Why may teenage girls persist in smoking?

Teenage girls often smoke cigarettes, recognizing that it protects them from the impulse to binge eat with its feared weight-gain consequences. Evidence is marshalled from our studies of a female eating-disordered population, teenage females (London, U.K. and Ottawa, Canada) and middle-aged women (London and rural England) in the general population. Teenage female data analysis reveals links between smoking and body-weight/shape concerns. Those who smoked were likely to be moderately overweight. Smoking was also related at all ages to being postmenarchal. Sensitivity to shape is largely and qualitatively prompted by the development of body fat in puberty. Smoking by the London schoolgirls in particular also independently revealed an association with greater weight loss since puberty. Smoking was powerfully linked with vomiting undertaken as another defence against weight gain and may also be further reinforced as a behaviour by it. The eating-disordered population showed these latter associations most strikingly. Since smoking amongst older women is associated with below average body-weight it may indeed be effective in curbing weight gain and therefore promoting desired weight loss. Our studies provide little evidence of association between smoking and generalized or social anxiety. We propose that preventative psychological approaches to teenage female smoking should include attention to these matters.

Smoking and pursuit of thinness in schoolgirls in London and Ottawa.

It has been proposed that teenage girls often smoke cigarettes to protect themselves from the impulse to binge eat, with its feared weight-gain consequences, particularly when other measures such as greater dietary restraint have failed. The present study looked at the relationship between body mass index and standardised questionnaire responses concerning smoking, alcohol consumption, moods, weight changes, attitudes to body weight and shape, dietary patterns and menstruation in 1936 British (London) and 832 Canadian (Ottawa) schoolgirls. Data analysis revealed links between cigarette smoking and body weight/shape concerns, and awareness by subjects of these links; there was also a tendency for smokers in these two populations to be overweight but not grossly obese. Smoking was also related at all ages to being postmenarchal. The London population in particular revealed an association between smoking and a weight loss of 7 kg or more at some stage since puberty. Smoking was also linked, in a minority, with regular vomiting undertaken as a further defence against weight gain when overeating had occurred. These associations existed alongside the major and predictable association found between alcohol consumption and smoking. Similarities between the British and Canadian schoolgirls were striking in respect of rank order of reasons given for smoking and consequences of giving it up. Since smoking amongst older women is reportedly associated with below-average body weight it may indeed be effective in helping to curb weight gain. Our study provided little evidence of association between smoking and generalised anxiety or social anxiety (in either population), or depression (in the British cohort). We suggest that any preventive psychological approach to teenage female smoking should include attention to weight gain anxiety and consequent pursuit of thinness.

Eating disorders--the role of the nurse.

The role of the nurse in the treatment of eating disorders requires a wide range of skills and confidence in managing high levels of anxiety. For many patients, the nurse is the key professional involved in their care and is in a position to offer significant treatment and support to individuals who suffer these distressing problems.

Outcome of outpatient psychotherapy in a random allocation treatment study of anorexia nervosa.

Ninety subjects with DSM-III-R anorexia nervosa were randomly allocated to four treatment options, one inpatient, two outpatient, and one comprising an assessment interview only. Twenty were thus offered a package of outpatient individual and family psychotherapy. At 2-year follow-up, 12 of the 20 were classed as well, or very nearly well, according to operationally defined criteria. Statistically significant improvements over time were obtained for weight, mean body mass index (BMI), and also for psychological, sexual, and socioeconomic adjustments. Weight and BMI changes were significantly better than for the assessment only group, some of whom had received extensive treatment elsewhere. The style of the outpatient therapy and compliance with it are described in some detail and prognostic indicators for the treated and untreated groups presented. Lower weights at presentation and vomiting were associated with poorer outcome, although age and length of history were not.

Sexual abuse as a factor in anorexia nervosa: evidence from two separate case series.

Evidence of a link between reported sexual abuse and specific eating behaviours was considered in 100 anorexic women. The women were assessed as part of case series from two separate clinics. There was a strong association between reported unwanted sexual experiences and purging behaviours (vomiting and/or abusing laxatives). Anorexics who did not purge had lower reported rates of abuse. The pattern and prevalence rate of reported abuse were similar in the two clinics, suggesting that the association was not a product of a specific method of eliciting disclosure. Implications for treatment and possible causal mechanisms behind this relationship are discussed.

Relationship between smoking, weight and attitudes to weight in adolescent schoolgirls.

A total of 1,932 schoolgirls aged 11-18 from seven schools in the South London area were surveyed using questionnaires which addressed eating patterns, body weight history, attitudes to body weight and shape, menstrual history and current smoking behaviour. They were also weighed and their height was measured. Twelve per cent of the girls were regular smokers and 10% smoked seven or more cigarettes over a 4 day period. Amongst girls aged 14 and over, 15% smoked regularly and a further 9% occasionally. A significant relationship was found between smoking and weight. Smokers were more likely to be moderately overweight in relation to their peers and to have been worried about their weight at some stage. There were differences between girls in state schools and those in independent schools with regard to smoking behaviour and weight. The findings have implications for anti-smoking strategies and health education generally.

Long-term mortality in anorexia nervosa. A 20-year follow-up of the St George's and Aberdeen cohorts.

Two cohorts of anorexia nervosa patients were followed up for a mean of 20 years. All except 4% of each cohort was traced. The crude mortalities were: St George's, 4%; Aberdeen, 13%. The SMRs were: St George's, 136; Aberdeen, 471. If the untraced were assumed to be dead, crude mortalities were 7.6% and 15.9% respectively, and SMRs were 276 and 592 respectively. Causes of death were complications of the illness and suicide. Medical treatment may reduce early mortality, while comprehensive medical and psychotherapeutic treatment may reduce late mortality.

A controlled study of the effect of therapies aimed at adolescent and family psychopathology in anorexia nervosa.

Ninety patients with severe anorexia nervosa fulfilling DSM-III-R criteria were assessed in depth in terms of their family developmental psychopathology and then randomly allocated to either one of three treatment groups or to no treatment. In three treatment regimes, a behavioural approach to diet and weight gain was coupled with individual and family psychotherapy directed at the adolescent maturational problems. All three treatment regimes were highly significantly effective at one year in terms of weight gain, return of menstruation, and aspects of social and sexual adjustment. Body weights above those at pubertal onset were achieved for the group mean maximum and one-year follow-up weights for all three treatment groups but not the control group.