Score all 8 items from 0 to 3, get the total out of 24 and compare it with published fall-risk cut-offs. For clinicians. Dynamic Gait Index used with permission of Anne Shumway-Cook, Ph.D. Source: Shumway-Cook A, Woollacott M. Motor Control: Theory and Practical Applications. Baltimore: Williams & Wilkins; 1995.Reviewed by Dr. Richard Holsman, PT, DPT, NCS · Last reviewed November 13, 2026
Score all 8 items from 0 to 3. The total out of 24 updates as you go. When you’re done, print it or save it as a PDF. It takes about 10 to 15 minutes, depending on the patient.
This supports clinical judgment. It is not a diagnosis. A DGI score can point to fall risk while walking, but it can’t say for sure whether a person will fall, or why. Read it together with the rest of the exam. Results should be interpreted by the treating physical therapist.
Higher is better (maximum 24). Fall-risk cut-offs, as listed by the Shirley Ryan AbilityLab Rehabilitation Measures Database unless noted:
| Group | Cut-off | What the research found |
|---|---|---|
| Community-dwelling older adults | < 19 / 24 | Indicative of increased fall risk; sensitivity 59%, specificity 64% (Shumway-Cook et al., 1997). Increased risk of falls; sensitivity 67%, specificity 86% (Wrisley & Kumar, 2010). |
| Vestibular disorders | ≤ 19 / 24 | People scoring 19 or less were 2.58 times more likely to report a fall in the previous 6 months (95% CI 1.47–4.53) (Whitney et al., 2000; 247 people, abstract). |
| Parkinson disease | < 19 / 24 | Separated fallers from non-fallers; sensitivity 0.64, specificity 0.85 (Dibble et al., 2008). A cut score of 18.5 had sensitivity 0.68, specificity 0.708 (Landers et al., 2008). |
| Multiple sclerosis | < 19 / 24 | Sensitivity 87%, specificity 51% (Forsberg et al., 2013). A stricter cut-off of < 12 had sensitivity 45%, specificity 80% (Cattaneo et al., 2006). |
No fall cut-off is listed for stroke; for that group the tool shows the total and change values only. The SRALab form itself states: “< 19/24 = predictive of falls risk in community dwelling elderly.”
| Group | Change value |
|---|---|
| Community-dwelling older adults | MDC 2.9 points (Romero et al., 2011); MCID 1.90 points (Pardasaney et al., 2012) |
| Vestibular and balance disorders | MDC 4 points (Marchetti et al., 2014); MDC 3.2 points, peripheral vestibular (calculated from Hall & Herdman, 2006) |
| Parkinson disease | MDC 2.9 points (Huang et al., 2011) |
| Multiple sclerosis | MDC 4.19–5.54 points (calculated from Cattaneo et al., 2007) |
| Stroke | MDC 4 points (Lin et al., 2010); 2.6 points in chronic stroke (calculated from Jonsdottir & Cattaneo, 2007) |
The Functional Gait Assessment (FGA) was adapted from the DGI to reduce its ceiling effect. The DGI has a ceiling effect in vestibular disorders (Wrisley et al., 2003) and in community-dwelling older adults (Pardasaney et al., 2012), and the FGA is recommended over the DGI in stroke (Lin et al., 2010). For standing balance, see the Berg Balance Scale; for a quick mobility screen, the Timed Up and Go; for combined balance and gait in older adults, the Tinetti POMA.
Reviewed by Dr. Richard Holsman, PT, DPT, NCS (Doctor of Physical Therapy; board-certified clinical specialist in neurologic physical therapy). Last reviewed: .
Who uses this tool. This is a clinician-administered test. A physical therapist (or other trained clinician) gives it, guards the patient, and interprets the result together with the rest of the exam. It is not a self-test and it does not diagnose.
Privacy. Scoring runs only in your web browser. Nothing you enter is saved, sent to us or shared.
How references are graded. [I] the instrument or score sheet itself; [P] a primary study whose abstract or full text we read; [S] a value taken from the Shirley Ryan AbilityLab Rehabilitation Measures Database summary of the cited study (we did not read that study itself); [T] a textbook or library catalog record used to confirm a citation.
Licensing. The Dynamic Gait Index item wording and scoring are reproduced on this page with the written permission of Anne Shumway-Cook, Ph.D. (granted November 9, 2026). Original source: Shumway-Cook A, Woollacott M. Motor Control: Theory and Practical Applications. Baltimore: Williams & Wilkins; 1995. The page layout, calculator code and explanatory text are © Holsman Physical Therapy. Please obtain your own permission before reproducing the DGI elsewhere.
Review policy. Every number on this page (cut-offs, change values, norms) is taken from the sources listed above, and we leave a value out when we cannot verify it from a source we have read. We re-check each tool page at least once a year and whenever a cited source is corrected or updated. To report an error, please contact us.
Change log.
Conflict of interest and advertising. This page is published by Holsman Physical Therapy, which provides physical therapy services. The tool is free, carries no advertising and no sponsored content, and we receive no payment from the instrument’s authors, publishers or any manufacturer. Holsman Physical Therapy is not affiliated with the authors, journals or organizations cited, and they do not endorse this website.
Next step: Holsman physical therapists see patients at clinics across North Jersey, the Bronx and Brooklyn, plus in-home visits. Refer a patient or call 1-855-465-7626.
The Dynamic Gait Index (DGI) is an 8-item clinician-scored test of how well a person adjusts walking to changes in speed, head turns, obstacles, pivot turns and stairs. It was developed by Anne Shumway-Cook and Marjorie Woollacott (Motor Control: Theory and Practical Applications, 1995) to assess the likelihood of falling in older adults. The Functional Gait Assessment was later adapted from it.
Each of the 8 items is scored 0 (severe impairment) to 3 (normal). The form says to mark the lowest category that applies. The items add up to a total out of 24. Higher scores mean better balance while walking.
The Shirley Ryan AbilityLab lists a score under 19 out of 24 as indicating increased fall risk in community-dwelling older adults (Shumway-Cook et al., 1997; Wrisley and Kumar, 2010). In people with vestibular disorders, those scoring 19 or less were 2.58 times more likely to report a fall in the previous 6 months (Whitney et al., 2000).
In community-dwelling older adults the minimal detectable change is 2.9 points (Romero et al., 2011) and the minimal clinically important difference is 1.9 points (Pardasaney et al., 2012). The minimal detectable change is 2.9 points in Parkinson disease (Huang et al., 2011), 4 points in balance and vestibular disorders (Marchetti et al., 2014) and 4 points in stroke (Lin et al., 2010).
The Functional Gait Assessment was developed from the DGI to reduce its ceiling effect. The DGI has a ceiling effect in people with vestibular disorders (Wrisley et al., 2003), and the FGA is recommended over the DGI in stroke (Lin et al., 2010), as summarized by the Shirley Ryan AbilityLab.
No. Everything runs in your web browser. Nothing is sent to us or stored.
Holsman physical therapists see patients at clinics across North Jersey, the Bronx and Brooklyn, plus in-home visits.
📞 Call 1-855-465-7626