← Clinical Evidence

SRC-507aeb58eeff Paul SS 2013

Evidence source and scope

Three simple clinical tests to accurately predict falls in people with Parkinson's disease.

This note indexes a cited or audit-added source. It may be an original study, review, guideline, form or methods reference. Treat the specific design and the evidence below as authoritative; a source note is not automatically a primary study.

DOI: 10.1002/mds.25404

Bibliographic record from the original reports

Access labels embedded in these original reference strings are historical. Current access and exact checked components are stated separately below.

Paul SS, Canning CG, Sherrington C, et al. Three simple clinical tests to accurately predict falls in people with Parkinson's disease. Movement Disorders. 2013;28:655–662. DOI 10.1002/mds.25404

Paul SS, Canning CG, Sherrington C, Lord SR, Close JCT, Fung VSC. Three simple clinical tests to accurately predict falls in people with Parkinson’s disease. Movement Disorders. 2013;28:655–662. Original article

Paul SS, Canning CG, Sherrington C, Lord SR, Close JC, Fung VS. Three simple clinical tests to accurately predict falls in people with Parkinson's disease. Mov Disord. 2013;28(5):655–62. DOI 10.1002/mds.25404

Conditions: COND-PD Parkinson’s disease

Scoped audit evidence

These are source- and component-level audit summaries as of 3 October 2026. Access and comparison scope can differ by report; none certifies every result or the underlying raw data.

Parkinson’s disease — gait

Latest listed audit check, current for its stated components. Later recovery/refinement here supersedes earlier access limitations only where explicitly stated; it does not resolve unrelated source contradictions.

GaitP32andTable1P56–58 updatedafterreadingcompletelicensedHTMLcapture:MethodsON,4mcomfortablegait,two trials/bestscore,monthlydiariesandcalls,205;ResultsandTable1explicitly120fallers(59%),whereasabstractprints125(59%). Table4counts8+36+76=120. OriginaldevelopmentAUC.80(.73–.86),zero-corrected.78,weights6/3/2 andcategories0,2–6,8–11,predictedrisks17/51/85;allmatch report headlines. Source locator: archived audit evidence Limit: The residual source abstract/body inconsistency is identified; preferResults/Table1120whilepreservingabstractdiscrepancy. Priorabstract-onlyaccessstatementshouldbeupdated.

Earlier audit check 1 (historical evidence record). Its access limitations are superseded wherever the latest check above explicitly closes them. Retained for the distinct components and provenance, not as a current access statement.

GaitP32/P56-58/P93:205participants;abstract125(59%)fallers;3predictorspriorfalls/freezing/speed<1.1;AUC.80(.73-.86);category17/51/85%. LaterDuncan120/205counts independentlyconfirmed. Theoriginalabstractitselfhasinconsistent125/205percentage. Source locator: EuropePMC/PubMedprimaryabstract PMID23450694,archived source evidence Limit: Fulloriginal unavailable;no adjudication ofcorrecteventnumber;125/205=60.98% not59%. Do not label125or120definitivelycorrect.

Limitations: Only scoped comparisons, not whole-paper certification.

Parkinson’s disease — strength

STR-63-03

STR-63-01

STR-63-02

Limitations: Only current scoped checks; archived captures not authenticated against live originals.

Parkinson’s disease — transfer

205 PD, including 133 trial controls; ON home baseline, monthly falls diaries/telephone over six months. Results give 120 fallers, 1,854 falls, 413 injurious. STS 14.5±6.3 versus 11.7±4.8 s; unadjusted OR 1.11 (1.04–1.18), adjusted 1.04 (.96–1.13), p=.38. STS retained in 33% of 1,000 bootstraps. AUC .83 (.77–.88), corrected .81; three-variable model excludes STS, AUC .80 (.73–.86), corrected .78; difference p=.14.

Limitations: Only the enumerated transfer-relevant components were adjudicated. Source availability does not imply every statement, supplement, figure or raw datum was checked.

Parkinson’s disease strength

Parkinson’s disease strength component — Body fall denominator, exact force odds ratio and protocol

Results and Table 1 give 205 participants, 120 fallers and 85 nonfallers over six months; the abstract says 125 (59%). Seated spring-gauge knee force used the best of three trials per leg, then the bilateral mean. Table 1 gives OR.97 [.94, .99] per kg, p=.008, and 35.2±11.1 versus 30.7±12.0 kg. The body denominator matches Paul 2014. The report correctly uses 120 and discloses the abstract discrepancy.

Source locator: https://doi.org/10.1002/mds.25404; Library original archived audit evidence Abstract, Methods, Results and Table1

Limitations: AI-assisted comparison with archived licensed capture; transcription accuracy was not independently verified against the live publisher original.

Parkinson’s disease strength component — Shared cohort, selected tool and internal validation scope

Published recruitment, baseline rows and outcomes match Paul 2014. The 2013 Discussion cites a related analysis underway as reference 48, titled “Freezing of gait, impaired balance and impaired cognition are important fall risk factors in people with Parkinson’s disease.” Methods describes 1000 bootstrap samples, backward selection and zero-corrected AUCs. The final three-test tool uses prior falls, freezing and gait speed<1.1 m/s. Full-model strength OR.98 [.95, 1.02], p=.34, appeared in 41% of bootstrap selections. This supports a shared cohort and internal validation, without verifying individual participant IDs or inferring overlap with the 31- and 82-person machine studies.

Source locator: https://doi.org/10.1002/mds.25404; Library original archived audit evidence Statistical Analyses, Table2, Clinical Prediction Tool, Discussion and reference48

Limitations: AI-assisted comparison with archived licensed capture; transcription accuracy was not independently verified against the live publisher original.

Linked audit findings

Reports citing this source

Update record

Updating the wiki · Evidence and status guide