Standing balance assessment and prognosis after total hip arthroplasty

Executive assessment

Standing balance after total hip arthroplasty (THA) is best assessed as several related abilities: maintaining a stance, redistributing load, controlling an intended movement, recovering from an unexpected disturbance and judging one's confidence in everyday activities. A person can perform well in one of these domains and remain limited in another. Quiet-standing center of pressure (CoP), one-leg stance duration, a multidirectional stepping time, a clinical balance scale and observed falls should therefore remain separate outcomes.

A practical assessment should combine a stage-appropriate stance task with a safe dynamic task, explicit recording of support and noncompletion, and a brief history of falls, circumstances and confidence. Clinical stepping tests have direct THA reliability evidence, including the Four Square Step Test (FSST), modified FSST, step test and newer Six-Spot Step Test. The reported error estimates differ materially by protocol, stage and design. None should be converted into a universal balance responder threshold. Several original studies remain available only as abstracts, limiting replication of their procedures. [1–4]

Instrumented studies document persistent group differences, especially under more demanding sensory conditions, but do not establish a universal postoperative CoP abnormality or future-fall threshold. Force-platform and clinical tests are only moderately related because they sample different tasks. A small camera-based study showing pre/postoperative changes is recovery evidence, not criterion validation of its balance algorithm. [5–11]

Actual falls require particular care. In a prospective cohort of 157 predominantly female primary THA patients, preoperative one-leg stance time did not distinguish later fallers from nonfallers. Preoperative abductor strength and prior falls were more informative, although the proposed strength threshold had only 50% specificity. The source's AUC of 0.702 is not a 70.2% probability that an individual will fall. A separate 2025 study's ABC confidence cutoff discriminated recalled falls reported in the same late postoperative survey; it did not prospectively predict future falls. [12–16]

For a rehabilitation tool, the immediate opportunity is a transparent, protocol-aware balance profile. Automated fall probabilities, definitive safe/unsafe labels and claims that changing a balance score prevents fractures require considerably more validation than the current assessment literature provides.

Scope and evidence approach

The main population is adults undergoing primary elective THA for hip osteoarthritis (OA). Preoperative observations are included when they establish baseline status or precede a later postoperative endpoint. Fracture-related THA, hemiarthroplasty, revision, resurfacing and simultaneous bilateral surgery are separate populations. A mixed THA/TKA study can inform measurement reasoning while remaining indirect for THA-specific numerical thresholds.

This is a critical narrative evidence review supported by an auditable search record. Three native PubMed and three native Scopus streams addressed measurement properties, prognosis/recovery and protocol/mechanism/technology. Exact PubMed streams were reconciled using official ESearch/EFetch after overlapping connector pagination; Scopus retrieval reached the provider totals. Original-source retrieval and bounded citation chasing supplemented those searches. Counts describe retrieval coverage, not eligible-study counts or independent studies. There was no registered protocol, exhaustive independent dual screening or formal systematic-review claim.

Define the population and assessment context

Surgical indication and approach

Record indication, primary versus revision status, operated side, approach, surgery date, current weight-bearing permission and precautions. These variables affect what can be tested and the interpretation of the result. An unassisted stance test from an uncomplicated elective OA cohort is not automatically appropriate after a fracture, abductor repair or restricted reconstruction. The treating team's current instructions determine eligibility; an approach name alone does not.

Surgical approach may influence early recovery, but observational comparisons are often confounded by surgeon, era, implant and soft-tissue management. Holnapy's perturbation study compared direct-lateral, anterolateral and capsule-preserving posterior procedures. That is not an isolated randomized comparison of incision location. Similarly, a study of large-head THA or resurfacing does not establish the same balance trajectory for all conventional primary THAs. [17, 18]

Contralateral disease matters for both performance and reference selection. A bilateral stance can look symmetrical because both limbs are limited. A nonoperated hip may have symptomatic or radiographic OA; contralateral knee disease and previous joint replacement can alter loading and support strategies. Record both sides and retain absolute values alongside any symmetry index.

Postoperative stage and selection

Use exact postoperative days in addition to broad categories such as inpatient, early outpatient and late follow-up. Rapid analgesic changes, fatigue, dizziness, aid use and wound discomfort can dominate an early test. Late studies commonly exclude people unable to stand unaided, those with neurological or vestibular disease, or those with other joint symptoms. Their apparently favorable scores describe selected completers rather than the entire THA population.

Pop and colleagues' study at 24–36 months included adults no older than 65, excluded severe obesity and several balance-affecting conditions, and required unaided standing. Rasch and colleagues excluded other lower-limb comorbidity; several participants could not complete preoperative gait or one-foot tests. These exclusions are central to interpretation, not minor demographic details. [5, 6]

Record pain, fatigue, recent falls, vision, footwear, medication-related symptoms, assistance and the reason for stopping. A test that is not attempted because of restrictions is different from an attempted task terminated by loss of balance. Neither should be silently converted to a zero time or omitted from an average.

What each balance measure captures

Quiet standing and weight distribution

A force platform measures the point of application of the resultant ground reaction force. CoP displacement is not identical to center-of-mass (CoM) motion, and a video-derived body-sway estimate is not automatically CoP. Different systems may use the same word, sway, for displacement variability, path length, velocity, area or a proprietary composite score. The exact output and unit must be preserved.

Weight distribution and postural control are also different. Equal bilateral loading can coexist with substantial sway, and deliberately shifting weight toward the operated limb changes the task. Talis and colleagues compared loading across quiet standing, walking and sit-to-stand, illustrating task specificity; the original body was not recovered, so precise protocol-dependent values are not transferred here. [9]

Reduced sway is not invariably better performance. A shorter recording, wider base, hand contact, reduced movement or a stiffening strategy can lower a metric without improving the ability to recover from a disturbance. Conversely, a person who explores a larger stability margin during a reaching task may have more excursion because the instruction demands it. Interpret the measurement in relation to the task, not a generic smaller-is-better rule.

One leg stance and clinical scales

One-leg stance provides a simple challenge to unilateral support but has prominent floor and ceiling effects. Cap duration, arm position, lifted-leg position, visual condition, footwear, assistance and termination criteria all matter. A 10-second pass/fail test, a 30-second capped best trial and a supported force-platform trial are different assays.

Ninomiya's preoperative protocol used hands on hips, a 30-second cap and the longer of two attempts on each side. Stopping criteria included movement of the stance foot, contact of the raised foot with the floor or supporting limb, or reaching the cap. In contrast, Rasch's instrumented unilateral trials allowed a rod in the contralateral hand and averaged the two best of three recordings. The latter cannot be presented as unsupported single-leg balance. [6, 12]

The Berg Balance Scale (BBS) samples multiple observed activities; the Activities-specific Balance Confidence scale (ABC) captures perceived confidence. A high BBS score may coexist with difficulty in rapid multidirectional stepping, and high confidence does not ensure safe balance. Total TUG time is a composite of transfer, walking and turning rather than a pure standing-balance measure. Preserve the clinical purpose of each instead of selecting whichever score correlates best with another.

Anticipatory and reactive balance

FSST and step tests involve planned transfers of weight and foot placement. They are not unexpected perturbation tests. A step initiated after a known instruction differs from compensatory stepping after a sudden external disturbance in timing, attention, force demand and perceived threat.

Reactive assessment should describe perturbation direction, magnitude, predictability, support surface, stance, harness or guarding, hand contacts, stepping permission and failure rules. Holnapy studied platform responses characterized by Lehr's damping ratio, with 72 THA patients across three approach groups and 45 controls. Group recovery differed by approach, but the original abstract does not provide a validated individual MDC, clinically important difference or prospective fall-risk cutoff. Full methods were unavailable, so this remains evidence that reactive behavior may differ, not a ready-to-deploy clinical rule. [18]

Clinical test reliability and interpretation

Four Square Step Test

Batting and Barker enrolled 58 people with moderate-to-severe hip OA scheduled for primary replacement. Their repeated-measures study assessed FSST before and after surgery, with BBS and Figure of 8 Walk comparators. The original abstract reports within-rater limits of agreement of −3.2 to 3.5 seconds before THR and −1.5 to 2.0 seconds after THR; between-rater limits were −2.2 to 3.4 seconds. Correlation was stronger with Figure of 8 Walk than BBS. [1]

Feasibility is as important as the correlation. Twenty-one participants failed the preoperative FSST, whereas only one was categorized as moderate risk by BBS and only one failed the Figure of 8 Walk. This suggests that a challenging stepping test can reveal limitations missed by an easier scale while also excluding many patients from a timed-score analysis. It does not establish that FSST failure prospectively predicts a fall. Timing after surgery, exact retest spacing and handling of unsuccessful attempts could not be fully verified without the article body.

Record successful and unsuccessful sequences, cane contacts, sequence errors, foot clearance and permitted support. Do not report only the successful time after repeated undocumented attempts. The original obstacle-based test and a tape-based modification should have distinct names and separate reference values.

Modified Four Square Step Test

Horata and colleagues studied a tape-based modified FSST in 29 primary THA recipients. The original abstract reports intrarater ICC 0.925, SEM 1.56 seconds and MDC95 4.32 seconds; associations with TUG and BBS were strong. These findings support reproducibility and convergent validity in the studied sample, with no demonstrated patient-valued change threshold. Postoperative timing, precise repetition structure and retest interval remained unverified because the original body was unavailable. [2]

Replacing canes with tape removes an obstacle-clearance demand, which can improve feasibility but also changes the task. The modified test's 4.32-second MDC should not be assigned to the original FSST. Correlations with BBS and TUG indicate shared performance features, not equivalence or a gold-standard diagnosis of balance impairment.

Step test

Özden and colleagues assessed 32 unilateral THA recipients, with mean age 75.4 years. The original abstract reports ICC 0.908 for the right-side step test and 0.846 for the left, with SEM 0.37 and 0.55 and MDC95 1.02 and 1.52, respectively. The reported side labels are right and left; they cannot be silently changed to operated and nonoperated. The abstract also repeats a right-side label in its validity results, creating an unresolved side-attribution ambiguity. [3]

Step height, duration, number of attempts, support allowance and whether the named side refers to the stepping or supporting limb are necessary for replication. Without those details, the error numbers are evidence that a particular protocol was studied, not a license to use the same threshold for any step-count test.

Six Spot Step Test

Aygun-Polat and colleagues' 2026 issue publication, available online in 2025, adds useful between-day evidence. Sixty-one patients at least six months after primary THA were tested within a day, after 5–7 days and between raters. The abstract reports ICCs 0.95, 0.94 and 0.95, respectively, and corresponding agreement intervals of −1.7 to 2.3, −2.3 to 2.0 and −2.0 to 1.9 seconds. [4]

The reported 10.7-second fall-risk cutoff had AUC 0.82, sensitivity 75% and specificity 73% in the development sample. However, the design is explicitly cross-sectional, and the accessible abstract does not fully identify the fall-status reference or ascertainment. This cutoff must not be presented as validated prediction of the next year's falls. The relatively high apparent discrimination is encouraging for further study, while external validation and full protocol appraisal remain necessary.

Compact measurement table

Table 1 Compact measurement table

An ICC describes relative ordering, not the size of an individual's expected error. Limits of agreement include the observed bias and spread of paired measurements. A confidence-based MDC is not a patient-valued MIC, and a group standardized response mean is not either of these. No verified anchor-based MIC for the specific THA balance protocols above was established in the appraised original sources. That is a bounded evidence statement, not proof that no relevant study exists anywhere.

Instrumented balance and technology

What original force platform studies show

Pop and colleagues compared 55 unilateral THA recipients with 48 controls at 24–36 months after a lateral approach. Testing used a calibrated platform at 62 Hz, barefoot self-selected stance, reproduced foot outlines, arms by the side, 30-second eyes-open and eyes-closed recordings in random order, and a visual target during eyes-open testing. Up to three attempts were allowed after balance loss. [5]

The THA group had greater mediolateral CoP velocity with eyes open; eyes-closed testing showed additional differences in anteroposterior velocity and path length, but not path area. For example, eyes-closed mean path length was 52.36 versus 40.32 cm. These are group observations from selected younger, uncomplicated patients. No prospective falls were assessed, and no individual error or diagnostic threshold was established. The within-group relation between time since surgery and sway is cross-sectional and cannot establish within-person recovery.

Eyes-closed testing challenges reliance on vision, but a larger change cannot specifically identify damaged hip proprioceptors. Vestibular function, distal sensation, attention, stance strategy and strength also contribute. Direct joint-position-sense measurement is a separate construct; balance is not a selective assay of hip proprioception.

Rasch's 20-patient longitudinal study found improved bilateral sway, particularly with eyes closed, while hip-abductor weakness persisted at two years. Its 30-second recordings, 20-cm foot gap, selection of the two best recordings and hand-supported unilateral protocol differ from Pop's setup. The fact that strength, quiet stance and gait recovered differently is more useful than trying to reconcile them into one global recovery score. [6]

Clinical scales versus force platform measures

Jogi and colleagues evaluated primary OA THA/TKA recipients after a five-week home rehabilitation program. Of 63 participants, 54 completed the clinical tests and 50 completed platform testing. The initial sample included 30 THA and 33 TKA participants; many reported correlations pooled both procedures. Force-platform recordings used two 10-second trials of comfortable standing, anterior lean and posterior lean, with eyes open and a nearby walker released for recording. [7]

BBS had moderate relationships with platform ellipse area, whereas confidence and self-reported function had weaker relationships. The reported correlation magnitudes do not quantify agreement or provide a conversion between scores and platform area; this report therefore does not use them as calibration constants. The central result is that clinical performance, confidence and a short laboratory stance do not provide interchangeable information. Pooling THA and TKA to meet a sample-size target further limits THA-specific inference.

Jogi's earlier comparison of original and reduced BBS versions included 26 THA and 28 TKA participants. Reduced scores were derived from the administered original scales, not independently administered instruments. Similar correlations or standardized response means can therefore support efficient score construction while failing to establish an independent administration's reliability or MIC. The full body was unavailable for further appraisal. [8]

Portable sensors and markerless cameras

Gras and colleagues studied 42 patients approximately three weeks after primary THA using a lumbar inertial sensor across different unstable surfaces. The measured challenge increased nonlinearly across foam conditions and did not simply track ability to complete a 20-second operated-limb stance. Device labels such as unstable or advanced are not calibrated doses of balance demand. This abstract-level study evaluated task challenge, not agreement of the inertial estimate with a simultaneous reference or prospective falls. [11]

Roustemis and colleagues used a RealSense camera with Kinetisense software in 20 OA patients before and one year after posterolateral THA. Single-leg stance lasted 20 seconds, eyes open, no external support, with the best of two attempts selected. The source reported reduced sway estimates after surgery. However, it did not compare these balance outputs with a simultaneous force-platform or motion-capture reference and did not establish test-retest error or clinically important change. Its reported comparison-group descriptions and some numerical summaries are inconsistent. The appropriate conclusion is feasibility and exploratory recovery description, not established clinical accuracy. [10]

For a tool, validate the precise variable intended for display. Camera position, frame rate, calibration, occlusion, clothing, sensor location, support contacts, algorithm version and invalid-recording rules belong in the protocol. CoM sway inferred from lumbar motion and CoP from a plate should have distinct labels. A model should be evaluated across the actual patients, devices, aids and stages in which it will be used.

Falls and prognosis

Prospective observation

Ninomiya and colleagues measured 157 primary unilateral OA THA patients one month before surgery and followed falls during the first postoperative year. Participants were 45–75 years old, predominantly women, operated through an anterolateral approach and discharged home within five days. Falls were collected using questionnaires alongside follow-up; the report does not describe the same monthly calendar method used by Hill. Thirty-two patients fell. [12]

Preoperative one-leg stance times did not distinguish fallers from nonfallers on either side: affected-side means were 14.48 and 13.06 seconds, respectively. This does not prove that balance never predicts falls. It means that this capped preoperative task did not discriminate in this selected cohort. Prior falls and lower abductor strength were more informative. The proposed 0.46 Nm/kg threshold had AUC 0.702, sensitivity 73.6% and specificity 50.0%, with unclear coding/scaling of the regression odds ratios. The article's discussion misinterprets the AUC as a probability of falling; that interpretation is not retained here.

Hill and colleagues provide a useful complementary ascertainment model: 167 adults aged at least 60 years after primary elective THR were followed with calendars and monthly calls for 12 months. The original abstract reports 67 fallers, reported as 42%, with 140 falls and nine fractures. The abstract does not explain the percentage denominator; 67 divided by the stated baseline 167 is 40.1%. The difference from Ninomiya's incidence should not be attributed solely to balance impairment: age, selection, exposure and ascertainment differ. Full methods were not retrieved, and the study does not validate one of the balance cutoffs above. [13]

Ikutomo and colleagues followed 162 women from an initial THA cohort of 286. Gait abnormality measured three weeks after surgery was associated with subsequent falls, with a reported hazard ratio of 2.91. This is a prospective cohort, but detailed gait-abnormality scoring and ascertainment remained unavailable in the original abstract. Whether falls in the first three postoperative weeks were excluded from the risk window after the index assessment also requires full-text confirmation. A hazard ratio is not an individual absolute probability and cannot be used as a balance-scale threshold. [14]

Retrospective and concurrent evidence

Konishi and colleagues surveyed patients more than three years after primary OA THA, with mean follow-up 6.9 years. Four hundred and one of 673 eligible patients responded, representing 438 hips. Preoperative physical tests preceded the later survey, but ABC confidence and prior-year falls were reported concurrently; fracture history covered the postoperative interval rather than the same one-year window. [15]

The ABC cutoff of 90.3 had AUC 0.703, sensitivity 61.6% and specificity 71.1% for identifying recalled falls. Fracture discrimination was weaker, AUC 0.599. A cluster with both falls and high confidence had more reported fractures, but the cluster used fall history itself and does not establish that overconfidence caused subsequent fractures. Falls are person-level events; analyses expressed in hips, including bilateral cases, require attention to within-person dependence. The available methods do not clearly establish patient-level clustering adjustment. Response selection, recall, missing outcomes and different exposure windows also limit prediction.

In the 12-patient THA subgroup studied by Lin and colleagues, current measurements at approximately one year were related to recalled falls over the preceding year. The high in-sample explained variance from strength and fall efficacy is not prospective prognosis. Fall-related changes in confidence or behavior could precede the measurement, and multivariable selection in 12 patients is highly unstable. [16]

Compact prognosis table

Table 2 Compact prognosis table

Practical protocol and decision framework

Table 3 Practical protocol and decision framework

Start with the least demanding assessment that answers the clinical question, and progress only within the person's restrictions and observed safety. Provide close guarding where appropriate; do not use unsupervised unsupported or eyes-closed testing merely to produce a more challenging score. A near fall, new inability or adverse symptom is clinically relevant even when no complete numerical result is produced.

Display the underlying score, side, support condition and exact date. Flag protocol changes rather than merging them invisibly into a trend. A capped stance time should be shown as at least the cap, not an exact maximum capacity. An unsuccessful FSST should remain an unsuccessful trial with its reason, not a time imputed from successful patients.

For a future predictive product, define index time, future outcome and horizon first. Retain complications, noncompleters and changing aids; collect falls prospectively using consistent definitions and repeated ascertainment; validate calibration and discrimination outside the development site. Compare added value against age, prior falls and routine clinical information. Repeated measurements and both hips from one person must remain together during data splitting and analysis.

Conclusions

THA balance assessment is useful now when the task, surgical stage and support conditions are explicit. Clinical stepping tasks have promising direct reliability evidence, while force platforms and portable technologies can reveal additional movement behavior. The limiting issue is not a lack of measurable signals but overextension of those signals into unverified clinical meaning. Measurement error, patient-valued improvement, future falls and fracture risk are separate questions and should remain visibly separate in clinical reporting and rehabilitation-tool design.

Primary study characteristics

Original studies and explicitly labeled contextual sources are grouped by question. Population, postoperative stage, protocol, endpoint, source access and analysis design constrain interpretation. Publications from the same cohort are not independent replications.

Framework and contextual evidence

Labanca L 2021

Study and population [17] Systematic review used for context and discovery. 41 studies; heterogeneous procedures, stages and balance constructs

Protocol and timing Review of static, dynamic, clinical-scale and proprioception assessments

Principal findings Residual deficits appear in several tasks; extensive heterogeneity

Interpretive limits Not direct validation of a universal balance/fall cutoff; reliability study excluded; score-defined risk is not observed falls

Source examined Original article body retrieved

Clinical measurement properties and important change

Batting M 2019

Study and population [1] Direct measurement study; original abstract. 58 people with moderate/severe hip OA scheduled for primary THR within four months

Protocol and timing FSST before/after THR; within/between-rater repeat; BBS and Figure of 8 Walk comparators

Principal findings Within-rater LoA −3.2 to 3.5 s preoperative and −1.5 to 2.0 s postoperative; between-rater −2.2 to 3.4 s; 21 preoperative failures

Interpretive limits Full timing/protocol and failure handling unavailable; does not establish prospective falls; obstacle test differs from tape version

Source examined Original abstract only; full article not retrieved

Horata ET 2024

Study and population [2] Direct measurement study; original abstract. 29 primary THA recipients; mean age 56.86 years; stage not verified

Protocol and timing Modified FSST uses tape instead of canes; intrarater repeat; TUG and BBS

Principal findings ICC .925; SEM 1.56 s; MDC95 4.32 s; correlations with TUG .881 and BBS −.718

Interpretive limits No MIC or prospective prognosis; full retest interval and procedures unavailable; modification changes obstacle demand

Source examined Original abstract only; full article not retrieved

Özden F 2020

Study and population [3] Direct measurement study; original abstract. 32 unilateral THA; mean age 75.4 years

Protocol and timing Two FTST and step-test repetitions; exact postoperative stage/step setup unverified

Principal findings Step ICC right .908/left .846; SEM .37/.55 and MDC95 1.02/1.52; 5STS ICC .987, MDC95 2.91 s

Interpretive limits Right/left are not operated/nonoperated; duplicated right-side validity label in abstract; no patient anchor

Source examined Original abstract only; full article not retrieved

Aygun Polat E 2026

Study and population [4] Direct measurement and cross-sectional discrimination; original abstract. 61 primary THA recipients at least six months postoperative

Protocol and timing Six-Spot Step Test twice within one day, after 5–7 days, and with a second rater

Principal findings ICC 0.95 within-day, 0.94 between-day and 0.95 interrater; corresponding LoA −1.7 to 2.3, −2.3 to 2.0 and −2.0 to 1.9 s; development cutoff 10.7 s, AUC 0.82

Interpretive limits Fall-status reference details unavailable; explicitly cross-sectional, not future-fall validation; issue publication 2026 and online publication 2025

Source examined Original abstract only; full article not retrieved

Prognostic questions and temporal ordering

Ninomiya K 2020

Study and population [12] Prospective falls cohort and threshold development. 157 primary unilateral OA THAs; 142 women; age 45–75; anterolateral approach; discharge home within five days

Protocol and timing One month preoperative; supine HHD 2.5 cm above malleolus, best of three, spina-malleolar lever and body-mass normalization; SLS hands on hips, 30-second cap, best of two; first-year falls questionnaire

Principal findings 32 fallers; affected abductor torque 0.45 versus 0.60 Nm/kg; SLS did not differ; cutoff 0.46 Nm/kg, AUC 0.702, sensitivity 73.6%, specificity 50%

Interpretive limits Selected cohort; unclear odds-ratio coding/scaling; discussion incorrectly calls AUC a 70.2% individual probability; no external validation; questionnaire not verified as monthly calendars

Source examined Original article body retrieved

Hill AM 2021

Study and population [13] Prospective falls cohort; original abstract. 167 adults aged at least 60 after primary elective THR; Perth; 54.4% female

Protocol and timing Predischarge baseline; 12-month falls calendars and monthly telephone follow-up

Principal findings 67 fallers, reported as 42%; 140 falls; nine fractures; the abstract does not explain the percentage denominator

Interpretive limits Not validation of a particular balance threshold; complete methods unavailable; ascertainment and selection differ from Ninomiya; 67/167 equals 40.1%

Source examined Original abstract only; full article not retrieved

Ikutomo H 2018

Study and population [14] Prospective falls association; original abstract. 162 women analyzed from 286 THA patients with severe hip OA

Protocol and timing Gait abnormality at three weeks; falls during the first postoperative year; Cox model

Principal findings 31.5% fell; gait abnormality hazard ratio 2.91, 95% CI 1.55–5.48

Interpretive limits Selection and scoring details require full text; whether falls in postoperative weeks 0–3 were excluded from the post-index risk window is unclear; hazard ratio is not individual absolute risk; not a clinical balance-scale cutoff

Source examined Original abstract only; full article not retrieved

Konishi T 2025

Study and population [15] Retrospective cohort with concurrent confidence discrimination. 401 of 673 patients responded, representing 438 hips; primary OA; posterolateral approach; mean 6.9 years postoperative

Protocol and timing Preoperative peak torque, best of two after practice; late survey of ABC, preceding-year falls and postoperative fracture history

Principal findings ABC falls cutoff 90.3%, AUC 0.703, sensitivity 61.6%, specificity 71.1%; fracture AUC 0.599; preoperative knee torque associated with fewer falls

Interpretive limits ABC and fall history concurrent; fractures use a longer exposure window; 60% response; bilateral hip-level dependence; clustering used falls itself; no external prediction validation

Source examined Original article body retrieved

Lin X 2022

Study and population [16] Concurrent/retrospective association, not prospective prognosis. 12 THA patients assessed around one year after surgery, plus OA-faller and healthy comparison groups

Protocol and timing Current muscle force, fall efficacy and treadmill gait related to recalled falls during the preceding year

Principal findings Strong in-sample relation of strength and fall efficacy to prior falls; high regression R²

Interpretive limits Tiny selected THA sample; backward regression, reverse causation and recall; not a model of the following year

Source examined Original article body retrieved

Instrumented assessment and recovery

Pop T 2018

Study and population [5] Cross-sectional case-control instrumented balance. Pop et al.; 55 unilateral OA THAs and 48 controls; 24–36 months; lateral approach; age at most 65; unaided standing required

Protocol and timing 62-Hz force platform; barefoot self-selected stance with traced feet and arms by sides; randomized 30-second eyes-open/closed trials; target two metres away; up to three attempts after balance loss

Principal findings Greater mediolateral velocity with eyes open; anteroposterior/mediolateral velocity and path length with eyes closed; eyes-closed path 52.36 versus 40.32 cm; area not different

Interpretive limits Cross-sectional, not longitudinal recovery or individual error/MIC; selected sample; visual-condition effect not specific to hip proprioception; no prospective falls

Source examined Original article body retrieved

Rasch A 2010

Study and population [6] Longitudinal strength/gait/balance cohort. 20 primary unilateral OA THAs; posterior approach; baseline, six months and two years; no other lower-limb comorbidity

Protocol and timing Fixed strain-gauge force testing; mean of two 3–5-second efforts; hip actions standing and knee actions seated; 30-second platform recordings; 20-cm foot gap; unilateral rod support; mean of two best of three

Principal findings Persistent 15% abductor-force deficit at two years; other muscle deficits narrowed; bilateral sway improved, particularly with eyes closed

Interpretive limits Force in N, not torque; hand-supported unilateral balance; preoperative noncompletion; averages of different muscle actions; no patient-level MIC or prognostic model

Source examined Original article body retrieved

Jogi P 2017

Study and population [7] Concurrent comparison of clinical and instrumented measures. Initially 30 THA and 33 TKA recipients; 54 clinical completers and 50 platform completers; primary unilateral OA; assessed after a five-week home rehabilitation program

Protocol and timing BBS, TUG, ABC and WOMAC; two 10-second eyes-open trials of comfortable standing and anterior/posterior lean; nearby walker released; mean of two

Principal findings Moderate BBS–platform relation and weaker confidence/self-report relation; measures capture different constructs

Interpretive limits Pooled THA/TKA; correlation is not agreement; no MDC, MIC or prospective falls; missing platform trials

Source examined Original article body retrieved; numerical table bodies incomplete in extraction

Jogi P 2011

Study and population [8] Responsiveness/score-reduction comparison; original abstract. 26 THA and 28 TKA recipients, before and after 5–7 weeks of home exercise

Protocol and timing Original BBS and WOMAC administered; reduced scores extracted from the original versions

Principal findings High original/reduced correlations and similar standardized response means

Interpretive limits Shared item data; shorter forms not independently administered; mixed joints; no patient-anchored MIC

Source examined Original abstract only; full article not retrieved

Talis VL 2008

Study and population [9] Cross-sectional task-specific loading comparison; original abstract. Unilateral THA recipients and controls; the primary abstract does not specify postoperative-stage distribution. The heterogeneous-stage description comes from the secondary review

Protocol and timing Quiet standing, walking and sit-to-stand bilateral force measurements

Principal findings Loading asymmetry differs by task

Interpretive limits Original body unavailable; no verified individual-change or future-harm threshold

Source examined Original abstract only; full article not retrieved

Roustemis AG 2024

Study and population [10] Exploratory longitudinal camera-based recovery. 20 OA patients; posterolateral THA; before and one year after surgery; comparison-group reporting inconsistent

Protocol and timing RealSense D435 camera and Kinetisense; 20-second eyes-open single-leg stance without external support, best of two; hip/spine ROM

Principal findings Camera-derived sway decreased after THA

Interpretive limits No simultaneous criterion-reference comparison or test-retest/MDC/MIC; inconsistent control descriptions and some numerical summaries

Source examined Original article body retrieved

Gras N 2024

Study and population [11] Cross-sectional task-challenge study; original abstract. 42 patients 23 ± 6 days after unilateral primary THA

Protocol and timing Lumbar inertial sensor; foam and oscillatory surfaces relative to hard standing; groups defined by 20-second single-leg stance

Principal findings Challenge varied nonlinearly across devices; no difference between groups defined by stance completion

Interpretive limits Task challenge, not criterion validation or outcome prediction; no demonstrated equivalence to CoP; full methods unavailable

Source examined Original abstract only; full article not retrieved

Holnapy Gergely 2013

Study and population [18] Longitudinal reactive-balance approach comparison; original abstract. 25 direct-lateral, 22 anterolateral and 25 posterior THA recipients; 45 controls

Protocol and timing Sudden unidirectional platform perturbation; Lehr damping ratio; bilateral, operated-side and nonoperated-side stance; baseline, six weeks, 12 weeks and six months

Principal findings Approach-associated response trajectories; posterior group approximated controls by six months

Interpretive limits Capsule preservation and approach confounded; no individual error, MIC or future-fall threshold; full protocol unavailable

Source examined Original abstract only; full article not retrieved

Standing balance methods and search appendix

This report is a critical narrative review, not a registered systematic review. Broad database retrieval was followed by targeted appraisal of original measurement, protocol, longitudinal recovery and prognostic evidence. Screening and extraction were not performed by two independent reviewers. No formal risk-of-bias score or GRADE certainty assessment was newly generated. Appraisal instead records population selection, task definition, reliability design, error/anchor interpretation, temporal ordering, model development and validation, and source-access limitations beside each source.

Primary elective THA for hip osteoarthritis is the principal population. Preoperative studies were eligible for baseline measurement and later postoperative endpoints. Revision, fracture-related replacement, hemiarthroplasty, resurfacing and bilateral surgery were treated as separate contexts. Mixed hip/knee or OA/osteonecrosis sources remain explicitly labeled. Treatment studies were considered only when they contributed relevant measurement or recovery information; this is not a treatment-effectiveness review.

Search date and retrieval coverage

The official PubMed union across the three streams was 489 unique records, while the Scopus union was 852. These overlap within and across databases. They must not be summed or presented as eligible-study counts. The smaller connector-only PubMed union (372) was superseded by official reconciliation.

Table 4 Search retrieval coverage

Review reference lists and bounded citation-network retrieval were used to identify additional relevant originals. These reference searches were not exhaustive citation graphs. Targeted web/publisher/repository retrieval was used to resolve known papers and source gaps; broad database queries were not rerun during manuscript preparation.

Exact executed queries

Standing balance measurement PubMed

("Arthroplasty, Replacement, Hip"[MeSH Terms] OR "total hip arthroplast*"[Title/Abstract] OR "total hip replacement*"[Title/Abstract]) AND ("Postural Balance"[MeSH Terms] OR (balance[Title/Abstract] OR "postural control"[Title/Abstract] OR "postural sway"[Title/Abstract] OR "postural stability"[Title/Abstract] OR "single leg"[Title/Abstract] OR "single-leg"[Title/Abstract] OR "one leg"[Title/Abstract] OR "one-leg"[Title/Abstract] OR "center of pressure"[Title/Abstract] OR "centre of pressure"[Title/Abstract] OR posturograph*[Title/Abstract] OR "functional reach"[Title/Abstract] OR "step test"[Title/Abstract] OR BESTest[Title/Abstract] OR Berg[Title/Abstract] OR "sensory organization"[Title/Abstract])) AND (reliab*[Title/Abstract] OR valid*[Title/Abstract] OR reproducib*[Title/Abstract] OR psychometr*[Title/Abstract] OR clinimetr*[Title/Abstract] OR agreement[Title/Abstract] OR "measurement error"[Title/Abstract] OR "standard error"[Title/Abstract] OR "minimal detectable"[Title/Abstract] OR "minimum detectable"[Title/Abstract] OR "smallest detectable"[Title/Abstract] OR "minimal important"[Title/Abstract] OR "minimally important"[Title/Abstract] OR "minimum important"[Title/Abstract] OR responsiv*[Title/Abstract] OR interpretabil*[Title/Abstract] OR "floor effect"[Title/Abstract] OR "ceiling effect"[Title/Abstract]) AND ("1800/01/01"[Date - Publication] : "2026/10/02"[Date - Publication])

Standing balance measurement Scopus

TITLE-ABS-KEY(("total hip arthroplast*" OR "total hip replacement*" OR THA OR THR) AND (balance OR "postural control" OR "postural sway" OR "postural stability" OR "single leg" OR "single-leg" OR "one leg" OR "one-leg" OR "center of pressure" OR "centre of pressure" OR posturograph* OR "functional reach" OR "step test" OR BESTest OR Berg OR "sensory organization") AND (reliab* OR valid* OR reproducib* OR psychometr* OR clinimetr* OR agreement OR "measurement error" OR "standard error" OR "minimal detectable" OR "minimum detectable" OR "smallest detectable" OR "minimal important" OR "minimally important" OR "minimum important" OR responsiv* OR interpretabil* OR "floor effect" OR "ceiling effect")) AND PUBYEAR BEF 2027

Standing balance prognosis PubMed

("Arthroplasty, Replacement, Hip"[MeSH Terms] OR "total hip arthroplast*"[Title/Abstract] OR "total hip replacement*"[Title/Abstract]) AND ("Postural Balance"[MeSH Terms] OR (balance[Title/Abstract] OR "postural control"[Title/Abstract] OR "postural sway"[Title/Abstract] OR "postural stability"[Title/Abstract] OR "single leg"[Title/Abstract] OR "single-leg"[Title/Abstract] OR "one leg"[Title/Abstract] OR "one-leg"[Title/Abstract] OR "center of pressure"[Title/Abstract] OR "centre of pressure"[Title/Abstract] OR posturograph*[Title/Abstract] OR "functional reach"[Title/Abstract] OR "step test"[Title/Abstract] OR BESTest[Title/Abstract] OR Berg[Title/Abstract] OR "sensory organization"[Title/Abstract])) AND (prognos*[Title/Abstract] OR predict*[Title/Abstract] OR longitudinal[Title/Abstract] OR prospective[Title/Abstract] OR cohort[Title/Abstract] OR "follow up"[Title/Abstract] OR "follow-up"[Title/Abstract] OR recovery[Title/Abstract] OR deteriorat*[Title/Abstract] OR fall*[Title/Abstract] OR "natural history"[Title/Abstract] OR "return to work"[Title/Abstract] OR discharge[Title/Abstract] OR "risk factor"[Title/Abstract]) AND ("1800/01/01"[Date - Publication] : "2026/10/02"[Date - Publication])

Standing balance prognosis Scopus

TITLE-ABS-KEY(("total hip arthroplast*" OR "total hip replacement*" OR THA OR THR) AND (balance OR "postural control" OR "postural sway" OR "postural stability" OR "single leg" OR "single-leg" OR "one leg" OR "one-leg" OR "center of pressure" OR "centre of pressure" OR posturograph* OR "functional reach" OR "step test" OR BESTest OR Berg OR "sensory organization") AND (prognos* OR predict* OR longitudinal OR prospective OR cohort OR "follow up" OR "follow-up" OR recovery OR deteriorat* OR fall* OR "natural history" OR "return to work" OR discharge OR "risk factor")) AND PUBYEAR BEF 2027

Standing balance protocol mechanism technology PubMed

("Arthroplasty, Replacement, Hip"[MeSH Terms] OR "total hip arthroplast*"[Title/Abstract] OR "total hip replacement*"[Title/Abstract]) AND ("Postural Balance"[MeSH Terms] OR (balance[Title/Abstract] OR "postural control"[Title/Abstract] OR "postural sway"[Title/Abstract] OR "postural stability"[Title/Abstract] OR "single leg"[Title/Abstract] OR "single-leg"[Title/Abstract] OR "one leg"[Title/Abstract] OR "one-leg"[Title/Abstract] OR "center of pressure"[Title/Abstract] OR "centre of pressure"[Title/Abstract] OR posturograph*[Title/Abstract] OR "functional reach"[Title/Abstract] OR "step test"[Title/Abstract] OR BESTest[Title/Abstract] OR Berg[Title/Abstract] OR "sensory organization"[Title/Abstract])) AND (protocol[Title/Abstract] OR biomechan*[Title/Abstract] OR kinematic*[Title/Abstract] OR kinetic*[Title/Abstract] OR symmetr*[Title/Abstract] OR asymmetr*[Title/Abstract] OR "weight bearing"[Title/Abstract] OR "weight-bearing"[Title/Abstract] OR "ground reaction"[Title/Abstract] OR "force plate"[Title/Abstract] OR "force platform"[Title/Abstract] OR sensor*[Title/Abstract] OR wearable*[Title/Abstract] OR inertial[Title/Abstract] OR acceleromet*[Title/Abstract] OR markerless[Title/Abstract] OR "motion capture"[Title/Abstract] OR camera[Title/Abstract] OR video[Title/Abstract] OR algorithm[Title/Abstract] OR electromyogra*[Title/Abstract] OR activation[Title/Abstract] OR "sampling frequency"[Title/Abstract] OR "filter cutoff"[Title/Abstract]) AND ("1800/01/01"[Date - Publication] : "2026/10/02"[Date - Publication])

Standing balance protocol mechanism technology Scopus

TITLE-ABS-KEY(("total hip arthroplast*" OR "total hip replacement*" OR THA OR THR) AND (balance OR "postural control" OR "postural sway" OR "postural stability" OR "single leg" OR "single-leg" OR "one leg" OR "one-leg" OR "center of pressure" OR "centre of pressure" OR posturograph* OR "functional reach" OR "step test" OR BESTest OR Berg OR "sensory organization") AND (protocol OR biomechan* OR kinematic* OR kinetic* OR symmetr* OR asymmetr* OR "weight bearing" OR "weight-bearing" OR "ground reaction" OR "force plate" OR "force platform" OR sensor* OR wearable* OR inertial OR acceleromet* OR markerless OR "motion capture" OR camera OR video OR algorithm OR electromyogra* OR activation OR "sampling frequency" OR "filter cutoff")) AND PUBYEAR BEF 2027

Original source access and appraisal

For balance, special checks included the distinctions between original and tape-modified FSST, right/left versus operated/nonoperated step labels, supported versus unsupported unilateral stance, CoP versus CoM, cross-sectional discrimination versus future prediction, and fall-window alignment. For strength, checks included force versus torque, normalization, 1RM versus power, RFD assay definition, approach/stage selection and whether a PROM anchor defined a strength MIC.

Numerical claims were tied to their original stage and protocol. Reported versus derived values are distinguished; no new pooled effect was calculated. The report does not treat a statistically significant group change as an individual response, an MDC as an MIC, a correlation as agreement, a regression association as causation, an AUC as a probability, or a study using previously recorded outcomes as prospective follow-up from the index test.

Limitations of this review

Access was incomplete for several primary psychometric and rapid-force papers. The absence of a verified threshold in the appraised sources is not proof that no such evidence exists. Some relevant abstracts were indexed later than their online publication year; citations preserve publication details from the original record. Source text and table discrepancies remain visible rather than being silently corrected. Selection of deeply appraised papers was purposive and clinically focused, so the bibliography is not an exhaustive inventory of all studies in the retrieval sets.

Standing balance source access appendix

Full-text retrieval used permitted literature-connector, open publisher, PMC/Europe PMC and institutional repository routes. Some requests returned unavailable content, provider errors, rate limits, or blocked pages. An unverified open-access URL was not treated as successful retrieval.

References

References are numbered in first citation order. Source descriptions identify the material examined and do not constitute a study quality rating. Each link identifies the original publication or explicitly named primary source version.

1. Batting M, Barker KL. Reliability and validity of the Four Square Step Test in patients with hip osteoarthritis before and after total hip replacement. Physiotherapy. 2019;105(2):244-253. DOI 10.1016/j.physio.2018.07.014 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-bf8a50ce5d1b Batting M 2019

2. Horata ET, Eken F, Yesil M, Ozcan O. Validity and reliability of the modified four square step test in total HIP arthroplasty. Journal of bodywork and movement therapies. 2024;40:345-349. DOI 10.1016/j.jbmt.2024.04.028 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-4d070e5d8e70 Horata ET 2024

3. Özden F, Coşkun G, Bakırhan S. The test-retest reliability and concurrent validity of the five times sit to stand test and step test in older adults with total hip arthroplasty. Experimental gerontology. 2020;142:111143. DOI 10.1016/j.exger.2020.111143 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-30ef0036351f Ozden F 2020

4. Aygun-Polat E, Polat Y, Ozbilen YE, Altun A, Acaroz S, Cirakli A. Reliability and validity of the Six-Spot Step Test in patients with total hip arthroplasty. Physiotherapy. 2026;130:101848. DOI 10.1016/j.physio.2025.101848 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-649d2d7ace8d Aygun-Polat E 2026

5. Pop T, Szymczyk D, Majewska J, Bejer A, Baran J, Bielecki A, et al. The Assessment of Static Balance in Patients after Total Hip Replacement in the Period of 2-3 Years after Surgery. BioMed research international. 2018;2018:3707254. DOI 10.1155/2018/3707254 Source examined: Original article body retrieved.

Source note: SRC-5d5f8c1f4242 Pop T 2018

6. Rasch A, Dalén N, Berg HE. Muscle strength, gait, and balance in 20 patients with hip osteoarthritis followed for 2 years after THA. Acta orthopaedica. 2010;81(2):183-8. DOI 10.3109/17453671003793204 Source examined: Original article body retrieved.

Source note: SRC-166903727563 Rasch A 2010

7. Jogi P, Overend T, Kramer J. Comparisons of clinically based outcome measures and laboratory-based outcome measure for balance in patients following total hip and knee arthroplasty. Orthopedic research and reviews. 2017;9:23-33. DOI 10.2147/orr.s125581 Source examined: Original article body retrieved; numerical table bodies incomplete in extraction.

Source note: SRC-0bb03569e0e8 Jogi P 2017

8. Jogi P, Spaulding SJ, Zecevic AA, Overend TJ, Kramer JF. Comparison of the original and reduced versions of the Berg Balance Scale and the Western Ontario and McMaster Universities Osteoarthritis Index in patients following hip or knee arthroplasty. Physiotherapy Canada. Physiotherapie Canada. 2011;63(1):107-14. DOI 10.3138/ptc.2009-26 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-f24c290c2816 Jogi P 2011

9. Talis VL, Grishin AA, Solopova IA, Oskanyan TL, Belenky VE, Ivanenko YP. Asymmetric leg loading during sit-to-stand, walking and quiet standing in patients after unilateral total hip replacement surgery. Clinical biomechanics (Bristol, Avon). 2008;23(4):424-33. DOI 10.1016/j.clinbiomech.2007.11.010 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-7a15c7b8fd2f Talis VL 2008

10. Roustemis AG, Gavriil P, Skouras AZ, Melissaridou D, Sioutis S, Trikoupis I, et al. Assessment of Hip and Lumbar Spine Range of Motion After Total Hip Arthroplasty Using a Single Camera Markerless System. Cureus. 2024;16(7):e65875. DOI 10.7759/cureus.65875 Source examined: Original article body retrieved.

Source note: SRC-34fb80e05d37 Roustemis AG 2024

11. Gras N, Brauner T, Wearing S, Horstmann T. Do increasingly unstable balance devices provide a graded challenge to bipedal stance in total hip arthroplasty patients? Gait & posture. 2024;108:9-14. DOI 10.1016/j.gaitpost.2023.11.004 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-090d0feb7462 Gras N 2024

12. Ninomiya K, Takahira N, Ikeda T, Suzuki K, Sato R, Hirakawa K. Predictors of falls in patients during the first year after total hip arthroplasty: A prospective cohort study. Health science reports. 2020;3(3):e184. DOI 10.1002/hsr2.184 Source examined: Original article body retrieved.

Source note: SRC-a75d340b9e57 Ninomiya K 2020

13. Hill AM, Ross-Adjie G, McPhail SM, Jacques A, Bulsara M, Cranfield A, et al. Incidence and Associated Risk Factors for Falls in Older Adults Postdischarge Who Undergo Elective Total Hip Replacement Surgery-A Prospective Cohort Study. The journals of gerontology. Series A, Biological sciences and medical sciences. 2021;76(10):1814-1820. DOI 10.1093/gerona/glaa283 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-eb3200138bfd Hill AM 2021

14. Ikutomo H, Nagai K, Tagomori K, Miura N, Nakagawa N, Masuhara K. Gait Abnormality Predicts Falls in Women After Total Hip Arthroplasty. The Journal of arthroplasty. 2018;33(10):3215-3219. DOI 10.1016/j.arth.2018.05.044 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-8b3544eb6f13 Ikutomo H 2018

15. Konishi T, Hamai S, Fujita T, Hara D, Kawahara S, Motomura G, et al. Falls and Fractures After Total Hip Arthroplasty: Associations with Preoperative Physical Function and Postoperative Balance Confidence, with Insights from Cluster Analysis. JB & JS open access. 2025;10(4):e25.00233. DOI 10.2106/jbjs.oa.25.00233 Source examined: Original article body retrieved.

Source note: SRC-e60af4e6eb44 Konishi T 2025

16. Lin X, Wu W, Weijer RHA, Prins MR, van Dieën JH, Bruijn SM, et al. Strong relationship of muscle force and fall efficacy, but not of gait kinematics, with number of falls in the year after Total Hip Arthroplasty for osteoarthritis: An exploratory study. Clinical biomechanics (Bristol, Avon). 2022;92:105551. DOI 10.1016/j.clinbiomech.2021.105551 Source examined: Original article body retrieved.

Source note: SRC-de68397ec95d Lin X 2022

17. Labanca L, Ciardulli F, Bonsanto F, Sommella N, Di Martino A, Benedetti MG. Balance and proprioception impairment, assessment tools, and rehabilitation training in patients with total hip arthroplasty: a systematic review. BMC musculoskeletal disorders. 2021;22(1):1055. DOI 10.1186/s12891-021-04919-w Source examined: Original article body retrieved.

Source note: SRC-0e667b2f6b3f Labanca L 2021

18. Holnapy Gergely, Kiss Rita M. Impact of the method of exposure in total hip arthroplasty on balancing ability in response to sudden unidirectional perturbation in the first six months of the postoperative period. Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology. 2013;23(3):727-733. DOI 10.1016/j.jelekin.2013.01.005 Source examined: Original abstract only; full article not retrieved.

Source note: SRC-4643698ff655 Holnapy Gergely 2013