

Free tools on this topic

Related Articles

Ready to simplify your scheduling?

Join home health agencies saving 8+ hours per week with Logicly. Start with a 14-day free trial — $24/month after that.