Walking While Talking Test

Community dwelling older adults without dementia: (Verghese et al, 2002; n = 60 individuals aged 65 to 98; mean age = 79.6 (6.3) years; Cut-off scores were chosen at one standard deviation from the group mean)

• 20 seconds or longer for WWT-simple (16.3 ± 4)
  • High Specificity 89.4%- for identifying falls
  • Modest Sensitivity 46.1%- for ruling out falls
  • Positive predictive value 54.5%
• 33 seconds or longer for WWT-complex (22 ± 11.4) 
  • High Specificity 95.6%- for identifying falls 
  • Modest Sensitivity 38.5%- for ruling out falls 
  • Positive predictive value 71.4%
• Specificity is the proportion of individuals without falls that have a negative result.
• Sensitivity is the proportion of individuals with falls who have a positive test result. 
• Positive predictive value is the proportion of individuals who have a positive test result have the disease. 

Healthy Community Dwelling Elderly:

(Verghese et al, 2012. 594 healthy CDOA mean age 79.9 ± 5.3yrs. Study outcomes were frailty, disability and all-cause mortality)

WWT Cutscores and Risk of Frailty and Disability (HR 95% CI and P-Value):

*WWT cutscores did not predict mortality 

*WWT speed of < 70cm/s had a 93% greater risk of developing frailty and an 82% greater risk of developing disability that individuals with a WWT speed > 70 cm/s.

Older adults without dementia:

(Verghese et al, 2007; n =189; mean age= 80.2 (4.9) years)

Community dwelling older adults without dementia:

(Verghese et al, 2002)

• Adequate reliability (r = 0.602, P < 0.001) on the WWT-simple test

Community dwelling older adults without dementia 

(Verghese et al, 2002)

• Poor performance on WWT-simple (OR = 7.02, 95% CI = 1.7-29.4) and WWT-complex tasks (OR = 13.7, 95% CI = 2.3-83.6) was strongly predictive of falls over 12 months.
  • WWT-simple: 55% of subjects scoring over the WWT-simple cutoff (20 seconds or longer ) fell, compared with 15% of older people who did not exceed this score.
  • WWT-complex: 71% of subjects scoring over the cutoff of the WWT-complex (33 seconds or longer ) fell, compared with 15% of older people who did not exceed this score. 
• Combining the Tinetti Balance and Mobility Scale with WWT-simple improves the sensitivity to 71% with specificity of 70%. 

(Verghese et al, 2008; n = 539 (60.5% female) Mean age 80.1 ± 5.2 yrs, participated in the Physical Performance Battery (PPB). The PPB was analyzed as an aggregate (including the WWT), and the WWT analyzed separately)

• WWT Velocity not associated with frailty (OR 1.00, 95% CI 0.99-1.01) 

(Verghese et al, 2012; n = 594 healthy CDOA (61% female), mean age 79.9 ± 5.3yrs. Study outcomes were frailty, disability and all-cause mortality)

• WWT Speed predicted frailty (n = 473) (HR per 10-cm/s change = 1.12, 95% CI = 1.06-1.18)
• WWT Speed predicted disability (n = 594) (HR = 1.13, 95% CI = 1.03-1.24)
• WWT is associated with the risk of mortality (n = 545) (HR = 1.14, 95% CI = 1.01-1.28)
• Each 10-cm/s change in WWT speed was associated with a 12% greater risk of developing frailty, 13% greater risk of developing disability, and 13% greater risk of mortality.

Community dwelling older adults without dementia:

(Verghese et al, 2002)

• Timed gait (same distance traversed by the subject as the WWT) was used in this study. The progressive increase in positive predictive value (PPV) from 42% for timed gait (Odds Ratio (OR) = 4.3) to 55% for WWT-simple (OR = 7), and 71% for WWT-complex (OR = 13.7), demonstrates the incremental validity of WWT test over timed gait.
• The authors felt that the process involved in the WWT was relevant to the causation of falls.

Community dwelling older adults:

(Hall et al, 2011; n = 77 community-dwelling older adults with a mean (SD) age of 75.5 (5.8) years )

• Regardless of the specific cognitive task, participants walked slower (ie, took longer and DTCs, therefore, were positive) under dual-task conditions than under single-task conditions.

Geriatrics:

(Liu-Ambrose et al, 2009; n = 140 healthy senior females aged 65-75 yrs (mean 69.6(3.0), Balance confidence assessed by ABC. Dual task gait performance assessed by WWT test under 2 conditions: simple (recite alphabet consecutively) and complex (recite alphabet alternately)

(Verghese et al, 2012)

• Moderate correlation between WWT and gait speed (r = 0.32) Low correlation between WWT and Short Physical Performance Battery (SPPB

Sedentary older adults: 

(Verghese et al, 2010, n = 24) 

• Participants in the intervention group (n = 10) demonstrated a change in WWT- complex over the 8 week period
• Change: 19.9 ± 14.9-2.5 ± 20.1cm/s, p = 0.05

Parkinson's Disease: 

(O’Shea et al 2002; n = 15 subjects with PD (12 M/3F, mean age 68.33 ± 6.59, range 52-76 yrs, H&Y not provided) and 15 Healthy controls (age sex and height matched- mean age 67.73 ± 6.97, range 52-79 yrs). Subjects walked along a 14m walkway (middle 10m used for data collection), spatiotemporal (time and distance) measures were recorded. Pressure sensors in footswitches. Secondary motor and cognitive task performed. Cognitive task: digit subtraction (count backward by 3’s from a randomly selected target number). No practice trials. 

Statistics for unitask and dual task (digit subtraction) walking in subjects with PD (n = 15) and controls (n = 15):

PD responded slower than controls in dual task than controls (t₂₈ = -4.41, P < 0.0001)PD committed more errors in dual task than controls (t ₂₈ =3.19, P < 0.005) 

(La Point et al, 2010; n = 25 PD (19M/6F, mean age 67.44 range 41-91 yrs, H&Y stage II: 18 ppl, stage III: 4 ppl, stage IV: 3 ppl) and 13 controls (mean age 68.07 yrs) completed gait tasks while conducting tasks of low (counting by ones) middle 9serial subtraction of 3’s) and high load (alpha-numeric sequencing)

• No significant difference between PD and controls for Stride length (F _1,38= 3.64, p = 0.065) or Step velocity (F1,38 = 4.09, p = 0.051) 
• Significant difference bet PD and control in double support time ((F _1,38 = 4.42, p = 0.04) with a significant interaction. *control group sig increased double support time while Pd did not. 

(Camicioli et al, 1998; n = 19 subjects with PD (Freezers(n = 9): 4M/5F, mean age 72 (4.7) mean dz duration 7.8 (4.1) yrs, UPDRS 9.2 (4.7)/44; Non-Freezers(n = 10): 5M/5F, mean age 67.3 (9.3) yrs, mean dz duration 11 (4.4) yrs, UPDRS 12.1 (5.8)/44) and 19 healthy controls: 11M/8F, mean age 71.7 (3.8)yrs) 

• PD-F significantly increased the number of steps taken (mean ± SD: PD-F, 4.2 ± 4.6; PD-NF, 0.11 ± 1/62 as compared to controls (1.53 ± 1.54; F[2,35] = 5.8, P = 0.007). The PD-F group changed more than both the PF = NF group (p = 0.002) and controls (P = 0.02). The PD-NF group did not differ from the controls (P = 0.20). 
• Increase in time (seconds) to walk 30 feet did not differ significantly between groups (PD-F, 2.00 ± 1.41 sec; PD-NF, 0.44 +/-1.51 sec; controls 1.53 ± 2.04 sec; F[2,35]+1.9, P = 0.16)
• With anti-PD medications, PD-F significantly decreased # steps (F = 14.1, P < 0.001), and decreased time (F = 10.6, P < 0.0002) but PD-NF did not sig decrease steps or time. 

Effect of Anti PD meds on # steps and time to walk 30 feet in PD-F:

Floor Effect: 

Acquired Brain Injury: 

(McCulloch et al, 2007; n = 18; aged 24-58; residents of a supported living environment (n = 14), community dwellers (n = 2), and patients in acute rehabilitation (n = 2), able to ambulate unaided (with or without an assistive device))

• Single task condition of WWT: All subjects completed the walking portion
• Single task condition of WWT: 87% of the subjects could perform the simple cognitive task of the WWT-simple
• Single task condition of WWT: 44% of the subjects could perform the complex cognitive task of the WWT-complex
• Dual task condition: Several subjects refused to complete the WWT-complex because it was too difficult to do