• Investigating the role of healthcare accessibility and socio-economic background on the decision to attend for screening for gestational diabetes mellitus in Ireland

      Owens, L; Cullinane, J; Gillespie, P; Avalos, G; O'Sullivan, EP; O'Reilly, M; Dennedy, C; Dunne, F; Department of Medicine, National University of Ireland, Galway 2Department of Economics, National University of Ireland, Galway (Diabetic Pregnancy Study Group, 2011)
      Gestational diabetes mellitus (GDM) is associated with increased maternal and neonatal morbidity and mortality. We investigated the role of healthcare centre accessibility on the decision to attend for screening, employing geographic information systems (GIS), econometric and simulation techniques. In particular, we focus on the extent to which ‘travel distance to screening site’ impacts upon the individual’s screen uptake decision, whether significant geographic inequalities exist in relation to accessibility to screening, and the likely impact on uptake rates of providing screening services at a local level. We also aimed to assess whether Irish women of lower socio-economic status were less likely to attend for screening for Diabetes in Pregnancy than their higher status counterparts. This study was completed through the Atlantic Diabetes in Pregnancy (DIP) partnership, which offers universal screening for Gestational Diabetes at 24-28 weeks gestation. Data was collected on all women who delivered in 5 antenatal centres along the Irish Atlantic Seaboard between 2007-2009. Patients were ‘geocoded’, in order to provide precise spatial (x,y) coordinates for their residential locations. This facilitates geographic information systems -based route analysis of travel distances for each individual to their nearest screening site. We then model the decision to attend for screening, where control variables include travel distance to screening site, a range of other site accessibilityrelated variables, as well as a number of individual-level variables relating to personal, socioeconomic, clinical and lifestyle characteristics. The socio-economic status is based on the deprivation score derived from the 2006 Census of Population for the Republic of Ireland. The Deprivation Index is constructed from a combination of various indicators; education, employment, percentage skilled/unskilled workers, demographic information, lone parents and number of persons/room. 9,043 pregnant women offered screening, 5,218 (58%) of whom participated in testing. The probability of attending for screening was reduced by 1.8% [95% CI: 1.3% to 2.3%] for every additional 10kms required to travel for screening (p=0.000). We also find significant variation in uptake rates across hospitals after controlling for travel distance and other factors, suggesting that accessibility and quality-of-service are also important determinants of overall uptake rates. Using the deprivation index 60% of those who scored 1 (most affluent) attended for screening, 58% in score 2, 56% in score 3, 53% in score 4 and 46% in the score 5 (most deprived) group attended, p=0.0001. This shows a clear decrease in attendance levels in those who are deemed to be more disadvantaged. The most disadvantaged women overall were 40% less likely to attend than their most affluent counterparts (OR0.6, 95%CI {0.55-0.71},p=0.001). Accessibility to healthcare centres and socio-economic background both affect the decision to attend for screening for Gestational Diabetes Mellitus in Ireland.
    • Investigating the role of healthcare centre accessibility on the decision to attend for screening for gestational diabetes mellitus in Ireland [presentation]

      Cullinan, John; Dunne, F; Gillespie, Paddy; Owens, Lisa; John Cullinan (National University of Ireland, Galway. School of Economics), Paddy Gillespie (National University of Ireland, Galway. School of Economics), Lisa Owens (National University of Ireland, Galway. School of Medicine) and Fidelma Dunne (National University of Ireland, Galway. School of Medicine) (2011)
      Gestational diabetes mellitus (GDM) is defined as any degree of glucose intolerance with onset or first recognition during pregnancy and is associated with several serious maternal and neonatal complications and conditions. Screening practices for GDM vary within and across European countries, with some offering universal screening to all pregnant women and others only to selective high risk groups. In Ireland, no single policy with respect to GDM screening is implemented nationally and a debate exists as to what form such a policy should take. Within this context, the Atlantic Diabetes in Pregnancy (ATLANTIC DIP) network was established in 2005 to provide robust information on pregnancy outcomes for women with diabetes. The network includes five healthcare centres along the Atlantic seaboard and provides testing for all pregnant women at 24-28 weeks using a 75g Oral Glucose Tolerance Test. The centres are linked using a clinical information system which allows for data to be captured within a central database, containing a comprehensive range of data on screening uptake rates, maternal characteristics, outcomes for mothers and infants, healthcare resource usage over the course of pregnancy, as well as the postal address of each individual. At present it contains observations on 9,043 pregnant women offered the screening, 5,218 (58%) of whom participated in testing. This paper uses the ATLANTIC DIP dataset to investigate the role of healthcare centre accessibility on the decision to attend for screening, employing geographic information systems (GIS), econometric and simulation techniques. In particular, we focus on the extent to which ‘travel distance to screening site’ impacts upon the individual’s screen uptake decision, whether significant geographic inequalities exist in relation to accessibility to screening, and the likely impact on uptake rates of providing screening services at a local level. In order to do so, the postal addresses contained within the ATLANTIC DIP dataset are first ‘geocoded’, in order to provide precise spatial (x,y) coordinates for patients’ residential locations. This facilitates GIS-based route analysis of travel distances for each individual to their nearest screening site. We then model the decision to attend for screening, where control variables include travel distance to screening site, a range of other site accessibility-related variables, as well as a number of individual-level variables relating to personal, socio-economic, clinical and lifestyle characteristics. Overall, our model suggests that after controlling for these different factors, the probability of attending for screening is reduced by 1.8% [95% CI: 1.3% to 2.3%] for every additional 10kms required to travel for screening (p=0.000). We also find significant variation in uptake rates across hospitals after controlling for travel distance and other factors, suggesting that accessibility and quality-of-service are also important determinants of overall uptake rates. The uptake model allows us to also estimate ‘two-part’ and ‘bivariate probit with sample selection’ models of the risk factors associated with GDM, which provide strong evidence of a socio-economic gradient in the prevalence of GDM in Ireland. Altogether, the findings have important implications for the provision of GDM screening services in Ireland.
    • Prostate cancer risk assessment tools in an unscreened population

      Lundon, D. J.; Kelly, B. D.; Foley, R.; Loeb, S.; Fitzpatrick, J. M.; Watson, R. W. G.; Rogers, E.; Durkan, G. C.; Walsh, K. (World Journal of Urology, 2014-08)