That albuminuria was a risk factor for home nursing
rather than peripheral neuropathy, a risk factor for diabetic
foot problems, requires consideration. Microvascular
disease often affects multiple vascular beds simultaneously
in diabetes, hence the continuous variable albuminuriacould be acting as a surrogate marker for neuropathy that
was dichotomised in our models. Aspirin use, Asian
ethnicity and greater income were associated with a
lower frequency of home nursing visits and, notably,
indigenous patients were non-users of home nursing
although their numbers were small in our sample. Aspirin
could be acting as a marker for vascular disease risk in this
situation (bias by indication). The role of cultural and
economic factors in influencing usage patterns of home
services might reflect family or other support and deserves
further study.