Quit methods compared: what the evidence shows for each approach

Anyone who decides to quit soon meets a crowded menu: patches, pills, vapes, apps, helplines, or simply stopping. The useful question is how many extra people each option helps to quit, and how sure researchers are about that. The largest recent comparison is a Cochrane network meta-analysis published in 2023, which pooled 319 randomised trials with 157,179 participants. In its news summary, Cochrane reported that around 14 in 100 people were likely to succeed long term with a nicotine e-cigarette, varenicline or cytisine, against around 6 in 100 who tried without any of the aids studied. This page walks through what each approach has shown, and where the evidence runs out. Related pages cover nicotine replacement therapy, prescription medicines and counselling and quitlines in more detail.

How to read the numbers

In these reviews, "quit" usually means not smoking at six months or longer, which is a stricter test than a few smoke-free weeks. Results are often reported as an odds ratio, which compares the odds of quitting with a treatment against the odds with a control group. A value above 1 favours the treatment. The range in brackets is a credible interval (a range that likely contains the true value), and the reviews also translate results into extra quitters per 100 people.

None of these numbers predict what will happen to one person. They describe averages across thousands of people in trials, and they say little about how much dependence, history or circumstances matter. Readers who want the broader picture of how often attempts succeed can start with quit attempts and success rates.

Medicines and e-cigarettes side by side

The 2023 Cochrane review compared each option with control groups that had no medicine or e-cigarette. The table gives its main results, all judged high-certainty evidence for the options listed.

The review concluded that nicotine e-cigarettes, varenicline and cytisine were the most effective, followed by combination nicotine replacement. It noted that the combination estimate was calculated from the effects of individual products rather than from direct trials of combinations, so it is less certain. Details on the products are on the nicotine replacement page, and on varenicline and bupropion on the prescription medicines page.

Cytisine and what is actually available

Cytisine looks strong in the numbers, but access varies. The Cochrane authors wrote that cytisine is not available in many countries, and a Cochrane news summary stated that it was not licensed or marketed in most countries outside central and Eastern Europe, including the UK and the US. Regulatory status can change, so a pharmacist or prescriber is the right source for what is sold where a reader lives.

E-cigarettes in outline

Cochrane found high-certainty evidence that nicotine e-cigarettes helped more people quit than control. E-cigarettes without nicotine did not clearly help (odds ratio 1.16, with an interval that spanned no difference and possible harm). The authors also said more evidence is needed on long-term harms. The US Preventive Services Task Force (2021) concluded that evidence was insufficient to weigh the benefits and harms of e-cigarettes for quitting in adults, and advised clinicians to direct people to other interventions with proven effectiveness and established safety. Those two positions describe different questions, and both are worth reading in full on the vaping and quitting smoking page.

Stopping without aids, or cutting down first

Some people prefer to stop all at once, often called going cold turkey. Others cut down first. A Cochrane review of 22 studies with 9,219 participants found no clear difference in long-term quit rates between reducing before a quit date and quitting abruptly (risk ratio 1.01, 95% interval 0.87 to 1.17). It rated the certainty moderate. That comparison is between two ways of quitting, so it does not show how either compares with using a medicine. The 6 in 100 baseline above is the most direct figure for attempts without the aids studied.

Counselling, text messages and apps

Cochrane's overview of behavioural support covered 33 reviews, 312 studies and 250,503 adults. It found that counselling, and money given for successfully quitting, helped more people quit for six months or longer, whether or not they also took medicines. Support by text message probably helped more than no support, and increasing the intensity of support modestly raised quit rates.

A separate Cochrane review of mobile phone programmes found moderate-certainty evidence that automated text messaging beat minimal support (risk ratio 1.54, 95% interval 1.19 to 2.00; 13 studies, 14,133 participants). For smartphone apps the evidence was of very low certainty and showed no benefit (risk ratio 1.00, 0.66 to 1.52), so the question of apps is open rather than closed. The counselling and quitlines page covers phone, text and group support.

Official guidance and combining approaches

The US Preventive Services Task Force (January 2021) gave a Grade A recommendation that clinicians offer behavioural interventions and FDA-approved medicine to adults who use tobacco who are not pregnant. For pregnant people its Grade A recommendation was for behavioural interventions. The Cochrane results fit that pairing: most of the strongest options in the table can be used together with counselling.

Planning matters as well as the choice of method, which is the subject of preparing to quit. Withdrawal symptoms in the first weeks are described on the withdrawal timeline page.

What the evidence cannot tell you

Trials average over many people, and the best option on paper may not suit a person with heart disease, a mental health condition, pregnancy or other medicines. The Cochrane authors flagged sparse data on serious side effects for several treatments. Anyone in these groups should check with a clinician or pharmacist before starting a medicine, and anyone who experiences severe symptoms while quitting should seek medical advice promptly.

The short version

Across the largest Cochrane comparison, the leading options (e-cigarettes, varenicline and cytisine) had odds ratios of 2.21 to 2.37 for quitting for six months or longer, and behavioural support helped on top of that. No option works for everyone. A clinician or pharmacist can help weigh pregnancy, heart disease, mental health conditions and other medicines before choosing.