1995 Health Ethnicity Poverty

7/30/2019 1995 Health Ethnicity Poverty

1/116

Long-term ill health, povertyand ethnicity


2/116

This publication can be provided in other formats,

such as large print, Braille and audio.

Please contact:

Communications, Joseph Rowntree Foundation,

The Homestead, 40 Water End, York YO30 6WP.

Tel: 01904 615905. Email: [email protected]


3/116

Long-term ill health, povertyand ethnicitySarah Salway, Lucinda Platt, Punita Chowbey, Kaveri Harrissand Elizabeth Bayliss


4/116

First published in Great Britain in 2007 by

The Policy Press

Fourth Floor, Beacon House

Queens Road

Bristol BS8 1QU

UK

Tel no +44 (0)117 331 4054

Fax no +44 (0)117 331 4093

Email [email protected]

www.policypress.org.uk

University o Sheeld and University o Essex 2007

Published or the Joseph Rowntree Foundation by The Policy Press

ISBN 978 1 86134 993 4

British Library Cataloguing in Publication Data

A catalogue record or this book is available rom the British Library.

Library o Congress Cataloging-in-Publication Data

A catalog record or this book has been requested.

Sarah Salwayis Principal Research Fellow in the Centre or Health and Social Care Research, Sheeld

Hallam University. Lucinda Plattis Senior Lecturer in Sociology in the Institute or Social and Economic

Research, University o Essex. Punita Chowbeyis a Research Associate in the School o Nursing and

Midwiery, University o Sheeld. Kaveri Harriss is a doctoral student in the Centre or Population Studies,

London School o Hygiene & Tropical Medicine. Elizabeth Bayliss is Executive Director at Social Action or

Health.

All rights reserved: no part o this publication may be reproduced, stored in a retrieval system, or transmitted

in any orm or by any means, electronic, mechanical, photocopying, recording or otherwise without the priorwritten permission o the publishers.

TheJoseph Rowntree Foundationhas supported this project as part o its programme o research and

innovative development projects, which it hopes will be o value to policy makers, practitioners and service

users. The acts presented and views expressed in this report are, however, those o the authors and not

necessarily those o the Foundation.

The statements and opinions contained within this publication are solely those o the authors and not

o the University o Bristol or The Policy Press. The University o Bristol and The Policy Press disclaim

responsibility or any injury to persons or property resulting rom any material published in this publication.

The Policy Press works to counter discrimination on grounds o gender, race, disability, age and sexuality.

Cover design by Qube Design Associates, Bristol

Printed in Great Britain by Latimer Trend, Plymouth


5/116

Contents

List o tables, gures and boxes viiAcknowledgements ixSummary x

1 Background and methodology 1Summary points 1Background to the study and research aims 1Conceptualising poverty, long-term ill health and ethnicity 3

Methods 5Overview o the report 10

2 Family, community and society: placing long-term health conditions in context 11Summary points 11Introduction 11Family and household 12Community 18Society and state 22Conclusions 24

3 Individual experiences o long-term ill health: barriers and supports to resilience 25Summary points 25Introduction 25Adjusting to the situation and nding meaning 27Achieving eelings o sel-worth 32Gaining control 33Conclusions 35

4 Long-term health conditions and employment 37

Summary points 37Introduction 37Illness trajectories and attitudes to employment 38Ill health, economic activity and employment 39Ill health, ethnicity or other actors? 41Barriers to employment 43Exploring the possibility o dierential eects o ill health 47The nancial rewards o work 48Conclusions 51

5 Long-term health conditions and welare benefts 52

Summary points 52Introduction 52Patterns o benet take-up 56Deciding to claim sickness benets 58

v


6/116vi

Families without jobs

Experiences o the claiming process: battling with the system 62Implications o receiving ill-health-related benets 64Conclusions 66

6 Long-term health conditions and social participation 67Summary points 67Introduction 67Social participation: networks and activities 68The impacts o long-term health conditions on social participation 70The nature, use and eectiveness o social support 78Conclusions 83

7 Relevance o fndings or national policy 84Introduction 84Benet entitlement and take-up 84

The appropriateness o work or those with long-term health conditions 86The appropriateness o work or those caring or someone with a long-termhealth condition 88Managing long-term ill health 88Tackling minority ethnic disadvantage 89

Reerences 91Appendix A: Qualitative feldwork 97Appendix B: Rates o ill health and caring 99Appendix C: Summary o DWP measures aimed at reducing the number 102

o sickness beneft claimants


7/116vii

Introduction

List o tables, fgures and

boxes

Tables

1 Household composition by ethnic group: households containing one or more 12people o working age with a long-term health condition (% o households)

2 Availability o potential carers: households containing one or more people o 16working age with a long-term health condition (% o households)

3 Prevalence o DLA receipt among working-age adults with long-term health 56condition and activity limitations by ethnic group

4 Percentage engaging in dierent orms o social activity among those o 71working age (18-59/64) by ethnic group and sex

5 Percentage o those with a long-term health condition engaging in dierent orms 71o social activity among those o working age (18-59/64) by ethnic group and sex

A1 Prole o individuals with long-term health conditions interviewed in depth (N=57) 98

Figures

1 Proportions reporting a long-term health condition by ethnic group 21

2 Proportions reporting an activity-limiting health condition by ethnic group 213 Employment, unemployment and economic inactivity by ethnic group and 40health status: men

4 Employment, unemployment and economic inactivity by ethnic group and 40health status: women

5 Predicted probabilities o employment or Pakistani men with an activity-limiting 42long-term health condition, and or the same men as i White British and withouta long-term health condition

6 Average hourly pay by ethnic group and health status: men 497 Average hourly pay by ethnic group and health status: women 498 Proportion o men with a long-term health condition/activity-limiting long-term 57

health condition in receipt o IB by ethnic group

9 Predicted probabilities o lack o participation on our measures, White British 75and Black Arican men compared

10 Predicted probabilities o lack o participation on our measures, women rom 76our ethnic groups compared

11 Predicted probabilities o lack o participation on our measures by ethnic group, 77sex and health status

B1 Percentage o individuals with poor health, England and Wales, London, and 99study area boroughs, 2001

B2 Percentage o individuals perorming unpaid care, England and Wales, London, 100and study area boroughs, 2001

B3 Proportion o individuals with long-term ill health, by age band and ethnic group, 100

2001: menB4 Proportion o individuals with long-term ill health, by age band and ethnic group, 101

2001: women


8/116viii

Long-term ill health, poverty and ethnicity

Boxes

1 Summary inormation regarding the main ill-health-related benets available to 53working-age adults at the time o the study in the UK

A1 Qualitative work: Phase One methods 97


9/116ix

Introduction

Acknowledgements

Numerous people have contributed to this piece o work. We would especially like tothank our team o community researchers or their hard work and insightul comments:Shareen Akhtar, Asma Akhter, Sam Amponsah, Scott Coleman, Wilma Cullen, Md. EnamulHaque, Amirul Islam, Evelynda Laryea, Malvinder Matharu, Arpana Rohatgi, SaleemMallik and Shahid Yasin. We would also like to thank our advisory team members ortheir many useul suggestions and guidance: David Drew, Brian Fisher, Stephen Jenkins,Saron Karlsen, Maura Farrelly, Sally Reeves, Tania Burchardt and Amanda McLeish. HelenBarnard at the Joseph Rowntree Foundation has also given important support to thisproject. We are indebted to a number o community-based organisations in London whohelped us publicise our work and build positive relationships with local people, althoughin the interests o condentiality we have decided not to name these organisationsindividually here. Amanda Baxter, Caroline Cross, Madhu Francine, Eyarun Chaudhury andRizwan Rasheed provided expert transcription and translation support. Crown copyrightmaterial is reproduced with the permission o the Controller o HMSO and the QueensPrinter or Scotland. We are grateul to the Home Oce or permission to use the 2001Citizenship Survey and to the Oce or National Statistics (ONS) or use o the LabourForce Survey. We are grateul to the UK Data Archive at the University o Essex or makingthe various datasets available. Neither ONS, the Home Oce nor the Data Archive,however, bear any responsibility or the analysis or interpretation oered here. Finally, wethank all our respondents in the qualitative component o the study or so willingly givingtheir time and sharing their experiences. Clearly, this work would not have been possible

without their valuable contribution.


10/116x


Summary

Long-term health conditions aect a large proportion o working-age adults in Britain,oten limiting daily activities, employment chances and incomes. This heavy burden olong-term ill health is now o major policy concern across the UK health and social welarearenas. Government interventions aim to increase sel-reliance among those with long-term ill health and reduce the costs to the state o such conditions. However, protectingindividuals and amilies who experience long-term ill health against poverty is currentlyless subject to direct attention.

Research has highlighted the association between long-term ill health and the risk opoverty, and the ar-reaching eects that such health conditions can have on individualsand their amilies. Less is known about the pathways that link ill health to poverty or howthey may be broken. In particular, little attention has been given to social participation,access to non-labour income, and other knock-on eects. In Germany and the US theconsequences o long-term ill health appear to be worse among minority ethnic groups,but little is understood about why this should be so, and these issues have not beenexamined in any detail or the UK.

The present study used a mixed-methods approach to explore the links between long-termill health and dimensions o poverty across our ethnic groups: Bangladeshis, Pakistanis,

White English (White British in the quantitative work) and Ghanaians (Black Aricans inthe quantitative work). Secondary analysis o national quantitative datasets (the LabourForce Survey and the Home Oce Citizenship Survey) was combined with new detailed

qualitative work (including locality-based rapid appraisal methods, 86 in-depth interviewsand eedback workshops).

The study investigates: the ways in which amily, community and wider society aectresponses to long-term ill health (Chapter 2); the ways in which individuals adapt tohaving a long-term condition (Chapter 3); and the links between long-term ill health andemployment (Chapter 4), access to welare benets (Chapter 5) and social participation(Chapter 6). The relevance o ethnicity is explored, although not assumed, across each othese areas.

The immediate amily is shown to be the primary resource or managing the impact

o long-term ill health across all groups. However, ethnic groups vary considerably intheir levels o ill health and household composition, with important implications or thecompletion o tasks, including caring. Family members are oten constrained in their abilityto shit roles in response to ill health. Caring or children alongside an adult with long-termill health can make paid work very dicult.

Positively adapting (developing resilience) in the ace o ill health involves bothmental adjustments and gaining control (coping strategies and experience). While somerespondents showed remarkable buoyancy and enthusiasm or lie, others were strugglingand showed signs o serious emotional upset. Pressures to be normal and hide ill healthexist across all our ethnic groups, with negative implications or seeking healthcare andsupport. While holding on to prior roles and activities is positive or some, or others

it leads to rustration and unhappiness. Achieving sel-worth was particularly dicultor younger people with long-term conditions. Pakistani and Bangladeshi women alsooten struggled to eel valued. The close association with health services experiencedby many respondents had discouraged a sense o control via: long delays to diagnosis/


11/116

treatment; lack o inormation and understanding; unpredictable symptoms; and requentappointments. Few respondents received any support rom external sources to positivelyadapt to their health condition.

Both minority ethnicity and long-term ill health are associated with greatly reduced

chances o employment or men and women. Respondents with long-term ill healthexpressed a commitment to paid work and recognition o its benets over and aboveincome. However, there was variation in the extent to which paid work is seen as apossibility or priority. While there were dierences in the ways in which relationshipsbetween paid work and ill health were discussed between the ethnic groups, quantitativeanalysis showed no ethnic dierences in the eect o ill health on employment. Ethnicpenalties in employment were, however, substantial and resulted in much lower rates oemployment or those rom the three minority groups, than among otherwise comparable

White British men and women. Respondents identied diverse barriers to employmentincluding: their inability ollowing ill health to return to ormer types o employment(particularly that which was physically demanding); the demands o hospital appointmentsand the experience o chronic pain and atigue; employers infexibility; and particularlyor those with mental health conditions, stigma and discrimination. Pay was also adverselyaected by having an activity-limiting health condition and by minority ethnicity.

Just 19% o those with an activity-limiting long-term condition received Disability LivingAllowance. Pakistani, Bangladeshi and Black Arican individuals had much lower receiptthan comparable White British. Our quantitative analyses also suggested striking dierencesin receipt o Incapacity Benet by ethnicity, with White British men with activity-limitingconditions having the highest proportions in receipt (55%) and Bangladeshi men the lowest(just 27%). For many o our respondents, the decision to claim sickness benets was nottaken lightly, and the claiming process was stressul and time-consuming. Respondentsacross all our ethnic groups held concerns about the undesirability o claiming ill-health-

related benets, which were related to the perceived negative implications o assumingthe identity o unt or work or disabled. Such concerns were particularly strong amongthe Ghanaians. However, limited access to inormation and specialist support appearedto be more signicant obstacles, and ones that were heightened or many o the minorityethnic individuals. Ethnically dominated social networks resulted in dierential access toinormation and know-how. Receiving Incapacity Benet/Income Support does not, inand o itsel, appear to exclude people rom the labour market. However, some recipientselt outside mainstream society and expressed strong preerences or (re)-entering the

world o work. Nevertheless, some Incapacity Benet/Income Support recipients had noplans to (re)-enter employment and were nevertheless leading productive and ulllinglives.

Lack o social participation is both a dimension o poverty in and o itsel and may alsoundermine access to other important resources. Most respondents identied maintainingsocial contact and accessing social support as major concerns since the onset o ill health.Most minority ethnic respondents socialised primarily within their ethnic group, andthe our groups showed some distinctive patterns in types o inormal and ormal socialactivity. Social relationships were oten important in acilitating access to more ormaltypes o support rom the state or voluntary agencies. Bangladeshi and Pakistani networkscarried particular reciprocal obligations. Bangladeshi, Pakistani and particularly Ghanaian

women were more constrained in their social contact than White English women or menrom their communities. Quantitative analyses suggested that such dierences were onlypartly explained by low income. Quantitative analyses also indicated that or men caring

was associated with lower social participation, but that ill health was not. For women, illhealth was linked to lower social participation, but caring was not. However, qualitativeanalyses suggest that many individuals experiencesocial participation less positively inthe ace o ill health, oten because social ties do not provide valued support. Social

xi

Summary


12/116xii


networks were oten a burden or Bangladeshi and Pakistani women with chronic illhealth. A complex o actors makes securing social support rom inormal networks dicultincluding: reluctance to seek help; ear o using up avours; the need to reciprocate;mistrust; and eelings o vulnerability. Formal groups and organisations are oten no betterat providing valued practical and material assistance. Many individuals were reluctant

to join ill-health-specic groups, particularly when the perceived ocus on talking aboutill health is not elt to be useul. Individuals with higher income appeared better able tomaintain social participation and to secure the types o social support that they need.Black Aricans (and Ghanaians in the qualitative work) had particularly high risks o socialisolation.

The study ndings highlight the need or the current UK welare reorm agenda to beaccompanied by an equal stress on entitlement and supporting people in ully claimingbenets or which they are eligible and which will at least partially ease some o thestresses associated with chronic ill health. Intervention aimed at increasing awareness,overcoming aversions to claiming and providing specialist support is particularly needed.There is also a need to increase the appropriateness o assessments o tness or work.Current approaches ail to recognise employability as a contingent and dynamic process,dependent not only on the condition, but also on the other characteristics/skills o thepotential employee and the labour market options available. The continuing use ocapacity or (any) work as a yardstick or assessing benet entitlement is overly narrow,potentially stigmatising and unair to those who, much as they would like work, ailto get it. For those with long-term health conditions, insucient consideration to theirexperiences o pain, atigue and the burden o hospital appointments may make tness-or-work assessments unreasonable.

Although employment remains a core part o identity to many with long-term ill health,these perceptions vary with stage o lie and with the type o work that can be carried

out or is anticipated. Job search support and help with the transition to employment mustbe appropriately targeted and consider the ull range o options, without alling back onstereotyping assumptions. An approach to job search and job training that links holisticallyto social support provision and carers assessments is recommended. Furthermore, moreattention must be given to the quality o employment that is on oer and the likelihoodo job retention, since poor work can exacerbate health conditions and is unlikelyto be sustainable. There should also be more recognition o the multiple obstaclesto employment aced by many individuals with long-term conditions and support oralternative positive outcomes, such as voluntary work, where appropriate and desired.Our study ndings thereore illustrate the need or the welare reorm agenda to addressthe question o potential inability o those deemed capable o work to actually nd a

job (or a stable job), and consider what levels o support might be appropriate in suchcircumstances rather than assuming the current overwhelmingly punitive stance.

The policy ocus on engaging employers and instituting appropriate workplace practicesto support those with long-term conditions beore they become economically inactiveis welcome. However, the tension between employers needs or a reliable workorceand the unpredictable and disruptive nature o much long-term ill health requires careulconsideration. At present, many individuals with long-term health conditions are reluctantto reveal their ill health and to claim their right to employment.

There is also a need or increased recognition that ill health aects whole amilies ratherthan simply the individual, and that competing priorities exist. In particular, households

oten combine both dependent children and adults with a chronic health condition. Insuch cases, work may not be regarded by those with multiple caring roles as the mostappropriate way o maintaining the welare o amily members. Employment policies needto be sensitive to the dierences in congurations o households, and, when engaging with


13/116

aims or abolishing child poverty, adequately consider how the overall welare o children,including older children, is best maintained.

In terms o support to individuals and amilies in managing and positively adaptingto long-term health conditions, it is clear that much needs to be done. Few o our

respondents were receiving any kind o ormal support to positively adapt to livingwith a long-term condition and many elt isolated, unsupported and lacking controlover their lives. Furthermore, or many respondents, concerns about nancial insecurityand hardship, immigration issues, or amily disputes, overshadowed ill-health-relatedsymptoms, and support services must thereore recognise the complex realities o peoplesdaily lives and oer support in a holistic ashion. It is clear that group-based provision willnot suit everyone and there is a need or diverse modes o support to those living withlong-term ill health. Particular attention must be given to supporting those who have long-term mental health conditions.

The particular sets o circumstances that characterise the experiences o those belongingto certain ethnic groups, and the tendency or individuals to regard seeking support andservices on the basis o shared ethnic identity as meaningul, suggest that the provisiono interventions in ethnically specic ways is appropriate. However, diversity within andbetween groups must be recognised, so that the relevance o ethnicity, i any, or particularoutcomes can be clearly understood. Furthermore, it is clearly important to meet theneeds o minority ethnic individuals without urthering their marginalisation and isolationrom mainstream services and opportunities, something that was keenly elt by manyparticipants.

xiii

Summary


14/116


15/1161

Background and methodology

Background andmethodology

SummaryPointS

Long-term health conditions aect a large proportion o working-age adults in the UK, otenlimiting daily activities, employment chances and incomes.

This heavy burden o long-term ill health is now o major policy concern across the UK healthand social welare arenas.

Government interventions aim to increase sel-reliance among those with long-term ill healthand reduce the costs to the state o such conditions.

However, protecting individuals and amilies who experience long-term ill health against povertyis currently less subject to direct attention.

Research has highlighted the association between long-term ill health and the risk o poverty,and the ar-reaching eects that such health conditions can have on individuals and their

amilies. Less is known about the pathways that link ill health to poverty or how they may be broken. In

particular, little attention has been given to social participation, access to non-labour income,

and other knock-on eects. In Germany and the US the consequences o long-term ill health appear to be worse among

minority ethnic groups, but little is understood about why this should be so, and these issues

have not been examined in any detail or the UK. The present study integrates secondary analysis o national quantitative datasets with new

in-depth qualitative data to explore the links between long-term ill health and dimensions opoverty.

The study investigates: the ways in which amily, community and wider society aect responsesto long-term ill health (Chapter 2); the ways in which individuals adapt to having a long-term

condition (Chapter 3); and the links between long-term ill health and employment (Chapter 4),access to welare benets (Chapter 5) and social participation (Chapter 6).

The relevance o ethnicity is explored, although not assumed, across each o these areas, with aocus on our diverse ethnic groups: Pakistanis, Bangladeshis, Ghanaians and White English.

Implications or policy and practice are discussed in Chapter 7.

Background to the study and research aims

Over the last century signicant changes in demographic and disease proles haveresulted in ageing populations and a rising prevalence o long-term health conditionsacross Western Europe. Improved medical care and treatment mean that large numbers opeople now live long lives with illnesses that were once thought o as killers; including

heart disease and cancer. At the same time, there has been an increase in other long-term conditions, particularly mental illness and those related to liestyle actors, includingdiabetes. Labour Force Survey data or September 2004, revealed that around 27% o

working-age adults had a long-term health problem with 16% regarding it as activity

1


16/1162


limiting.1 This heavy burden o long-term ill health is now o major policy concern acrossthe health and social welare arenas in the UK. The 2006 Welare Reorm Green PaperAnew deal or welare: Empowering people to workdevoted specic attention to helping illor disabled people. This paper announced changes to the rules and procedures or receipto sickness benets2 as well as new initiatives aimed at increasing the numbers o people

who remain in, or re-enter, employment despite long-term health conditions (DWP, 2006a).At the same time, the UK public health agenda places increasing emphasis on patientchoice and individual responsibility or health and the need to reduce the burden o long-term care to the NHS; priorities are laid out specically in the Public Health White PaperChoosing health: Making healthy choices easier(DH, 2004a) and the White Paper Ourhealth, our care, our say: A new direction or community services(DH, 2006).3

Current government interventions, particularly the expansion in lay-led sel-managementprogrammes (or the Expert Patient Programme)4 and welare-to-work initiatives (such asthe Pathways to Work pilot or Incapacity Benet recipients5), aim to achieve increasedsel-reliance among those with long-term health conditions and at the same time reducethe costs o such conditions to the state. However, other ways in which individuals andamilies experiencing long-term health conditions might be protected against (or lited outo) poverty are currently receiving less attention. It can be argued that the current policyemphasis on paid work as a dening characteristic o citizenship or as the major routeout o social exclusion (Craig, 2004, p 98) gives insucient attention to (a) the need orsecurity or those who cannot work, (b) the act that many orms o employment oergood returns neither in terms o pay nor in terms o participation in society, and (c) theact that there may be other orms o participation that are more valuable and relevant orsome individuals.

Recent research has drawn attention to the way in which long-term health conditionsmay contribute to the risks o poverty. An analysis o poverty dynamics using the British

Household Panel Survey indicated that 8% o cases where an individual moves intopoverty are triggered by a rise in the number o household members with limiting illness

1 UK national surveys have employed a range o similar, although not identical, questions to ascertain levels o long-

term health conditions. The terms long-term, chronic and longstanding are commonly used interchangeably in

the UK research and policy literature, and only in some circumstances is a time period explicitly stated. These terms

indicating an enduring nature are variously combined with health condition, health problem, illness, illness or

disability and illness, impairment or disability. In addition, questions about the extent to which conditions limit

day-to-day activities are also oten used to gain a measure o limiting long-term illness, although again there

is variation in the exact measures used in dierent surveys and studies. In the present study we opt to use the

terms long-term health condition and long-term ill health interchangeably to reer to illnesses and impairments

that have an onset during adult lie and are ongoing without the prospect o a cure. We describe our operational

approach below.2 See Chapter 5 or details o sickness benets available in the UK in 2006.3 The present project also has relevance or the Department or Work and Pensions Public Service Agreement

target on minority ethnic groups employment, and the related work o the (cross-departmental) Ethnic Minority

Employment Task Force; the carers agenda, specically the 2004 Carers (Equal Opportunities) Act; and the disability

rights agenda and the 1995 and 2005 Disability Discrimination Acts, as discussed in Chapter 7.4 In 1999 the Expert Patients Task Force was set up to design a programme or people with long-term health

conditions that would support people to increase their condence, improve their quality o lie and better manage

their condition. It is the central plank o the Department o Healths agenda or increasing the capacity o patients

to sel-care (DH, 2005).

5 Pathways to Work was introduced in October 2003 as a pilot package o measures aimed at increasing the numbero Incapacity Benet recipients who move into paid work. The scheme has been gradually expanded and is planned

to be delivered by private and voluntary sector providers as well as Jobcentre Plus so that by April 2008 it should be

available to all Incapacity Benet claimants.


17/1163


and 16% by a rise in the number with poor mental health (Jenkins and Rigg, 2001). Childpoverty is particularly aected by adult limiting illness (Adelman et al, 2003), and recoveryrom ill health is also identied as an important route out o poverty (Jenkins and Rigg,2001). Other recent analyses o dynamics have highlighted the complexity and diversityo relationships between income, employment and ill health (Burchardt, 2000a; 2000b).

However, little is known about the pathways that link ill health to poverty or how theymay be broken. The coping strategies6 used by individuals and households living withlong-term health conditions and the ways in which these might be supported in orderto buer against a decline in living standards, or oer an escape rom poverty, are alsopoorly understood.

Evidence rom the UK and other settings suggests that the impact o long-term healthconditions on individuals and their amilies may be diverse and ar-reaching. Both care-giving and the extra costs associated with long-term illness and impairment have beeninvestigated (Baldwin, 1985; Martin and White, 1987; Matthews and Truscott, 1990;Berthoud et al, 1993; Ahmad, 2000; Zaidi and Burchardt, 2004; Young et al, 2006).However, less attention has been given to other issues including social participation, accessto non-labour income, and other types o knock-on eects or household members.

While there is evidence rom Germany and the US that the consequences o long-termhealth conditions are not uniorm across population subgroups, with higher risks onegative outcomes among minority ethnic groups (Arrow, 1996; Bound et al, 2003), littleis understood about why this should be so, and these issues have received relatively littleattention in the UK.

The research objectives o the present study were, then:

To describe the relationships between long-term health conditions in working-age

adults and indicators o well-being, poverty and social exclusion o individuals andhouseholds.

To document the range o social, economic and cultural resources that individuals andhouseholds draw on in responding to long-term health conditions.

To identiy routes via which individuals and households experiencing long-term healthconditions can be better supported in order to protect and improve living conditionsand well-being.

The study also aimed to explore whether and in what way ethnicity was relevant to eacho these domains.

Conceptualising poverty, long-term ill health and ethnicity

The broad project aim was to inorm policy and programme responses to the growingnumbers o individuals experiencing long-term health conditions (and associatedimpairment) in adult lie. Although we acknowledge Ahmads (2000) observation thatthe distinction between disability and chronic illness (p 4) is oten blurred in reality, theacademic literature and policy discourses relating to these two areas oten remain distinct.

We thereore draw primarily on the large body o work that seeks to understand rom

6 In the caring and long-term illness literature, a lack o conceptual clarity has been noted, with the term coping

being used to reer both to the ways in which individuals and amilies deal with dicult situations and the

outcome o these eorts, that is, how successul they are (Nolan et al, 1996). We preer to consider coping as the

strategies and activities pursued in response to stress or diculty, and this is consistent with much o the literature

on poverty and social exclusion.


18/1164


patients points o view what it means to live with a long-term health condition, much owhich has explored particular illnesses (such as diabetes).

Although the emphasis o this work has undergone a number o shits (Lawton, 2003; Buryet al, 2005), a common theme has been to recognise that the experience o long-term ill

health is not merely medical, but rather social.

Medical models o illness and disabilityocus on the impairment or condition as the problem and direct intervention towardsthis dimension. In contrast, social models highlight the ways in which society respondsto illness and impairment, oten rendering them disabling. (See urther the discussions inBraddock and Parish, 2001, and Swain et al, 2004.)

Negotiations and social interactions, particularly with amily members and care-givers, arecentral to the experience o living with a long-term condition (Charmaz, 2000). Much othis work has been preoccupied with the degree to which individuals manage to positivelyadapt to living with long-term ill health and similar conditions have been shown tohave diverse implications or dierent individuals (Bury et al, 2005). The implications oindividual and amily responses to long-term ill health or poverty and well-being have,however, received less attention.

Our approach to the present study was infuenced by this body o literature in that wegave particular attention to examining (a) the role o people other than the individual withthe long-term condition, (b) diversity in peoples responses to ill health, including bothnegative and positive implications o the condition, (c) the signicance o peoples liestories and their ideas about who they are, and (d) the infuence o wider community andsocietal actors beyond the amily.

In relation to point (d) above, we were also infuenced by so-called livelihoodsapproaches, which understand poverty to relate to more than just material resources and

income. Instead, individuals and households are seen as actively managing complex setso interrelated resources (Wood and Salway, 2000). Room (2000) suggests that householdscan be thought o as being endowed, to a greater or lesser extent, with resources,relationships and welare entitlements. Resources include not only material capital(current income, assets, stores, savings, gits) but also human capital (skills, educationand, importantly, health status). Relationships include networks within and beyondthe household, links within the local neighbourhood, religious and ethnic communities,and also a persons relationship to the labour market. Relationships give individualsaccess to various resources, but these arrangements are inormal and thereore may beunpredictable. Welare entitlements reer to the various types o support provided bypublic and private organisations, which are based on ormal entitlements underpinned by

state legislation. In the UK context, the tax and benet system should play a major role incushioning the eects o ill health. However, in practice access to benets may be dicult,particularly or certain groups, and the amounts o benets received may be inadequate toensure a decent standard o living.

This way o thinking about the processes that create and perpetuate poverty led usto pay particular attention to the ways in which (a) social networks infuenced, and

were infuenced by, access to material resources, (b) people prioritised dierent partso their lives and held diverse opinions as to which outcomes were most important,(c) individuals actively developed strategies or coping with long-term ill health andthe extent to which these involved relationships within and beyond the household, and(d) wider societal actors (such as labour markets, housing markets, transport networks,

education system, benet system) provided opportunities to, or constraints on, individuals.

Finally, it is important to note that the relevance o ethnicity was considered, although notassumed, throughout our analyses. It is increasingly recognised that ethnic dierentials


19/116


20/1166


were overtly Ghanaian, although churches proved useul. Interestingly, outsiders, such asproessionals working in the area, were also less able to reer suitable respondents, withindividuals o diverse Arican heritage being sent in our direction. Finally, seeking outsuitable White English respondents proved time-consuming. Individuals themselves didnot readily sel-identiy on the basis o ethnicity, there were no community organisations

that were overtly or White English people, and proessionals working in the area wereuncomortable with the act that our work was seeking to single out this group (seeMcLean and Campbell, 2003, or similar research experiences).

Three phases o qualitative data collection were conducted during our eldwork period,which lasted rom March 2005 to May 2006. First, a phase o rapid assessment was carriedout by the research team and a group o trained community researchers. A geographicalcommunity was identied or each o the our ethnic groups through consultation, anda series o inormal conversations, mapping/ranking exercises, key inormant interviews,group discussions, ethnographic interviews and observations were then employed (see

Appendix A). The aim o this phase was to gain a broad overview o the patterns o social,economic and cultural resources available to members o the our communities as well asan understanding o how long-term ill health is perceived and the prominence it has inpeoples everyday lives. This phase also inormed interview guideline development and theidentication o suitable participants or Phase Two.

In Phase Two, in-depth interviews were conducted with (a) working-age adults with along-term health condition and (b) individuals living with someone who had a long-termhealth condition, usually an adult child or spouse.

Although in practice many long-term health conditions bring impairment we neverthelessexplicitly avoided using the term disabled. The intention was to include a broad rangeo conditions (rather than to restrict the sample to those who sel-identied as disabled).

A requirement o our sample was that the health condition should be ongoing and onsetshould have been during adult lie. In practice, our eldwork took a very open-ended,fexible approach, although clearly identication o potential respondents required usto use some terms to indicate who we were looking or. We tended to use the termslong-term health problem, long-term health condition and long-term illness (and theirequivalents in local languages where available).7 On occasion, we also oered exampleso conditions that might be considered as long term, although no precise denitionso degree o severity, duration or intererence with daily activities were employed.This approach allowed us to explore subjective interpretations o ill health and itsconsequences, as well as the extent to which individuals adopted a disabled sel-identity(see Chapter 3).

A total o 86 detailed interviews were available or analysis: 57 o these were interviewswith individuals with a long-term health condition (47 completed in Phase Two and10 completed in Phase One), and 29 were interviews with amily members. Overall,22 interviews were completed with Bangladeshis, 20 with White English, 22 with Pakistanisand 22 with Ghanaians. We ensured that our respondents were mixed with respect tohealth condition, age, sex, migration and employment characteristics. However, ourintentional ocus on East London communities meant that our respondents were, by andlarge, rom more deprived socioeconomic groups. Respondents reported a very wide rangeo health conditions. The majority had conditions that were primarily associated with a

7 Among Bengali/Sylheti speakers the term chirourughi was oten used as the appropriate translation, although the

term literally means permanent illness. Among the Urdu/Punjabi speakers we ound no equivalent term but instead

used the expression lambe arse ki bimari, that is, illness o long duration.


21/1167


long-term illness,8 the most common o which were heart disease, diabetes, depression andarthritis, although there were some who had suered accidental injury. Many respondentsreported more than one condition and experienced pain, atigue and unpredictability osymptoms as core elements o their health condition. The characteristics, including allthe health conditions that were reported, or our interview respondents are presented in

Appendix A. It should be remembered that we gathered inormation rom many morepeople than were interviewed in depth through the use o additional data collectionmethods in Phases One and Three.

Interviews, while being open ended and fexible, included history-taking methods toexplore individual and household trajectories over time. Interviews were conducted inthe language o respondents choice by members o the research team and, subject torespondent approval, were tape-recorded and transcribed. The research team includedfuent speakers o Bengali, Sylheti, Urdu, Punjabi, Hindi, Twi and Ga. Over hal othe interviews with Pakistanis and Bangladeshis were conducted in languages otherthan English, although among the Ghanaians only two respondents chose not to beinterviewed in English. In a small number o interviews it was elt that the respondenthad chosen English but would have been better able to express themselves in anotherlanguage. Translations into English were carried out by research team members and otherexperienced translators with whom we were in close contact. The ocus was on retainingconceptual consistency (Atkin and Chattoo, 2006) and in many cases we opted or thetransliteration o signicant words and phrases so as not to lose meaning.

Analysis and interpretation o ndings were ongoing during data collection. Researcherskept eld diaries and held regular meetings and analysis workshops in which emergingndings and ideas were shared. Part-way through the second phase o data collectionthe inormation gathered was reviewed and some adjustments were made to the datacollection tool to ensure that urther detail could be gathered on issues o interest. A

coding scheme was developed or use with the sotware package Nvivo through aniterative (three-stage) process involving line-by-line blind coding o a subsample otranscripts by three qualitative researchers. Once nalised, the coding structure wasapplied to the interview transcripts and, through multiple search-and-retrieve actions,inormation rom across the range o respondents was brought together or urther themebuilding. This code and compile approach was complemented by detailed memo writingor each respondent using a consistent guideline in which themes running through eachstory, contextual inormation and more interpretive comments (or instance regardinginconsistency between the individual with long-term ill healths reports and those o theother amily member, or obvious omissions) were noted.

Finally, a phase o community eedback and consultation was undertaken during whichemerging ndings were shared via a series o ve inormal meetings. This allowed studyparticipants and other community members to consider the validity and useulness o studyndings, and to comment on the ways in which they represent their community. Detailedeld notes were taken and the salient points integrated with earlier ndings. The ndingspresented in the chapters that ollow draw on evidence generated during all three phaseso data collection, although quotations rom the second phase interviews with individuals

with a long-term health condition and amily members are most commonly used toillustrate the points being made.

8 We use the term illness to reer to respondents own perceptions o having abnormal bodily unctioning. In act,

most respondents also reported medical diagnosis o one or more particular diseases, where a disease can be

dened as the pathological condition o a part, organ or system o the body or mind resulting rom various causes,

such as inection, genetic deect or environmental stress, and characterised by an identiable group o signs or

symptoms.


22/1168


Quantitative

Three quantitative datasets were used. First, we used the Labour Force Survey (LFS), anationally representative survey covering around 60,000 individual respondents eachquarter. We used this survey to examine the relationship between long-term ill health and

(a) labour market status and (b) use o benets. The LFS is a semi-panel survey in whicheach respondent is surveyed at ve consecutive quarters. We pooled multiple quarters toensure sucient numbers o minority ethnic respondents: 12 quarters rom 2002-05 oranalysis o benet use and labour market position and 16 quarters (using wave 1 responsesonly) rom 2001-05 or analysis o pay. We used the question: Do you have any healthproblems or disabilities that you expect will last or more than a year? We exploited thepanel element o the survey to remove some o the noise associated with the healthquestions by requiring that, to be counted as having a long-term health condition, therespondent had to have answered positively to the question in the subsequent quarter as

well as the current one. Thus, we eectively excluded all wave 5 responses and includedan additional quarter o data or the construction o this variable. We also made use othe question (addressed to those who had already positively answered the rst question):Do these health problems or disabilities, when taken singly or together, substantially limit

your ability to carry out normal day to day activities? We reer to those who respondedpositively to this latter question (and who were already included in our long-term healthcondition measure) as those with an activity-limiting health condition. In addition, weused number o conditions reported to approximate severity o ill health.

We used the Oce or National Statistics (ONS) question on ethnic group to carry outanalysis or the our groups that corresponded most closely to our qualitative sample, thatis Bangladeshi, Pakistani, Black Arican and White British (see also discussion on pages 9-10 on integrating approaches). We employed descriptive statistics or the topics o interestand binary and multinomial logistic regression models (as appropriate). The regression

models allowed us to investigate whether apparent ethnic dierences could be explainedby dierences in other characteristics relevant to our outcome variables. Moreover,

we specically explored whether there was any evidence or ethnic dierences in theassociations between long-term ill health and employment (see Chapter 4).

Second, we used the household le o the LFS (HLFS)to investigate the structure andcomposition o households containing working-age individuals with long-term ill health.The household le covers approximately 55,000 households at each quarter and isavailable or the spring and autumn quarters o each year. We pooled six quarters rom2002-04 to ensure sucient numbers or analysis, resulting in around 334,000 cases,representing 190,000 unique households. We used this household dataset to describe

patterns o employment and caring in households with a member with a long-term healthcondition. We used the ONS question on the ethnic group o the household reerenceperson to explore variation in these associations by ethnicity.

Third, we used the 2001 Home Oce Citizenship Survey (HOCS)9 to investigate therelationships between long-term health conditions and social participation. The HOCS isa biennial cross-sectional survey designed to capture ethnic dierences in citizenship andcommunity experiences. It is specically employed to monitor aspects o inclusion andcommunity cohesion in relation to the UKs reporting responsibilities under the EuropeanUnion Lisbon goals. It includes a booster sample o 5,000 members o minority ethnicgroups as well as the 10,000-person main sample. This survey included the question:Do you have any long-term illness, health problem or disability which limits your daily

activities or the work you can do?, which we used to identiy individuals with a limiting

9 We had some unresolved questions about the quality o the ethnicity data or 2003, so we thereore used the 2001

sweep instead.


23/1169


long-term health condition. In addition, we examined outcomes or individuals whoreported regular caring responsibilities or someone within the household or elsewhere.Patterns o social participation were explored by ethnic group (again using the ONScategories), health status and caring status, using descriptive statistics and ordered probitregression models (see Chapter 6).

Throughout the elaboration o the quantitative analysis, we ocus on discussion o theresults. In particular, we summarise results o regression models rather than supplyingtables o coecients, which are not always straightorward to interpret. We ocus ondiscussion o results that are statistically signicant at the 5% level (and in the LFS analysisthis is having adjusted standard errors or repeat observations on individuals). We alsoemploy graphical illustrations o relevant results, including predicted probabilities rom theregression models, where it claries the argument. Full regression tables or all the analysesare available rom the authors.

Integrating approaches

The above description o methods has indicated a number o ways in which the samplesor the qualitative and quantitative components o the study diered. Our qualitative workcan be considered to have generated data relating to a subset o the sample included inthe quantitative datasets or three reasons.

First, the way in which individuals were categorised as having a long-term healthcondition diered slightly between the datasets. Our qualitative work was designedspecically to inorm policy and programme responses to the growing numbers oindividuals experiencing long-term health conditions (and associated impairment) in adultlie and we thereore intentionally excluded individuals who had had health conditions

since birth or childhood, and also avoided using the term disabled during recruitmento participants. However, our quantitative work utilised existing datasets, which, as notedabove, grouped together individuals reporting health problems or disabilities that youexpect will last or more than a year in the case o the LFS, and long-term illness, healthproblem or disability in the case o the HOCS. Clearly, then, these survey data will includeindividuals who have been experiencing their health condition since childhood. It is alsolikely that the survey data include some individuals who regard themselves as having adisability but not a health problem, health condition or illness; individuals who wouldnot have been included in our qualitative work. Clearly, the distinction between disabilityand long-term illness is not clear cut and it is unlikely that the dierent denitionsemployed by the qualitative and quantitative approaches compromise their compatibility.

We thereore opt to use the terms long-term health condition and long-term ill healthinterchangeably to encompass the range o conditions that were included in the study.

The second area o dierence between our data sources relates to geographical spread.Whereas the LFS and HOCS data derive rom a national sample, the qualitative work tookplace primarily in East London. Exploration o Census data indicates that levels o long-term ill health were higher and socioeconomic conditions poorer in our study areas thannationally (see Chapter 2). However, since the objective o the research project was toexplore the links between long-term health conditions and various dimensions o poverty,such a geographical and socioeconomic ocus was warranted.

Third, it is important to consider the dierences in the ways in which ethnic group

membership was operationalised in the two approaches and in the ethnic categoriesavailable or analysis. As noted above, our qualitative work ocused on our contrastingethnic groups, namely Bangladeshi, Pakistani, Ghanaian and White English and allowedus to examine the extent to which such categories were meaningul and the implications


24/11610


they had or individuals lives. In contrast, our quantitative analyses were restricted to usingthe ONS question on ethnic group, and there was less scope or exploring the relevance othese categories. We concentrated on the our groups that mapped most closely onto thosein our qualitative sample: Bangladeshi, White British, Pakistani and Black Arican. ClearlyBlack Arican is a much more heterogeneous category than our Ghanaian sample (itsel

containing dierent linguistic and ethnic groups). However, it is likely to include the vastmajority o those who dene themselves as Ghanaian and it is the closest approximation

we could arrive at to our Ghanaian sample. The White British category is also somewhatbroader than our White English sample, but the disparity is much smaller. It will includemost o those who dene themselves as White English, who numerically dominate thecategory, but it will also include White Scottish, White Welsh and some White Irish.

Notwithstanding these dierences, the integration o qualitative and quantitative methodsprovided a much more comprehensive investigation o the links between long-termill health and poverty across diverse ethnic groups than either one approach wouldhave provided. The qualitative and quantitative approaches were considered to becomplementary (Adamson, 2005), allowing us both to explore in detail the ways in

which individuals and amilies experience and respond to long-term ill health and alsoto describe the aggregate levels and dierentials in access to key resources: employment,

welare benets and social participation. The research team included individuals with bothqualitative and quantitative skills and the two pieces o work were guided by a commonset o theoretical assumptions and research objectives. During the course o the projectthe qualitative and quantitative work proceeded side by side with regular eedback oemerging ndings between the two. The writing-up process was also conducted in anintegrated ashion with particular team members taking responsibility or considering thequalitative and quantitative evidence on particular topics and bringing the two together toprovide an integrated account.

Overview o the report

We start our discussion o the studys ndings in Chapter 2, by setting out the ways inwhich amily, community and society structure the resources that individuals with long-term health conditions and their amily members can draw upon in responding to illhealth. In Chapter 3, we present ndings relating to ways in which individuals adapt toliving with long-term ill health. These two chapters orm the backdrop or the subsequentthree chapters 4, 5 and 6 which look at the implications o long-term ill health orthree key areas that impact upon individual and household poverty: employment, welarebenets and social participation. Chapter 7 considers the relevance o the studys ndings

to policy and practice across the health and social welare arenas.


25/11611

Family, community and society: placing long-term health conditions in context

Family, community andsociety: placing long-termhealth conditions in context

SummaryPointS

The immediate amily is the primary resource or managing the impact o long-term ill

health. Ethnic groups vary considerably in household composition, with important implications or the

completion o tasks, including caring. Family members are oten constrained in their ability to shit roles in response to ill health. Caring

or children alongside an adult with long-term ill health can make paid work very dicult. Levels o long-term ill health dier signicantly between ethnic groups.

Community membership aects the impact o long-term ill health, via both the social networksit coners and the resources available locally.

Social relationships are oten important in acilitating access to more ormal types o support

rom the state or voluntary agencies. For the minority ethnic respondents, membership o an ethnic community was meaningul

and brought access to opportunities and support, but also expectations, obligations and

constraints. Although experiences o direct racist discrimination were rarely reported, general eelings o

exclusion rom mainstream White society and services were widespread among minority ethnicrespondents.

Recent policy emphases and wider public opinion can aect the way individuals eel aboutthemselves and their condition, as well as their entitlements rom the state.

Introduction

Family, community and wider society infuence the social and economic consequences olong-term health conditions in complex ways. Family and household structure, particularlythe absence o potential adult workers and presence o dependent children, have beenshown to exacerbate the impact o long-term conditions (Parker, 1993; Drewett et al, 1994;Mutchler et al, 1999). Existing literature has also described the importance o the amilyas a resource in managing the impact o ill health. In the short term, the amily is the rstreserve or dealing with symptoms and diagnosis (Bury, 1991). In the longer term, thepractical limitations brought on by the health condition oten imply a transormation oamily roles and relationships. Prior research has emphasised that the capacity to weather

these problems depends on the pre-existing quality o amily lie and the personalities othe individuals involved (Parker, 1993). Some amilies do break down, more commonlysoon ater onset rather than ater a long period o strain (Locker, 1983). At the communitylevel, there has been some interest in inormal community care and the involvement o

2


26/11612


riends, neighbours and acquaintances beyond the amily (see, or example, Drewettet al, 1994). However, the implications o belonging to a particular community, such asaccess to resources and expected patterns o behaviour, have been given less attention.Furthermore, the wider social climate impacts upon individuals, and this should also beexamined. Here we draw on both qualitative and quantitative data to describe some o the

processes and pathways through which amily, community and society appear to infuencethe experiences o living with a long-term health condition among our our ethnic groupso ocus: Bangladeshis, Pakistanis, White English and Ghanaians (or Black Aricans in thequantitative analyses).

Family and household

Important commonalities were ound across the our ethnic groups in the impact o long-term health conditions on the amily. Furthermore, within each ethnic group, there was alot o variation in the ease with which amilies made adaptations. However, amily size,structure and the ways in which amily roles are organised varied to some extent across theethnic groups, and these dierences had implications or the ways that amilies responded.

Family members, care and support

The amily members who were at hand within the household varied substantially.Restricting our analysis to households containing at least one person o working age witha long-term health condition, Table 1 illustrates the diering composition o householdsacross the ethnic groups. Among the Black Arican group, almost a quarter were single-person households that is, by denition, people living on their own with a long-termcondition and thus with no potential co-resident carers or alternative workers.

Table 1 also shows the very high proportion o lone-parent households among BlackArican households. In these households, the demands on the lone parent will besubstantial: not only caring or dependent children but also attending to their ownhealth (or, in some cases, that o a non-dependent child); as was the case or a numbero our respondents in the qualitative work. Conversely, non-dependent and even

Table 1: Household composition by ethnic group: households containing one or more people oworking age with a long-term health condition (% o households)

White BlackBritish Pakistani Bangladeshi Arican

Single person 17.4 6.3 3.3 24.4

Couple without children 28.7 6.9 3.5 7.8Couple with dependent children 25.7 47.3 51.6 21.0

Couple with non-dependent children only 10.6 5.6 4.4 1.5Lone parents with dependent children 7.1 9.9 11.2 27.4

Lone parents with non-dependent children only 4.2 2.3 2.2 4.6Complex (multiple-amily) households 6.1 21.7 24.1 13.3

Notes:Complex households include multiple-amily households (including where the amily consists o a single

person), and couple or lone-parent amilies where people other than (dependent and non-dependent) children are

present. Same-sex couples have been excluded due to the very small proportions, particularly rom the minority

groups. Ethnic group is attributed to the household on the basis o the ethnicity o the household reerence person.

Source:HLFS 2002-04, authors analysis


27/11613


dependent children may be providing care and support to their parent or taking on therole o caring or younger children. By contrast, the high number o complex (multiple-amily) households among Pakistanis and Bangladeshis (22% and 24% respectively) maymaximise the possibility o available carers or those able to take on alternative roles thatthe individual with a long-term condition can no longer manage (although it may o

course imply other problems, particularly overcrowded living conditions). Around 50% oBangladeshi and Pakistani households with a long-term ill member consisted o couples

with dependent children, compared with just 26% o the White British group, amongwhom couples living alone were the most likely household composition (29%). Thesehousehold-level data can, o course, tell us nothing about the extent to which there areamily members living close by, who provide important care and support, a pattern that

was commonly revealed in the qualitative work.

Despite important diversity in household composition, our qualitative work indicatedstrong areas o commonality between the ethnic groups in the involvement o amilymembers in care and support. First, it was evident that support is requently a source otension across all the ethnic groups. In common with other work (or example Katbamnaet al, 2000) our ndings dispel the stereotype o cohesive South Asian amilies withunproblematic support structures.

Second, there were also strong commonalities in the types o kin who the respondents eltthey could turn to in practice or support and care, most commonly spouses/partners andadult children. Similarly, across all our groups, daughters were more likely to be drawninto supportive roles than sons. The respondents also held common understandings aboutthe desirability o children being sheltered rom the impact o ill health.

Most respondents had contact with amily living outside the household, including relativesin other cities and other countries in the case o our minority respondents, and received

various orms o support rom them. Childcare provided by local grandparents wasparticularly valued in reeing up parents to do other work. However, members o theextended amily did not generally become involved in the everyday management o thehealth condition (such as maintaining medication regimes or dietary requirements) and itsknock-on eects (such as increased household maintenance work). There was oten a gapbetween the expectations people held and the provision o support rom relatives beyondthe household, particularly among the Pakistanis and Bangladeshis who commonly eltthat certain amily members were obligated (culturally and religiously) to provide certainkinds o support. In all groups, the provision o support rested critically on the strength oaection between specic amily members, although this was most pronounced among the

White English respondents. The eeling that children today ailed to respect and support

their parents was common traditional expectations about amily caring were in a state ofux and being replaced with an individualism in which people had their own lives or didwhat they liked.

Respondents stories revealed variation in the extent to which amilies engaged incollective planning, managing and sharing o income and other resources. Although

variation within the ethnic groups should not be ignored, collective activity was mostevident among the Pakistanis and Bangladeshis, oten extending across the memberso multiple households, and usually involving individuals related through the male line(although this was not absent among the other groups). For instance, purchasing ohouses, educational ees and trips back home would all commonly be unded througharrangements among multiple adult members o an extended amily. In contrast, although

some White English respondents reerred to relatives as sources o nancial support andadvice in times o need, the management o amily resources and the long-term well-beingo amily members were the concern and responsibility o the nuclear amily, primarilythe husbandwie couple. Among our Ghanaian respondents, within the nuclear amily,


28/11614


separate money management by spouses/partners was ound to be common, with womenand men directing their income to dierent expenditures, particularly their own parentsand siblings who remained in Ghana with whom some degree o sharing o resources

was common. Clearly, such dierential patterns o income sharing have implications orthe resources over which amily members with long-term health conditions might have a

legitimate claim, and the individuals who can be called upon in times o need.

Adaptations to amily roles and responsibilities

In all amilies, individuals other than the person with the long-term health condition havea stake in how the condition and its knock-on eects (or instance withdrawal rom paidemployment) are managed. The extent to which amily members adjust their roles dependson perceived options and their desirability; actors that are in turn dependent on a range oexternal actors such as the local labour market, and socially expected behaviour, as wellas personal preerences. In general, we ound much evidence o infexibility surroundingthe roles that individuals played in amilies.

The eect o an individuals health condition on the likelihood o another amily memberadjusting their pattern o employment was varied. Some amily members, particularlyspouses, but also adult children, had to take time o work to provide direct care or, morecommonly, to take on childcare or be a companion or medical appointments. The latter

was a particular issue or Bangladeshis and Pakistanis who oten lacked the necessaryEnglish language skills and condence to negotiate the health service alone and there wasgreater expectation that individuals would be accompanied than among the White Englishor Ghanaians. Taking time o work or the health needs o another amily member was

very dicult and had led in some cases to leaving employment completely.

They used to call me to work they need me, but I had personal problem athome. So beore I try the sick note, you know, they dont accept. Then I go or

without pay leave or three months. Ater that I let the job [in 1992]. (Bangladeshiman, amily member, 55-59 years)

In other cases, amily members had altered their working patterns to accommodate theneeds associated with the health condition, or instance by working part time, taking workcloser to home, or taking a night shit, and thereby managed to remain in employment.

However, in some cases across all the ethnic groups it appeared that individuals prioritisedtheir income-earning role and made ew allowances or the health condition o the

amily member. This was particularly true o men (although not absent among women)and seemed oten to relate to a combination o reusing to recognise their wies healthcondition, a reluctance to take up a caring or home-maker role and the importance givento their sel-identity as income-earner.

The departure rom employment o the main amily income-earner due to ill health is notnecessarily a trigger or others to start work; in act it can have quite the opposite eect.In some cases, caring requirements limited the potential employability o alternative

workers ollowing the onset o the breadwinners condition, although it was clear thatperceptions o need and demand or care varied greatly. Caring oten took a toll onparticular individuals, usually women, who could even eel exploited. Many o ourBangladeshi and Pakistani emale respondents reerred to social pressures to be seen to be

caring appropriately or their husbands with long-term conditions, making the uptake o aworking role more dicult and a lower priority.


29/11615


In cases where the health condition had become all-consuming because o its severityand/or the emotional response to it (see Chapter 3), amily members elt overwhelmed

with day-to-day household tasks and taking up employment had not even beenconsidered.

No, I am busy with children, looking ater children. Now, Im busy with myhusband and children. How could I work? Hes ill and so. Because he is myhusband I have to look ater him. I cant just leave him behind and go awayanywhere. I have to be with him all the time. (Bangladeshi woman, amilymember, 30-34 years)

The value attached to amily maintenance also tended to restrict fexibility in householdroles, particularly among the minority ethnic groups, or whom cooking ood rom backhome was an important and morally loaded symbol o cultural belonging and o womencorrectly ullling their expected emale roles.

There was some evidence that amily roles were more fexible among the Ghanaiansthan in the other communities, with caring and productive roles less clearly opposed toone another, particularly or women. Furthermore, the high prevalence o lone-parenthouseholds clearly precluded the division o types o work between men and women inmany cases. However, across all our groups there was more evidence o women assumingincome-generating roles than o men picking up household tasks. Nevertheless, womenacross all our groups gave their mothering (and grandmothering) roles high priorityand in a number o cases sought to emphasise these when long-term health conditionscompromised their ability to perorm paid employment. Furthermore, when husbands orpartners experienced health conditions, women were not always willing to sacrice theirchildcare role and go out to work.

Ive still be really wanting to stay at home and look ater my children. Obviously Icould have just said right hes not working I need to go and get a job and have mychildren looked ater. But I really didnt want to. So I basically had to put up withjust been living on benets which obviously isnt very much money. So I justeel that I my children get a good quality o lie rom actually having a lot otime with us but then maybe they dont have like the holidays abroad or the latesttoys and things. (White English woman, amily member, 40-44 years)

Furthermore, in some cases the health condition was such that husbands were unable toadequately perorm household work or childcare, urther limiting the possibility o role-switching between spouses. Negative social attitudes towards men perorming household

work (evident to a greater or lesser extent across all the communities, but particularlymarked in some Pakistani, Bangladeshi and Ghanaian amilies) urther restricted thefexibility o household roles.

Notwithstanding these commonalities across the communities, there was evidence amongsome o the Pakistani and Bangladeshi households (although by no means all) that men

with long-term conditions were reluctant or their wives to work because it was seen asurther undermining their role as head and provider or the household. Some Pakistani andBangladeshi women in turn actively clung onto their responsibility or household workand childcare, regardless o their husbands health, oten refecting their limited alternativesources o recognition and status within the amily.

Furthermore, quantitative analyses showed signicant variation between the ethnic groupsat the aggregate level in the extent to which childcare and caring roles exist alongside eachother, which may make substitution o roles, and taking up employment more complicated,as our respondents indicated. Seventy-nine per cent o Bangladeshi households containing


30/11616


a working-age adult with a long-term health condition also contained a dependent child,compared with 66% o Pakistani households, 49% o Black Arican households and just30% o White British households.

Moreover, i we consider caring and employment as potentially competing with each other,

we see that the likelihood o there being potential carers in a household (dened asworking-age adults who do not have a long-term health condition and are not working)also varied substantially by ethnic group (Table 2); rom 13% o White British householdsto 64% o Bangladeshi households. Moreover, the presence o available carers may beparticularly important in households with children, both to provide care or them and toprotect them rom taking on caring roles themselves (although such responsibilities arenot necessarily experienced negatively). Again, a much higher proportion o Bangladeshiand Pakistani households (at 66% and 60%) had such available carers than Black Aricanor White British households (at 35% and 19%). This dierence partly refects ethnicdierences in amily composition; but the qualitative ndings discussed above suggest thatdierent approaches to prioritising caring versus breadwinning, a reluctance to relinquishchildcare roles, the perceived demands o caring, as well as the perceived unavailabilityand low rewards o employment, may discourage people rom being economically active.

Long-term health conditions are also not evenly distributed across households. Instead,particular households may have several members with long-term conditions while otherhouseholds are ree rom ill health. Examining those households with at least one working-age adult with a long-term condition, analysis o the HLFS showed that 44% o Bangladeshihouseholds in act contained two or more people (including older people) with such acondition, compared with 39% o Pakistani households, 28% o White British householdsand 15% o Black Arican households. Although these ethnic dierentials again in partrefect the diering household structures, nevertheless, the co-existence o several healthconditions within one household is likely to create added strain and urther restrict the

ability o household members to adapt. Our qualitative work included several amilieswith more than one member experiencing long-term ill health. In such cases, caring andsupporting between these amily members was oten complex and reciprocal.

Clearly, complex caring demands co-exist in some households meaning that amilymembers interests may not necessarily be best served by adults/parents (with and withoutlong-term health conditions) being in paid work, particularly when the economic returns tosuch work are not high (see Chapter 4); a point not lost on many o our respondents.

Table 2: Availability o potential carers: households containing one or more people o workingage with a long-term health condition (% o households)

Households with potential carers

None Yes, all working Yes, all not working Yes, some not working

White British 44.8 41.7 8.5 4.9Pakistani 24.5 23.5 29.2 22.8

Bangladeshi 19.9 15.8 39.5 24.9Black Arican 53.4 18.3 20.7 7.6

Source:HLFS 2004-06, authors analysis


31/11617


Multiple household stresses and fnancial hardship

Long-term health conditions are oten just one o many stresses that individuals and amilymembers have to deal with. Our respondents oten talked about their health condition asinherently connected to other problems in their lives including marital instability, amily

confict or domestic violence, worklessness or nancial insecurity. In addition, many oour minority ethnic respondents aced problems related to their country o origin and, orsome, their status as migrants. For Bangladeshis and Pakistanis, transnational dicultiescommonly related to the need to nancially support elderly relatives back home, thehigh costs o international travel and health conditions making such travel impossible. Forseveral more recent Ghanaian migrants, their immediate amilies were split between Ghanaand the UK, causing much distress and meaning limited networks o support locally. Long-term health conditions complicate and in turn are complicated by these other problems, sothat stresses begin to reinorce one another and may make managing extremely dicult.

Across all our ethnic groups, the link between long-term health conditions and nancialhardship was a prominent one. A decline in earnings was the most common reason orthis link. However, respondents also spoke o the increased expenditures associated withlong-term health conditions, particularly in relation to special dietary needs and the addedcosts associated with impaired mobility. Respondents described not only having to go

without and needing to budget careully, but also the mental stress that living on a lowincome brings. Many respondents spoke o savings being exhausted and accumulatingdebt. As one key inormant in Phase One noted: People eel overwhelmed by debt, theydont open their letters, bills. Respondents ound not being able to aord things or theirchildren particularly upsetting. Financial hardship was also reported to lead to amilytension and confict by many respondents, although some individuals did seem to nd iteasier than others to come to terms with their limited resources.

Beore I just used to buy it and not have to worry. Not like Id have a great bigwage packet and lots o money, but just to be able to, you know, not always haveto think about it all the time. Not to have to work it out i you spend money onthat then youre not going to be able to do all these things. (White English woman,amily member, 40-44 years)

Respondents whose only income was welare benets oten elt particularly vulnerable,and many respondents reported getting into nancial diculties due to benets beingdelayed, stopped or reduced without warning.

Families were ound to adopt various strategies not only to reduce expenditure but also

to make keeping track o bills and expenses easier and less stressul. Nevertheless, it wasapparent that over the long term, living on a low income had signicant implications orquality o lie as things wear out and break down and homes all into disrepair. Costs otransport seemed to be a particular issue or many with long-term conditions. Giving up acar to reduce expenditure had let some housebound. Buses were oten inconvenient anddicult to manage, while taxis were prohibitively expensive.

Although these issues aected individuals across all the ethnic groups, the Ghanaians stoodout as being most at risk o nancial hardship because o their oten weak support romscattered amily members and their lack o entitlement to welare benet (due to recentmigration) and generally poor knowledge o the system. In contrast, it was evident thatseveral o our Bangladeshi and Pakistani individuals with long-term health conditions were

nancially supported by their adult children, although this clearly depended upon lie-cycle stage and was not true across the board. The White English experience o nancialsupport within amilies was mixed and highly dependent upon the closeness o particularrelationships. Several o our White English respondents were entirely dependent on benet


32/11618


income, although some who had worked in the past had occupational pensions andsavings as added buers against nancial hardship.

Community

Membership o a community has relevance or the experiences o individuals living withlong-term health conditions and their amilies in two related ways: rst, via resourcesheld at the community level, which may be drawn upon; and second, via the options andconstraints placed on individuals by community norms and expectations.

Meanings o community: ethnicity and location

The majority o respondents elt that it made sense to speak o a community in their localarea and most expressed a sense o belonging to such a community. However, the ways in

which ideas about community related to location and ethnicity were complex and contextspecic.

For the Ghanaians, Pakistanis and Bangladeshis it was clear that ethnicity (understoodprimarily as a shared country o origin and common culture) was a salient part o sel-identity (that is, the way people elt about themselves) and that their ethnicity impliedbeing part o an ethnic community. Despite evidence o diverse liestyles within thesecommunities (particularly according to socioeconomic status and generation), respondentsnevertheless talked about shared ethnicity bringing a sense o eeling easy or having acommon understanding. Importantly also, shared ethnicity entailed expectations andinormal claims to support, advice, inormation and advocacy (although such claims wererequently compromised by health conditions, as discussed in Chapter 6). In our Pakistani

and Bangladeshi eldwork areas, dense patterns o settlement meant that daily interactionspredominantly involved other individuals rom the same ethnic group. Furthermore, amilynetworks, including links to Bangladesh or Pakistan, remained active or many, binding themembers o these communities together both socially and economically.

The Pakistanis and Bangladeshis were particularly reliant on intra-ethnic contacts or workand support, including support with negotiating state agencies. A reliance on inclusion ininormal networks opens up possibilities and accesses certain resources, but also entailsexclusion rom others, and a narrowing o what is known and possible.

I used to eel very scared and earul going up to the doctors I had nobody to

support me. My parents were there but they are old and dont speak English. Ihad no Bengali person who could support me I had nobody to guide me or letme know that I had other options. (Bangladeshi woman with long-term condition,35-39 years)

Having noted the signicance o being Ghanaian, Pakistani or Bangladeshi, it isimportant to recognise that subdivisions were evident within these groups, and thatthese divisions had some implications both or the types o behaviour and attitudes thatpeople adopted and, importantly, the networks o resources and support that were opento individuals. Such subdivisions were most evident among the Ghanaians, refecting thediverse linguistic, cultural, religious and socioeconomic prole o this group.

A strong sense o belonging to a particular ethnic community was requently combinedwith eelings o alienation rom mainstream White society and its acilities. Althoughour minority ethnic respondents did not oten talk about racism in the orm o activeexclusion, the ear o racism could be a powerul deterrent to the take-up o services and


33/11619


opportunities. Furthermore, many minority ethnic respondents recounted the extremediculties they had aced in accessing services and support rom varied governmentagencies. Although similar problems were also mentioned by White English respondents,such experiences seemed to exacerbate minority e

Documents

1995 Health Ethnicity Poverty