Insomnia is a very common presenting complaint. UK data shows that it affects about one third of the population in any given year.1 Hypnotics are one strategy for the management of insomnia. Medicines for the management of insomnia that are available in New Zealand include zopiclone, triazolam, temazepam and nitrazepam. These are funded by district health boards (DHBs) in accordance with the Pharmaceutical Schedule.2
Aim—The aim of this study is to describe the community dispensing trends in all 21 New Zealand DHBs of these hypnotics for the ten financial years of 2000/01 to 2009/10.
Method—Data for all subsidised community dispensing of zopiclone, triazolam, temazepam and nitrazepam between the financial years of 2000/01 and 2009/10 were extracted from the data warehouse managed by New Zealand Health Information Service. Population data from Statistics New Zealand were combined with this database to enable population analysis.3Defined daily doses (from the WHO database) were used to determine the number of daily doses dispensed each financial year.4 These data were analysed on an Excel™ spreadsheet.
Results—Zopiclone dispensing increased from 216,069 defined daily doses (DDD) in 2000/01 to 719,957 DDD in 2009/10. Triazolam dispensing increased from 86,428 DDD in 2000/01 to 100,114 DDD in 2009/10. Temazepam dispensing decreased from 81,593 DDD in 2000/01 to 64,538 DDD in 2009/10. Nitrazepam dispensing decreased from 31,904 DDD in 2000/01 to 23,818 DDD in 2009/10
When corrected for population, DDD/person were: zopiclone - 0.056 in 2000/01 and 0.167 in 2009/10; triazolam - 0.022 in 2000/01 and 0.023 in 2009/10; temazepam - 0.021 in 2000/01 and 0.015 in 2009/10; and nitrazepam - 0.008 in 2000/01 and 0.006 in 2009/10.
Discussion—There was evidence of considerable regional variation in the 2009/10 data. While most DHBs share of the dispensing of hypnotics reflected their share of the population, there were some outliers. Zopiclone dispensing was 2.0 times the national mean DDD/person in Wairarapa DHB; 1.5 times the mean in MidCentral DHB and 1.4 times the mean in Hawke's Bay DHB. Triazolam dispensing was 1.7 times the mean DDD/person in Canterbury DHB; 1.4 times the mean in South Canterbury DHB and Hawkes Bay DHB. Temazepam dispensing was 2.6 times the mean DDD/person in South Canterbury; 2.1 times the mean in Nelson-Marlborough DHB and 1.6 times the mean in Canterbury DHB. Nitrazepam dispensing was 5.4 times the mean DDD/person in Whanganui and 4.0 times the mean in Otago DHB.
The overall decline in dispensing of DDDs over the ten year period of temazepam and nitrazepam is most likely due to concerns about benzodiazepine addiction. In contrast, the use of zopiclone has increased by over 300%. This may reflect patients being switched to zopiclone from other benzodiazepines in combination with an overall increase in number of patients being treated with it.
There is evidence to suggest that zopiclone is a relatively safe hypnotic and it does not have the same dependence potential of benzodiazepines.5,6 Despite this, zopiclone does act on benzodiazepine receptors and is associated with a similar adverse effect profile as benzodiazepines (i.e. dependence and withdrawal effects on discontinuation).6 Prescribers may have increased their use of zopiclone over time, as it is a safer alternative to the benzodiazepines, in the context of insomnia being a common presenting complaint in general practice.
Triazolam was withdrawn in the UK in 1993, following evidence of rebound phenomena on discontinuation and withdrawal symptoms of: amnesia, delirium, psychosis and behavioural disturbance. These were thought to be due in part to its very short half life.7 We would have expected to see a downward dispensing trend in New Zealand given these concerns rather than staying constant as seen in DDD/person of 0.022 in 2000/01 and 0.023 in 2009/10.
Benzodiazepine hypnotics and zopiclone are associated with a range of adverse effects. A meta-analysis looking at the usage of hypnotics in older people found that although hypnotics did improve sleep with a number needed to treat (NNT) of thirteen (i.e. for every thirteen patients treated, one patient would experience improved sleep), they were associated with a range of adverse effects including ataxia, falls, daytime fatigue and memory impairment. The number needed to harm (NNH) for a hypnotic when compared with a placebo agent was six (i.e. for every six patients treated, one would experience an adverse reaction).8
The risk of a patient on hypnotics (i.e. zopiclone and benzodiazepines) having a road traffic accident was increased in a Norwegian study and a British study. 9,10 These observations suggest that the benefits of hypnotics particularly in the elderly barely outweigh the risks.
This study has some limitations. Data are of dispensings and compliance rates are unknown. Indication and duration of therapy were unavailable as were the numbers of patients. Finally, there is no outcome data easily available to clarify whether increased usage of zopiclone and triazolam and decreased usage of temazepam and nitrazepam resulted in better clinical outcomes.
Conclusion—The increased use of zopiclone is a worrying trend in light of its potential to cause harm. This may be partially balanced out by the decrease in the use of temazepam and nitrazepam and the resultant decrease in harm potential. Similarly the increase in triazolam use is of concern in light of its withdrawal elsewhere. Hypnotic agents should be prescribed for the shortest period of time and at the lowest dose possible. Patients on long term hypnotics should be reviewed regularly with a view to reducing and stopping the hypnotic where possible. The reasons behind why some DHBs prescribe particular hypnotics at higher rates than the rest of the country is worth examining in terms of clinical outcomes.
Pharmacy Department, Hillmorton Hospital
Drug Utilisation Pharmacist
Clinical Pharmacology Department, Christchurch Hospital