Objective To determine whether attempts to prevent weight gain will increase success rates for stopping smoking.
Design 16 week, open, randomised study with 1 year follow up.
Setting Obesity unit.
Subjects 287 female smokers who had quit smoking before but started again because of weight concerns.
Intervention Combination of a standard smoking cessation programme with nicotine gum and a behavioural weight control programme including a very low energy diet. A control group was treated with the identical programme but without the diet.
Main outcome measure Sustained cessation of smoking.
Results After 16 weeks, 68/137 (50%) women had stopped smoking in the diet group versus 53/150 (35%) in the control group (P = 0.01). Among these women, weight fell by mean 2.1 (95% confidence interval 2.9 to 1.3) kg in the diet group but increased by 1.6 (0.9 to 2.3) kg in the control group (P [is less than] 0.001). After 1 year the success rates in the diet and control groups were 38/137 (28%) and 24/150 (16%) respectively (P [is less than] 0.05), but there was no statistical difference in weight gain.
Conclusions Combining the smoking cessation programme with an intervention to control weight helped women to stop smoking and control weight.
A meta-analysis has established that nicotine gum is better than placebo in achieving smoking cessation. Overall success rates, however, remain modest. About 80% of people gain weight after stopping smoking. Although the mean weight gain has been described as modest, some people gain substantial weight.[3,4] In some studies,[3-5] but not all, the weight increase has been greater in women. Women are generally more concerned about weight gain, and this seems to play an important part in relapse to smoking.[7,8]
Changes in basal metabolic rate,, altered food preferences, or food as a substitute for the psychological effects of tobacco consumption might explain the weight increase. The fall in basal metabolic rate that occurs after stopping smoking cannot be controlled by dietary recommendations, but the other factors could be controlled. However, studies so far have failed to show any benefit of concurrent weight interventions in connection with stopping smoking.[12,13]
Our obesity unit has developed several programmes that show promising results with weight control in obesity.[14,15] Intermittent very low energy diets used over a year have had similar results to continuous very low energy diets.[16-18] The Smoking Cessation Clinical Practice Guideline Panel and staff recommended "that smokers not take strong measures to counteract weight gain during a quit attempt."[19,20] We conducted a randomised trial to determine whether intermittent use of a very low energy diet to improve weight control affected the success of a smoking cessation programme.
Participants and methods
We invited 25 female smokers to focus group discussions to describe their problems with weight control while attempting to stop smoking. These discussions formed the background for the study design.
We conducted an open, randomised study of a smoking cessation programme with nicotine gum (Nicorette 2 or 4 mg) and moderate behavioural advice in combination with a behavioural weight control programme and intermittent very low energy diet (Nutrilett 1.76 MJ/day) as total food replacement. The control group followed an identical programme but did not receive the very low energy diet. The study comprised 11 sessions during 16 weeks (weeks 0, 1, 2, 3, 4, 6, 8, 10, 12, 14, and 16). The programme included three group sessions with a dietician and also standardised written information. All sessions (45 minutes) were in groups of 10 to 15 women, with group members all individually randomised to the same treatment. Additional follow up visits were made after 21, 26, 39, and 52 weeks. Participants were offered 2 mg nicotine gum. If their daily consumption exceeded 20 pieces, they could switch to 4 mg. Participants were given free nicotine gum for 3 months and thereafter recommended to taper consumption, but on request additional nicotine gum was supplied up to 12 months. All participants were recommended a standardised balanced diet of about 6.7 MJ/day. The group meetings were designed as conventional moderate behaviour modification sessions emphasising techniques for stopping smoking and providing support for weight control. The very low energy diet was given free of charge for three, two week periods (weeks 1 and 2, 7 and 8, and 13 and 14). Subjects were recommended not to eat anything else during these periods but to increase their intake of energy free drinks.
Characteristics of participants
Participants were female smokers aged 30 to 60 who wanted to stop smoking and maintain their weight. To be eligible women had to have a body mass index of 23-31, smoke at least 10 cigarettes a day, have smoked for at least three years, and have made at least one serious attempt to stop and restarted because of weight gain. Exclusion criteria were cardiovascular disease in the past 6 months, clinically important renal or hepatic disease, participation in any other clinical study in past 6 months, pregnancy or lactation, lactose intolerance, alcohol or other drug misuse, use of any form of smokeless tobacco or nicotine replacement therapy, gout, acute porphyria, diabetes mellitus type 1, vegetarian diet, or any serious metabolic or malignant disease likely to interfere with compliance. Exclusion criteria were based on safety recommendations for very low energy diets[16,20] and nicotine replacement treatment.
An advertisement resulted in 547 responses. A total of 438 women answered a postal questionnaire about admission criteria and brief demographics; 361 were eligible and given a consecutive number in the order their answers were received at the clinic. They were allocated to one of the treatment groups according to the corresponding number in the randomisation list. All women were invited to an information meeting, and they were then divided into groups of 10-15. Seventy four did not attend the clinic so 287 women were finally included in the study. As this dropout was expected, the protocol defined the intention to treat analysis to comprise women coming to the first visit and receiving treatment.
Data collection, power calculation, and statistics
Based on results from previous studies and the fact that weight conscious women would be a difficult group, we expected about 20% of women in the control group and 35% in the diet group successfully to stop smoking after 16 weeks. Given these assumptions, a sample size of 135 women in each arm of treatment was needed for a power of 80% and an Alpha of 0.05.
No formal adjustment for multiplicity was made, but P values are presented for each test. We used Pearson's [chi-square] test for unordered categorical or binary variables and the Mann-Whitney U test for ordered categorical and all continuous variables.
Successful smoking cessation was defined as self reported complete abstinence from week 2 to week 16, verified by a carbon monoxide concentration less than 10 ppm (New Smokerlyzer, Bedfont). Two missed visits between weeks 2 and 16 were allowed. If the week 2 visit was missing, the woman had to be abstinent from week 1 until endpoint. Women had to attend the week 16 visit to be eligible for the success criteria.
Compliance with the very low energy diet was based on self reported adherence and verified through ketone body analysis in urine, sampled by the women at home. Women with missing samples were asked to provide a urine sample after the very low energy diet period. Compliance with nicotine gum treatment was based on self reported daily number of gums and verified by saliva cotinine concentration.
Withdrawal symptoms, including desire or urge to smoke; irritability, frustration, or anger; restlessness; difficulty concentrating; anxiety; dysphoric or depressed mood; insomnia; and increased appetite, together with six fasting related symptoms (headache, constipation, fatigue, nausea, hair loss, and chilliness) and adverse events were recorded at each visit.
The study was performed according to the principles outlined in the Declaration of Helsinki and approved by the ethics committee at the Karolinska Hospital. All participants gave informed consent.
Table 1 shows the characteristics of the participants. Mean number of daily cigarettes smoked was 19.5 (95% confidence interval 18.8 to 20.2) for 28.8 (27.9 to 29.6) years. Subjects were moderately nicotine dependent. Almost three quarters (210) of the women had used nicotine replacement therapy before, and 199 (70%) had tried to stop smoking three times or more. In all, 206 (72%) had gained 3-10 kg and 56 (19%) more than 10 kg last time they tried to quit. Two thirds had tried to lose weight during a previous attempt to stop smoking.
Table 1 Baseline demographic characteristics including smoking status and smoking history. Values are mean (SD) unless stated otherwise
The smoking cessation rate in the diet group was 68/137 (50%) after 16 weeks compared with 53/150 (35%) in the control group (P = 0.01; table 2). The slight increase in success rate from week 10 to 12 in the control group and 14 to 16 in both groups is probably due to women being allowed to miss up to two visits without disqualification from the analysis.
Table 2 Number of women who had completely and continuously stopped smoking from week 2 onwards
(*) Pearson's [chi-square] test.
Figure 1 shows the mean weight change in women who had successfully stopped smoking. The diet group had lost 2.1 (2.9 to 1.3) kg at 16 weeks whereas the control group had gained 1.6 (0.9 to 2.3) kg (P [is less than] 0.001). No body mass index dropped below 20.
[Figure 1 ILLUSTRATION OMITTED]
Mood items in the withdrawal symptoms questionnaire (irritability, anxiety, poor concentration, restlessness and depression; score 0-15)[22 23] were analysed in continuous abstainers up to week 16. The diet group reached the highest score after 1 week whereas the control group peaked after 2 weeks, when the difference between the group medians was 1.5 units (0.6 to 2.4). The score of the diet group was significantly lower than that of the control group at weeks 2, 3, and 4 (fig 2).
[Figure 2 ILLUSTRATION OMITTED]
Rhinitis and headache were the most common adverse events, reported by 170 (59%) and 105 (37%) women during the study. No significant differences were found between the groups, although headache was reported by 58 (42%) women in the diet group and 47 (31%) in the control group (P = 0.053, 95% confidence interval for difference -0.1 to 22.1). Eighty six women withdrew from treatment during the 12 months. None of the withdrawals were because of adverse events.
Long term success
After 12 months, 35 and 51 women had dropped out from the diet and control groups respectively. Thirty eight (28%) of the diet group were still not smoking compared with 24 (16%) in the control group (P = 0.02). Among non-smokers, the mean weight increase was 2.5 (0.78 to 4.3) kg in the diet group and 3.8 (2.5 to 5.1) kg in the control group (P = 0.61). The overall weight increase irrespective of smoking status was 1.2 (0.32 to 2.1) kg in the diet group and 2.3 (1.5 to 3.0) kg in the control group (P = 0.13).
Women who successfully stopped smoking in the diet group used 7.8 gums/day and those in the control group used 8.3/day during the first 16 weeks. Fifteen subjects used 20 or more gums a day and switched to 4 mg gum. Ninety six of the women who came to the clinic after one year (n = 201) were still using the gum. Ninety four used 5.4 (SD 4.1) 2 mg gums/day and two women used 7.0 4 mg gums/day.
In all, 103/137 (75%) women reported total compliance with the very low energy diet during the first period. During weeks 6 to 8 and 12 to 14 complete compliance was reported by 42 (31%) and 25 (18%) respectively. These figures were confirmed by the ketone body analysis.
Our study shows that smoking can be stopped for up to one year with acceptable weight control in a group of women selected for their previous weight control problems when attempting to stop smoking. This study was open, and all subjects received nicotine replacement therapy. There is no practical way to blind very low energy diets, and there is ample evidence that nicotine replacement improves abstinence from smoking.
It is reasonable to assume that programmes to stop smoking have lower success in weight conscious women than in the general smoking population. Generally, the success rate after 12 months is 19%, compared with our results of 28% in the diet group and 16% in controls. Women were given the opportunity to drop out of the treatment programme between randomisation and the first visit, and this probably resulted in a more motivated study population.
Role of very low energy diets
While on the very low energy diet women experienced less craving for cigarettes and increased appetite scores. This may reflect diet induced acidosis, as ascertained by measurement of urinary ketone bodies. Although clinical experience suggests that weight loss is facilitated by acidosis, which is associated with fat catabolism, no clear proof exists.
Very low energy diets have not been previously reported in programmes to stop smoking. Weight gain when stopping smoking has, however, been prevented by giving an anorectic drug, but when drug treatment was discontinued, weight was regained almost up to the control group level.
Combining smoking cessation and weight control produced only modest side effects. Headache was more common in the diet group (although not significantly), and this can partly be explained by the fact that headache is a common side effect both of nicotine withdrawal and supplemented fasting. As expected, compliance with the very low energy diet fell over time. However, the immediate differences in stopping smoking between the two groups were already evident after 2 weeks. Thus the first diet period may be the critical phase. Focusing on weight, in an obesity unit, probably helped the women stop smoking by providing more weight control information than they had previously received.
Our study was designed with power to determine smoking cessation rates and weight change up to week 16. Although significantly improved smoking cessation rates were found after one year in the diet group, the study did not have power to demonstrate significant differences in weight gain between the groups. The weight gain in both groups was less than reported by most women in previous attempts at stopping smoking. Nicotine gum may therefore depress and postpone weight gain.
Contributors: TD designed the trial, interpreted the results, and prepared the manuscript. SR was principal investigator and responsible for clinical conduct, interpreted results, and wrote the manuscript. AW preformed the statistical analyses. Urbain Sawe initiated the trial and contributed with ideas and valuable professional input. Nurses and dieticians at the obesity unit at Karolinska and Huddinge Hospitals collected all study data.
Funding: Swedish Peoples' Health Institute, Pharmacia and Upjohn Consumer Healthcare, and Nycomed Pharma.
Competing interests: TD and AW are employed by Pharmacia and Upjohn Consumer Healthcare. SR has been reimbursed by Pharmacia and Upjohn and Nycomed Pharma, the manufacturer of Nutrilett, for attending conferences and received financial support from Nycomed Pharma for clinical research.
 Silagy C, Mant D, Fowler G, Lodge, M. Meta-analysis on efficacy of nicotine replacement therapies in smoking cessation, Lancet 1994;343: 139-42.
 US Public Health Service. The health benefits of smoking cessation. A report of the surgeon general. Washington, DC: Department of Health and Human Services, 1990:469-515. (DHHS publication No (CDC) 90-416).
 Williamson DF, Madans J, Anda RF, Kleinman JC, Giovino GA, Byers T. Smoking cessation and severity of weight gain in a national cohort. N Engl J Med 1991;324:739-45.
 Flegal KM, Troiano RP, Pamuk ER, Kuczmarski RJ, Campbell SM. The influence of smoking cessation on the prevalence of overweight in the United States. N Engl J Med 1995;33:1165-70.
 Klesges RC, Klesges LM, Meyers AW. Relationship of smoking status, energy balance, and body weight: analysis of the second national health and nutrition examination survey. J Consult Clin Psychol 1991;59:899-905.
 Tuomilehto J, Nissinen A, Puska P, Salonon J, Jalkanen L. Long-term effects of cessation of smoking on body weight, blood pressure and serum cholesterol in the middle-aged population with high blood pressure. Addict Behav 1986;11:1-9.
 Leischow SJ, Stitzer ML. Smoking cessation and weight gain. Br J Addict 1991;86:577-81.
 Norregaard J, Tonnesen P, Petersen L. Predictors and reasons for relapse in smoking cessation with nicotine and placebo patches. Prey Med 1993;22:261-71.
 Hofstetter A, Schutz Y, Jequier E, Wahren J. Increased 24-hour energy expenditure in cigarette smokers. N Engl J Med 1986;314:79-82.
 Grunberg NE. The effect of nicotine and cigarette smoking on food consumption and taste preferences. Addict Behav 1982;7:317-31.
 Spring B, Pingitore R, Kessler K. Strategies to minimise weight gain after smoking cessation: psychological and pharmacological intervention with specific reference to dexfenfluramine. Int J Obes Relat Metab Disord 1992;16(suppl 3):19-23.
 Hall SM, Tunstall CD, Vila K, Duffy J. Weight gain prevention and smoking cessation; cautionary findings. Am J Public Health 1992;82:799-803.
 Pirie P. Effects of nicotine gum on post-cessation weight gain. Health Psychol 1992;11(suppl):53-4.
 Ryttig KR, Rossner S. Weight maintenance after a very low calorie diet(VLCD) weight reduction period and the effects of VLCD supplementation. A prospective, randomised, comparative, controlled long-term trial. J Intern Med 1995;238:299-306.
 Ryttig KR, Flaten H, Rossner S. Long-term effects of a very low calorie diet (Nutrilett[R]) in obesity treatment. A prospective, randomised comparison between VLCD and a hypocaloric diet plus behavior modification and their combination. Int J Obes Relat Metab Disord 1997;21:674-9.
 National Task Force on the Prevention and Treatment of Obesity. National Institute of Health: very low-calorie diets. JAMA 1993;27 (1:967-74.
 Rossner S, Flaten H. VLCD and LCD in long term treatment of obesity. Int J Obes Relat Metab Disord 1997;21:22-6.
 The Agency for Health Care Policy and Research Smoking Cessation Clinical Practice Guideline. JAMA 1996;275:1270-80.
 Fiore MC, Bailey WC, Cohen SJ, Dorfman SF, Goldstein MG, Gritz ER, et al. Smoking cessation clinical practice guideline. Rockville, MD: Agency for Health Care Policy and Research, Public Health Service, US Department of Health and Human Services, 1996. (AHCPR publication No 96-0692.)
 Wadden TA, Van Itallie TB, Blackburn GL. Responsible and irresponsible use of very-low-calorie-diets in the treatment of obesity. JAMA 1990;263:83-5.
 Galeazzi RL, Daenens P, Gugger M. Steady-state concentration of cotinine as a measure of nicotine-intake by smokers. Eur J Clin Pharmacol 1985;28:301-4.
 American Psychiatric Association. Diagnostic and statistical manual of mental disorders, DSM-IV. 4th ed. Washington, DC: APA, 1994:244-7.
 Russell MAH, Stapleton JA, Feyerabend C, Wiseman SM, Gustavsson G, Sawe U, et al. Targeting heavy smokers in general practice: randomised controlled trial of transdermal nicotine patches. BMJ 1993;306:1308-12.
 Silagy C, Mant D, Fowler G, Lancaster T The effect of nicotine replacement therapy on smoking cessation. In: Cochrane Collaboration. Cochrane Library. Issue 2. Oxford: Update Software, 1997.
 Spring B, Wurtman J, Wurtman R, el-Khoury A, Goldberg H, McDermott J, et al. Efficacies of dexfenfluramine and fluoxetine in preventing weight gain after smoking cessation. Am J Clin Nutr 1995;62:1181-7.
 Atkinson RL. Medical evaluation and monitoring of patients treated by severe caloric restriction, treatment of the seriously obese patient. London: Guilford Press, 1989:273-89.
(Accepted 30 April 1999)
Obesity Unit, Karolinska Hospital, S-171 76 Stockholm, Sweden
Tobias Danielsson clinked research manager
Stephan Rossner professor of health behaviour research, Karolinska Institute
Pharmacia and Upjohn Consumer Healthcare, S-251 09 Helsingborg, Sweden
Ake Westin biostatistician
Correspondence to: T Danielsson, Obesity Unit, Huddinge University Hospital, S-141 86 Huddinge, Sweden tobias.danielsson@ medhs.ki.se
Commentary: results are unlikely to be as good in routine practice
General practitioners and primary care physicians of all types are used to being harangued about the issue of stopping smoking. Although stopping smoking is probably the largest beneficial health decision any smoker can make, it is important to determine how far each individual patient should be pushed on this issue. It is part of health promotion folklore that 5% of smokers will stop merely on the advice of their general practitioner; the use of nicotine replacement therapy increases the proportion of successful quitters.
Danielsson et al have provided a well conducted study which suggests that the use of very low energy diets together with nicotine gum boosts one year cessation rates from 16% to 28%. On the face of it, this is a dramatic and meaningful increase that deserves to be considered in routine practice. However, a number of factors must be reflected on before this initiative is taken up more widely. The first of these concerns the nature of the sample of patients in this study. The women entered were all in the normal or overweight category, wanted to stop, and had failed to stop before because of unacceptable weight gain. Careful patient selection would be necessary before entering subjects routinely into this programme.
Secondly, in the research setting about a fifth of those enrolled did not turn up for the sessions--in service use this proportion could easily be greater. Thirdly, the intensity of the behavioural programme must be remembered. Eleven, 45 minute group sessions (10-15 women each) were held over 16 weeks. This level of intervention is not only expensive but nearly impossible to provide in routine practice. Furthermore, the very low energy diet was provided free for the participants in this research. This would be unlikely outside the research setting.
Although the study was done well, it did not include any data on cost effectiveness. It would have been interesting to know a cost per successful quitter. Readers of this paper might be tempted to try a version of the method for their own patients, probably using the diet alone without the group sessions. It is unlikely that results would come near those achieved in the research setting, and thus the case for dietary restriction as an adjunct to nicotine replacement therapy when stopping smoking remains unproved for the moment.
Department of Primary Health Care, School of Health Sciences, Medical School, Newcastle upon Tyne NE2 4HH
Kevin Jones senior lecturer
COPYRIGHT 1999 British Medical Association
COPYRIGHT 2000 Gale Group