Oral health behaviors and bacterial transmission from mother to child: an explorative study

The study targeted mothers with children under three years attending public child
health clinics in Finland. A self-administered anonymous questionnaire assessed mothers’
health behavior and background information. The Ethical Committee of Human Sciences
at the University of Oulu approved the study.

The study population comprised of mothers with children under three years attending
two municipal child health clinics in Southern Finland 14], 15]. The public health clinics with free-of-charge services were from two middle-sized
towns (50,000 inhabitants) with similar socio-economic and ethnic background; main
source of livelihood in the towns is service trade and industry. A four to six month
period was estimated to achieve a representative sample of mothers (ca. 330) from
the health clinics. During the mothers’ routine visits to the clinics, health nurses
distributed the questionnaires to all and collected them immediately after the mothers
completed them. Nearly all invited mothers (95 %) participated in the voluntary surveys
and the final sample included 313 mothers 14], 15].

The question inquiring about the frequency of tooth brushing offered four answer options,
later dichotomized to twice daily or less frequently. Based on the question “Do you
smoke?”, the mothers were dichotomized as either smokers (daily, occasionally) or
non-smokers (never).

Questions related to the potential transmission of oral bacteria from mother to child
were as follows: “Do you share a spoon when feeding your child?”, “Do you kiss your
child on the lips?”, “Do you clean the pacifier with your own mouth before returning
it to your child?”; respondents answered each question with either “yes” or “no”.
In addition, the mothers were asked to provide their opinion to the statement “Bacterial
transmission from the mother’s mouth to the child’s mouth is impossible”; their answers
ranged on a five-point Likert scale from total agreement to total disagreement, later
dichotomized to those stating that such transmission is either possible or impossible.

The mothers’ background information included age in years (25, 25-29, 30-34, 35-39,
40+), later categorized into four by combining the two oldest groups into one (35+).
The mothers’ level of education was recorded as basic (compulsory?=?9 years) education,
vocational or professional education, or higher (polytechnic, university).

Statistical methods

We applied multiple correspondence analysis (MCA) 16], 17] in order to explore and illustrate the relationships between the mothers’ health
behaviors (tooth brushing and smoking), two of their health practices with their young
children (sharing a spoon with their child and kissing the child on the lips), and
the background variables (mother’s age and education level). MCA, an exploratory method
often used to generate hypotheses, generalizes the simple correspondence analysis
of frequency tables. MCA reveals the multidimensional structure inherent in the data
based on pairwise frequency tables of the variables. The principal result of MCA is
a graphical display called a biplot, most often a two-dimensional map of the categories
and their relationships 16], 17]. In addition to the graphical displays, we produced a numerical summary of the MCA
results describing the characteristics of the categories 18].

To further analyze the MCA findings, we used separate, unconditional, binary, multivariable
logistic regression models: one of the health practices served as a dependent variable,
while the other served as one of the explanatory variables along with the background
variables. The results of the logistic regression analyses were presented as odds-ratios
(OR) and their 95 % confidence intervals (95 %CI).