Skip to main content

Table 1 Descriptive summary and association with validity of self-reported smoking status, 45+ year old "true" smokers, United States, 1988–1994.

From: Sociodemographic characteristics and diabetes predict invalid self-reported non-smoking in a population-based study of U.S. adults

  Invalid Self-Report Valid Self-Report  
Explanatory Variables 112 (n) 5.8% 1371 (n) 94.2% ORCrude 95% CI
Gender      
   Female 52 5.7% 614 94.3% 1.00
   Male 60 5.9% 757 94.1% 1.05 (0.65–1.70)
Age (years old)      
   45–64 48 3.6% 962 96.4% 1.00
   65+ 64 12.6% 409 87.4% 3.88 (1.95–7.73)*
Race/Ethnicity      
   Non-Hispanic White 48 5.8% 615 94.2% 1.00
   Non-Hispanic Black 48 8.5% 427 91.5% 1.53 (1.04–2.26)*
   Mexican-American 15 5.1% 279 94.9% 0.87 (0.42–1.81)
Income      
   Less than $20,000 69 6.6% 738 93.4% 1.31 (0.74–2.33)
   $20,000 or more 39 5.1% 602 94.9% 1.00
High School Graduate      
   Yes 39 4.4% 671 95.6% 1.00
   No 73 8.6% 688 91.4% 2.04 (1.35–3.08)*
Diabetes      
   Yes 19 15.0% 118 85.0% 3.26 (1.38–7.72)*
   No 93 5.1% 1251 94.9% 1.00
Myocardial Infarction      
   Yes 10 6.4% 97 93.6% 1.12 (0.42–2.99)
   No 102 5.8% 1264 94.2% 1.00
Hypertension      
   Yes 42 7.0% 395 93.0% 1.35 (0.59–3.08)
   No 69 5.3% 970 94.7% 1.00
  1. Note, unweighted number with weighted percent. Bold font identifies statistically significant findings (*p < 0.05).
  2. All adults did not currently use smokeless tobacco, pipe, cigars, or nicotine gum and had serum cotinine data.
  3. "True" smokers were defined as having 15+ng/ml serum cotinine. Invalid self-report was defined as "true" smokers self-reporting as non-smokers. Valid self-report was defined as "true" smokers self-reporting as smokers.
  4. ORCrude = Unadjusted odds ratio for the association between invalid self-report and the potential explanatory variable.