The dependent variable in this study was late receipt of DTP 1, DTP3 and the combination of four DTP, three polio and onel MMR vaccines at three points in time: 3 months, 7 months and 19 months of age respectively. A child was considered late if they received DTP1 at 3 months (>/= 91 days) and/or if they received DTP3 at 7 months (>/= 213 days) and/or if they had not received at least 4 DTP, 3 polio and 1 MMR vaccines by 19 months of age. When vaccination dates couldn’t be determined they were left as missing data. In addition, in some cases children were classified as late for a vaccination by “default.” For example, if the date of immunization for DTP3 was missing and DTP2 was received very late (when DTP3 should have been given), then DTP3 was entered as being late.
The authors used a 30-day grace period in determining lateness for DTP1, DTP3 and completion of the 4:3:1 series at 19 months based on typical clinic practice. At the time of this study, the Recommended Childhood Immunization Schedule, which is approved by the Advisory Committee on Immunization Practices, American Academy of Pediatrics and the American Academy of Family Physicians, recommended that the first doses of DTP and polio be given at 2 months, the third doses of DTP and polio be given at 6 months and by 18 months, a child should have received four DTP, three polio and one MMR vaccines. These guidelines were used in this study. kamagra oral jelly uk
The rationale for the choice of vaccine compo¬nents included as dependent variables in this study was as follows. The choice of DTP and MMR was influenced by the fact that there was a measles outbreak in the early 1990s and the fact that the incidence and consequences of pertussis is highest among children < 6 months. Pertussis and measles are two childhood diseases that children were dying from in the 1990s. Polio was chosen as a dependent variable because it is part of the 4:3:1 series. The varicella and pneumococcal vaccines were not included in this study because the varicella vaccine was only recommended but not mandatory at the time the study was conducted and the pneumococcal vaccine was not yet licensed.
The survey and immunization record data were joined together to create a combined data set for analysis. The analysis of the data included the following. First, in order to identify the underlying dimensions of parental beliefs about immunization and to reduce the number of predictive variables, a series of factor analyses (principal component extraction with varimax orthogonal rotation) was carried out in each of the question content domains. Factor analysis is a statistical technique used to identify a relatively small number of underlying factors that can be used to represent the associations among a larger set of interrelated variables. In this case, meaningful factors were found for questions focusing on the domains of parental beliefs regarding disease transmission, disease prevention, immunization effectiveness and the side effects of immunizations. Scales for these factors were constructed as the mean of the questions in a domain having a factor score of 0.50 or higher (the factor score measures the strength of the relationship of each variable with the underlying dimension). These scales were then utilized in the bivariate and multivariate analyses. canadian antibiotics
Second, the bivariate relationships between the independent variables and the three dependent variables were examined. In some cases, a new independent variable was created to summarize parental answers to a question when multiple responses were possible. For example, participants were asked two questions which measured their correct knowledge of symptoms of whooping cough and measles. Based on their responses, two, new yes/no variables were created called: “correct knowledge of symptoms of whooping cough” and, “correct knowledge of symptoms of measles”. If a respondent was able to name at least three out of five correct symptoms for whooping cough and, three out of four correct symptoms for measles they were coded as having correct knowledge of whooping cough and measles respectively. These two, new independent variables were then used in the bivariate analyses for each of the three dependent variables-late for DTP1, late for DTP3 and late for 43-1 at 19 months.
Finally, logistic regression was used to examine the independent effects of each variable on the immunization outcome measures. In order to address the possibility of racial differences, interaction terms were included in the three logistic regression models for the significant variables in each model. None of these interaction terms were statistically significant, suggesting that there are no interaction effects by race in immunization behavior. However this result must be viewed with some caution because of the small sample size and should be examined in subsequent studies.
Tables summarizing the results of the bivariate analyses were not included in this manuscript because of the volume of data involved. The survey instrument consisted of 51 questions, 12 of which, had subparts for a total of 138 data points. Each of these 138 independent variables was included in bivariate analyses by the 3 dependent variables (i.e. late for DTP1, late for DTP3, and late for 4-3-1 at 19 months). It was not practical to display all of the significant and non-significant bivariate findings in table format.