Dynamic Gait Index

Community Dwelling Elderly:

(Romero et al 2011; n = 42; mean age = 75.6 (range 59-88) years, Community Dwelling Elderly) 

• SEM = 1.04 points

Community Dwelling Elderly: 

(Romero et al, 2011, Community Dwelling Elderly) 

• MDC = 2.9 points

(Pardasaney, 2012; n=111; mean age=75.9(7))

Total sample

• MID (0.3xSb) = 1.23
• MID (0.5xSb) = 1.90

Baseline DGI score <21/24

• MID (0.3xSb) = 1.16
• MID (0.5 x Sb) = 1.80

Baseline DGI score >21/24

• MID (0.3xSb) = 0.40

MID (0.5xSb) = 0.60

Community-dwelling Older adults:

(Pardasaney et al, 2012; n = 111 aged 65 or older; mean age 75.9 + 7 years, 68.5% female)

• 1.90 points for total sample
• 1.80 for subjects with DGI scores < 21/24
• 0.60 for subjects with DGI scores > 21/24

Community Dwelling Elderly:=

(Shumway-Cook et al,1997; n = 44; age > 65, Community Dwelling Elderly)

• < 19 indicative of increased fall risks (Sensitivity 59%, Specificity 64%) 

Community-Dwelling Older Adults:

(Wrisley and Kumar, 2010; n = 35, mean age 72.9 + 7.8 years followed for prospective falls)

• < 19 indicates increased risk of falls (Sensitivity 67%, specificity 86%)

Community Dwelling Older Adults with Baseline Impairment:

(Jønsson et al, 2011; n = 24 (Hospital), mean age = 79.4 (6.8) years; n = 26 (outpatient rehabilitation), mean age = 76.8 (6.4) years; assessed at a 1.5 hour interval, rated by 3 PT's at each site; Danish language sample, Community Dwelling Older Adults)

Hospital:

• Excellent intrarater reliability (ICC = 0.90)
• Excellent interrater reliability (ICC = 0.92)

Outpatient Rehabilitation:

• Excellent intrarater reliability (ICC = 0.89)
• Excellent interrater reliability (ICC = 0.82)

Community Dwelling Older Adults: 

(Shumway-Cook et al, 1997; n = 44; aged 65 to 94 years, Community Dwelling Older Adults)

• Excellent, Balance Self-Perceptions Test and DGI (r = 0.76)
• Excellent, Berg Balance Scale and DGI (r = 0.67)
• Adequate, Assistive Devices History and DGI (r = -0.44)
• Adequate, History of imbalance and DGI (r = -0.46)

Community-dwelling male veterans:

(Chiu et al, 2006; n = 84, 64-88 years, mean age 75 + 6.47 years)

• Eash original 4 rating scale categories distinctly identified subjects at different ability levels
• Clear difficulty hierarchical order
• 8 items on the DGI appear to represent a single construct

Community-dwelling older adults: 

(Pardasaney et al, 2012)

• No floor effect
• Poor ceiling effect
  • 40% at maximum score at baseline
  • 70% at maximum score after intervention

Community-dwelling older adults:

(Pardasaney et al, 2012)

• Poor responsiveness
  • Effect size 0.27
  • Only people in the lower balance subgroups demonstrated change scores > MCID

Total Sample:

• Standardized Response Mean (SRM95=0.45 (0.28-0.61)

Stroke:

(Jonsdottir & Cattaneo, 2007, n = 25; mean age = 61.6 (13.1) years; mean time since stroke = 4.2 (7.5) years, range 0.5-35.3 years; Italian sample, Chronic Stroke)

• SEM for inter-rater reliability = 0.97 points 
• SEM for interrater reliability = 0.94 points

Patients with Chronic Stroke

(Alghadir, 2018; n=56; age=58.6 (9.8); mean time since onset=22.2 (18.3) months))

• Standard Error of Measure (SEM=0.71)

Chronic Stroke:

(calculated from Jonsdottir and Cattaneo, 2007)

• MDC (calculated from SEM) = 2.6 points

Stroke:

(Lin et al, 2010; n = 45; mean age = 60.7 (12.2) years; median time since stroke = 9 months (range 3 to 36 months); Taiwanese sample, Stroke) 

• MDC = 4 points 
• Percent change = 16.6%

Patients with Chronic Stroke

(Alghadir, 2018; n=56; age=58.6 (9.8); mean time since onset=22.2 (18.3) months))

•             Minimal Detectable Change (MDC=1.9)

Stroke:

(Lin et al, 2010, Acute and Chronic Stroke)

(Hwang et al, 2010; n = 11; mean age = 48.09 (5.85) years; mean time post stroke = 24.36 (10.84) months, Chronic Hemiparetic Stroke)

Stroke:

(Lin et al, 2010; tested in separate sample of 48 patients with chronic stroke; mean age = 54.9 (10.2) years, Acute and Chronic Stroke)

• Excellent test-retest reliability (ICC > 0.94; 0.91-0.97)

(Jonsdottir & Cattaneo, 2007; n = 25 mean age = 61.6 (13.1) years, Chronic Stroke)

• Excellent total score test re-test reliability (ICC = 0.96)
  • Individual items varied from 0.56 (gait and pivot turns) to 1.00 (stair climbing)

Patients with Chronic Stroke

(Alghadir, 2018; n=56; age=58.6 (9.8); mean time since onset=22.2 (18.3) months))

• Excellent test-retest reliability (ICC=0.98)

Stroke:

(Jonsdottir & Cattaneo, 2007, Chronic Stroke)

• Excellent interrater reliability (ICC = 0.96)

Patients Poststroke

(Alghwiri, 2014, n = 51 age = 64 (12)

• Excellent for both interrater (ICC = 0.98) and intrarater (ICC = 0.97) reliability

Stroke:

(Jonsdottir and Cattaneo, 2007, Chronic Stroke)

• Excellent concurrent validity with Berg Balance Scale (r = 0.83)
• Excellent concurrent validity with the ABC (r = 0.68)

(Lin et al, 2010, Acute and Chronic Stroke)

• Excellent concurrent validity amongst DGI, 4 itme DGI, and FGA (r > 0.91)

Convergent Validity

Acute and Chronic Stroke:

(Lin et al, 2010) 

• Excellent with the 10 meter walk (r = -0.68, -0.87, -0.83) test and Postural Assessment Scale (r = 0.85, 0.76, 0.83) for Stroke at 1st week of PT, 2 months after PT, and 5 months after PT

Patients Poststroke

(Alghwiri, 2014, n = 51 age = 64 (12)

• Correlated moderately but significantly with GCS (rho = 0.39 p=0.01), Beck Depression Inventory (rho = 0.50 p <0.01)

Discriminant Validity

Patients Poststroke

(Alghwiri, 2014, n = 51 age = 64 (12)

• The A-DGI discriminated between subjects who were at risk of fall and those who were not (U = .00, P < .01), younger and older adults (U = 195, P = .03), side of hemiplegia (U = 218, P = .04), and mild and moderate (U = 105.5, P = .02) as well as mild and severe stroke severity (U = 30, P < .01)

Stroke:

(Lin et al, 2010, Acute and Chronic Stroke)

• Excellent indicates minimal floor effects
• Adequate indicates relatively little ceiling effects

Stroke:

(Lin et al, 2010, Acute and Chronic Stroke) 

• Moderate responsiveness in depicting change at 2 months and 5 months after therapy (effect size 0.56, 0.62 respectively)

Patients with Chronic Stroke

(Alghadir, 2018; n=56; age=58.6 (9.8); mean time since onset=22.2 (18.3) months))

• Effect size= 0.50 (0.31-0.73)
• Standardized Response Mean= 0.89 (0.47-1.30)
• DGI demonstrated better responsiveness than TUG & BBS

Parkinson’s Disease:

(Huang et al, 2011, n = 72; mean age = 67.5 (11.6); range of disease duration = 2 months-15 years; Hoehn-Yahr range I-III(Stage I = 17, II = 33, III = 22), Taiwanese sample)

• MDC = 2.9 points 
• MDC% = 13.3%

Parkinson’s Disease:

(Dibble et al, 2008; total n = 70, mean age = 73.9(6.45), mean J&Y=2.5 (range 1-4), 36 fallers & 34 nonfallers; fallers (n = 36, mean age = 75.25(5.15)), mean duration of PD = 8.32 (8.67), mean modified H&Y=2.5, 28 male/8 female); and non-fallers (n = 34, mean age = 72.50(7.4)), mean duration of PD = 6.49(4.93), mean modified H&Y = 2.0

• Adequate discriminative ability between fallers and non-fallers based on cutoff score < 19 (sensitivity = 0.64, specificity = 0.85, +LR = 4.26, -LR = 0.42)

(Landers et al 2008, JNPT) (Total n = 49, mean age = 70.9(8.9) yrs, 20 female/29 male; 25 of participants were fallers (mean duration of diagnosis = 75.6(66.9) months, mean UPDRS = 41.8(9.6), mean H&Y = 3.0(0.55)) and 24 participants were non-fallers (mean age = 70.1(6.9) yrs, mean duration since Dx = 45.4(36.9) months, mean UPDRS = 26.8(10.0), mean H&Y = 2.1(0.61)} 

• Adequate discriminative ability between fallers and nonfallers using cut score = 18.5 (AUC = 0.758, Sensitivity = 0.68, Specificity = 0.708, +LR = 2.33, -LR = 0.45; post-test probability = 0.708, Odds ratio 5.20(1.54-17.56) 95%CI)

Parkinson’s Disease:

(Cakit et al, 2007; n = 31; mean age 71.8 (6.4) years, Parkinson's Disease)

Parkinson’s Disease:

(Huang et al, 2010, Parkinson's Disease)

• Excellent test-retest reliability (ICC = 0.84)

(Kadivar et al, 2011) (n = 16, age range 59-81 yrs, 5 female/11 males, H&Y stage (range 2-4)) 

• Excellent test retest reliability (ICC > 0.99)

Parkinson’s Disease:

(Cakit et al. 2007; n = 31, mean age = 71.8(6.4), mean duration of PD = 5.58(2.9), mean UPDRS motor subscale = 18.4(9.32)) 

• Adequate concurrent validity with UPDRS-motor subscale scores (r = -0.567, p < 0.001)
• Excellent correlation with history of falls in PD (r = 0.643, p < 0.01)

Parkinson’s Disease:

(Landers et al, 2008) 

• Significant difference in total DGI scores between fallers (16.1, SD = 3.4) and non-fallers (19.6, SD = 2.6) in PD cohort, p < 0.01

(Dibble et al, 2006; n = 45 with idiopathic PD, mean age = 69.94(11.8), mean duration of PD = 7.43(5.62) years, mean H&Y level = 2.6(0.66); fallers n = 25 and non-fallers n = 20;  faller subgroup characteristics (mean age = 73.25(8.58), mean duration of PD = 9.07(6.13), mean H&Y = 2.95(0.62)); non-faller subgroup characteristics (mean age = 65.91(9.89), mean duration of PD = 5.43(4.25), mean H&Y = 2.17(0.42)) 

• Significant difference in DGI scores between fallers {17.92 (4.36) 95% CI = 16.23-19.62} and non-fallers {21.82 (3.42) 95%CI 21.02-22.63 
• Adequate discriminative ability between fallers and non-fallers based on cutoff score = 22 {AUC = 0.84, Sensitivity = 0.89, Specificity = 0.48, +LR = 1.86 (1.38-2.55), -LR = 0.22 (0.10-0.49)
• Strongest positive LR as compared to Functional Reach test, Berg Balance Test and TUG. Author recommended DGI be administered as first test to detect fall risk in community dwelling persons with PD

Parkinson’s Disease:

(Cakit et al, 2007) 

• Was moderately responsive to change (mean change = 4.72 SD = 0.88) in training group following locomotor treadmill training intervention compared to control group (p < 0.05) 

(Kadivar et al, 2011) 

• Moderate responsiveness of DGI scores to cued step training with significant differences between the cued and noncued groups (effect size in cued RAS training group = 1.77, in non-cued group = 1.05)

(Landers et al, 2008) 

• Sensitivity = 0.680 
• Specificity = 0.708

Multiple Sclerosis:

(Cattaneo et al, 2007; n = 25, mean age = 41.7 (12.5) years; mean time since onset = 8.7 (8.8) years; Italian sample, MS)

• SEM inter-rater reliability = 1.51 points 
• SEM intra-rater relibaility = 2.00 points

Multiple Sclerosis:

(Cattaneo et al, 2007, MS)

• MDC (Calculated from SEM) = 4.19-5.54

Multiple Sclerosis:

(Cattaneo et al, 2006; n = 51; relapsing-remitting or secondary progressive MS; mean age 45.3 (18.1) years; mean time since onset 15.6 (7.6) years, MS)

• < 12 indicative of fall-risk (Sensitivity 45%, Specificity 80%)

(Forseberg, 2013, n=81; age=49 (11); mean time since diagnosis= 12 (8) years))

• Adequate discriminative ability between fallers and non-fallers based on cutoff score <19 (sensitivity=87%, specificity=51%, LR+ 1.77, LR- 0.26, Positive predictive=50%, Negative predictive=87%)

Multiple Sclerosis:

(Cattaneo et al, 2007, Multiple Sclerosis)

• Excellent test-retest reliability (ICC = 0.85)

(Bennett, 2017; n=50; age=53.2 (9.23); time since diagnosis=13.29 (10.8) years))

• Excellent test-retest reliability (ICC = 0.955)

Multiple Sclerosis:

(McConvey and Bennett, 2005, MS)

• Excellent interrater reliability (ICC = 0.983)
• Range from r = 0.910-0.976 for individual test terms
• Excellent intrarater reliability (ICC = 0.760-0.98)

Multiple Sclerosis:

(Cattaneo et al, 2006; Cattaneo et al, 2007, MS)

• Excellent concurrent validity with the Berg Balance Scale, TUG, DI, and ABC respectively (Spearman coefficient = 0.78, -0.80, -0.54)
• Poor concurrent validity with the DHI (Spearman coefficient = -0.39)

Multiple Sclerosis:

(Cattaneo et al, 2006) 

• Statistically significant difference of three points noted between fallers and non-fallers (P = 0.025)

Convergent Validity:

(Forseberg, 2013, n=81; age=49 (11); mean time since diagnosis= 12 (8) years))

• Moderate to Strong Convergent Validity ((ρ=−.582 to −.778)

(Mehta et al, 2019, n = 75 age = 49.81 (10.27)

• In Patient Determined Disease Steps ≤ 2, the difference in the convergence of Berg Balance Scale with Timed Up and Go (TUG) and DGI with TUG was −0.123 (95% CI: −0.280, −0.012).

Multiple Sclerosis:

(Cattaneo et al, 2006, MS)

• Adequate ceiling effects (7.8%)

Multiple Sclerosis:

(Cattaneo et al, 2006, MS)

• Scores of less than 12 discriminated between fallers and non fallers, however, Cattaneo (2006) reported that the BBS and DGI measures were not as good at discriminating between fallers and non-fallers compared to the Equiscale Test.

(Bennett, 2017; n=50; age=53.2 (9.23); time since diagnosis=13.29 (10.8) years))

• DGI discriminated between fallers (mean score=13.3) and non-fallers (mean score=16.9) in persons with MS
• DGI discriminated better than the Berg Balance Scale, but not as well as the Activities-Specific Balance Confidence Scale or the Dizziness Handicap Index

Peripheral Vestibular Disorders: 

(Calculated from Hall & Herdman, 2006; n = 16; age 51.8 (13.4) years, diagnosis of unilateral or bilateral vestibular hypofunction)

• SEM calculated = 2.8 points

Vestibular Disorders:

(Calculated from Hall & Herdman, 2006, Peripheral Vestibular Disorders) 

• MDC calculated = 3.2 points

Patients with Balance and Vestibular Disorders

(Marchetti, 2014; n=326; age=60 (18.3); median symptom duration= 4 months (1-462))

• MDC = 4

Vestibular Disorders:

(Whitney et al, 2000; n = 247, age 14-95 years, mean 62.4 years, central and peripheral vestibular dysfunction)

• Subjects with scores of < 19/24 are 2.58 times more like to have reported a fall in the previous 6 months than subjects with scores above 19

Vestibular Population:

(Hall et al, 2006; n = 16 patients with vestibular disorders; mean age = 51.8 (13.4) years)

• Excellent test-retest reliability (ICC = 0.86 (total score) (individual items ranged from 0.04-0.90))

Vestibular:

(Wrisley et al, 2003; n = 30, age range 27-88 years referred for vestibular rehabilitation)

• Adequate inter-rater reliability (k = 0.64)

Patients with Balance and Vestibular Disorders

(Marchetti, 2014; n=326; age=60 (18.3); median symptom duration= 4 months (1-462))

• Excellent internal consistency of baseline score (coefficient alpha=0.85)

Vestibular Population:

(Hall & Herdman, 2006; n = 16 patients with confirmed peripheral vestibular disorders; mean age = 51.8 (13.4) years, Vestibular Population)

• Berg Balance Scale and DGI scores agreed 63% of the time on fall risk criteria
• DGI appears to be more sensitive than the Berg Balance Scale in identifying fall risk in patients with vestibular disorders

(Whitney et al. 2000 (n = 30 patients (aged 27 to 88) diagnosed with vestibular disorders, Vestibular Population)

• Excellent concurrent validity (r = 0.71) between the DGI and the Berg Balance Scale

Vestibular Disorders: (Dye, 2013; n=117 (Dx. of Dizziness n=77; Dx. of Imbalance n=40); age=60.2 (17.1)

• No relationship between sex and clinical presentation (X2=1.29, P=0.294)
• No relationship between sex and fall history  (x2=1.67, P=0.294)
• No relationship between clinical presentation and fall history (x2= 2.95, P=0.094)
• Negative association between DGI person ability measures and age (r=-0.45, P<0.001)
• Mean person ability was 1.41 logits higher for participants <60 years of age (t=2.10, p<0.001, 95% CI=0.69-2.13)
  • Imbalance (r=-0.56, p<0.001)
  • Dizziness (r=-0.29, p=0.01)
• Mean person ability measures were 1.31 logits higher for participants who were categorized as dizzy compared with those with imbalance (t=3.35, p=0.001, 95% CI= 0.54 to 2.08)
• Mean person ability measures were 1.51 logits lower for participants with significant fall history (t=-3.82, p=0.001, 95% CI=0.73 to 2.31)

Patients with Balance and Vestibular Disorders

(Marchetti, 2014; n=326; age=60 (18.3); median symptom duration= 4 months (1-462))

• Poor Ceiling effect - 50% of subjects received optimal scores at discharge from a physical therapy exercise program

Vestibular Disorders: (Dye, 2013; n=117 (Dx. of Dizziness n=77; Dx. of Imbalance n=40); age=60.2 (17.1)

• Modest ceiling effect (ceiling effect=16.9)

Patients with Balance and Vestibular Disorders

(Marchetti, 2014; n=326; age=60 (18.3); median symptom duration= 4 months (1-462))

• Index of responsiveness was greater for FGA (1.25) than for DGI (0.72)

Normal Pressure Hydrocephalus sample

(Chivukula, 2015; n=170; mean age=72.7 (10.2))

• Baseline score of >7 was found to have significant ability in selecting patients for permanent CSF diverting shunt surgery
• Sensitivity=84.2% (75.6-90.2%)
• Specificity=80.6% (70-88%)
• Diagnostic odds ratio of 22.1 (9.9-49.3)
• AUROC= 77.4% (69.6-85.2%)

Brain Injury:

(Medley, A. et al, 2006; n = 26; mean age = 41.9 (12.4) years, Brain Injury)

• Found that a person who scores 19 out of 24 points on the DGI has a 28% probability of falling. A person who scores 24 out of 24 points would have a 6% chance of falling and a person who scores 0 out of 24 points would have a 100% change of falling

(McConvey and Bennet, 2005)

• Excellent concurrent validity with 6.1m timed walked test (r = -0.81)

Healthy Adults:

(Vereeck et al, 2008; n = 318; mean age = 49.2(18.7), Healthy Adults)