UK National Health Service Data Show that NRT is No More Effective than Unassisted Quitting on a Population Level

A report released yesterday by the Health and Social Care Information Centre of the UK National Health Service assesses the smoking cessation rates of clients offered services through the NHS Stop Smoking Services during the period April 2011 through March 2012. The cessation rates refer to 4-week quit rates (defined as not having smoked for 2 weeks at the 4 week mark post-quit date). Data are based on 400,955 successful quitters at the 4-week mark among 816,444 people who set a quit date with NHS Stop Smoking Services.

Because some of the clients were offered and received pharmacotherapy and others received no medication, we are able to compare 4-week cessation rates of the clients with and without pharmacotherapy. Here are the results, showing 4-week quit rates by type of treatment:

Drugs: 49.6%
No drugs: 50.5%

Based on this finding, then, there was no improvement in the effectiveness of smoking cessation treatment with the use of medication.

The results by type of medication varied slightly but the overall patterns showed little difference between various types of drugs and unassisted quitting:

NRT: 45.6%
Zyban: 52.9%
Chantix: 60.2%
NRT and Zyban: 44.8%
NRT and Chantix: 46.6%

This finding that smoking cessation drugs do not improve smoking cessation rates on a population level in the setting of a national smoking cessation service provider is further supported by data on trends in 4-week cessation. Although the proportion of clients who were prescribed drugs increased from 84% in 2001/02 to 93% in 2011/12, the rate of successful quitting actually decreased slightly from 53% to 49%.

In 2006/07, when the use of NRT peaked at 83%, the proportion of 4-week quitters was 53%, no different from the start of the study period when only 63% of clients received NRT.

Prior to the availability of Chantix, the 4-week cessation rates ranged from 53% to 57%. Since the availability of Chantix, when use rates were as high as 26%, the 4-week cessation rates have ranged from 49% to 52%.

The Rest of the Story

These data indicate no benefit from offering smokers drugs in an effort to help them quit smoking in the setting of provision of national smoking cessation services on a population basis.

The data conflict with those from clinical trials, where smokers are generally offered a smoking cessation drug or a placebo. There is evidence that these trials are not successfully blinded. Smokers who receive placebo can often tell that they are not receiving a study medication. This could well explain the lower observed rates of smoking cessation with placebo and could indicate that the observed "effectiveness" of these drugs in clinical trials is a result of what one might call the opposite of a placebo effect.

While these results should not be interpreted as suggesting that NRT or other drugs should not be prescribed by physicians as treatment for specific patients who are trying to quit, they do suggest that on a population basis, in the setting of a national smoking cessation strategy, pharmacotherapy simply does not appear to be effective in increasing smoking cessation rates.

Keep in mind as well that these are only 4-week success rates. The long-term effectiveness of unassisted quitting might be expected to exceed that of drug therapy based on these short-term findings, as the benefits of medication often diminish over time and NRT in particular is only intended to be taken for a limited period of time.

This report demonstrates why it is so problematic that anti-smoking organizations such as the North American Quitline Consortium have severe financial conflicts of interest with pharmaceutical companies. Quitline service decisions should be based on an objective view of the scientific evidence, but this is not possible when the group making these decisions is financially conflicted.