Skip to main content
. 2018 May 23;133(4):379–384. doi: 10.1177/0033354918774064

Table 2.

Adjusted differencesa between Appalachian and non-Appalachian respondents (n = 14 451) on body mass index, beliefs about obesity, and health behaviors, HINTS 4, cycles 1-4, 2011-2014b

Continuous Variable No.c Weighted Meansd Be (95% CI) P Value
Appalachian Non-Appalachian
Body mass index, kg/m2 11 929 28.6 27.7 0.75 (0.08 to 1.42) .03
Daily cups of fruit and vegetables 12 095 5.0 5.1 0.01 (–0.26 to 0.29) .92
Weekly minutes of physical activity 6497 179.7 205.6 5.12 (–56.54 to 66.77) .87
Weekly frequency of strength training, d/wk 9031 0.9 1.2 –0.21 (–0.40 to 0.02) .03
Daily hours of sedentary behavior 11 639 3.8 3.8 –0.02 (–0.33 to 0.30) .90
Weighted Percentages
Categorical Variable No. Appalachian Non-Appalachian OR (95% CI) P Value
Think that behaviors influence obesity 6018 0.69 (0.49-0.97) .03
 Not at all 4.9 2.6
 A little 5.9 4.4
 Somewhat 20.1 14.4
 A lot 69.1 78.7
Think that genes influence obesity a lot 6013 0.90 (0.71-1.15) .40
 Not at all 7.5 7.3
 A little 24.4 23.0
 Somewhat 42.2 41.4
 A lot 26.0 28.3
Confidence in ability to take care of health 12 210 1.04 (0.82-1.32) .74
 Not at all confident 1.0 1.4
 A little confident 4.6 4.7
 Somewhat confident 27.9 26.0
 Very confident 44.1 45.1
 Completely confident 22.4 22.9
Frequency of (non-diet) soda consumption 5575 1.29 (0.85-1.95) .28
 I do not drink any regular soda or pop 35.2 39.0
 <1 day a week 18.4 22.6
 1-2 days a week 13.1 14.7
 3-4 days a week 6.9 8.6
 5-6 days a week 8.4 4.2
 Every day 18.0 10.8
Have smoked ≥100 cigarettes 12 155 45.7 40.3 1.09 (0.85-1.38) .50

Abbreviations: HINTS, Health Information National Trends Survey; OR, odds ratio.

a Results of separate weighted regression analyses for each variable using listwise deletion. Continuous variables were submitted to linear regression analyses; categorical variables were submitted to ordinal regression analyses.

b Data source: HINTS.9,10

c Differences in sample size across models are primarily due to the fact that not all variables were included in each of the 4 cycles of HINTS 4. See the Methods section for more information on measure repetition.

d Weighted means are not adjusted for sociodemographic variables or general health.

e Unstandardized regression coefficients, odds ratios, and 95% confidence intervals refer to differences in mean values between Appalachians and non-Appalachians, controlling for sociodemographic variables (age, sex, race/ethnicity, education, household income, and rural-urban designation) and general health. P < .05 was considered significant.