Am I losing muscle? A four-part sarcopenia self-test you can score at home

A client of mine, 56, came in a few months ago after a checkup. Bloodwork was fine. Blood pressure was fine. But his doctor had watched him get out of the waiting-room chair and asked, almost offhand, whether he'd noticed his legs getting weaker. He hadn't. Then he thought about it on the drive over and realized he'd started putting a hand on the armrest to stand up sometime in the last year without ever deciding to.

That's how muscle loss shows up after 50. Not as a feeling. As a small adjustment you make without noticing, and then another one.

The clinical term is sarcopenia, and it has something most vague health worries don't: actual cutoffs. Researchers have spent two decades arguing about how to define it, and the useful byproduct of that argument is a handful of screening tests with published thresholds that you can run at home with a chair, a tape measure, and a hallway. This post walks through four of them, what the numbers mean, and what to do about a low score depending on which part of the system is the problem.

This isn't medical advice. If you've had a recent fall, unexplained weight loss, or a diagnosis that affects your muscles or nerves, take these results to a doctor rather than to a training program.

Why muscle loss after 50 is a test result, not a feeling

The reason people don't notice sarcopenia is that the body is very good at routing around it. Your quads get 15 percent weaker over a decade and you compensate with a hand on the armrest, a slightly wider stance on the stairs, a shorter stride. None of those feel like weakness. They feel like nothing.

The standard clinical definition, from the European Working Group on Sarcopenia in Older People (EWGSOP2, published in Age and Ageing in 2019), has three layers. Low muscle strength makes sarcopenia probable. Low muscle mass, measured with a DEXA scan or similar, confirms it. Low physical performance, usually walking speed, makes it severe. Notice the order. Strength comes first, because strength predicts outcomes better than mass does, and because a grip dynamometer or a chair is a lot cheaper than a scan.

That ordering is also why a self-test is possible at all. Three of the four layers below have home versions with real thresholds behind them. The fourth, mass, has a tape-measure proxy that's rougher but still worth running.

Before the numbers, a caveat. Nearly all of these cutoffs were developed in people over 60, and a lot of them in people over 70. If you're 50 and hit a cutoff, that's a loud signal, louder than it would be at 75. If you're 50 and clear a cutoff easily, that's not a clean bill of health. It means you haven't declined to the level where a geriatrician would get worried, which is a low bar for someone with 30 more years of living to do.

Test 1: the SARC-F questionnaire

Start here because it takes 60 seconds and it's what a clinician would hand you first.

SARC-F was published by Malmstrom and colleagues in the Journal of Cachexia, Sarcopenia and Muscle in 2016. Five questions, each scored 0, 1, or 2. Answer them honestly, which mostly means answering how you actually do things now, not how you'd do them if you tried hard.

Strength. How much difficulty do you have lifting and carrying 10 pounds? None (0), some (1), a lot or unable (2).

Assistance walking. How much difficulty do you have walking across a room? None (0), some (1), a lot, use aids, or unable (2).

Rising from a chair. How much difficulty do you have transferring from a chair or bed? None (0), some (1), a lot or unable without help (2).

Climbing stairs. How much difficulty do you have climbing a flight of 10 stairs? None (0), some (1), a lot or unable (2).

Falls. How many times have you fallen in the past year? None (0), one to three (1), four or more (2).

A total of 4 or more is a positive screen. That's the threshold the guideline uses to send someone on for the physical tests.

The sales pages that reproduce this questionnaire never mention what the number is actually worth. A 2021 systematic review in the Journal of the American Medical Directors Association pooled 29 studies and 21,855 people and found SARC-F has a sensitivity between 29 and 55 percent depending on which sarcopenia definition you compare it to, with specificity between 69 and 89 percent. So if you score 4 or more, you very likely have a problem. If you score under 4, the test missed roughly half the people who do. It's built to catch the obvious cases, not to reassure you.

So for someone in their 50s who's still active, a score of 0 or 1 tells you very little. That's why the next three tests exist.

Test 2: calf circumference, and why the cutoff differs by sex

This is the mass proxy. It's rough, but it's the only part of the definition you can approximate without a scan, and adding it to SARC-F measurably improves the screen.

Stand with your feet shoulder-width apart and your weight even. Wrap a soft tape measure around the widest part of your calf, snug against the skin without compressing it. Measure both sides and record the larger one. Do it in the morning if you can, since calves swell through a day of sitting.

Cutoffs: under 34 cm for men, under 33 cm for women.

Those numbers come from Kawakami and colleagues in Geriatrics & Gerontology International in 2015, who measured 526 Japanese adults aged 40 to 89 and compared calf circumference against appendicular muscle mass on DEXA. The correlation was strong, 0.81 in men and 0.73 in women, and the 34 cm cutoff in men caught 88 percent of true low-mass cases while correctly clearing 91 percent of the rest. In women the cutoff was less accurate, 76 and 73 percent.

Two caveats that matter. First, that's a Japanese cohort with a lower average body size than most American adults, and the EWGSOP2 authors themselves offer a more conservative 31 cm as an alternative. Second, a tape measure can't tell muscle from fat. A heavier person can carry a 37 cm calf with less muscle in it than a lean person's 33. If your BMI is well over 30, read this test as unreliable in your case and lean on the other three.

The combined version, called SARC-CalF, adds 10 points to your SARC-F total if your calf is under the cutoff, with a positive screen at 11 or more. A 2025 meta-analysis in The Gerontologist of 18 studies and 10,367 older adults found the combined tool had pooled sensitivity of 53 percent and specificity of 87 percent. Better than the questionnaire alone, still missing close to half of true cases. Screening narrows the search. It doesn't close it.

Test 3: five-times sit-to-stand against the 15-second line

Now the part that actually measures something.

Use a standard chair, seat around 43 to 45 cm off the floor, no arms, backed against a wall so it can't slide. Sit with your back against the backrest, feet flat, arms crossed over your chest. On go, stand fully upright and sit back down five times as fast as you safely can. The timer starts on go and stops when your backside touches the seat on the fifth rep. Standing fully means hips and knees straight, not a half-rise. Sitting means you actually make contact, not a hover.

Have someone else time you, or use a phone video and read the timestamps. Self-timing while doing the reps reliably shaves a second off.

Over 15 seconds is the EWGSOP2 marker for low muscle strength.

11.4 seconds is the line for worse-than-average performance in people aged 60 to 69, from a 2006 meta-analysis by Bohannon that pooled 13 papers on adults over 60. The equivalent lines were 12.6 seconds for people in their 70s and 14.8 for people in their 80s.

Under 10 seconds is where I'd want a healthy 50-year-old who trains. Under 8 is comfortable.

What this test reads is leg strength relative to your bodyweight, plus the speed you can produce it. Don't skip past the second half of that. A lot of people over 50 have enough strength to stand up but have lost the ability to do it fast, and power drops faster with age than strength does. If your time is fine but the first rep is noticeably slower than the last four, that's the power side talking. Write it down for the plan later. And if you want to turn the same five reps into an actual power score with its own pass and fail lines, the muscle power test runs your seconds through the validated equation.

This is a cousin of the floor-based sitting-rising test, which scores support and balance rather than speed and starts from the ground rather than a chair. The two disagree in interesting ways. Someone with stiff ankles and hips can post a fast chair time and a poor floor score. Someone flexible and unsteady can do the reverse.

Test 4: the four-metre walk, and the 1.0 and 0.8 m/s thresholds

Measure out 4 metres (about 13 feet) on a hallway or driveway with clear space before and after. Start walking a couple of steps before the line so you're at your normal pace when you cross it, and keep going a couple of steps past the far end. Time from the moment your first foot crosses the start to the moment your first foot crosses the finish. Walk at the pace you'd use to get to the mailbox. Not a stroll, not a march. Do it twice and take the faster.

Divide 4 by your time in seconds. That's your gait speed in metres per second.

0.8 m/s or slower (5 seconds or more for 4 metres) is the EWGSOP2 threshold for low physical performance, the marker that makes sarcopenia severe.

Under 1.0 m/s (over 4 seconds) is the more common clinical flag for higher risk.

Over 1.2 m/s (under 3.3 seconds) is where most healthy adults under 65 land, and I'd expect anyone in their 50s who trains to clear it without trying.

The number everyone quotes on this comes from Studenski and colleagues in JAMA in 2011, who pooled individual data from nine cohort studies covering 34,485 community-dwelling adults aged 65 and over, followed for 6 to 21 years. Every 0.1 m/s of usual walking speed was associated with a 12 percent lower mortality risk, and the association held across the whole range of speeds rather than flattening at some safe plateau. At age 75, predicted ten-year survival ranged from 19 percent to 87 percent in men depending on gait speed alone.

I'm always careful with this one because it gets quoted as if walking faster makes you live longer. It doesn't, any more than a fast sit-to-stand does. The mean age in that cohort was 73.5, and the mean gait speed was 0.92 m/s. Gait speed predicts survival because it's a composite. Leg strength, balance, joint health, heart and lung capacity, and how much pain you're in all show up in how fast you naturally walk, which is exactly what makes it worth measuring and also why a slow score needs decomposing rather than a walking program.

What a low score actually means: strength, power, balance, or mass

Four tests, four somewhat different questions. Here's how I read the combinations with clients.

Slow sit-to-stand, fine walk. Leg strength relative to bodyweight is the deficit, and it's usually a training-history problem rather than a medical one. This is the most common pattern I see in desk workers over 50, and the most fixable. If you also fail a grip strength test, which is the other strength marker the guideline uses, the pattern is systemic rather than lower-body specific.

Sit-to-stand time is fine, but the first rep is slow and the last is fast. Power. You have the strength; you've lost the rate of producing it. This is the deficit that shows up on a curb you misjudged or a bus you had to jog for, and it's the one most standard strength programs don't address because they never ask you to move quickly.

Fine sit-to-stand, slow walk. Something other than leg strength is limiting the walk. Balance, a painful joint, cardiovascular capacity, or a stiff hip and ankle shortening your stride. A single-leg balance check sorts the first of those quickly. If balance is fine, look at the joints.

Calf under the cutoff with everything else passing. Either you're naturally lean and the tape is misreading you, or you've kept strength on a shrinking base and the mass loss hasn't reached the performance tests yet. That second case is the one that rewards early attention. Protein intake is nearly always part of the answer here, and I've written a full post on how much after 40.

SARC-F of 4 or more, plus any physical fail. That's a positive screen by the guideline's own logic, and it's worth a conversation with a doctor before you start a program, mainly to rule out the things training can't fix.

Everything passes comfortably. Good. Write the numbers down anyway. The value of this battery for someone in their 50s isn't the pass; it's the retest a year from now.

A 12-week plan, sorted by cause

Different lives need different versions of this, and someone who's never lifted needs a different first month than someone coming back after a decade off. But the components map cleanly onto the causes above.

If strength is the deficit. Two lower-body sessions a week, built around a squat pattern and a hinge pattern, with loads heavy enough that the last two reps of each set are genuinely hard. Sets of 6 to 10. Goblet squats, box squats, Romanian deadlifts, step-ups. The single most useful cue for this population is to lower slowly and stand up with intent. Progress the load every week or two. Strength returns before mass does, because the first six weeks of gains are mostly the nervous system relearning how to recruit what you have, so don't judge the program by the tape measure early.

If power is the deficit. Add speed to what you're already doing rather than adding a separate day. Once a week, take a light load, around half of what you'd lift for strength, and do sets of 3 to 5 reps moving up as fast as you safely can, with full rest between sets. Sit-to-stands from a chair with a jump at the top, kettlebell swings, medicine ball throws against a wall, quick step-ups. The goal is speed, not fatigue. If you're breathing hard you've turned it into conditioning.

If balance or joint range is the deficit. Work single-leg. Split squats, single-leg Romanian deadlifts with a hand on a wall, step-downs from a low step. Add the hip and ankle drills from the relevant self-tests. This is also where the movement screen earns its keep, since it scores six areas at once and will tell you which joint is actually the one shortening your stride.

If mass is the deficit. Everything from the strength plan, plus protein at roughly 0.7 to 1 gram per pound of bodyweight per day, spread across three or four meals with 30 to 40 grams in each. Older muscle needs a larger dose per meal to trigger the same synthesis response as younger muscle, which is why "I eat plenty of protein" and "I eat 20 grams at dinner" can both be true.

Whatever the cause, keep one aerobic session a week you don't count as training. A brisk 30-minute walk. Gait speed is partly a cardiovascular number, and it's the one you can improve without lifting anything.

Retesting, and when to take the results to a doctor

Retest all four at 12 weeks, same chair, same hallway, same time of day. What you're reading is direction. A sit-to-stand that goes from 12.5 to 10 seconds is a real change. A calf that goes from 33.5 to 34 might be breakfast.

At 12 weeks I'd expect strength to have moved clearly, power to have moved some, and mass to have barely moved, in that order. Mass takes six months to show on a tape measure for most people over 50, and a year to be unambiguous.

Take the results to a doctor if: your SARC-F is 4 or higher and you failed a physical test; you've lost weight without trying; you've fallen in the past year; or you retest at 12 weeks after consistent training and the numbers went the wrong direction. That last one is the useful failure. A reasonable program that produces no change is information, and it usually points to something other than the program.

If you want to know where to start before committing to 12 weeks, the free movement screen is the same assessment I run with new clients. Five minutes, six areas scored, no email required, and it'll tell you whether the limiting factor is the joint, the strength, or the control.

And if you'd rather have someone build the program around your numbers than around a template, that's what one-on-one training is for. I've logged more than 12,000 sessions with adults around San Jose and the South Bay, and a good share of them started with exactly this conversation: a doctor's offhand comment, a hand on an armrest, and a number that turned out to be very fixable.

My 56-year-old client posted 13.8 seconds on the sit-to-stand the first time. Twelve weeks later, 9.1. He stopped using the armrest around week five, and he noticed that one.