Quitting Smoking: Reasons for Quitting and Predictors of Cessation among Medical Patients CAROL L. DUNCAN, RD, MPH, STEVENR. CUMMINGS,MD, ESTHER SlD HUDES, PhD, MPH, ELAINEZAHND, PhD, THOMAS J. COATES, PhD

Objective: To describe why medical patients quit smoking a n d the methods they use. Design: Cross-sectional a n d prospective cohort design. Patient smokers were enrolled in a study o f physician counseling about smoking. One y e a r later, 2,581 o f thepatients were asked about quit attempts a n d methods used. O f those, 2 4 5 f o r m e r smokers whose quitting had been biologically validated were interviewed about why a n d h o w they had quit. Setting: Offices o f internists and f a m i l y practitioners in private practice a n d a health maintenance organization. Subjects: Consecutive sample o f ambalatory patients who smoked. Measurements and m a i n results: B a s e l i n e questionnaires included demographic data, smoking history, a n d symptoms a n d diagnoses related to smoking. After one year, subjects were interviewed about smoking status a n d methods u s e d in attempting to quit. Cessation was conf i r m e d by biochemical testing. Those who had quit were asked about reasons f o r quitting. Seventy-seven p e r c e n t o f successfid quitters gave health-related reasons f o r quitting a n d the quitters ranked "ffJarmful to health" us the most important reason f o r quitting. I n a multivariate analysis, those who had a college education, who had social pressures to qui~ a n d who had greater confidence in being able to quit were more likely to have quit smoking one y e a r later, while those who smoked theirf i r s t cigarette within 15 minutes o f awakening a n d who had more diagnoses reiated to smoking were less likely to have quit smoking one y e a r later. Participation in a treatment p r o g r a m a n d having been counseled by a physician o r nurse practitioner were positively related to successful quittin~ while use o f filters o r mouthpieces was negatively related. Conclusions: Concerns about health are the most common reason patients give f o r quittin~ a n d addiction is the most important barrier to quitting. Education, social pressure, p r o v i d e r advice, and f o r m a l programs, but not over-thecounter devices, appear to increase the chances that smokers will quit. Key words: smoking cessation; counseling. J GEN INTERN MED 1992;7:398- 404.

Received from the Division of General Internal Medicine, Department of Medicine (CLD, SRC, EZ, TJC), and the Department of Epidemiology, International Health (SRC, ESH, TJC), University of California, San Francisco, San Francisco, California. Supported by grant #CA38374 from the National Cancer Institute. Dr. Cummings' work is supported in part by the HenryJ. Kaiser Faculty Fellowship in General Internal Medicine. Address correspondence and reprint requests to Ms. Duncan: Prevention Sciences Group, 74 New Montgomery Street, Suite 600, San Francisco, CA 94105.

398

IN THE PAST 25 years 43 million Americans have quit smoking, t Those w h o have quit rate health as the most important reason. 2 Since 70% of all smokers see their physicians each year, physicians have the o p p o r t u n i t y to reinforce a potentially p o w e r f u l motivation for quitting) Familiarity w i t h the characteristics, motivations, and methods associated w i t h smoking cessation can guide physicians in their counseling efforts. Reasons for quitting given b y self-reported quitters in a variety of settings have b e e n described; however, reasons given by medical patients w h o s e quitting has b e e n bioc h e m i c a l l y validated have not b e e n previously examined. 4"6 In a large-scale r a n d o m i z e d trial in w h i c h w e evaluated the effectiveness of a training p r o g r a m for providers in smoking cessation counseling, w e gathered information from patient smokers a b o u t their smoking backgrounds, health status, attitudes toward quitting, and reasons for wanting to quit. One year later w e asked the patients a b o u t quit attempts and methods used. We validated self-reported successful quit att e m p t s and asked the successful quitters a b o u t their reasons for quitting. We report these results below, as w e l l as the result of c o m p a r i n g the characteristics, motivations, and m e t h o d s of successful quitters w i t h those of continuing smokers.

SUBJECTS AND METHODS Providers We recruited 44 private practice internists and 81 internists and 12 registered nurse practitioners (RNPs) from K a i s e r - P e r m a n e n t e Medical Centers in the San Francisco Bay Area. The r e c r u i t m e n t p r o c e d u r e for the counseling intervention has b e e n described in detail elsewhere. 7, ~ We r a n d o m l y assigned providers to an e x p e r i m e n t a l g r o u p designated to receive training and office support and to provide counseling to patients, or to a usual-care control g r o u p designated to provide usual care only.

Patients We enrolled 916 patients from private practices and 2,354 patients from four Kaiser sites. We enrolled

JOURNALOF GENERALINTERNALMEDICINE, Volume 7

patient smokers before they saw their providers. A smoker was defined as anyone w h o had s m o k e d a tob a c c o cigarette, even a puff, in the past seven days. All smokers w h o visited a participating provider w e r e app r o a c h e d until either a m a x i m u m of 30 smokers per provider was enrolled or a m i n i m u m of 15 smokers was enrolled after m o r e than six weeks. At the time of enrollment each patient c o m p l e t e d a questionnaire that asked information a b o u t basic d e m o g r a p h i c characteristics, smoking b a c k g r o u n d with s y m p t o m s that might be related to smoking, attitudes toward quitting, and reasons for wanting to quit. Each patient read a description of the study and was given the option not to participate. The research p r o t o c o l was a p p r o v e d by the Committees on H u m a n Research at b o t h the University of California, San Francisco, and K a i s e r - P e r m a n e n t e Medical Center of Northern California. A t e l e p h o n e interview within seven days of the patient's e n r o l l m e n t asked m o r e information about smoking history and about diagnoses that might be related to smoking.

Procedures Research staff called patients one year after their e n r o l l m e n t dates to d e t e r m i n e their current smoking status. Interviewers asked subjects w h e t h e r they had a t t e m p t e d to quit and, if so, what methods they had used. Those w h o r e p o r t e d not smoking a cigarette during the past seven days w e r e defined as self-reported quitters. Interviewers offered $ 25 to self-reported quitters for breath and saliva samples. Concentrations of carbon m o n o x i d e (CO) in expired air w e r e analyzed using an Ecolyzer® Model 211 Carbon Monoxide Monitor (National Draeger, Pittsburgh, PA). Salivary nicotine concentrations w e r e d e t e r m i n e d by a modification of Jacobs' m e t h o d of gas chromatography. 9 The m e t h o d was modified for simultaneous determinations of nicotine and cotinine using a capillary column. Patients w h o s e saliva concentrations e x c e e d e d 30 n g / m L w e r e classified as smokers unless they r e p o r t e d using nicotine gum. Those w h o w e r e using nicotine g u m w e r e also classified as smokers if their partial pressure of expired CO e x c e e d 16 parts per million. ~° Those w h o p r o v i d e d samples for b i o c h e m i c a l validation w e r e asked to c o m p l e t e a questionnaire a b o u t their reasons for quitting. The first item instructed subjects to write in their o w n words w h y they had quit smoking. From a list of potentially smoking-related s y m p t o m s and diagnoses, subjects marked all those they had e x p e r i e n c e d during the past year. Subjects w e r e asked to rate the i m p o r t a n c e of c o m m o n reasons for quitting on a scale from 1 (not important) to 6 (very important). Likewise, they w e r e asked to rate the importance of the s y m p t o m s and diagnoses they had experienced to their decisions to stop smoking.

(July/August),

1992

399

Data Analysis We found no significant difference b e t w e e n experimental g r o u p patients and control g r o u p patients in reasons for quitting or in rates of smoking cessation, so in subsequent analyses we included all validated quitters as a single study group. 7' 8 Since randomization was p e r f o r m e d at the provider level rather than at the patient level, w e p e r f o r m e d a logistic regression analysis taking clustering into consideration. 11 This analysis, however, did not reveal any clustering effect, hence w e report the results of standard logistic regression analyses. We first c o m p a r e d d e m o g r a p h i c characteristics, smoking histories, and motivational differences bet w e e n smokers and those w h o had s u b s e q u e n t l y quit. Numbers of social pressures w e r e a c c u m u l a t e d from responses to five separate questions on the baseline questionnaire. Having a spouse w h o did not smoke, having family w h o wanted the smoker to quit, having no or few friends w h o smoked, having friends w h o wanted the smoker to quit, and having co-workers w h o objected to the smoker's smoking each c o u n t e d as one social pressure. To test the significance of differences b e t w e e n the smoker and quitter groups for these baseline characteristics, w e used the chi-square test for proportions and the t-test for means. 12 We sorted validated quitters' reasons for quitting into 16 groups, then consolidated those with a comm o n theme. We calculated frequencies by type of reason and by respondent. To c o m p a r e rankings on the six-point i m p o r t a n c e scale w e calculated means and 95% confidence intervals. Types of reasons and mean i m p o r t a n c e ratings w e r e c o m p a r e d by gender, race, age, education, marital status, e m p l o y m e n t status, time off cigarettes, mean n u m b e r of symptoms, and mean n u m b e r of diagnoses. Because individuals tend to agree w i t h questionnaire items to the extent that calculated mean scores do not accurately reflect true relative values, we adjusted for this tendency by c o m p u t i n g each individual's mean response score and subtracting it from each rating that person had given. The resultant standardized ratings w e r e c o m p a r e d b e t w e e n groups using the t-test. Major significant differences found in baseline data by c o m p a r i n g quitters and smokers w e r e entered into a m u l t i p l e logistic regression analysis w i t h validated smoking cessation as the d e p e n d e n t variable. Odds ratios and confidence intervals w e r e calculated for each variable's contribution to smoking cessation. Methods used for quit attempts w e r e c o m p a r e d by gender, race, age, and education. Finally, these methods w e r e added to baseline differences in the multiple regression analysis with validated smoking cessation as the d e p e n d e n t variable. Odds ratios and confidence intervals w e r e calculated for individual

400

Duncaneta/.. SMOKINGCESSATION REASONSAND PREDICTORS TABLE 1

Baseline Characteristicsof the Subjects* Validated Quitters

Smokers

(N = 245*)

(N = 2,652*)

Gender--female Age--mean Race White Nonwhite Educationt High school graduate or less Some college Collegegraduate or more Number of cigarettes/day-- meant Desire to quit smoking-mean, 1O-point scaler Confidence in quitting-- mean, 1O-point scaler Number of diagnoses-- meant Number of symptoms-- meant Tried to quit beforet Smoke within 15 rain of awakeningt Want to quit becausesmoking is harmful to healtht Want to quit becauseof dependencet Number of socialpressuresto quit--meant

59% 43.7 years

58% 44.7 years

69% 31%

63% 37%

27% 38% 35%

40% 37% 23%

16.3

18.6

7.6

6.9

6.3 0.82 1.47 84%

5.1 1.08 1.68 76%

25%

40%

84%

76%

51%

44%

2.70

2.35

*Nranges from 2,209 to 2,652 for smokers and from 220 to 245

for validatedquitters due to missing data. tp < 0.05 for comparisonof smoker and quitter groups by t-test for continuousdata and chi-squaretest for categoricaldata.

methods. Computations were done using the Statistical Analysis System (SAS) 12 running under CMS on an IBM 4341 computer.

RESULTS Subject Characteristics We contacted 2,581 (79%) participants one year after enrollment. Three hundred thirty-four (10%) of those enrolled said they had quit smoking. Two hundred forty-five (73%) of the 334 self-reported quitters consented to, and passed, our validation procedures. Those w h o had quit had been abstinent for a median of 129 days by self-report. A majority of subjects were women, although quitters did not differ significantly from smokers by gender (Table 1). Subjects w e r e predominantly white. More quitters had c o m p l e t e d college, smoked fewer cigarettes, had greater desire to quit smoking, and had greater confidence that they w o u l d quit than did continuing smokers. Those w h o had subsequently quit reported fewer smoking-related diagnoses and fewer symptoms than did those w h o con-

tinued to smoke, and quitters were no more likely to attribute their diagnoses and symptoms to smoking. More quitters than continuing smokers had attempted to quit before. However, the mean n u m b e r of quit attempts did not differ significantly between the two groups. Fewer quitters than continuing smokers smoked their first cigarette within 15 minutes of awakening. This is consistent with the fact that quitters were less likely to say that they w e r e "addicted to smoking" than were smokers (68% vs. 80%, p = 0 . 0 2 3 ) . Subjects differed in reasons for wanting to quit in that those w h o went on to quit were more likely than those w h o did not to say they wanted to quit because smoking was harmful to their health and because they didn't like feeling dependent on cigarettes. Quitters also reported more social pressures to quit than did continuing smokers.

Reasons for Quitting "Own-word" Reasons f o r Quitting. The 245 validated quitters gave 453 reasons for quitting. The numbers of reasons ranged from zero ( " n o n e " ) to four per person. Most reasons were health-related; 60% of reasons related to personal health and an additional 4% related to others' health. Of personal health reasons, 74% were c o n n e c t e d with improving health, while 24% were about preventing ill health. The second largest category o f reasons related to social concerns, such as family pressure to quit. Since subjects gave different numbers of reasons, the n u m b e r of respondents w h o gave each type o f reason better describes group tendencies than does the n u m b e r of reasons in each category. Seventy-seven percent gave at least one health reason for quitting (Table 2). The largest proportion o f respondents, 25.7%, cited one or more symptoms as their reason for quitting. Over one-fifth cited diagnoses. Over one-fifth gave reasons related to general health, often by saying that smoking is "bad for my health," and 22% m e n t i o n e d fear of illness. Twenty-eight percent gave one or more social reasons for quitting. The next largest proportion, 16%, cited aesthetic reasons, most commonly, " b a d smell." Slightly fewer subjects specified physician's or nurse practitioner's advice than mentioned social pressure. Fewer than 10% of subjects quit to relieve feelings of dependence. Importance Ratings o f Reasons. Ratings of the importance of c o m m o n reasons for quitting are presented in Figure 1. "It's harmful to my health" had the highest rating; feeling d e p e n d e n t and reluctance to expose c h i l d r e n / l o v e d ones ranked second and third, respectively. Provider advice was in the mid-range of importance. "Smoking is no longer p o p u l a r " was judged the least important reason.

JOURNALOF GENERAL INTERNALMEDICINE, Volume 7

(July/August), 1992

401

It was harmful to my health Didn't like feehng dependent •

-.,

:. .;,.

.:..-~.:.=.~x.-.;...'-~.:.....~.~:,

•

Didn't want to expose family

.!"~.x:~. - -.~.-,-~.-.:.,,.-4.,.

~''~ .~. . . . ~ It was harmful to others

FIGURE 1. Mean ratings of importance of reasons for quitting on a scale from 1 (not important) to 6 (very important). Bars indicate 9 5 % confidence intervals for the mean. RNP = registered nurse practitioner.

MD/RNP advised quitting ~-"..'... ". . . . . . .

Friends wanted me to quit

~

~$~i~ : ~ : ~ . ' - ~

:: "

Smoking was too expensive •~

Knew smoker who became fll or died ~ . . . . ~

.

,................

,...~. ,..~

..~:.: ~ :~ #.:~. '.~:~:~.... : ~

/

"'.. . . .

Smoking was no longer popular

~

• .x.~:.(.$?..~:~..:,:e....~.,

I

I

I

I

I

!

I

0

1

2

3

4

5

6

Not important

Symptoms and Diagnoses. Table 3 reports the prevalence of smoking-related symptoms and diagnoses among validated quitters. Ninety percent had one or more smoking-related symptom. Respiratory symptoms predominated; over half of quitters reported having experienced either a cough or shortness of breath during the previous year. Shortness of breath achieved the highest mean importance rating (4.6 on the sixTABLE 2

Reasons for Quitting* % o f Respondentst Health Because of symptoms Because of diagnoses For better health; to feel better; because it's bad for my health Fear of illness or desire to prevent illness in myself To improve fitness

77.0

Social Social pressure Family pressure Health of family/friends (e.g., " m y children's health") Media influence

28.0

25.7 23.3 23.3 22.0 3.3

12.7 10.6 7.3 2.0

Aesthetic reasons (e.g., "Everything smelled of ashes.")

16.3

Other (e.g., timing, expense, religion)

13.1

Doctor/nurse practitioner's advice

12.0

Didn't like feeling dependent

7.3

*Responses to " I n your own words, briefly explain why you quit smoking." t N = 453 reasons given by 245 respondents. Percentages add to more than 1O0 because respondents gave from one to four responses.

Very important

TABLE 3

Importance of Smoking-related Symptoms and Diagnoses as Reasons for Quitting % Reporting System or Diagnosis*

Mean Importance-6-point Scale

95% Confidence Interval

Symptom Shortness of breath Chest pain Cough Sore throat Leg cramps Heartburn Facial wrinkles Cold hands or feet

53 29 62 49 22 38 21 44

4.6 3.8 3.6 3.3 2.8 2.6 2,5 1.9

4 . 3 - 4.9 3.3 - 4.3 3.3-3.9 2 . 9 - 3.7 2 . 3 - 3.4 2.2 - 3.0 2.0 - 3.0 1.6-2.2

Diagnosis Angina Asthma Heart attack Emphysema Bronchitis Cancer Peptic ulcer Osteoporosis

6 6 1 5 15 5 1 4

5.4 5.2 5.0 4.9 4.6 3.0 3.0 2.0

4.8-6.0 4 . 5 - 5.9 t 4 . 0 - 5.8 4 . 0 - 5.2 1.4-4.6 f f

*n = 245. tNumber reporting diagnosis is too small for sample estimates to be meaningful.

point scale), followed by chest pain (3.8) and then cough (3.6). Most quitters tended to report more than one smoking-related symptom; a mean of 3.2 symptoms were reported by the 90% who had symptoms. Almost half of quitters (47%) reported having one or more of the diagnoses in Table 3. Subjects rated diagnoses as more important than symptoms in their decisions to quit, although symptoms were more common. Angina had the highest rating among diagnoses, but only 6% of patients reported having angina.

402

DuncanetaL, SMOKINGCESSATION REASONSAND PREDICTORS

TABLE 4

Predictors of Smoking Cessation

Predictors of Smoking Cessation*

Odds

95% Confidence

Ratiot

Interval

Some college or higher education (yes/no)

1,52

1.1 O- 2.10

Number of cigarettes/day (per 10 cigarettes)

1.01

0 . 9 7 - 1.04

Number of diagnoses (per diagnosis, 8 maximum)

0.80

0.69-0.93

Number of symptoms (per symptom, 8 maximum)

0.92

0.83- 1.02

Desire to quit smoking ( 1 point on 1O-point scale)

1,01

0.94- 1.08

Confidence in quitting (1 point on 1O-point scale)

1,12

1,06-1.19

Having tried to quit before (yes/no)

1.43

0.97 - 2.11

Smoking within 15 minutes of awakening (yes/no)

0.58

0.41 - 0.81

Wanted to quit because smoking is harmful to health (yes/no)

1.47

0.95-2.27

Wanted to quit becauseof dependence(yes/no)

1,11

0 . 8 2 - 1.50

Number of social pressure to quit (per pressure, 5 maximum)

1.13

1.02- 1.27

When baseline differences b e t w e e n those w h o had subsequently quit and those w h o c o n t i n u e d to smoke w e r e entered into a m u l t i p l e logistic regression model, college or higher education, greater confidence in b e i n g able to quit, and m o r e social pressures to quit w e r e all associated with a significantly greater chance of quitting. More diagnoses and smoking the first cigarette within 15 minutes of awakening (a sign of addiction) w e r e associated w i t h a significantly decreased chance of having quit one year later (Table 4). We also p u t age, race, and gender in the m o d e l and they w e r e not significant and did not substantially change the other associations in the model. Cessation Methods

*N = 2,292 in analysis. tOdds ratio per unit change.

Demographic Differences. Men and w o m e n did not differ significantly in the types of reasons they gave for quitting. There was no significant difference between racial groups in reasons for quitting. However, the t e n d e n c y for Hispanics to give e x p e n s e a higher i m p o r t a n e e rating than did whites persisted w h e n w e controlled for differences in educational levels ( m e a n i m p o r t a n c e = 4.1 vs. 2.8, p = 0.003). Quitters older than the median age of 41 years rated their providers' advice as m o r e important than did y o u n g e r quitters ( m e a n i m p o r t a n c e = 3.8 vs. 3.1, p = 0 . 0 0 2 ) . On the other hand, y o u n g e r quitters rated concern about exposing their c h i l d r e n / l o v e d ones to smoke m o r e highly than did older quitters ( m e a n i m p o r t a n c e = 4.2 vs. 3.5, p = 0 . 0 0 1 ) . These differences also r e m a i n e d w h e n w e controlled for education. Those w i t h some high school or less education gave significantly f e w e r health reasons for quitting than did m o r e highly educated quitters ( m e a n n u m b e r = 0.86 vs. 1.1, p = 0 . 0 2 5 ) .

Methods used for quit attempts are r e p o r t e d in Table 5. Choice of m e t h o d s differed b y race, age, and educational level. Nonwhites w e r e less likely to have gotten a prescription for nicotine g u m (15% vs. 21%, p = 0 . 0 0 4 ) , to have tried b o o k s or p a m p h l e t s (14% vs. 28%, p < 0 . 0 0 1 ) , and to have tried a g r o u p or p r o g r a m (3% vs. 12%, p < 0 . 0 0 1 ) than w e r e whites. More older patients than y o u n g e r ones had tried books or p a m p h lets (26% vs. 18%, p = 0 . 0 0 6 ) , had relied on counseling from their providers (24% vs. 19%, p = 0 . 0 3 6 ) , and had used filters or m o u t h p i e c e s (11% vs. 6%, p---0 . 0 0 1 ) . These differences persisted after controlling for differences in educational levels. Patients with some college education w e r e m o r e likely to have gotten a prescription for nicotine g u m (21% vs. 15%, p = 0 . 0 1 9 ) , to have relied on h e l p from family and friends (33%vs. 26%, p = 0 . 0 0 5 ) , and to have tried a g r o u p or p r o g r a m (11% vs. 6%, p = 0 . 0 0 3 ) than w e r e those w h o had not attended college. Quitters differed from continuing smokers in that they had more often used a stop-smoking g r o u p or program and had less often used special filters or smoking devices. When methods w e r e added to the m u l t i p l e logistic regression w i t h validated smoking cessation as the d e p e n d e n t variable, those w h o had used a g r o u p or p r o g r a m and those w h o had b e e n c o u n s e l e d by their medical providers w e r e m o r e likely, w h i l e those w h o had used filters or devices w e r e less likely, to have s u c c e e d e d in quitting smoking.

DISCUSSION Seventy-seven p e r c e n t of smokers r e p o r t e d that they had quit, at least in part, because of concerns a b o u t the effects of smoking on health. They also gave health the highest i m p o r t a n c e rating a m o n g reasons for quitting. Studies in other p o p u l a t i o n s have f o u n d health to be an important motivator of smoking cessation, s, 6, 13

JOURNALOFGENERALINTERNALMEDICINE,Vo/ume 7 (July/August), 1992

and the larger majority of those w h o had quit for health reasons r e p o r t e d here is p r o b a b l y due to the fact that our subjects w e r e medical patients. However, it is striking that in this medical setting only 12% of those w h o had quit smoking said that their providers' advice was important. More constant aspects of patients' lives, such as health concerns and the urging of family and friends, may be stronger motivating factors. In an earlier review, Pederson c o n c l u d e d that the m o r e severe the disease and the m o r e i m m i n e n t the danger from c o n t i n u e d smoking, the m o r e likely patients are to c o m p l y with advice to quit. 14 However, w e found that patients with m o r e smoking-related diagnoses w e r e less likely to quit. Perhaps these patients did not perceive themselves to be in i m m i n e n t danger, or p e r h a p s having m o r e than one smoking-related disease indicated that these patients w e r e m o r e severely addicted. Nicotine d e p e n d e n c e is regarded as a major obstacle to quitting a m o n g the current p o p u l a t i o n of smokers.iS, 16 Our finding that those w h o s m o k e d within 15 minutes of awakening w e r e less likely to quit was consistent with this. In fact, smoking within 15 minutes of awakening was the strongest p r e d i c t o r of o u t c o m e a m o n g baseline variables. Patients with m o r e diagnoses might also have b e e n m o r e depressed. Recent studies have presented evidence to suggest that a larger p r o p o r t i o n of smokers than persons in the general p o p u l a t i o n have a history of depression and that the depression itself makes quitting m o r e difficult. 17, 18 We w e r e not able to study this possibility because w e did not measure depression.

403

Those with some college w e r e m o r e likely to quit. This is consistent with current trends in smoking cessation a m o n g the population. National Health Interview data indicate that b y the year 2000 major inequities in smoking prevalence will o c c u r a m o n g educational categories, with three times as m a n y smokers in the lower educational level. 19 Since better educated quitters gave more health reasons for smoking, it may be that education increases motivation to quit by increasing awareness of the health risks of smoking. The challenge to providers will be to use a p p r o a c h e s that are m o r e effective with patients w h o have lower educational levels. A large b o d y of literature suggests that personal self-efficacy increases chances for success in behavioral change, 2°,2~ and, accordingly, researchers have reported that smokers w h o believe they will be successful are m o r e likely to succeed in quit attempts. 22-24 That o u r subjects w h o w e r e m o r e confident in being able to quit w e r e more likely to quit adds to this evidence. Several studies have found that social s u p p o r t is important in smoking cessation and maintenance, 25,26 and Cohen et al. have speculated that social norms in the e n v i r o n m e n t may be the most important social determinant of smoking behavior. 26 We also found that the m o r e social pressures our subjects had, the m o r e likely they w e r e to quit. These pressures included n o n s m o k i n g status of their spouses or partners, family m e m b e r s ' desire that they quit, n o n s m o k i n g status of friends, friends' desire that they quit, and co-workers' objection to their smoking. Our results suggest that the

TABLE 5 Methods Used for Quit Attempts* Predictors of Quitting

Quitters (%) Smokers (%) Odds Ratiot for Quitting 95% Confidence Interval

Help from family or friends

26.9

30.8

0.83

0 . 5 6 - 1.22

Counseling from physician or registered nurse practitioner

22.4

22.2

1.53

1.03-2.28

Educational books or pamphlets

21.6

24.0

0.85

0.55-1.31

Nicotine gum prescription*

18.4

18.4

1.03

0 . 6 5 - 1.63

Stop-smoking group or program*§

12.7

8.0

2.22

1.29-3.81

Sedatives

2.9

4.4

1.04

0.40-2.71

Special filters, smoking devices§

2.0

9.9

O. 13

0.04 - 0.44

Clonidine

0.4

0.5

¶

23.7

24.5

1.08

Other methods volunteered by respondents, such as chewing gum

0.71 - 1.63

* Nranges from 243 to 245 for validated quitters and from 949 to 1,003 for smokers who attempted to quit due to missing data. Percentages add to more than 1O0 because of multiple responses. All predictors are yes/no items. tOdds ratios are from logistic regression analysis including these and all predictors from Table 4. *These items were based on a response to a direct inquiry (Did you fill the prescription?; Did you attend a group or program?), whereas the other items were identified from a list read to subjects. §p < 0.05 for comparison of quitter and smoker groups by chi-square test. ¶Number reporting use of method is too small for sample estimates to be meaningful.

404

Duncan eta/., SMOKINGCESSATIONREASONSAND PREDICTORS

emphasis on smoke-free environments and the declining prevalence of smoking may increase quitting among smokers. C h a p m a n r e p o r t e d t h a t s m o k i n g c e s s a t i o n prog r a m s c o n t i b u t e d l i t t l e to t h e o v e r a l l d e c l i n e in s m o k i n g c e s s a t i o n in t h e U n i t e d K i n g d o m . 27 F i o r e e t al. in t h e U n i t e d States a n a l y z e d data f r o m t h e 1 9 8 6 A d u l t U s e o f T o b a c c a S u r v e y to c o n c l u d e that c e s s a t i o n p r o g r a m s s e r v e a small, b u t i m p o r t a n t , p o p u l a t i o n o f s m o k e r s t h a t i n c l u d e s h e a v i e r s m o k e r s . 2s W e f o u n d t h a t u s e o f a s m o k i n g c e s s a t i o n g r o u p o r p r o g r a m w a s r e l a t e d to a s o m e w h a t greater chance of quitting. This does not p r o v e that s u c h p r o g r a m s are e f f e c t i v e ; it is also possib l e that m o r e m o t i v a t e d p a t i e n t s are m o r e l i k e l y to j o i n t i m e - c o n s u m i n g p r o g r a m s . N o n w h i t e a n d less e d u c a t e d p a t i e n t s w e r e less l i k e l y to h a v e t r i e d a g r o u p o r p r o g r a m . F i o r e has p o i n t e d o u t that s u c h p r o g r a m s are l i k e l y to b e less a c c e s s i b l e to p o o r a n d m i n o r i t y g r o u p s b e c a u s e o f c o s t o r c u l t u r a l biases. W e also f o u n d t h a t c o u n s e l i n g b y t h e i r m e d i c a l p r o v i d e r s p r e d i c t e d s m o k i n g c e s s a t i o n in this p a t i e n t p o p u l a t i o n . O u r p r o v i d e r s w e r e s e l f - s e l e c t e d to p a r t i c i p a t e in a test o f t h e e f f e c t i v n e s s o f t r a i n i n g p r o v i d e r s t o counsel patients for smoking cessation. However, w e c o n t r o l l e d f o r t r e a t m e n t g r o u p in t h e m u l i t i p l e l o g i s t i c r e g r e s s i o n , a n d t r e a t m e n t g r o u p w a s n o t in i t s e l f a sign i f i c a n t p r e d i c t o r , n o r d i d it affect t h e r e l a t i o n s h i p bet w e e n p r o v i d e r advice and successful quitting. This suggests that provider's advice was s o m e w h a t effective, w h i l e f u r t h e r t r a i n i n g a b o u t s m o k i n g c e s s a t i o n d i d litt l e to i n c r e a s e t h e e f f e c t i v e n e s s o f that a d v i c e . 7, 8 T h i s s u p p o r t s t h e c o n c e p t that s i m p l y g i v i n g a d v i c e t o q u i t m a y b e m o r e i m p o r t a n t t h a n h o w t h e c o u n s e l i n g is d o n e . 29 W e c o n c l u d e that c o n c e r n s about health are the most c o m m o n reason patients give for quitting smoking a n d that a d d i c t i o n is t h e m o s t i m p o r t a n t b a r r i e r t o q u i t ting. E d u c a t i o n , s o c i a l p r e s s u r e , p r o v i d e r a d v i c e , a n d f o r m a l p r o g r a m s , b u t n o t o v e r - t h e - c o u n t e r d e v i c e s , app e a r to i n c r e a s e t h e c h a n c e s that s m o k e r s w i l l q u i t .

The authors acknowledge George Stone, PhD, for his advice and assistance, and Robert Richard, MA, for statistical consultation.

REFERENCES 1. Office on Smoking and Health. Tobacco use in 1986--methods, & basic tabulations from adult use of tobacco survey. DHHS Publication No. OMM 90-2004. Bethesda, MD: National Centers for Disease Control, 1986. 2. Green DE. Psychological factors in smoking. In: Jarvick ME, Cullen JW, Gritz ER, Vogt TM, West CG (eds.). Research on smoking behavior (NIDA Research Monograph 17). DHRW Publication No. ADM 78-581. Washington, DC: U.S. Government Printing Office, 1977. 3. Ockene JK. Smoking intervention: the expanding role of the

physician. Am J Public Health. 1987;77:782-3. 4. Barnes GE, Vulcano BA, Greaves L. Characteristics affecting successful outcome in the cessation of smoking. Int J Addict. 1985;9:1429-34. 5. Feldman BM, Richard E. Prevalence of nurse smokers and variables identified with successful and unsuccessful smoking cessation. Res Nurs Health. 1986;9:131-8. 6. Hammond EC, Percy C. Ex-smokers. N Y State J Med. 1958;58:2956-9. 7. Cummings SR, Coates T, Richard RJ, et al. Training physicians in counseling about smoking cessation: a randomized trial of the "Quit for Life" program. Ann Intern Med. 1989;110:641-7. 8. Cummings SR, Richard RJ, Duncan CL, et al. Training physicians about smoking cessation: a controlled trial in private practices. J Gen Intern Med. 1989;4:482-9. 9. Jacob P, Wilson M, Benowitz NL. Improved gas chromatographic methods for the determination of nicotine and cotinine in biologic fluids. J Chromatogr. 1981 ;221:61-70. 10. Cummings SR, Richard RJ. Optimum cut-offpoints for biochemical validation of smoking status. Am J Public Health. 1988;78:574-5. 11. Zeger SL, Liang K-Y. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121-30. 12. SASInstitute, Inc. SASuser's guide: statistics, version 5 edition. Cary, NC: SASInstitute, Inc., 1985. 13- Fletcher C, Doll R. A survey of doctors' attitudes to smoking. BrJ Prey Soc Med. 1969;23:145-53. 14. Pederson LL. Compliance with physician advice to quit smoking: a review of the literature. Prev Med. 1982; 11:71-84. 15. Hughes JR. Clonidine, depression and smoking cessation. JAMA. 1988;259:2901-2. 16. Hughes JR, Gust SW, Pechachek TF. Prevalence of tobacco dependence and withdrawal. Am J Psychiatry. 1987;144(2): 205-8. 17. Glassman AH, Stemer F, Walsh BT, et al. Heavy smokers, smoking cessation, and clonidine. JAMA. 1988;259:2863-6. 18. Hall SM, Mufioz RF, Reus VI. Depression, dysphoria and smoking cessation. In: Scientific meetings of the Committee on the Problems of Drug Dependence. Richmond, VA: National Institute on Drug Abuse, 1992 (In press). 19. Pierce JP, Fiore MC, Novomy TE, Hatziandreu EJ, Davis RM. Trends in cigarette smoking in the United States. Projections to the year 2000. JAMA. 1989;261:61-5. 20. Bandura A, Adams NE. Analysis of self-efficacy theory of behavior change. Cogn Ther Res. 1977;1:287-310. 21. Williams SL, Watson N. Perceived danger and perceived self-efficacy as cognitive determinants of acrophobic behavior. Behav Ther. 1985;16:136-46. 22. Russel MA, Armstrong E, Patel UA. Temporal contiguity in electric aversion therapy for cigarette smoking. Behav Res Ther. 1976;14:103-23. 23. Mothersill KJ, McDowell I, Rosser W. Subject characteristics and long term post program smoking cessation. Addict Behav. 1988;13(1):29-36. 24. Ockene JK, Benfari RC, Nuttall RL, Hurrwitz I, Ockene JS. Relationship of psychosocial factors to smoking behavior change in an intervention program. Prev Med. 1982; 11:13-26. 25. Lichtenstein E, Glasgow RE, Abrahms DB. Social support and smoking cessation: in search of effective interventions. Behav Ther. 1986;17:607-1926. Cohen S, Lichenstein E, Mermelstein RJ, McIntyre-Kingsolver KO, BaerJS, Kamarck TW. Social support interventions for smoking cessation. In: Gottlieb BH (ed.). Marshalling social support: formats, processes and effects. New York: Sage, 1988:211-40. 27. Chapman S. Stop-smoking clinics: a case for their abandonment. Lancet. 1985;1:918-20. 28. Fiore MC, Novomy TE, Pierce JP, et al. Methods used to stop smoking in the United States: do cessation programs help?JAMA. 1990;263:2760-5. 29. Kottke TE, Battista RN, DeFriese GH, Brekke ML. Attributes of successful smoking cessation interventions in medical practice. JAMA. 1988;259:2883-901.