A quasi-experimental pretest-posttest design was concerned to examine 116 mothers with toddlers referred to four comprehensive health centers affiliated with Hamadan University of Medical Sciences, Hamadan, Iran, in 2018. Mothers were non-randomly assigned to the experimental and control groups (58 persons in each group). A theory –based education of planned behavior (four 45-60 minute training sessions) was run for the experimental group; however, the control group (socio-demographically homogenous to the experimental group) received usual educational program. Our main outcome included preventive home accident practices by mothers with toddlers:
Population of the Study:
In this study, 116 mothers with toddlers (90% power with a 2-sided significance level of α = 0.05) participated. Inclusion criteria were living in Hamadan, having at least one young child, referring to comprehensive health centers regularly, having elementary reading and writing literacy, being at an acceptable mental, physical, and cognitive health status for mothers and their children, not working at health centers for mothers or close relatives, and not being pregnant. Exclusion criteria were being absent more than one training session and changing the place of residence.
The cluster sampling method was used to select mothers with regard to the inclusion criteria of the study. Then mothers were assigned into two groups with 58 persons per group. The experimental group was trained in accordance with the theory–based education of planned behavior, and the control group received usual education. The two groups were homogenous in terms of demographic data.
In order to collect data, a researcher-made questionnaire was used, which consisted of two sections addressing demographic information of mothers and children as well as questions about knowledge and constructs of planned behavior theory (i.e., attitude, perceived behavioral control, behavioral intention, and subjective norms) and preventive practices adopted by the mothers for toddlers’ home accidents.
Knowledge was assessed using 23 items scored based on a two-point scale ranging from 0 (incorrect answer) to 1 (correct answer). Intra-class correlation reliability coefficient (ICC) of this scale was estimated to be 0.9.
Attitude was measured using 16 items scored based on a four-point scale ranging from 1 (strongly disagree) to 4 (strongly agree). Intra-class correlation reliability coefficient (ICC) of the scale was estimated to be 0.85.
Perceived behavioral control was assessed using 10 items scored based on a four-point scale ranging from 1 (little control) to 4 (complete control). Intra-class correlation reliability coefficient (ICC) of this scale was 0.83.
Behavioral intention was assessed using nine items scored based on a four-point scale ranging from 1 (strongly disagree) to 4 (strongly agree). Intra-class correlation reliability coefficient (ICC) of this scale was calculated to be 0.92.
Subjective norm was measured using seven items scored based on a four-point scale ranging from 1 (strongly disagree) to 4 (strongly disagree). Intra-class correlation reliability coefficient (ICC) of this scale was 0.82.
Mothers’ preventive practices for toddlers’ home accidents was assessed using 35 items scored based on a two-point scales ranging from 0 (no preventive practice) to 1 (preventive practice). Intra-class correlation reliability coefficient (ICC) of this scale was 0.91.
To determine the validity of the questionnaire, qualitative content validity method was used. First, the content of the questions was determined based on the relevant books and articles (15, 16); then the questionnaire was submitted to 10 faculty members and 12 mothers. Finally, their comments were applied in the questionnaire.
After conducting a preliminary study and approving the study plan, the researcher referred to the comprehensive health centers in Hamadan and identified mothers with toddlers, who met the inclusion criteria. All the mothers were informed of the methods and the goals of the study, and their written consent forms were received. To limit the effect of the intervening variables, the two groups were supposed to be homogeneous in terms of demographic variables for toddlers and mothers.
At the beginning of the study, the self-reported questionnaires were submitted to the both groups of mothers and were completed. In the experimental group, the TPB was implemented by the researcher during four 45-60 minute training sessions. The sessions were run in the comprehensive health centers with regard to the TPB constructs over four weeks through using lecture, questions and answers, group discussion, and presentation of pamphlets and educational booklet in groups consisting of 10-15 mothers. The control group had usual training by family health-care personnel and received pamphlets and educational booklets at the end of the study. Finally, after two months, the data were recollected from the two groups and analyzed using Kolmogorov-Smirnov, Wilcoxon, Mann-Whitney, and Kruskal-Wallis tests by SPSS software version 16.