Statistical Analysis
Continuous data are shown as means with standard deviation and
categorical data as frequencies and percentages. Group comparisons were
made using the chi squared test and t test or Kruskal-Wallis test for
discrete and continuous variables, respectively, where appropriate. We
used repeated measures analysis to perform univariate and multivariable
regression to determine the outcome across the visits. While the primary
endpoint for GOAL Canada (11) was the proportion of
patients achieving the recommended LDL-C level, the purpose of this
analysis was a comparison between specialists and PCPs with respect to
any differences in the primary endpoint and the use of additional
recommended lipid lowering therapies.
Multivariable logistic regression model was developed to assess factors
independently associated with LDL-C achieving target of ≤2.0 mmol. The
following variables were considered: variables in Table 1 with
p<0.05 and specialist or PCP group. To account for the
clustering of patients within visits, we performed a generalized
estimating equations (GEE) model. The working correlation structure
selected was based on its lowest quasi-likelihood under the independence
model criterion (QIC). Adjusted odds ratio (OR) with 95% confidence
intervals (CI) are presented. A value of P<0.05 was considered
significant for all tests. All statistical analyses were performed in
SAS software version 9.4 (SAS Institute, Cary, NC).