Background

Cross-sectional studies consistently demonstrate that people who smoke cigarettes (“current smokers”) eat less nutritious diets than do people who used to smoke regularly and quit (“former smokers”) and persons who never became established smokers (“never smokers”)^1,2. More specifically, fruit and vegetable consumption (FVC) is consistently lower among current smokers than among former and never smokers. This behavioral pattern might simply indicate a stronger health orientation among nonsmokers than among current smokers. The relationship might also be biological. Data from cross-sectional studies do not allow us to determine the direction of the relationship between FVC and cigarette smoking. For example, smokers who quit might switch to eating more fruits and vegetables after they quit compared to those who continue smoking. Alternatively, smokers who eat more fruits and vegetables might be more likely to quit. To better understand the nature of the known cross-sectional relationship, we conducted a longitudinal study to determine if FVC actually predicts cessation among a cohort of 751 current smokers².

The longitudinal study

We surveyed (by telephone) 1,000 current smokers throughout the United States who were at least 25 years old and asked them about their smoking habits, levels of nicotine dependence, frequency of consumption of fruits and vegetables (using the questions from the 2003 Behavioral Risk Factor Surveillance System), and other important health behaviors (such as how much they exercised, whether they drank alcohol excessively, and whether they used illicit drugs). We also asked them to tell us their age, gender, race/ethnicity, educational attainment, and household income. We looked at fruit consumption, vegetable consumption, and total FVC, dividing each of these measures into quartiles. We re-contacted 751 of our 1,000 baseline respondents (75.1% response rate) 14 months later to determine if they were abstinent from cigarettes and all other tobacco product for at least 30 days.

Fruit and vegetable consumption and nicotine dependence

Indicators of nicotine dependence included smoking at least 20 cigarettes each day, smoking the first cigarette of the day within 30 minutes of awakening, and a score of at least nine on a modified version of the Nicotine Dependence Syndrome Scale. Those in the highest three quartiles of fruit consumption were less likely to exhibit all three indicators of nicotine dependence than those in the lowest quartile. Those in the highest quartile of vegetable consumption were less likely to exhibit any indicator of dependence than those in the lowest quartile. And those in the highest two quartiles of FVC scored lower on all three indicators of dependence than those in the lowest quartile of FVC.

Fruits and vegetable consumption and quitting smoking

We assessed the probability of quitting smoking in a multivariable analysis that statistically controlled for age, gender, race/ethnicity, education, income, heavy drinking, street drug use, and exercise. Smokers who at baseline ate the most fruits and vegetables (Quartile 4) were more likely to be abstinent from cigarettes and all other tobacco products than were those who ate the fewest servings of fruits and vegetables (Quartile 1). The adjusted odds ratio was 3.05 (p < 0.01).

Possible explanations

Several biological mechanisms might explain our findings. One is palatability. Previous research indicates that high FVC may worsen the perceived taste of cigarettes3. Another is via reward systems in the brain. Fruit sugars may stimulate dopamine levels and thus reduce the perceived need for a cigarette (nicotine also stimulates dopamine). A third potential mechanism is that high levels of FVC might reduce levels of depression, which has been associated with relapse to smoking among persons attempting to quit. A fourth possible explanation involves satiety. Feelings of hunger for food may be confused with cigarette cravings. Since FVC is a major determinant of satiety, higher levels of satiety might reduce cigarette cravings. Finally, high FVC (and water) reduce constipation, which is a withdrawal symptom for some people trying to quit smoking.

Future work

Additional cohort studies are needed to determine if these findings replicate. They should incorporate other food groups as well. In addition, experimental studies are needed to rule out the possibility of unmeasured factors influencing our observational findings.