Cotinine

INVESTIGATORS have evaluated smoking behaviors in epidemiological studies in North America and northern Europe by obtaining subjects' responses to a standard questionnaire. (1) Coultas et al., (2) however, reported that the true smoking prevalence is underestimated in some populations when such a questionnaire is used. Cotinine, a metabolite of nicotine with a longer half-life than nicotine (i.e., 7-40 hr), is an objective measure of tobacco exposure, (3,4) and it can be evaluated in different body fluids, including blood. (5,6)

The European Community Respiratory Health Survey (ECRHS) was an international, 2-stage survey of the general population that measured variations in the prevalence of asthma, asthma-like symptoms, treatment for asthma, and exposure to known and suspected risk factors for asthma. (7) Smoking behavior was evaluated via a standardized clinical interview that addressed past and present tobacco use and exposure to environmental tobacco smoke (ETS).

At the time of the current study, epidemiological studies of the agreement between reported tobacco smoke exposure and the value of a biochemical marker of tobacco exposure had not been conducted in Italy. In the current study, we sought to validate the results of the ECRHS questionnaire with respect to smoking habits and ETS exposure in Italy. We measured serum cotinine levels in a randomized sample of young adults who were invited to the chest clinic at Verona--1 of the 3 Italian cities that were included in the ECRHS.

Materials and Method

Study design and subjects. We invited 504 randomly selected Caucasian subjects to complete a standardized respiratory interview, which included a description of their smoking habits, and each subject was asked to provide a blood sample for serum cotinine measurement. All smokers were asked about the amount of time that had passed since they had smoked their last cigarette and if they had inhaled the smoke. All subjects who reported regular exposure to ETS answered the following questions:

(1) Not counting yourself, how many people in your household smoke regularly?

(2) Do people smoke regularly in the room where you work?

(3) How many hours per day are you exposed to other people's tobacco smoke?

We divided subjects into 4 groups, according to smoking habits: (1) never smokers ([a] subjects who had not smoked for at least the past 1 yr, [b] subjects who had not smoked more than 20 packs of cigarettes [i.e., 360 gm of tobacco] during their lifetimes, or [c] had not smoked more than 1 cigarette/day or 1 cigar/wk for the past 1 yr); (2) exsmokers (i.e., subjects who had smoked for at least 1 yr but not during the most recent month); (3) smokers (i.e., subjects who had smoked for at least 1 yr and during the most recent month); and (4) regularly ETS-exposed subjects (i.e., subjects who had been exposed regularly [that is, during most clays and/or nights] to ETS during the most recent 12 mo).

Occupation and education were evaluated by the subject's responses to the following questions: "Are you currently employed or self-employed?" and "At what age did you complete full-time education?" Education was categorized into 2 levels: (1) subjects who had attended compulsory school (i.e., [less than or equal to] 8 yr), and (2) subjects with higher (i.e., > 8 yr) education levels.

Interviews and blood sampling. A clinical interview was performed and venous blood drawn during the same visit, which occurred at some point in time between February 1992 and June 1993 for each subject. Serum was frozen at-70[degrees]C and stored for subsequent analysis in batches. All samples were placed with dry ice and sent to the laboratory of the National Institute of Health in Rome. Analyses were completed within 3 yr of serum collection. Serum cotinine levels were measured in duplicate with radioimmunoassay, as described by Van Vunakis et al., (3) who used tritiated cotinine, antiserum raised in rabbits, and goat anti-rabbit gamma globulin to separate antibody-bound cotinine from free analyte. To quantify cotinine, the laboratory used standards of 0.2-20 ng/ml. The limit of detection was 0.2 ng/ml, and the interassay coefficient of variation was 6-10%.

The level of cross-reactivity of the cotinine antibody with other nicotine metabolites was less than 5%, with the exception of trans-3'-hydroxycotinine, which inhibited the reaction by 34%. (8) Assays were performed without knowledge of questionnaire responses. We considered a cotinine level [greater than or equal to] 14 ng/ml to be indicative of "biochemical" smoking and, therefore, such levels were not considered as an outcome of ETS exposure. (9)

Statistics. We analyzed the data with SPSS 7.5 for Windows (SPSS, Inc. [Chicago, Illinois]) and Egret (SERC [Seattle, Washington]). Comparisons of proportions were made with chi-square tests, and means were compared with Student's t test. Agreement between smoking behavior(s) and classification of subjects according to serum cotinine levels was tested by Cohen's [kappa]. (10)

Results

Questionnaires were completed by 375 (74.4%) of the 504 randomly selected Caucasian subjects: 129 (34.4%) were smokers, 79 (21.1%) were exsmokers, and 167 (44.5%) were never smokers. The demographic features of the sample--stratified by age, sex, education level, and smoking habits--are shown in Table 1.

All smokers were cigarette smokers, except 1 subject who smoked only a pipe (50 gm/wk); 2 subjects smoked cigarettes and a pipe (50 gm/wk) and cigarettes and cigarillos (1/day), respectively. The mean number of cigarettes smoked was 13.7 (standard deviation = [+ or -] 8.0) in males and 11.3 ([+ or -] 8.6) in females. The smoking prevalence rate was higher in males (p < .05) and in subjects who reported < 8 yr of education (odds ratio [OR] = 1.8; 95% confidence interval [CI] = 1.0, 2.7; p < .05). Never smokers were significantly younger than exsmokers and smokers (p < .001 [Table 1]), and males who reported smoking less than 10 cigarettes/day were younger than "hard" smokers who smoked > 20 cigarettes/day (32.7 yr vs. 37.1 yr, respectively; p = .02). Regardless of gender, smokers and exsmokers started smoking at the mean age of 16.7 yr ([+ or -] 3.1 yr); among hard smokers, there was a significant difference (males vs. females: 16.1 [+ or -] 1.5 yr and 19.5 [+ or -] 5.0 yr, respectively; p < .001).

ETS was reported by 216 (57.6%) of 375 subjects, and the mean exposure was 3.8 hr/day ([+ or -] 3.4 hr/day [Table 2]). The higher exposure rate was reported by smokers (99/129 subjects, 76.7%; p < .05), whereas exsmokers reported a lower rate (43/79 subjects, 54.4%), and never smokers reported the lowest exposure (74/167 subjects, 44.3%). The exposure rate was similar at home (20.2%) and at work (22.5%) in smokers, but it was significantly higher at work in exsmokers (26.6%, p < .05). Never smokers had similar exposures while at work, home, and "other" places. The workplace was, therefore, considered the setting in which exposure occurred most frequenty, and more than one-half of our sample (135/240; 56.3%) was employed in such a setting. The mean exposure duration was not related to smoking habits but varied with the exposure setting (i.e., home, 2.3 hr/day; work, 4.4 hr/day (p < .001); both home and work, 6.5 hr/day (p < .001); and "other" places, 2.2 hr/day. "Other" place exposures could be evaluated only in regularly ETS-exposed subjects (n = 30) who did not report exposure at home or work.

Serum cotinine measurements were performed for 368 of 375 subjects who completed the questionnaire; measurements were unavailable for 7 subjects. When we used a serum cotinine cut-off of 14 ng/ml to distinguish smokers from nonsmokers, 11 subjects who reportedly were nonsmokers (i.e., 5 never smokers and 6 exsmokers) were actually "biochemical smokers" ([greater than or equal to] 14 ng/ml), and 11 subjects who reportedly were smokers were "biochemical nonsmokers" (Table 3). The agreement between self-reported smoking habits and the biochemical definition of smoking habits was very good (Cohen's [kappa] = 0.93).

The mean serum cotinine level was 2.2 ng/ml ([+ or -] 5.7 ng/ml SD) among nonsmokers who were unexposed to ETS; no differences existed between males and females. In nonsmokers exposed to ETS, the mean cotinine level was 11.7 ng/ml ([+ or -] 32.8 ng/ml), which was significantly higher than the mean level detected in nonsmokers unexposed to ETS (p < .02). Following exclusion of "biochemical nonsmokers" in the statistical analysis, the mean serum cotinine value was 1.7 ng/ml ([+ or -] 2.1 ng/ml) and 2.6 ng/ml ([+ or -] 2.6 ng/ml), respectively (p < .002). In nonsmokers, serum cotinine was strictly related to the duration (i.e., number of hours) of ETS exposure ([R.sup.2] = .136, p < .05).

We detected no differences in mean serum cotinine levels in subjects (a) unexposed to ETS, (b) exposed only at work, (c) exposed only at home, and (d) exposed "other" places. Among individuals who were regularly exposed to ETS at home and at work, serum cotinine values were significantly higher than in subjects exposed in 1 setting only (4.3 [+ or -] 3.5 ng/ml vs. 2.0 [+ or -] 1.9 [mean [+ or -] SD], p < .05). Finally, mean serum cotinine levels of individuals who were regularly exposed to ETS at home only were not related to the number of smokers residing in the home.

As expected, mean serum cotinine levels in smokers were significantly higher (273.3 ng/ml [[+ or -] 229.9 ng/ml]) than in any other group, and these levels were directly related to the number of cigarettes smoked daily ([R.sup.2] = .6693, p < .001 [Fig. 1]); however, there were no differences between males and females or between inhalers and noninhalers (7 subjects). Eleven of the 124 self-reported smokers (9 females, 2 males) had serum cotinine levels < 14 ng/ml. Nine of the 11 individuals were "light" smokers (i.e., < 5 cigarettes/day), and most individuals had quit smoking at least 20 hr prior to phlebotomy.

[FIGURE 1 OMITTED]

Discussion

To our knowledge, this is the first study conducted in Italy in which the investigators considered the validation of a standardized questionnaire with respect to smoking behavior and ETS exposure, settings, and duration of smoking/exposure. Our data indicate that the ECRHS questionnaire is a fairly accurate estimator of both smoking habit and ETS exposure in young adults from Verona, Italy.

The rates of smokers, never smokers, and exsmokers were similar to the mean rates observed in the ECRHS. (11) It appears, therefore, that being a smoker did not influence participation in the clinical phase of the survey--at least in Verona.

In our study, we considered nonsmokers with serum cotinine levels [greater than or equal to] 14 ng/ml to be "misclassified"; with this in mind, the mean misclassification rate of smoking status in our survey was approximately 3%--which is similar to rates reported in other studies. (12-14) This misclassification can result from data collection errors; however, in our study, we collected data in accordance with standard quality control procedures, and no data entry errors were found. Misclassification can also occur when a subject misunderstands smoking-habit questions. All of our subjects were Caucasian, all spoke Italian, and misclassified subjects had the same education levels as did other subjects. Perhaps "occasional" smokers (i.e., < 1 cigarette/day) might have answered "no" when asked whether they smoked regularly, or "regular" smokers might have cut back or quit smoking after the initial telephone contact; however, we were unable to assess these possibilities. Finally, misclassification could have resulted from the use of an inappropriate cotinine cut-off point.

In our sample, smoke inhalation did not appear to influence serum cotinine levels; cotinine levels were no different in noninhalers or in inhalers who smoked the same number of cigarettes/day, and values were strictly linked with the number of cigarettes. Perhaps this lack of difference resulted from rapid nicotine absorption by the mucosa of the mouth and by the upper airways or from partial, unnoticed inhalation. Additional studies that include a larger number of noninhaler smokers are necessary to elucidate this point.

An interesting result in our study was the relationship between ETS exposure and serum cotinine levels. We showed that, after the elimination of "biochemical smokers" from the analysis, there was a significant difference between the mean serum cotinine values in ETS-exposed and nonexposed subjects, with the former having levels that were approximately 2 times greater than the latter. Compared with the serum cotinine values of smokers, regularly ETS-exposed individuals had cotinine levels similar to those found in our subjects who smoked less than 5 cigarettes/day and who had quit smoking at least 20 hr prior to blood sampling; these results comported with those provided by other investigators. (13,15) The absence of an established cut-off value of serum cotinine that we could have used to differentiate the ETS-exposed subjects from the non-ETS-exposed subjects prevented us from (1) classifying each subject in his or her group and, therefore, (2) measuring the possible misclassification rate for the reported regularly ETS-exposed subjects.

A detectable serum cotinine level in the majority of subjects in our survey confirmed the widespread reported ETS exposure found in our sample; in most cases, exposure occurred at work or at home--despite the fact that exposure at "other" places most likely was underestimated. Additional study of the duration of exposure in each setting would have allowed us to assess the role of each type of exposure in modifying serum cotinine levels. The detection of cotinine in nonsmokers who were not exposed "regularly" could have resulted from occasional ETS exposure. Finally, we cannot exclude the fact that serum cotinine could result, in part, from dietary intake inasmuch as minute quantities of nicotine have been detected in some foods, (16) but Tunstall-Pedoe et al. (17) indicated that the contribution of dietary sources of nicotine to serum cotinine levels is small, compared with ETS exposure.

In nonsmokers, the mean cotinine level was significantly higher in subjects who declared ETS exposure than in subjects who did not report exposure--even when we excluded "biochemical" smokers from the analysis. In addition, we found a dose-response relationship between mean serum cotinine level and duration of reported exposure to other individuals' smoke; the average cotinine level was increased significantly in subjects exposed both at home and at work. Furthermore, our analysis documented that the main source of ETS exposure was the workplace, which accounted for approximately 70% of the total exposure. In Italy, tobacco smoking is forbidden by law in many places (e.g., classrooms, hospitals, cinemas, public offices), but such prohibition is not specifically enforced in the workplace. Our data suggest that smoking tobacco in the workplace should also be prohibited.

In summary, serum cotinine levels confirmed the validity of the answers to the questions of the ECRHS questionnaire about smoking and ETS exposure. The standardized ECRHS questionnaire revealed a widespread exposure to tobacco smoke in our sample.

Submitted for publication September 28, 2000; revised; accepted for publication January 11, 2001.

Requests for reprints should be sent to Dr. Mario Olivieri, Servizio di Medicina del Lavoro, Policlinico "G. Rossi," Piazza L.A. Scuro, 10, 37134 Verona, Italy.

E-mail: mario.olivieri@univr.it

References

(1.) Hammond E. Smoking in relation to the death rates of one million men and women. Natl Cancer Inst Monogr 1966; 19: 127-204.

(2.) Coultas DB, Howard CA, Peake GT, et al. Discrepancies between self-reported and validated cigarette smoking in a community survey of New Mexico Hispanics. Am Rev Respir Dis 1988; 137: 810-14

(3.) Van Vunakis H, Gijka HB, Langone JJ. Radioimmunoassay for nicotine and cotinine. In: O'Neil IK, Brunnemann KD, Dodet B, Hoffman D (Eds). Environmental Carcinogens--Methods of Analysis and Exposure Measurement. Vol. 9. Lyon, France: International Agency for Research on Cancer, 1987; pp 317-30.

(4.) Rosa M, Pacifici R, Altieri I, et al. How steady state cotinine concentrations in cigarette smokers is directly related to nicotine intake. Clin Pharmacol Ther 1992; 52:324-29.

(5.) Watts RR, Langone JJ, Knight GJ, et al. Cotinine analytical workshop report: consideration of analytical methods for determining cotinine in human body fluids as a measure of passive exposure to tobacco smoke. Environ Health Perspect 1990; 84:173-82.

(6.) Pacifici R, Altieri I, Gandini L, et al. Nicotine, cotinine and trans-3'-hydroxycotinine levels in seminal plasma of smokers: effects on sperm parameters. Ther Drug Monit 1993; 15:358-63.

(7.) Burney PGJ, Luczynska C, Chinn S, et al. The European Community Respiratory Health Survey. Eur Respir J 1994; 7:954-60.

(8.) Zuccaro P, Pichini S, Altieri I, et al. Interference of nicotine metabolites in cotinine determination by RIA. Clin Chem 1997; 43:180-81.

(9.) Cummings SR, Richard RJ. Optimum cut-off points for biochemical validation of smoking status. Am J Public Health 1988; 78: 574-75.

(10.) Altman DG. Some common problems in medical research. In: Altman DG (Ed). Practical Statistics for Medical Research. London, UK: Chapman & Hall, 1992; pp 396-439.

(11.) Kogevinas M, Anto JM, Sunyer J, et al. Occupational asthma in Europe and other industrialised areas: a population-based study. Lancet 1999; 353:1750-54.

(12.) Perez-Stable EJ, Marin G, Marin BV, et al. Misclassification of smoking status by self-reported cigarette consumption. Am Rev Respir Dis 1992; 145:53-57.

(13.) Jarvis N, Tunstall-Pedoe H, Feyerabend C, et al. Biochemical markers of smoke absorption and self-reported exposure to passive smoking. J Epidemiol Community Health 1984; 38:335-39.

(14.) Wagenknecht LE, Burke GL, Perkins LL, et al. Misclassification of smoking status in the CARDIA study: a comparison of self-report with serum cotinine levels. Am J Public Health 1992; 82:33-36.

(15.) Wald N, Boreham J, Bailey A, et al. Urinary cotinine as a marker of breathing other people's tobacco smoke. Lancet 1984; 1: 230-31.

(16.) Sheen SJ. Detection of nicotine in foods and plant materials. J Food Sci 1988; 53:1572-73.

(17.) Tunstall-Pedoe JJ, Woodward M, Brown CA. Tea drinking, passive smoking, smoking deception and serum cotinine in the Scottish heart health study. J Clin Epidemiol 1991; 44:1411-14.

COPYRIGHT 2002 Heldref Publications
COPYRIGHT 2003 Gale Group