Association Between Smoking Intensity and Duration and Tooth Loss Among Finnish Middle-aged Adults

The Northern Finland Birth Cohort 1966 Project

Toni Similä; Jorma I. Virtanen


BMC Public Health. 2015;15(1141) 

In This Article


Table 1 shows the basic characteristics of the study population and the mean number of missing teeth as well as the proportion of those who were missing at least one tooth according to the categories of study variables and smoking status. The mean number of missing teeth in the entire study population varied with education level (ranging from 1.1 at the higher level to 2.2 at the basic level), tooth brushing frequency (1.7 in those who brushed their teeth once daily or less and 1.2 in those who brushed their teeth at least twice daily), and diabetes status (1.9 in those who had diabetes and 1.4 in those who did not have diabetes). The proportions of those with a basic or secondary education, a habit of alcohol use and poor tooth brushing frequency were higher among current and former smokers than among never smokers. Table 2 shows the distribution of pack-years and years of smoking as well as the association with the number of missing teeth in a similar manner to Table 1.

Table 3 shows the unadjusted and adjusted relative risks (RR) for the average increase in the number of missing teeth for the entire study population. Smoking 21 or more pack-years associated with tooth loss (adjusted RR = 1.78, CI = 1.36–2.33). Similarly, 31 or more years of smoking associated with the outcome (adjusted RR = 1.72, CI = 1.20–2.45). From other study variables, education in particular associated with the outcome: adjusted RRs for basic and secondary education were 1.81 (CI = 1.34–2.46) and 1.48 (CI = 1.27–1.72), respectively. In addition, poor tooth brushing frequency associated slightly with the outcome (adjusted RR = 1.19, CI = 1.02–1.39).

Table 4 shows the relation of smoking to tooth loss stratified according to gender, education and tooth brushing frequency. The RRs revealed an exposure-dependent association between pack-years and years of smoking and tooth loss among all the strata, except for women. Adjusted RRs, however, revealed similar, though statistically non-significant, associations among women. In each stratum, former or occasional smokers showed no essential differences from never smokers.

The model-based estimate of the number of missing teeth among smokers according to pack-years appears in Fig. 1. This visualization shows the possibility of an exponentially increasing probability to lose teeth with increasing pack-year values.

Figure 1.

Expected (line) and observed (dots) number of missing teeth by pack-years among smokers. The estimation is based on an unadjusted negative binomial model