Skip to main content

Table 2 Summary of clinical measurement properties of MNA, SGA and PG-SGA tools

From: Clinical measurement properties of malnutrition assessment tools for use with patients in hospitals: a systematic review

Measurement property

Patient population applied and measure extracted

MNA

SGA

PG-SGA

Diagnostic accuracy

Geriatric patients [36]:

End-stage renal failure [37]:

Post-appendectomy surgery [38]:

− ROC = 0.85 ± 0.05 (95% CI 0.75 to 0.96, p < 0.0001); sensitivity: 83.30%; specificity: 74.40%

− SGA-C can discriminate patients who are not malnourished (specificity: 0.93-0.94), but cannot accurately identify patients with severe malnutrition (sensitivity: 0.05-0.14)

− Cut-off score = 7

 

− ROC = 0.98; sensitivity: 88.60%; specificity: 100%

Parkinson’s disease [33]:

Gynecological cancer [39]:

− Cut-off score = 22.5

− SGA-B is not accurate enough to identify mild-moderate malnutrition (specificity: 0.61–0.65; sensitivity: 0.68–0.59)

− ROC = 0.92 (95% CI, 0.84 to 1.01, p < 0.001)

− ROC =0.71 (p < 0.001); sensitivity: 58%; specificity: 82%

Pre-frail elderly [34]:

Cancer inpatient [35]:

− Cut-off score = 25.5

− Cut-off score = 9

− ROC = 0.83 (95% CI 0.79 to 0.87); sensitivity: 72.10%; specificity: 91.20%

− sensitivity: 98%; specificity: 82%

Frail elderly [34]:

− Cut-off score = 25.5

− ROC = 0.90 (95% CI 0.88 to 0.93); sensitivity: 72.10%; specificity: 91.20%

Test-retest reliability

Stroke [32]:

None

Palliation [40]:

− ICC = 0.91 (95% C, 0.84 to 8.94)

− r = 0.87

Internal consistency

Parkinson’s disease [33]:

None

Hemodialysis [41]

− Cronbach’s α = 0.70 (95% CI, 0.62 to 0.77)

− Cronbach’s α = 0.73

Geriatric patients [34]:

Palliation [40]:

− Cronbach’s α = 0.70

− Cronbach’s α = 0.72

Inter-rater reliability

None

General medicine [42]:

None

Patients rated by two dietitians (Cohen’s kappa = 0.96, p < 0.001) versus dietitian and a trained allied health assistance (Cohen kappa = 0.84, p < 0.001)

Gynecological cancer [43]:

Patients rated by dietitian and nurse (Cohen’s kappa = 0.82)

End-stage renal failure [37]:

Patient rated by two dietitians (Cohen’s kappa = 0.79, 95% CI 0.67 to 0.92) versus dietitian and nephrologist (Cohen kappa = 0.60)

Construct validity

Dialysis [44]:

Digestive disease [45]:

Patient with cancer had undergone radiation treatment [46]:

Moderately correlated with a female’s body weight (r = 0.64) and BMI (r = 0.62)

Weak correlation with BMI (r = − 0.36), percentage of arm muscle circumference (r = − 0.33), percentage of triceps skin fold thickness (r = − 0.26), albumin (r = − 0.41), total cholesterol (r = − 0.16), and lymphocyte count (r = − 0.24).

Moderately correlated with percentage weight loss (r = 0.53), global quality of life at the baseline (r = − 0.66), and after 4 weeks of radiotherapy (r = − 0.61)

Live failure [47]:

Moderately correlated with hand grip strength (r = 0.63) and albumin (r = 0.55)

Cancer inpatient [35]:

Cancer [48]:

Mild correlation with weight loss in the past 6 months (r = 0.56)

Highly correlated with PG-SGA at the baseline (r = − 0.76), at the 4–6 weeks follow up (r = − 0.73) and at the 8–12 weeks follow up (r = − 0.83).

Hemodialysis [41]:

Moderately correlated with percentage weight loss in the past 6 months (r = 0.56)

Parkinson’s disease [33]:

Mildly correlated with weight (r = 0.43), BMI (r = 0.35), mid-arm circumference

(r = 0.268), calf circumference (r = 0.29)

Stroke [32]:Mildly correlated with quality of life (r = 0.32)

Predictive validity

Dialysis [44]:

Digestive disease [45]:

Cancer inpatient [35]:

Moderately correlated with the duration of dialysis (r = − 0.53)

Weak correlation with the length of hospital stays (r = 0.29)

Weak correlation with the length of stays (r = 0.36)

Post-appendectomy surgery [38]:

Weak predictive validity for the length of stays (r = 0.38)

Responsiveness to change and minimal detectable change (MDC)

Stroke [32]:

None

Patient with cancer had undergone radiotherapy [46]:

MDC = 2.1

a moderate correlation between the change in PG-SGA score and change in global quality of life after 4 weeks of radiotherapy (r = − 0.55)