Table 3– Biases in prediction due to age truncated to whole years and/or height biased by +1%; combined results across all age ranges
Bias % (minimum–maximum)
MalesFemales
Truncated ageBiased heightTruncated age and biased heightTruncated ageBiased heightTruncated age and biased height
FEV1
 Crapo et al. [13]0–12–42–40–22–52–6
 ECSC [14]0–12–32–40–22–52–6
 Falaschetti et al. [15]-2–21–60–60–222–4
 GLI 2012 [2]-7–12-5–3-8–12-6–4
 Hankinson et al. [11]-3–22–4-1–5-5–31–3-2–5
 Knudson et al. [12]3–22–161–16-3–22–6-1–6
 SAPALDIA [16]-1–11–20–30–222–4
 Stanojevic et al. [10]-5–12-3–4-4–22-4–3
FVC
 Crapo et al. [13]0–12–42–40–12–52–6
 ECSC [14]0–12–32–40–12–42–5
 Falaschetti et al. [15]-3–11–30–4-3–21–40–5
 GLI 2012 [2]-8–12-6–3-9–12negative –3
 Hankinson et al. [11]-3–22–4-1–6-5–32–6-1–7
 Knudson et al. [12]-3–22–132–13#-3–72–40#0–40#
 SAPALDIA [16]-2–01–21–2-2–01–21–3
 Stanojevic et al. [10]-6–13-3–4-5–12-3–4
  • Biases are expressed as per cent of the accurate prediction. The values shown represent the range across all the various prediction equations relative to the Global Lungs Initiative (GLI) data. Note: analysis has been limited to those equations which include both height and age as predictors. FEV1: forced expiratory volume in 1 s; ECSC: European Community for Steel and Coal; FVC: forced vital capacity. #: the large errors occur in short people older than 70 yrs.