Searches identified 14 studies investigating effects of reducing cigarette consumption on lung cancer,
CVD, COPD or FEV1 decline. Three were case-control studies, six cohort studies, and five follow-up studies
of FEV1. Six studies consistently reported lower lung cancer risk in reducers. Compared to non-reducers,
meta-analysis (random-effects) showed significantly lower risk (RR 0.81, 95% CI 0.74–0.88 for any reduction,
and RR 0.78, 0.66–0.92 for the greatest reduction), with no between-study heterogeneity. Four
cohort studies presented CVD results, the combined RR for any reduction being a non-significant 0.93
(0.84–1.03). An effect of reduction was not consistently seen for COPD or FEV1 decline. Four cohort studies
presented all-cause mortality results, the combined RR of 0.92 (0.85–1.01) being non-significant. The
RR of 0.95 (0.88–1.02) for total smoking-related cancer, from three studies, was also non-significant. The
evidence has various weaknesses; few studies, few cases in reducers in some studies, limited dose–
response data, incomplete adjustment for baseline consumption, questionable accuracy of the lifetime
smoking history data in case-control studies, and bias in cohort studies if reducers are likelier than
non-reducers to quit during follow-up. Also, the variable definitions of reduction make meta-analysis
problematic. Though the results suggest some benefits of smoking reduction, more evidence is needed.