Skip to main content
. 2023 Sep 21;46(12):2171–2179. doi: 10.2337/dc23-1034

Table 3.

Adjusted geometric means of markers of glucose metabolism and insulin resistance according to quintiles of phasor magnitude and phasor angle in adults who did not take medications to lower blood glucose, NHANES 2011–2014

Adjusted geometric mean (95% CI)a,b
Fasting glucose, mg/dL (n = 2,796) Fasting insulin, μU/mL (n = 2,704) HOMA-IR (n = 2,704) 2-h glucose from OGTT, mg/dL (n = 2,603)
Phasor magnitude
 Q1 100.24 (97.56, 103.00) 10.93 (9.54, 12.51) 2.69 (2.34, 3.13) 120.82 (114.35, 127.65)
 Q2 100.52 (98.05, 103.06) 10.56 (9.43, 11.83) 2.61 (2.32, 2.97) 109.18 (103.06, 115.67)
 Q3 99.75 (97.68, 101.86) 10.14 (9.3, 11.05) 2.51 (2.27, 2.75) 108.59 (104.14, 113.25)
 Q4 99.28 (96.62, 102.01) 9.54 (8.48, 10.72) 2.34 (2.05, 2.66) 109.5 (102.97, 116.44)
 Q5 99.43 (96.96, 101.96) 8.90 (7.84, 10.11) 2.20 (1.90, 2.53) 104.88 (98.92, 111.19)
P for trend 0.21 0.001 0.002 0.0007
Phasor angle
 Q1 99.09 (96.70, 101.53) 9.25 (8.33, 10.27) 2.27 (2.01, 2.56) 110.31 (104.5, 116.44)
 Q2 99.17 (96.48, 101.92) 9.33 (8.40, 10.36) 2.29 (2.03, 2.59) 106.75 (100.06, 113.90)
 Q3 100.66 (97.94, 103.43) 10.24 (9.12, 11.49) 2.53 (2.23, 2.89) 110.06 (103.25, 117.32)
 Q4 99.74 (97.36, 102.18) 10.12 (8.76, 11.68) 2.48 (2.14, 2.92) 109.01 (102.83, 115.54)
 Q5 100.41 (97.78, 103.11) 10.64 (9.38, 12.07) 2.64 (2.29, 3.03) 113.0 (107.04, 119.28)
P for trend 0.20 0.02 0.02 0.35
a

Obtained from multiple linear regression models adjusted for age (continuous), sex (male, female), race and ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, other), education (less than high school, high school graduate, some college, college graduate or above), household income (<$20,000, $20,000–44,900, $45,000–74,900, ≥$75,000), marital status (married, not married), smoking (current, former, never, or <100 cigarettes in life), alcohol consumption (<1 drink/week, 1 drink/week to <1 drink/day, ≥1 drinks/day), and total energy intake (continuous).

b

Raw marker levels were log-transformed for regression model analyses, and model-derived least squares means were back-transformed.