The Lifeways Cross-Generation Study: design, recruitment and data management considerations.

The Lifeways Cross-Generation Cohort Study was first established in 2001 and is a unique longitudinal database in Ireland, with currently over three and a half thousand family participants derived from 1124 mothers recruited initially during pregnancy, mainly during 2002. The database comprises a) baseline self-reported health data for all mothers, a third of fathers and at least one grandparent b) clinical hospital data at recruitment, c) three year follow-up data from the families' General Practitioners, and d) linkage to hospital and vaccination databases. Data collection for the five-year follow-up with parents is underway, continuing through 2007. Because there is at present no single national/regional health information system in Ireland, original data instruments were designed to capture data directly from family members and through their hospitals and healthcare providers. A system of relational databases was designed to coordinate data capture for a complex array of study instruments and to facilitate tracking of family members at different time points.

Dietary habits of pregnant women in Ireland.

This analysis of the Lifeways Cohort study mothers during pregnancy (n = 1124), utilises information from a standard food frequency questionnaire completed at baseline recruitment during early pregnancy. We demonstrate that 76% of women achieved recommended intakes of 5 plus portions of fruit and vegetables daily, though this is strongly socially patterned, inversely associated with age and positively associated with level of education. Achievement of the other recommended shelf intakes of the Food Pyramid is much lower, ranging from 12% achieving the recommended sparing intake of foods high in fat, salt or sugar, to 45% consuming the recommended 3 portions per day of meat and poultry. General medical services eligible respondents are generally less likely to achieve recommended intakes. While 61% of women under 25 years old stopped drinking during pregnancy, this dropped to 38% of expectant mothers over 35 years. Less than half (45%) of those (n = 860) who responded specifically to the question reported peri-conceptual folate supplement intake, again strongly socially patterned. These findings both provide important prevalence data and highlight the need for more concerted and supportive health promotion interventions during pregnancy.

The risk factor profile of grandparents.

In the Lifeways Cross Generation Cohort Study, mothers were asked to recruit at least one of 4 potential living grandparents to the study, and 1177 grandparents became active participants who either completed a health status questionnaire only (n = 707), or subsequently underwent a cardiovascular risk assessment examination at home (n = 958). Mean age of grandfathers at baseline was 61.5 years (SD 10.3), of grandmothers 59.2 years (SD 9.1), with a range of 40-83 years, 21% of grandmothers and 16% of grandfathers were third level educated. Risk factor profile of grandparents tended to be more adverse than the general population as assessed by the standard cardiovascular risk factor SCORE. Grandparents' socio-demographic characteristics were similar, whether maternal or paternal in origin. Predictors of positive self-rated health were non smoking (OR 1.5, p = 0.06) and non GMS eligibility (OR 1.99, p < 0.001). At four year follow-up, complete general practice data were available for 285 of 488 respondents with full recruitment data (58.4%). Increased GP utilisation pattern was predicted by baseline morbidity characteristics, though heavier male drinkers were less likely to attend.

Primary care utilisation rates in pre-school children.

A key objective of the Lifeways cross generation cohort study is to examine health and healthcare, according to socio-economic indicators, during the first five years of life. GP contact details were available for 1032 children. 772 GPs in 589 practices were approached and data were obtained on 640 children (59% of original cohort). The mean follow-up time was 3.0 years (95% CI 2.5-3.5). 20.5% of children had a medical card. The mean GP consultation rate was 5.5 visits per child per year, 6.6 visits for children with a medical card and 5.1 for those without (95% CI 1.1 to 1.9) p = 0.001. 68 had a diagnosis of asthma; 19.10% with a medical card and 8.9% without (95% CI 3.0-17.5) p = 0.0001. 138 children had ever been admitted to hospital; 26.2% with a medical card and 21.0% without (95% CI -3.3% to 13.6%) p = 0.2. Incremental increases in parental income significantly decreased both the risk of asthma (p = 0.02) or hospital admission (p = 0.008).