Lay Summary: Evaluation of the safety of bupropion (Zyban)

Background

Bupropion (Zyban) was the first new treatment to be launched to help people stop smoking since nicotine substitutes (e.g. patches, gum etc.) were introduced. In 2000, the DSRU conducted a study to measure the frequency of known side effects and to identify any previously unknown side effects.


Methods

The research method used is known as Prescription Event Monitoring (PEM). Patients prescribed bupropion were identified from data on dispensed prescriptions supplied, in confidence, by the Prescription Pricing Authority (PPA) in August 2000. This is a processing centre to which pharmacists send dispensed prescriptions for payment purposes.

After approximately 6 months, GPs were asked to provide information on any medical events that patients experienced since bupropion was first prescribed. A medical event was defined as anything that was important enough to have been recorded in the medical notes regardless of whether or not it was thought to be related to treatment. The information received was then analysed and used to measure the frequency of known side effects and identify previously unknown effects of bupropion.

All prescription information received from the PPA and anonymised medical information received from GPs is treated in strict confidence by the DSRU, which is allowed under the Data Protection Act of 1998 to hold such information for research purposes.


Results

Information was received relating to 11,735 patients prescribed bupropion. The average age of the patients was 47 years. Approximately half were male patients ranging from 16 to 88 years old and half were female patients ranging from 16 to 87 years old.

There were 566 events reported by GPs to be adverse drug reactions to bupropion in 350 patients (some patients had more than one reaction). The most commonly reported side effects and reasons for stopping treatment were nausea/vomiting, insomnia, dizziness, headache/migraine and a general feeling of being unwell.

Bupropion was reported to have been taken by 12 women in the first three months of pregnancy. Five of these pregnancies resulted in live births with no known birth defects, two pregnancies resulted in abortions for medical reasons and one baby died in the womb. Details were not available for the other four pregnancies.

Fourteen patients died within the first 12 weeks of starting bupropion. This number was no higher than the number of people who would be expected to die, from any cause, in comparison with smokers who were not prescribed bupropion.


Conclusion

Most adverse events reported during bupropion treatment were expected based on previous studies. A small number of unexpected events reported as adverse drug reactions to bupropion included a general feeling of being unwell, lack of energy, diarrhoea, unsteadiness, paranoid feelings and heart attack. Most of the events described were also potential features of withdrawal from nicotine smoking, making it difficult to determine whether the effects were caused by bupropion. There was no indication of an increased risk of death associated with use of bupropion. Our results add to the emerging safety picture on the use of bupropion to help stop smoking and, overall, they are reassuring.