Handheld Dynamometer Hip Abduction Test Protocol: Make vs. Break Technique for Reliable Readings

A physio and an athletic trainer test the same athlete's hip abductors in the same week, both using a calibrated handheld dynamometer, and the numbers come back 18% apart. Nobody moved the equipment or miscounted reps. The gap traces back to two decisions that never get written into anyone's protocol notes: which technique was used, make or break, and whether the hip sat in the exact same position, at the exact same pad distance, both times.

Both tests measure the same muscle group, but they load the tissue differently and drift apart fast the moment stabilization isn't identical trial to trial. ACL return-to-sport panels, hip impingement rehab, and gluteal tendinopathy monitoring all lean on this number to make real decisions, so the fix that closes most of that gap is worth locking down before the next assessment day, not after the third mismatched retest.

In a make test, the dynamometer stays put, belt-fixed to the table or braced by the examiner's own body, and the athlete pushes into it, ramping up over roughly two seconds and holding a maximal effort for three to five. Nothing moves. The reading is a plateau, and it depends far more on the athlete's effort than on the examiner's own strength.

In a break test, the athlete builds to that same hard isometric hold, and the examiner applies smoothly increasing counterforce until the limb yields into adduction, reading the display at the precise instant that happens. That single requirement is where reliability quietly falls apart: the examiner has to out-force the athlete within a narrow window and catch a moving number at the right split second, not a beat after the limb has already started to collapse.

Test an 82kg academy soccer player with the make technique and a belt-fixed unit, and she posts 195N. Test her the same day with an examiner half her size running a break test, and the reading drops to roughly 165-170N, not because her hip weakened but because that examiner can't out-force a strong hold before their forearm fatigues. Neither number is wrong; they answer two different questions. Mixing them across a block is the single most common way a score looks like it changed when nothing in the hip actually did.

Position drives more variance here than most staff expect. The athlete lies on the non-test side, hips stacked, bottom knee bent for a stable base, top leg straight and held in a neutral line with the trunk, not drifted into flexion or rotated in or out. Even 10-15 degrees of extra hip flexion recruits tensor fasciae latae over gluteus medius and inflates the reading in a way that has nothing to do with true abductor strength. The pad sits on the lateral leg, about 5cm proximal to the lateral malleolus for the standard long-lever placement, and a non-elastic belt looped under the table anchors the unit so the examiner isn't fighting to hold it still with one hand while reading a display with the other.

A pad that sits 3cm closer to the knee on a retest changes the torque calculation even when the athlete's raw force output hasn't moved at all.

1. Setup (2 minutes): Position the athlete side-lying as above, secure the belt if using one, and mark the pad location for reuse on future sessions.
2. Familiarization: Two submaximal trials at roughly 50-70% effort using whichever technique you plan to score. This matters more for the break test, since athletes who've never felt a hip pushed into adduction tend to guard the first time.
3. Pick one technique per athlete and keep it: Decide make or break before the session starts and use it for every retest of that individual. Never compare a make-test score from one visit against a break-test score from the next.
4. Make test execution: With the dynamometer belt-fixed or braced firmly, cue the athlete to push out, building over two seconds and holding their hardest effort for three to five. Record the plateau value, not the initial spike.
5. Break test execution: The athlete builds to a maximal hold over two seconds. The examiner then applies smoothly increasing counterforce, timed to overcome the athlete's resistance within three to five seconds, and reads the number at the exact instant the limb starts to move, not a beat later.
6. Trials: Three maximal trials per leg with 30-45 seconds of rest. Decide in advance whether you're scoring the mean of three or the single highest trial, and apply that rule identically every time; switching between the two shifts the final number by 5-10% on its own.
7. Test order: Uninvolved or non-dominant limb first as a within-session reference, then the involved or weaker side.

A force reading in newtons only means something once it's tied to where on the leg it was taken. Torque reflects the muscle's actual work: torque (Nm) = force (N) x lever arm (m), the lever arm being the distance from the greater trochanter to the pad. Normalizing to body mass, index (Nm/kg) = torque / body mass (kg), lets you compare across athletes of different sizes or track one athlete across a season without a weight change muddying the trend.

Worked example: a 68kg athlete produces 145N at a measured lever arm of 0.38m. Torque is 145 x 0.38 = 55.1 Nm, a normalized index of 55.1 / 68 = 0.81 Nm/kg. The other hip, same pad placement, produces 132N: torque of 50.2 Nm, normalized to 0.74 Nm/kg. The limb symmetry index comes out to (0.81 - 0.74) / 0.81 = 8.6%, under most flagged thresholds but worth watching on the next retest.

Comparing raw force bilaterally on the same athlete at an identical lever arm skips the torque math with no loss of accuracy, since the lever arm cancels out on both sides. Comparing raw force across two different athletes, or across sessions where the pad landed at a different distance, does not cancel out and will misrepresent the real difference.

Widler, Glatthorn, Bizzini, Impellizzeri, Munzinger, Leunig, and Maffiuletti (2009), in the Journal of Bone and Joint Surgery (American volume), tested hip abductor strength in side-lying, supine, and standing positions, cross-checking each against concurrent gluteus medius EMG to see which position actually reflects true abductor output rather than compensation. Side-lying came out ahead on every metric: a coefficient of variation of 3.7%, against 6.1% supine and 4.2% standing, and an ICC of 0.90, against 0.83 supine and 0.88 standing. Their stated limitation matters: the comparison ran in a single-session design on healthy, uninjured adults, so whether side-lying keeps its edge in a painful or post-surgical hip isn't something the study addresses.

Schmidt, Iverson, Brown, and Thompson (2013), in Physiotherapy Theory and Practice, ran the more direct comparison this protocol leans on: 39 subjects aged 21-70, tested by two examiners using both make and break technique, both legs, each combination performed twice, for 16 ratings per participant, with pelvic stabilization and a long lever arm. Both techniques came back highly reliable, ICC above 0.87 across the board, with the make test edging ahead statistically. Their own read on that gap is worth carrying forward: statistically the make test wins, but clinically the break test is the more practical, faster option, and their data doesn't support dropping it. Their stated limitation: only two examiners were used, so the finding doesn't establish how far apart two testers of very different strength would land using break technique on the same strong athlete.

Absolute normative torque values vary enough by age, sex, sport, and equipment that quoting a single healthy range invites more confusion than clarity. What holds up better across settings is the limb symmetry index between an athlete's own two hips, calculated the same way as the worked example above.

Track the trend line more than any single session. A hip sitting at 12% asymmetry for three consecutive monthly retests is a different situation than one that was at 6% last month and jumps to 12% right after a spike in single-leg cutting load; the second pattern tends to precede a problem, not just describe an existing baseline.

An asymmetry in the borderline or flagged range is a training target, not a red card. Side-lying weighted abduction, banded lateral walks progressing to single-leg bridges with an abduction hold, and clamshells loaded past bodyweight all target the muscle group the test isolates, and a 4-6 week block focused on the weaker side closes a meaningful gap in most athletes without an underlying joint pathology driving the deficit.

Retest on the same cadence, 4-6 weeks, same technique, same position, ideally the same examiner or at least the same belt-fixed setup. Retesting sooner mostly captures day-to-day noise rather than real adaptation, and switching examiners mid-block without standardizing the setup is how a genuine improvement gets misread as no change at all. For hip abduction specifically, technique choice and position standardization matter more than almost any other variable in the protocol, which is exactly why they're worth deciding before the first athlete ever lies down on the table.