PT - JOURNAL ARTICLE AU - I. Roberts TI - Smoke alarm use: prevalence and household predictors. AID - 10.1136/ip.2.4.263 DP - 1996 Dec 01 TA - Injury Prevention PG - 263--265 VI - 2 IP - 4 4099 - http://injuryprevention.bmj.com/content/2/4/263.short 4100 - http://injuryprevention.bmj.com/content/2/4/263.full SO - Inj Prev1996 Dec 01; 2 AB - OBJECTIVE: To determine the prevalence of smoke alarm use among families with children and to identify household factors that predict the absence of a smoke alarm. DESIGN: Cross sectional analysis of data collected in the September and November 1995 Omnibus Survey, conducted by the Office of Population Censuses and Surveys in the UK. SUBJECTS: A random sample of British households. Interviews were completed with 4,043 householders. The response rate was 78%. RESULTS: 29% of British households do not have a smoke alarm and smoke alarms were absent in 20% of households with children under 15 years. A smoke alarm was absent in 41% of privately rented homes compared with 17% of owner occupied homes. Living in private rental accommodation was the strongest household predictor of the absence of a smoke alarm (odds ratio = 3.25, 95% confidence interval 1.94 to 5.42). Householders who had heard of National Fire Safety Week or the TV smoke alarm advertising campaign were significantly more likely to have a smoke alarm. The apparent effect of these campaigns was greatest in families with children. CONCLUSIONS: Smoke alarm use has continued to increase but a substantial proportion of British homes still do not have smoke alarms. Homes at greatest risk of residential fire are the least likely to have an alarm. Health professionals may be able to increase smoke alarm use among families with children, by counselling families about the benefits of smoke alarms. They may also be effective in this regard by lobbying local councils, houseing associations, or private landlords to install alarms in all properties and by advocating for national legislation.