Table 2

Crude and adjusted associations of NTLI with CVD risk factors among APCAPS adults, 2010–2012 (n=5937)

NModel predicted crude mean (95% CI) at the lowest NTLI (61.7 (4.1 on the log scale))Model predicted crude mean (95% CI) at the highest NTLI (1081.1 (7.0 on the log scale))P valueModel predicted age and gender adjusted mean change (95% CI) with increasing NTLI*Model predicted fully adjusted mean change (95% CI) with increasing NTLI*†§
Nβ (95% CI)P valueNβ (95% CI)P value
Systolic blood pressure (mm Hg)‡5752117.2 (115.1 to 119.4)121.6 (119.2 to 123.9)0.0457471.8 (0.4 to 3.2)0.0155141.5 (0.3 to 2.6)0.01
Body mass index (kg/m2)592420.3 (19.8 to 20.8)21.5 (21.0 to 22.1)0.0159200.4 (0.1 to 0.8)0.0156820.3 (0.03 to 0.7)0.05
Fasting serum LDL (mmol/L)52872.4 (2.2 to 2.6)2.7 (2.5 to 2.9)0.0652850.1 (0.01 to 0.2)0.0450510.1 (−0.003 to 0.2)0.06
Fasting plasma glucose (mmol/L)§53285.1 (5.0 to 5.3)5.2 (5.0 to 5.4)0.6153260.04 (−0.1 to 0.2)0.525094−0.002 (−0.1 to 0.1)0.97
  • Model predicted means (95% CIs), β coefficients (95% CIs) and p-values were obtained from multilevel linear regression models with clustering by household and village (using individual-level outcome data).

  • Participants were excluded if medicated for hypertension‡ or diabetes§.

  • *Mean change per integer increase in log transformed NTLI.

  • †Adjusted for age, gender, caste, religion, marital status and survey season.

  • β, beta-coefficient; APCAPS, Andhra Pradesh Children and Parents Study; CVD, cardiovascular disease; LDL, low-density lipoprotein; NTLI, night-time light intensity.