Extra healthy years or just extra years? What do we know about the - PowerPoint PPT Presentation

Extra healthy years or just extra years? What do we know about the gap between life expectancy and healthy life expectancy on the island of Ireland? 7 March 2013 Kevin P Balanda, Lorraine Fahy, Safa Abdalla, Steve Barron Institute of Public Health in Ireland (IPH)

Increasing life expectancy McGill 2012

Programmes for Government – Health and Well-being goals Department of Health – RoI Future Health: A Strategic Framework for Reform of the Health Service 2012-2015 – There will be a new focus on the need to move away from simply treating ill people to a new concentration on keeping people healthy. Future Health recognises the need for a whole-of-government approach to addressing health issues and commits to the development of a comprehensive Health and Wellbeing Policy Framework and the establishment of a Health and Wellbeing Agency. DHSSPS – NI Priority areas related to health and well-being 1. Creating Opportunities, Tackling Disadvantage and Improving Health and Well-Being 2. Protecting Our People, the Environment and Creating Safer Communities 3. Delivering High Quality and Efficient Public Services

Key questions regarding health expectancy Health expectancy measures – Life expectancy (LE) • how long we can expect to live – Disability-free life expectancy (DFLE) • how long we can expect to live without limitations on our daily activities – Healthy life expectancy (HLE) • how long we can expect to live in (self-rated) good health • Life expectancy increasing – Is healthy life expectancy also increasing? – Are the extra years being spent in good or bad health/ with or without activity limitations? • How long we live and how healthy we are? – Use Health expectancy measures to explore this

Exploring distribution of health expectancy across the island of Ireland

Study aims Description of health expectancies across island of Ireland (DFLE & HLE): – Over time – By geography • National level • Sub-national variation • N/S comparisons – Gender differences – Socio-economic variation • Investigate whether compression or expansion of morbidity occurring Comparable with international series • Describe health expectancies in a way comparable with international series e.g. ONS, EU-SILC • Compare our findings with international series Using three (very different) data sources: – Morbidity (SLÁN; NIHSWB); Mortality; Population

Initial findings • Following initial analysis: – significant differences were noted (by ourselves and others) between the values and patterns of health expectancy derived from different data sources • Tried to understand the differences • Forensic investigation into methodological issues and differences – understand how differences and limitations in data sources affected the findings – Understand the importance of these differences

IPH experience • Differences in methodology – Lead to different conclusions on whether or not compression of morbidity occurring • Within different surveys and series – Lots of different ways health questions are asked – Can have serious effects and implications • Overall findings • Policy development • Will show you there is a problem and how this happened via some case studies • Case studies examined – What health expectancy estimates we would expect from different methodologies – Whether health expectancy estimates were real or because of methodological differences

Three Case Studies

Case study 1

Case study 1 – questions SLÁN Surveys: EU-SILC Surveys: Repeated independent cross- Routine follow-up of a rotating sectional samples panel over a long period In general would you say your How is your health in general? health is: • • Excellent Very Good • • Very Good Good • • Good Fair • • Fair Bad • • Poor Very Bad

Case study 2

Case study 2 – questions NIHSWB survey 1997, 2001, SLÁN Surveys (RoI) 1998, NIHSWB survey 1997 2005 2002, 2006: (HLE2 – NIHSWB): (HLE1 – NIHSWB): Independent cross-sectional Independent cross-sectional Independent cross-sectional samples. samples. sample. Face-to-face interviews. Self-administered postal Face-to-face interviews. questionnaire 1998 & 2002; face-to-face interviews 2006. Over the last 12 months In general, would you say In general, would you say would you say your health your health is: your health is: • • has on the whole been: Excellent Excellent • • • Good Very Good Very Good • • • Fairly good Good Good • • • Not good Fair Fair • • Poor Poor

Proportion of Male Life Expectancy at age 65 with and without Disability NI 2005 NI Disability free 2001 With NI disability 1997 NI 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% RoI 2006 RoI Disability free 2002 With RoI disability 1998 RoI 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Methodological differences

Consequence of methodological differences • Methodological differences undermine validity: – International comparisons – North-South comparisons – Changes over time – Assessment of gender differences and socio-economic variation – Examination of whether compression or expansion of morbidity occurred • Effects of biases can misguide policy decisions

Methodological checklist • Mixed stories resulting from different methodologies: – How the question is phrased? • Time reference; normative reference – How the responses are formatted? • Number of responses; positive or negative responses – How the questions are administered? • Self administered vs. face-to-face – Survey design • Cross-sectional; longitudinal rotating panel – Break in time series • changes to questions in different waves of the same survey – Socio-economic classifications used

Take home messages • Same name for two measures – does not mean they measure the same aspect of health or health care . • Many different aspects of health and well-being – need to identify health measures that are relevant to different policy & service areas – different health expectancy measures lead to different findings – different findings could affect how health budgets are distributed

Take home messages • Working with existing data – Ensure limitations don’t mislead decision-making (e.g. policy) – Don’t take just one data source on which to base findings – Be aware of methodological differences between and within surveys – Read the fine print – If there are differences between data sources then drill down to explain the differences – Need to assist people with the data that we currently have • Longer term – Greater harmonisation between different survey series

Don’t be “stung” by differences… lorraine.fahy@publichealth.ie