Provision of Disability Adaptations to the Home: Analysis of Household Survey Data

Abstract

The move towards greater provision of healthcare at home has been a significant policy intention for the past two decades [Ham, C., Dixon, A., & Brooke, B. (2012) Transforming the Delivery of Health and Social Care: The Case for Fundamental Change (London: Kings Fund)]. Key to this ambition is the need to provide suitable accommodation for disabled households by installing a range of possible adaptations. Using data from English Housing Surveys of 2003/2004 and 2009/2010, we compare levels of the provision of adaptations with a number of socio-cultural variables, and report on some significant correlations. This includes most importantly, bias against non-white disabled households and younger disabled households, a significant link between rented accommodation and disabled households, and a worrying increase in the proportion of adaptations deemed by the householders to be ‘not needed’, from 7 to 25 per cent, over that 6-year time period. We discuss the context of these results and conclude with an outline plan for future research, which is urgently needed to verify and understand the issues raised.

Keywords::

• Adaptations
• disability
• discrimination
• housing need
• welfare state
• survey data analysis

Acknowledgements

The authors gratefully acknowledge the very helpful comments and advice given by the anonymous reviewers.

Notes

¹ Email: [email protected].

² Although the degree of subjectivity was limited by specific pointers such as amount of litter and presence of graffiti, these are themselves fairly subjective measures.

³ Questions about adaptations are included on a biennial basis, and the survey data-sets published the following year. The 2007 data were not published following an error in survey methodology.

⁴ EHS2009 and EHCS2003 differ slightly, but the lists were rearranged in such a way as to make them commensurate. Table 8 shows the list of 24 adaptations used.

⁵ For the remainder of the paper, the use of the term ‘household’ refers to those that fall within this definition, rather than all of the households in the surveys.

⁶ Correlations of statistical significance were confirmed by using the chi-square test, where the null hypothesis was that there is no significant difference between the expected and observed values of numbers of households with zero adaptations when related to the various social measures. A value of p < 0.05 was deemed to refute this hypothesis and thus confirm a non-coincidental relationship.

⁷p = 0.26.

⁸p = 0.26.

⁹p = 9.4E − 215.

¹⁰ 10. p = 7.7E − 17.

¹¹ Sixty-five might have been a more useful distinction as the common age of retirement with all its related changes, but the data gathered in these surveys has the closest category break at 60 years of age.

¹²p = 0.13.

¹³p = 0.13 for EHCS2003; p = 0.26 for EHS2009.

¹⁴ Despite the many shortcomings discussed in the text, one advantage of large-scale surveys is that by maintaining a consistent methodology, error rates remain constant. Issues such as interviewee misunderstandings are therefore equally likely between the two surveys used here, so that any observable patterns of change are likely to be statistically significant and not unduly skewed by individual cases.

¹⁵ Further details are available at www.lifetimehomes.org.uk (accessed October 2014)

¹⁶ Other forms of assistive technology based on ICT could be seen as more flexible, but suffer with the inflexibility of their infrastructure (Barlow et al., 2007).

Additional information

Funding

This research was supported by funding from the Engineering and Physical Sciences Research Council (EPSRC) ‘Health and Care Infrastructure Research and Innovation Centre’ (HaCIRIC: EP/I029788/1)