An exploratory non-randomized study of a 3-month electronic nicotine delivery system (ENDS) intervention with people...

Smoking is endemic amongst people accessing homeless services, and they are disproportionately affected by smoking-related diseases. This paper reports on the results of a 3-month small scale intervention which explored the efficacy, challenges and opportunities of using electronic nicotine delivery systems (ENDS) to support cessation of tobacco smoking with people accessing an Irish supported temporary accommodation (STA) homeless service. It considers the results of this intervention with reference to the balance of harms between the use of vaping to support smoking cessation and continued smoking.

Twenty-three participants were recruited. Demographic data, carbon monoxide (CO) measurements, homelessness status and smoking history were recorded. Participants were given an ENDS device and two 10-ml bottles containing e-liquid available in several flavours and at several strengths. Participants could pick up new bottles on a weekly basis. At weeks 1, 4, 8 and 12, the Fagerström Test and Mood and Physical Symptoms Scale (MPSS) were administered.

Over 75% of the residents in the participating hostel were recruited (23/30). However, there was a substantial loss to follow-up (n = 14) as a result of data protection issues, the transient nature of the population of interest and non-compliance with the intervention. Self-reported reductions in cigarette consumption were found to be statistically significant (p < 0.001). However, reductions in carbon monoxide measurements were not statistically significant. Decreases in Fagerström Nicotine Dependence Test were statistically significant (p = 0.001), but decreases in MPSS "urge to smoke" and "strength of urges" composite scores were not. Reported side effects included coughing, runny nose, bleeding nose, slight sweating, dizziness, increased phlegm and a burning sensation at the back of the throat. Barriers to engagement were peer norms, vaping restrictions in accommodation and adverse life events. Positive effects reported included increased energy, less coughing, better breathing and financial benefits. An improvement in the domain "poor concentration" was also found to be statistically significant (p = 0.040).

ENDS-based interventions may be effective with this population. Future research should aim to improve follow-up, consider including behavioural components and monitor health effects in relation to ongoing concerns around risks and the balance of harms.

Registered retrospectively ISRCTN14767579.