The effects of patient characteristics on outcome (coefficients of factors) are comparable using both the single and the multilevel models, but multilevel SEs result greater than the others.
For each health outcome, coefficients of the logistic model were used to calculate the predicted probability of illness of each individual, using his or her unique values for all covariates except for the study group.
We found that adequate and marginal health literacy levels were significantly associated with good glycemic control outcomes (coefficient for marginal and adequate levels were 0.69 and 0.57, respectively).
However, the effect of adequate health literacy was not significantly associated with glycemic outcomes (coefficient: 0.46; 95% confidence limits: -0.03 ~ 0.95).
In model 2, we found that patients with marginal health literacy and better perceived understanding of health education and instructions were significantly associated with good glycemic outcomes (coefficient: 0.62 and 0.46; 95% confidence limits: 0.11 ~ 1.14 and 0.01 ~ 0.91; respectively).
They concluded that the explanation predominantly lay in differing levels of key determinants (covariate effects) rather than differences in the strength of influence of covariates on nutrition outcomes (coefficient effects).
If the mechanism underlying the local average model is also relevant for the outcome, the coefficient on the predicted value will be statistically different from 0. The converse can also be done to test the relevance of the second model (the local aggregate model in our case) (see Liu et al. (2014a) for more details).
A higher proportion of cases survived that received any antiviral treatment compared to those that did not receive antivirals (67% [8/12 patients] vs 7% [1/14 patients], p = 0.003), with a positive correlation between antiviral therapy and disease outcome (Gamma coefficient = 0.664, p = 0.005).
Haemoglobin levels at W2 were associated positively with the infant outcome (regression coefficient of 1.32, 95% CI 0.30 to 2.34).
For both outcomes, covariates' coefficient > 0 or odds ratio (OR) > 1 predicted favorable response.
