Background Irish Travellers have a lower life expectancy than the general Population. Objective health-status indices are consistently poorer in Traveller and Gypsy populations than in other comparably deprived social groups. Self-rated Health (SRH) is established as a valid indicator of objective health status.

Methods As part of an all-Ireland census of 10 618 Traveller families both North and South (response rate 80% overall) employing a novel audio-visual computer interview methodology with peer researchers to overcome literacy barriers, a personal, structured interview was conducted with a random 20% sub-sample of adults (n=2065, 43.5% male). This analysis describes predictors of SRH in those adults, within health domains such as socio-demographic, environmental, lifestyle and psychosocial factors (including those related to culture/identity, and to social capital such as experiences of discrimination). Six models were constructed initially for each domain separately and then a final backwards stepwise logistic regression model was chosen that included 36 potential predictors from all domains of health, with self rated health as a outcome variable dichotomised into categories good (excellent, very good or good) vs fair or poor.

Results SRH was rated as excellent, good or very good by 82.6% of respondents and was age-related. Those who reported limiting disability/long-term illness were excluded. Each separate domain model showed variables predictive of SRH. On adjustment for age-group and sex in the social capital model for instance, experience of discrimination (OR 0.5, p=0.02), low levels of trust (OR 0.6, p=0.04) and worry about getting fair treatment (OR 0.6, p=0.007) were independently inversely associated with good SRH. In the final model, independent positive predictors of good SRH were having a flush toilet (OR 2.2, p=0.02), considering where you live to be healthy (OR 1.9, p=0.02), going “on the road” at least twice in the last year (OR 2.3, p=0.03) and a brisk walk at least once in the last week (OR 2.4, p=0.000). Good SRH was inversely associated with increasing age-group (p=0.000), smoking (OR 0.6, p=0.03), activity-limiting ill health for ≥1 day in the last month (OR 0.4, p=0.001), chronic health condition diagnosed by the GP (OR 0.4, p=0.002), and having been screened for hypertension, high cholesterol or diabetes (OR 0.6, p=0.03).

Conclusion There is a clear relationship between established health determinants and SRH in the Traveller community, suggesting the importance of both material and psychosocial factors. The directionality of the relationship between cultural factors and lifestyle cannot be inferred in these models.