Test-Retest: r=0.95 ((Gouvier et al., 1987); TBI).

Interobserver Reliability: Inter-rater correlations ranged from 0.97-0.98 (p<0.01) (Rappaport et al., 1982), average r=0.98 (Gouvier et al., 1987)(TBI), correlations between observer ratings ranged from 0.75-0.89 (Fleming & Maas, 1994).

Internal Consistency: Item to item correlations ranged from 0.23 to 0.95, item-to-total correlations ranged from 0.54 (eye opening) to 0.96 (feeding) (Rappaport et al., 1982).

Construct Validity: DRS ratings at admission correlated with evoked brain potential abnormality scores (r=0.78, p<0.01)(Rappaport et al., 1982). Correlations between DRS and scores in cognitive testing in intellectual, executive, academic and visuoperceptual domains ranged from -0.17 to –0.37 (p<0.05), suggesting that better levels of function as assessed by the DRS is associated with better performance in a given cognitive domain ((Neese et al., 2000); TBI).

Construct Validity (Known Groups): DRS could discriminate between groups of patients who had received cognitive rehabilitation or not (Fryer & Haffey, 1987) (TBI).

Concurrent Validity: Admission DRS scores correlated with initial Stover & Zeiger (S-Z) ratings (r=0.92), discharge DRS scores correlated with discharge SZ scores (r=0.81), GOS scores (0.80) and EGOS scores (0.85) (Gouvier et al., 1987). DRS ratings were significantly correlated with FIM motor, FIM cognition, FIM+FAM motor and FIM+FAM cognition scores (r=0.641, 0.728, 0.680, 0.746, respectively, all p<0.05), DRS rating also correlated with LCFS ratings (r=0.708) (Hall et al., 1993). GOS scores correlated with DRS at admission (r=0.50, p<0.01) and discharge from rehabilitation (r=0.67, p<0.01) (Hall et al., 1985) (TBI).

Predictive validity: Initial DRS scores correlated with discharge SZ scores (0.65), GOS scores (0.62) and expanded GOS scores (0.73). DRS scores at admission and discharge from rehabilitation were both significantly related to employment status at one year post-injury (Cifu et al., 1997) (TBI). Initial DRS ratings correlated with DRS ratings at 12 months post-injury (r=0.53, p<0.01) (Rappaport et al., 1982) (TBI). Initial DRS score correlated with length of hospital stay (r=0.50, p<0.01) and with discharge DRS scores (r=0.66, p<0.010, stroke) (Eliason & Topp, 1984). Via growth curve modeling, flatter rates of recovery on the DRS recovery curve were associated with higher rates of reported cognitive difficulties, as well as severity of affective/neurobehavioural disturbance and severity and burden of physical dependence at 6 months post-injury as reported by significant others (McCauley et al., 2001) (TBI). Initial DRS score and rate of recovery accounted for 62% of variance in discharge DRS scores (p<0.00, TBI) (Fleming & Maas. 1994). (Fryer & Haffey, 1987) (TBI) reported DRS at admission to rehabilitation was significantly predictive of need for supervision and return to work 1 year post injury (r=0.77, p<0.001). Initial and discharge DRS scores were significantly related to vocational status (p<0.007) (Rao & Kilgore, 1992).

Ceiling effects reported that DRS scores do not discriminate effectively among patients scoring in the upper categories of the Extended Glasgow Outcome Scale (Wilson et al., 2000) (TBI). Rasch analysis demonstrated that a wide range of difficulty is reflected in scale items from very simple functioning to very complex with less sensitivity at the high end (Hall et al., 1993). DRS had a 6% ceiling effect at discharge, 47% at 1-year post injury and 54% at year 2, when ceiling effect is defined as scoring in the top 10% of the scale as noted by Hall et al. (1996b)

From admission through discharge and follow-up, DRS scores rated by family members demonstrated significant change over time (p<0.0001), with level of disability decreasing over the duration of rehabilitation and from rehabilitation discharge to follow-up at 3 months post-discharge (Novack et al., 1991) (TBI). Significant differences were reported between DRS ratings at discharge from rehabilitation and at one-year follow-up (p<0.001, TBI) (Hammond et al., 2001). From admission to discharge from rehabilitation, improvement shown by the DRS was significantly greater than that shown by the GOS (71% versus 33%, p<0.01) (Hall et al., 1985).