Email updates

Keep up to date with the latest news and content from BMC Public Health and BioMed Central.

Open Access Research article

Causal assessment of smoking and tooth loss: A systematic review of observational studies

Takashi Hanioka1*, Miki Ojima2, Keiko Tanaka3, Keitaro Matsuo4, Fumihito Sato4 and Hideo Tanaka4

Author Affiliations

1 Department of Preventive and Public Health Dentistry, Fukuoka Dental College, Fukuoka, Japan

2 Department of Preventive Dentistry, Graduate School of Dentistry, Osaka University, Osaka, Japan

3 Department of Public Health, Faculty of Medicine, Fukuoka University, Fukuoka, Japan

4 Division of Epidemiology and Prevention, Aichi Cancer Center Research Institute, Nagoya, Japan

For all author emails, please log on.

BMC Public Health 2011, 11:221  doi:10.1186/1471-2458-11-221

Published: 8 April 2011



Tooth loss impairs oral function. The aim of the present review was to evaluate the causal association between smoking and tooth loss on the basis of high-quality studies.


Relevant literature was searched and screened, and the methodological quality was assessed. Information on the strength of the association between smoking and tooth loss, the dose-response relationship and natural experimental data was collected and evaluated with respect to consistency and study design.


Our literature search yielded 496 citations, and 6 cross-sectional and 2 cohort high-quality studies examining 58,755 subjects in four countries. All studies reported significant associations, although the strength of the association was usually moderate. Four studies reported dose-response relationships between exposure to smoking and the risk of developing tooth loss. A decrease in the risk of tooth loss for former smokers was evident in six studies. Interpretation of evidence for each element was consistent, despite some shortcomings regarding study type and population.


Based on the consistent evidence found with the existing biological plausibility, a causal association between smoking and tooth loss is highly likely. Further studies using a cohort design and different populations are necessary to confirm this association.