This last variable, living conditions prior to stroke, is not coded uniformly across centers and hence we accordingly use only categories that are available.
Another variable, "living in rural regions" (Table 3) was protective for the lower socioeconomic population.
Missing values for the variable "Living with persons under 18 years" were placed in the category "not living with persons under 18".
The authors prefer this model to analyzing mortality as a binary variable (lived versus died) using binary data analysis techniques such as logistic regression.
For all imputation analyses, imputation models included the complete outcome variable (living at home at wave 9) and the complete auxiliary variable (GHQ at wave 9, included as a 4-level ordinal variable).
Would you say you are… 1 Living comfortably, 2 Doing alright, 3 Just about getting by, 4 Finding it quite difficult or 5 finding it very difficult?' Data were recoded into a binary variable; living comfortably/doing alright v not.
These variables along with those where p < 0.20 (i.e. total Barthel score as adults continuous variable, living alone versus with others, and post-operative weight bearing) and with season when 25-OHD measure was taken (which is thought to influence 25-OHD levels though not found to be significant on its own) were examined using multivariate regression modeling.
The latter item was quantified using a dichotomous proxy variable – living in a brick house (versus shack, hut, backyard dwelling, "wendy house" [simple wooden one-room structure], or apartment) – because it stratifies relative higher income (who could afford to rent or own a brick house) and lower income families within the population surveyed in this study.
In multivariate analyses for work/societal limitations, we subdivided the living arrangement variable into living with another adult, living with a child, or living with both (with living alone as the reference category --rather than simply "living alone" versus "not living alone"--to ensure that category --rathere parenthood were not disproportionately responsible for missed work.
