To investigate the variation between practices with respect to the proportions of women willing to delay, we calculated a weighted mean proportion across practices (random-effects meta-analysis), which yielded a similar result (39%95%5% CI 27-52%)).
The model assumes that the observations are conditionally independent at the lowest level given the higher level GP and practice random effects and the predictor variables.
The use of practice random effects ensures that adjustments are limited to differences observed within practices and are not biased by the confounding of overall practice performance with between-practice differences in patient characteristics.
If the fixed effect for women (vs men) is a and the random effect for women by practice random effects has a variance of b, approximately 95% of practices will have women (vs men) coefficients between a-1.96×sqrt(b) and a+1.96×sqrt(b).
30 Random effects Poisson regression models were estimated to relate the number of psychiatric admissions per practice to its QOF achievement, conditioning on potential confounding factors and a normally distributed GP practice random effect with zero mean and constant variance.
In all multilevel analyses, we adjusted for age, gender and educational level of the patient (all fixed effects) and GP-practices (random effect) and the patient group without comorbidity was used as reference group.
To account for clustering of individuals within practices, we included practice level random effects.
Hierarchical linear regression models with random-effects levels for practice, individuals within practices and repeated measurements per individual (slope and intercept) will be used to estimate the effects of the intervention on SF-36 and MLHFQ scores for the post-baseline time points.
As the practices were not selected at random, and were a stratified sample, fitting practice as random effects variable was ruled out.
We will compare the probability of contamination in samples that are retrieved via a 'clean catch' method with those using nappy pads, controlling for patient and practice factors in a two level random-effects logistic regression model (objective 4).
Secondary outcomes will be analysed using a mixed modelling strategy, accounting for between practice variability by defining practices as random effects.
