Skip to main content
. Author manuscript; available in PMC: 2016 Feb 3.
Published in final edited form as: J Am Geriatr Soc. 2015 Oct 27;63(11):2247–2259. doi: 10.1111/jgs.13788

Table 4.

Association, Discrimination and Reclassification for FNIH Alternative Sarcopenia Definitions with Falls, Functional Limitation, Hip Fractures and Mortality in Older Men

Recurrent falls FNIH Alternative Sarcopenia Classification
Definition #1 Definition #2
Association and discrimination
 OR (95% CI) in model with age alone 2.33 (1.32, 4.10) 2.37 (0.79, 7.07)
 Change in AUC (95% CI) compared to model with age alone 0.003 (−0.001, 0.007) 0.000 (−0.001, 0.001)
Reclassification
 Among those with event (N=694)a
  Appropriately reclassified 18 (3%) 5 (1%)
  Inappropriately reclassified 105 (15%) 40 (6%)
  No change 571 (82%) 649 (94%)
  NRI c (95%CI), events 0.13 (0.16, −0.10) 0.05 (0.07, −0.03)
 Among those without event (N=5134)b
  Appropriately reclassified 433 (8%) 159 (3%)
  Inappropriately reclassified 45 (1%) 11 (0%)
  No change 4656 (91%) 4964 (97%)
  NRI (95%CI), non-events 0.08 (0.07, 0.08) 0.03 (0.02, 0.03)
 Overall NRI (95% CI) 0.05 (0.08, −0.02) 0.02 (0.04, 0.00)

Hip Fractures
Association and discrimination
 HR (95% CI) in model with age + BMD 1.68 (0.74, 3.81) 5.46 (1.71, 17.46)
 Change in C-statistic (95% CI) compared to model with age + BMD −0.001 (−0.001, 0.000) 0.000 (−0.001, 0.000)
Reclassification
 Among those with event (N=207)a
  Appropriately reclassified 6 (3%) 3 (1%)
  Inappropriately reclassified 11 (5%) 17 (8%)
  No change 190 (92%) 187 (90%)
  NRI c (95%CI), events −0.02 (−0.06, 0.01) 0.07 (0.11, −0.03)
Among those without event (N=5727)b
  Appropriately reclassified 185 (3%) 213 (4%)
  Inappropriately reclassified 60 (1%) 53 (1%)
  No change 5482 (96%) 5461 (95%)
  NRI (95%CI), non-events 0.02 (0.02, 0.03) 0.03 (0.02,0.03)
 Overall NRI (95% CI) 0.00 (−0.04, 0.04) 0.04 (0.08, 0.00)

Functional Limitations
Association and discrimination
 OR (95% CI) in model with age alone 1.01 (0.81, 1.23) 1.30 (1.04, 1.65)
 Change in AUC (95% CI) compared to model with age alone 0.000 (−0.001, 0.001) 0.001 (−0.005, 0.007)
Reclassification
 Among those with event (N=590)a
  Appropriately reclassified 38 (6%) 133 (23%)
  Inappropriately reclassified 0 (0%) 236 (40%)
  No change 552 (94%) 221 (37%)
  NRI c (95%CI), events 0.06 (0.04, 0.08) 0.17 (0.24, −0.11)
 Among those without event (N=3136)b
  Appropriately reclassified 1 (0%) 982 (31%)
  Inappropriately reclassified 147 (5%) 495 (16%)
  No change 2988 (95%) 1659 (53%)
  NRI (95%CI), non-events 0.05 (0.05, −0.04) 0.16 (0.13, 0.18)
 Overall NRI (95% CI) 0.02 (0.00, 0.04) −0.02 (−0.09, 0.05)

Mortality
Association and discrimination
 HR (95% CI) in model with age alone 2.03 (1.51, 2.73) 3.49 (2.01, 6.05)
 Change in C-statistic (95% CI) compared to model with age and BMD 0.002 (0.001, 0.003) 0.001 (0.000, 0.002)
Reclassification
 Among those with event (N=2003)a
  Appropriately reclassified 52 (3%) 22 (1%)
  Inappropriately reclassified 551 (28%) 306 (15%)
  No change 1400 (70%) 1675 (84%)
  NRI c (95%CI), events 0.25 (0.27, −0.23) 0.14 (0.16, −0.13)
 Among those without event (N=3866)b
  Appropriately reclassified 641 (17%) 222 (6%)
  Inappropriately reclassified 123 (3%) 130 (3%)
  No change 3102 (80%) 3514 (91%)
  NRI (95%CI), non-events 0.13 (0.12, 0.15) 0.02 (0.01, 0.03)
 Overall NRI (95% CI) 0.12 (0.14, −0.09) 0.12 (0.14, −0.10)

Bold text indicates p<0.05.

For recurrent falls, AUC for model with age alone was 0.577.

For hip fractures, C-statistic for model with age and BMD was 806.

For functional limitations, AUC for model with age alone was 0.632

For mortality, C-statistic for model with age alone was 0.684.

FNIH alternative definition #1: Presence of both weakness (grip<26 kg) and low lean mass (ALM<19.75 kg)

FNIH alternative definition #2: Presence of slowness (gait≤0.8 m/s), weakness (grip<26 kg) and low lean mass (ALM<19.75 kg)

FNIH = Foundation for NIH Sarcopenia Project, ALM=appendicular lean mass, AUC=area under the curve, BMD=bone mineral density, NRI = net reclassification improvement

a

For events, appropriate reclassification occurs when estimated risk increases when the additional factor is added to the model; inappropriate reclassification occurs when estimated risk decreases when the additional factor is added to the model.

b

For non-events, appropriate reclassification occurs when estimated risk decreases when the additional factor is added to the model; inappropriate reclassification occurs when estimated risk increases when the additional factor is added to the model.