Self-reported health and Healthy Life Expectancy: The impact of question change in the GHS .

• Slide 1

Self-reported health and Healthy Life Expectancy: The impact of question change in the GHS http://www.statistics.gov.uk/downloads/theme_health/HSQ41.pdf Slide 2 Background In 2005 an EU-SILC harmonised general health question was introduced to the GHS This would sit alongside the original general health question for 3 years until replacing it completely in 2008 Slide 3 Background Original: Over the last twelve months would you say your health has on the whole been.. Good? Fairly good? Not Good? Dichotomised Good health EU-SILC: How is your health in general? Is it.. Very good? Good? Fair? Bad? Very bad? Dichotomised Good health Slide 4 Background Self-reported general health contributes to estimates of healthy life expectancy (HLE) - periods of life in Good and not good health Good measure of health inequality HLE at 65 PSA National indicator 137 - Annex C3: Adult Health & Well-being and Tackling Exclusion & Promoting Equality 39 Slide 5 HLE since 1981 Healthy life expectancy (HLE) at birth, Great Britain 19812006 Slide 6 Question What will general health prevalence and healthy life expectancy look like under the new question? Slide 7 First things first Original & EU-SILC questions are asked in succession does this give rise to significant question order or exposure effects? Are responses to the EU-SILC question reliable? Slide 8 GHS Vs. HSE Table 1 Comparison of age-standardised 1 self-reported health rates between the GHS (England) and HSE in 2005-06, all ages Response categoryPer cent GHS (EU-SILC) How is your health in general? Is it... Very good41.9* Good39.6* Fair13.6 Bad3.9 Very bad0.9* Dichotomised 'Good' health 81.5 Sample size n= 25986 HSE** How is your health in general? Would you say it was... Very good40.1* Good41.1* Fair13.9 Bad3.7 Very bad1.1* Dichotomised 'Good' health 81.2 Sample size n = 30922 * = Significant difference between GHS and HSE ** Source: Dept of Health Slide 9 Original Vs. EU-SILC Table 2 Comparison of responses to original and EU-SILC general health questions; England, 2005- 06. All ages Original Over the last 12 months, would you say your health has on the whole been... Response category Per cent Good 64.8 Fairly good 23.6 Not good 11.6 Dichotomised 'Good' health 88.4 EU-SILC How is your health in general? Is it... Very good 39.7 Good 39.5 Fair 15.2 Bad 4.5 Very bad 1.0 Dichotomised 'Good' health 79.2 Sample sizen = 25981 Slide 10 Relationship to functional health Table 3 Comparison of proportional reporting of a limiting long-standing illness (LLSI) by self-reported health status in original and EU-SILC general health questions in England, 2005-06. All ages Original Over the last 12 months would you say your health has on the whole been... Response categoryPer cent also reporting LLSI Good5.4 Fairly good28.4 Not good74.0 Dichotomised 'Good' health11.5 EU-SILC How is your health in general? Would you say it was... Very good3.3 Good12.5 Fair50.4 Bad86.4 Very bad91.5 Dichotomised 'Good' health7.9 sample sizen = 25961 Slide 11 Translation Table 4 Translation of health reporting from original to EU-SILC general health questions Response to original general health question Subsequent reporting of health in response to EU-SILC general health question (per cent) Very goodGoodFairBadVery bad Good59.538.81.60.0 Fairly good7.954.037.10.90.0 Not Good2.08.640.639.98.9 Slide 12 Simulation by age and sex; males, all ages Slide 13 Factors affecting health reporting Table 5Variables associated with increased risk of reporting Not good health VariableOdds ratio95% CI Age (year-on-year increase)1.035*(1.033-1.036) Male1.000 Female1.131*(1.054-1.215) HRP NS-SeC 1 managerial and professional occupations 1.000 HRP NS-SeC 2 Intermediate occupations1.381*(1.247-1.529) HRP NS-SeC 3 Routine and manual occupations1.954*(1.803-2.119) * Significant difference Slide 14 Adjustment factors, males GB Table 6Adjustment factors for males by condensed HRP NS-SeC (GB) Age- band (years) Class 1 HRP NS-SeCClass 2 HRP NS-SeCClass 3 HRP NS-SeC * Good to Fair/Bad/Very bad health (per cent) * Fairly good to Good/Very good health (per cent) * Not good to Good/Very good health (per cent) * Good to Fair/Bad/Very bad health (per cent) * Fairly good to Good/Very good health (per cent) * Not good to Good/Very good health (per cent) * Good to Fair/Bad/Very bad health (per cent) * Fairly good to Good/Very good health (per cent) * Not good to Good/Very good health (per cent) 0-40.072.333.31.673.00.00.771.814.3 5-90.769.820.00.689.50.00.965.116.7 10-140.771.725.00.076.550.02.067.39.1 15-190.078.822.22.590.516.70.567.623.1 20-241.876.733.31.669.20.02.062.530.8 25-292.573.540.03.565.428.61.679.120.8 30-342.075.323.52.264.320.03.568.810.0 35-391.673.628.60.970.66.71.760.315.2 40-440.573.613.05.362.518.82.666.323.5 45-493.360.235.73.571.99.12.659.814.3 50-542.663.215.62.261.514.37.656.84.6 55-592.955.814.91.956.15.11.248.56.1 60-642.858.916.16.062.23.04.442.011.0 65-694.563.221.63.061.40.04.048.19.0 70-742.750.64.72.661.916.76.547.26.3 75-791.459.38.14.350.05.011.348.711.1 80-846.843.212.95.361.57.15.638.98.2 85+3.461.910.510.057.18.34.351.410.0 Total1.764.617.52.465.09.12.556.111.0 Slide 15 HLE since 1981, simulated EU-HLE 2000-02 to 2004-06, EU-HLE 2005-07 At birth, Great Britain 19812006 Slide 16 Findings Responses to EU-SILC general health question appear reliable cannot discount the possibility of minor exposure and order effects Reduction of around 9% in Good health - likely driven by exclusion of fair category and lack of timescale EU-SILC question more closely related to functional health status e.g. LLSI Slide 17 Findings Age, sex and HRP-NSSEC affect self- reported health Simulation indicates a loss of 7-8 years of HLE at birth and 3-3.5 years at age 65 Trend largely consistent with original measure & first *real* EU-HLE measure is consistent with simulation Slide 18 Slide 19 Thanks Questions/comments? For the latest in ONS health statistics please visit: http://www.statistics.gov.uk/hsq/hsqissue/