Table 5

Determinants of weight gain in the intervention group; candidate variables stratified by weight gain (≥1 kg of baseline weight)

Univariable analysisMultivariable model*
Weight gain (≥1 kg)
(n=100)
No weight gain
(n=180)
OR (95% CI)P valueOR (95% CI)P value
Demographics
 Age (<65 years)81 (81.0)124 (68.9)1.93 (1.08 to 3.54)0.031.47 (0.78 to 2.78)0.23
 Sex (female)20 (20.0)45 (25.0)0.75 (0.41 to 1.34)0.34
 Ethnicity (non-Caucasian)10 (10.0)6 (3.3)3.22 (1.16 to 9.74)0.032.77 (0.90 to 8.55)0.08
 Level of education (lower)†62 (62.0)113 (62.8)0.97 (0.59 to 1.61)0.90
 Having a partner83 (83.0)151 (83.9)0.94 (0.49 to 1.84)0.85
Baseline lifestyle-related risk factors
 Smoking
 Smoking at baseline‡29 (29.0)28 (15.6)2.22 (1.23 to 4.02)0.012.70 (1.38 to 5.27)0.004
 Quit smoking before or during hospitalisation§34 (34.0)31 (17.2)2.48 (1.41 to 4.38)<0.013.21 (1.70 to 6.08)<0.001
 BMI, kg/m2 31.5±4.031.7±3.30.98 (0.91 to 1.05)0.62
 Inadequate physical activity¶63 (63.0)109 (60.6)1.11 (0.67 to 1.85)0.69
 Established diabetes mellitus type 216 (16.0)32 (17.8)0.88 (0.45 to 1.68)0.71
Participation in lifestyle programmes
 Smoking cessation16 (16.0)14 (7.8)2.26 (1.05 to 4.91)0.04
 Physical activity49 (49.0)110 (61.1)0.61 (0.37 to 1.00)0.05
 Weight loss54 (54.0)128 (71.1)0.48 (0.29 to 0.79)<0.010.59 (0.34 to 1.02)0.06
 Partner participating in WW21 (44.7)60 (55.6)0.65 (0.32 to 1.28)0.21
 Motivated for WW75 (78.9)141 (84.9)0.66 (0.35 to 1.29)0.22
  • Values presented as mean±SD or counts (%).

  • *Variables selected with backward elimination (pp<0157) in 20 imputed datasets and estimates pooled using Rubins’ rule.

  • †Lower education: all except universities of applied sciences and research universities.

  • ‡Baseline urine–cotinine level >200 ng/mL.

  • §Smoking cessation ≤6 months before or during hospitalisation.

  • ¶Self-reported based on WHO recommendations.

  • BMI, body mass index; WW, weight loss programme.