Table 4

Adjusted logistic regression for likelihood of acquiring e-cigarettes from different sources in UK adolescents (n=608)

SupermarketSmall shopOnlineAcquired other
Characteristicn (%)aOR
(95% CI)
n (%)aOR
(95% CI)
n (%)aOR (95% CI)n (%)aOR (95% CI)
Total sample80 (12.3)259 (42.82)102 (16.8)324 (55.3)
Survey year
 201914 (9.3)Ref44 (34.2)Ref32 (25.6)Ref76 (56.7)Ref
 202016 (10.8)1.15 (0.51, 2.58)50 (39.9)1.32 (0.75, 2.31)16 (11.7)0.36** (0.18, 0.76)75 (61.4)1.21 (0.70, 2.10)
 202117 (15.4)1.65 (0.70, 3.87)39 (40.0)1.31 (0.75, 2.32)26 (22.9)0.78 (0.41, 1.48)58 (51.8)0.83 (0.46, 1.48)
 202233 (13.4)1.31 (0.64, 2.70)126 (51.2)2.02** (1.24, 3.29)28 (11.3)0.32*** (0.17, 0.58)115 (52.78)0.88 (0.55, 1.43)
Gender
 Male76 (56.3)RefRefRefRefRef161 (52.9)Ref
 Female75 (61.4)0.97 (0.59, 1.60)124 (41.6)0.82 (0.58, 1.18)45 (15.3)0.87 (0.55, 1.40)163 (58.3)1.33 (0.93, 1.92)
Age
 1840 (18.8)Ref91 (43.0)Ref35 (16.9)Ref86 (41.4)Ref
 16–1727 (10.5)0.54* (0.31, 0.95)106 (40.0)0.92 (0.63, 1.36)44 (16.7)1.02 (0.61, 1.71)139 (56.5)1.77** (1.20, 2.60)
 14–157 (6.8)0.33* (0.14, 0.77)39 (41.1)0.96 (0.58, 1.59)16 (16.7)1.04 (0.52, 2.08)76 (74.9)4.28*** (2.45, 7.47)
 11–136 (12.6)0.59 (0.23, 1.48)23 (55.4)1.82 (0.85, 3.91)7 (17.3)0.89 (0.37, 2.13)23 (50.4)1.44 (0.70, 2.98)
Social grade
 Middle class60 (12.6)190 (43.9)79 (17.9)241 (55.6)
 Working class20 (11.6)0.95 (0.53, 1.70)69 (39.8)0.93 (0.62, 1.41)23 (13.8)0.74 (0.41, 1.31)83 (54.4)0.86 (0.57, 1.32)
Country
 England70 (12.3)235 (45.3)92 (17.6)268 (53.8)
 Wales4 (13.7)1.05 (0.32, 3.41)5 (22.1)0.29* (0.10, 0.87)5 (16.8)0.89 (0.32, 2.51)15 (57.2)1.37 (0.57, 3.32)
 Scotland6 (11.7)1.06 (0.42, 2.65)19 (30.2)0.50* (0.26, 0.94)5 (8.5)0.49 (0.18, 1.33)41 (69.2)1.95* (1.06, 3.59)
E-cigarette use in the household
 No19 (6.7)86 (34.3)26 (10.2)163 (63.9)
 Yes61 (16.4)2.58** (1.47, 4.54)173 (49.2)1.69** (1.18, 2.43)76 (21.6)2.64*** (1.58, 4.41)161 (48.9)0.56** (0.39, 0.82)
  • N (%)=number and per cent of e-cigarette users who obtained cigarettes through this source within each stratum of the sociodemographic variable. E-cigarette sources are not mutually exclusive, therefore percentages do not add to 100% across sources.

  • *P<0.05, **p<0.01, ***p<0.001.

  • aOR, adjusted OR.