The arealevel association between hospitaltreated deliberate self harm, deprivation and social fragmentation in Ireland

×
Rating : Rate It:
 
Embed :   
Post a comment
    Post Comment on Twitter
Comments:  



  Notes
 
 
Slide 1 : The area-level association between hospital-treated deliberate self harm, deprivation and social fragmentation in Ireland Paul Corcoran1, Ella Arensman1, Ivan J Perry2 1National Suicide Research Foundation, Cork, Ireland 2Dept. of Epidemiology & Public Health, University College, Cork, Ireland (Journal of Epidemiology and Community Health 2007; 61(12): 1050-5)
Slide 2 : Background Systematic review of area-level association between socio-economic characteristics and suicide (Rehkopf & Buka, 2006) showed that socio-economically poorer areas generally have higher suicide rates Durkheim (1897/1951) developed the concept of ‘anomie’ which relates to a lack of social regulation and integration. Areas of greater anomie have also been shown to have higher suicide rates. Census-based index of anomie or social fragmentation (Congdon, 1996) Relatively few ecological studies of non-fatal suicidal behaviour and area characteristics (e.g. Congdon, 1996; Hawton et al., 2001) Effect modification has been found by geographic area, gender and age
Slide 3 : Aims To describe geographic variation in the rate of hospital-treated deliberate self harm in Ireland To investigate the area-level association between deliberate self harm, deprivation and social fragmentation
Slide 4 : Setting Republic of Ireland: 3,917,203 (Census 2002) 3,422 electoral divisions (EDs) Dublin county: 1,122,821 (29%) Other cities: 287,511 (7%) Urban districts: 437,084 (11%) Rural districts: 2,069,787 (53%) Dublin Galway Limerick Cork Waterford
Slide 5 : National Registry of Deliberate Self Harm Systematic monitoring of attendances to hospital accident and emergency (A&E) departments Data collected for 2002-2004 from almost all of the 40 Irish hospitals with A&E departments (34+3 in 2002, 37 in 2003, 38 in 2004) Identification of deliberate self harm presentations in accordance with an internationally-recognised definition (Platt et al., 1992) Addresses (90%) geocoded to electoral division (ED) level
Slide 6 : Measures Irish National Deprivation Index for Health and Health Services Research based on the 2002 National Census (Kelly & Teljeur, 2004) 1st component from principal component analysis of unemployment, social class, type of household tenure, car ownership, overcrowding Social fragmentation (Congdon, 1996) sum of z-scores of % persons (15+) unmarried, % of single person households, % of persons private renting, % of persons at a different address one year ago EDs assigned a score and categorised into lower, middle and upper tertile (nationally and by type of area)
Slide 7 : Data analysis Annual age-standardised rate (per 100,000) of persons in an area who presented to A&E following self harm with 95% confidence intervals (CIs) One-way ANOVA to compare ED levels of deprivation and social fragmentation by type of area (Dublin, other city, urban district, rural district) Small area data analysis limited to persons aged 15-64 years (96% of cases) Negative binomial regression models estimated the effects of deprivation and fragmentation on ED rates of self harm Adjustment for spatial autocorrelation by indicating that EDs clustered by county Likelihood ratio tests used to test effect modification by type of area, gender and age Effects reported as incidence rate ratios (IRRs) with 95% CIs
Slide 8 : Incidence of deliberate self harm, 2002-2004 Persons 25,797 Presentations 32,777 Rate* - persons 204 (95% CI: 201-206) - men 171 (95% CI: 168-175) - women 237 (95% CI: 233-241) Gender difference +38% * Annual age-standardised rate of hospital-treated individuals per 100,000
Slide 9 : Deliberate self harm, deprivation and social fragmentation by area type % diff from All1 national rate Deprivation2 Fragmentation2 Dublin 224 +10% 0.30 3.86 Other cities 301 +48% 1.87 6.01 Urban districts 330 +62% 1.50 3.87 Rural districts 139 -32% -0.25 -0.69 Ireland 204 -0.08 0.00 1 Annual age-standardised rate of hospital-treated individuals per 100,000 2 One-way ANOVA: df=3,3237, p<0.001
Slide 10 : Effects on ED rate of self harm by persons aged 15-64 years IRR1 (95% CI) Deprivation 1st 1.00 tertile 2nd 1.10* (1.01-1.20) 3rd 1.86*** (1.59-2.17) Fragmentation 1st 1.00 tertile 2nd 1.07 (0.96-1.20) 3rd 1.52*** (1.34-1.73) 1 Mutually adjusted effects of deprivation and social fragmentation * p<0.05, ** p<0.01, *** p<0.001
Slide 11 : Evidence of effect modification Effect of deprivation was modified by: Type of area (chi-square=17.79, df=3, p<0.001). Effect was weakest in rural EDs. Gender (chi-square=5.69, df=2, p=0.058). Greater effect on male rates. Effect of social fragmentation was modified by: Type of area (chi-square=15.02, df=3, p=0.002). Social fragmentation was associated with an increased rate of self harm but not in Dublin. Age (chi-square=12.84, df=2, p=0.002). Stronger effect on the rate of self harm by older adults.
Slide 12 : Explanation of area effects Variable Category IRR1 IRR2 IRR3 Area Dublin 1.38** type Other cities 2.42*** Urban districts 2.83*** Rural districts 1.00 Age 15-39yrs 40-64yrs Gender Male Female Deprivation 1st tertile 2nd 3rd Fragmentation 1st tertile 2nd 3rd 1 crude effects of area type; 2 effects adjusted for main effects of other variables; 3 effects adjusted for main effects of other variables & significant interaction effects
Slide 13 : Explanation of area effects Variable Category IRR1 IRR2 IRR3 Area Dublin 1.38** 1.24** type Other cities 2.42*** 1.81*** Urban districts 2.83*** 1.85*** Rural districts 1.00 1.00 Age 15-39yrs 1.00 40-64yrs 0.50*** Gender Male 1.00 Female 1.46*** Deprivation 1st 1.00 tertile 2nd 1.18*** 3rd 1.77*** Fragmentation 1st 1.00 tertile 2nd 1.07 3rd 1.31*** 1 crude effects of area type; 2 effects adjusted for main effects of other variables; 3 effects adjusted for main effects of other variables & significant interaction effects
Slide 14 : Explanation of area effects Variable Category IRR1 IRR2 IRR3 Area Dublin 1.38** 1.24** 1.20** type Other cities 2.42*** 1.81*** 1.02 Urban districts 2.83*** 1.85*** 1.34* Rural districts 1.00 1.00 1.00 Age 15-39yrs 1.00 1.00 40-64yrs 0.50*** 0.41*** Gender Male 1.00 1.00 Female 1.46*** 1.57*** Deprivation 1st 1.00 1.00 tertile 2nd 1.18*** 1.10 3rd 1.77*** 1.52*** Fragmentation 1st 1.00 1.00 tertile 2nd 1.07 1.01 3rd 1.31*** 1.34*** 1 crude effects of area type; 2 effects adjusted for main effects of other variables; 3 effects adjusted for main effects of other variables & significant interaction effects
Slide 15 : Summary of findings and implications Marked geographic differences in the incidence of deliberate self harm presentations to hospital A&E departments in Ireland. Self harm presentations primarily an urban phenomenon. Deprivation, rather than social fragmentation, had the stronger independent effect on small-area rates of self harm and particularly so in large urban areas Their effects were modified by gender, age and area type. The findings of this and other similar studies may not generalise to other regions and countries. The high rates in Dublin and the very high rates in other cities and urban districts were largely explained by the small-area distribution of deprivation, social fragmentation, age and gender and interactions between these factors. Community-based interventions aimed at reducing deliberate self harm should be targeted at deprived urban areas. There is a need to monitor changes in deliberate self harm rates, deprivation and other area characteristics over time.
Slide 16 : Study limitations Only composite measures of deprivation and social fragmentation were investigated. Other measures warrant consideration, e.g. social capital. We made no adjustment for proximity to hospital A&E departments in the study. The analysis was only at the are-level. Multilevel data analysis is needed to simultaneously assess the effects of individual- and area-level characteristics on rates of self harm. EDs and the Dublin/Other city/Urban/Rural district stratification are not ideal. There is a need to improve Ireland’s geographic information systems (GIS) infrastructure for statistical and research purposes. The geocoding of addresses relied on the Central Statistics Office Townland/ Street Index and the register of electors which have limitations. GIS resources need to be more available to health services and health services research.
Slide 17 : For more information… Paul Corcoran National Suicide Research Foundation 1 Perrott Avenue, College Road, Cork, Ireland www.nsrf.ie and paul.nsrf@iol.ie

 


Related 

 
Free Powerpoint Templates
Add as Friend paul.nsrf@iol.ie     5 Years ago.
785 Views, 0 favourite
PowerPoint Slide Presentation on
More By User

Flag as inappropriate





Browse | Powerpoint Templates | Tags | Contact | About Us | Privacy | FAQ | Blog

© Slideworld