Mentioning smoking on death certificates: health benefits of quitting in the Hong Kong mortality case-control study

Abstract

PURPOSE: Evidence on the benefits of stopping smoking in developing countries to help curb the growing tobacco epidemic is scarce and urgently needed. Mortality case-control study has yielded timely evidence on the hazards of smoking. We aimed to investigate its application on quitting. METHODS: The Lifestyle and Mortality study was a case-control study that included 81% of all deaths age 30+ years in 1998 in the death registries of Hong Kong by interviewing relatives. Cases were deaths from smoking-related causes (N=19,526) and controls were deaths from non-smoking related causes (N=6,076) following Sitas et al 2013. Unconditional logistic regression yielded adjusted odds ratios (AORs) of all smoking-related causes and cancers by duration of quitting and age at quitting (both compared with continued smoking), adjusting for sex and age at death. RESULTS: For duration of quitting, the AORs (95% CI) for all smoking-related causes, were 0.73 (0.58-0.93) in quitters who had stopped smoking for 5-9 years, 0.71 (0.60-0.84) for 10+ years, and 0.49 (0.44-0.56) in never smokers. The corresponding figures for cancers were 0.68 (0.52-0.90), 0.65 (0.54-0.80) and 0.36 (0.32-0.42). For age at quitting, the AORs for all smoking-related causes were 0.80 (0.66-0.95), 0.77 (0.52-1.15) and 0.49 (0.43-0.56) for quitting at the age of 45-64 years, 25-44 years and never smoking. The corresponding figures for cancers were 0.73 (0.59-0.89), 0.67 (0.43-1.03) and 0.36 (0.31-0.42). CONCLUSIONS: Graded benefits of quitting were observed using the mortality case-control study, which could be a quicker and cheaper alternative to cohort studies in examining the hazards of smoking and benefits of quitting. Smoking history should be recorded during death registration for long-term sustainable monitoring. This change is unlikely to be adopted by governments, unless this is recommended by WHO. Strong advocacy is needed.