Copyright Sociological Research Online, 1997


Ong, B. N. & Jordan, K. (1997) 'Health Experiences of Elderly People in an Outer London Area'
Sociological Research Online, vol. 2, no. 1, <>

To cite from articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 9/9/96      Accepted: 28/2/97      Published: 31/3/97


This paper addresses the use of a combined quantitative and qualitative methodology to assess the needs of elderly people within an outer London area. The study was in two stages; firstly, a postal survey of people aged 65 and over using the health profile tool, the Short Form-36 (SF-36), alongside a demographic questionnaire. This was followed by a small number of qualitative interviews on a subset of people who scored at the worst end of the SF-36. These interviews followed themes arising from answers to the SF-36 survey.

The SF-36 was able to pick up inequalities in health within the elderly population; with increasing age, even amongst the elderly, being an important factor in worsening SF-36 scores. Locality also appeared to influence the scoring although this was less conclusive.

The qualitative interviews were able to extract more information and greater understanding of the health of the elderly. Whilst the SF-36 could detect problems on a number of important dimensions such as functional status and well-being, the qualitative interviews addressed individual experiences of ill-health and other issues, such as environmental factors, which affected the health of an individual.

The mix of quantitative and qualitative methodologies gave a broader understanding of the health experience of elderly people to a greater extent than either could on its own. It is important for health policy to address the needs of populations whilst also accommodating individual needs.

Elderly Population; Health- Experiences; Needs; Quantitative-Survey; Qualitative-Interviews; Short Form-36 (SF- 36)


The considerable body of research concerned with inequality in health has been largely based upon its relationship to indicators of social deprivation (Townsend et al, 1989; Feinstein, 1993). These indicators tend to be collected and collated on a geographical basis, and often can be correlated with data on health and social services provision. The Health and Lifestyle studies carried out by Cox (1987) and Cox et al (1993) corroborate the findings from earlier studies that geographical inequalities continue to persist with, for example, men living in the north and north west 1.5 times more likely to die than men in the south and east. However, Illsley (1990) has pointed out that the administrative boundaries used for data collection do not necessarily overlap with meaningful social areas, which may be more heterogeneous than the aggregated figures suggest. Furthermore, individual behaviour is determined by an interplay of variables such as the area in which one lives and the opportunities provided within that environment, but also by social and cultural factors influencing choices and options (Illsley and Baker, 1991). MacIntyre et al (1993) argue that in relation to understanding mortality and morbidity this heterogeneity has to be explored in more detail. They propose that a more systematic examination of 'those characteristics of areas which might influence the physical or mental health of their residents' (p. 220) would assist in forming a better understanding of people's experiences of living in different areas. One of the issues they highlight is the need to gain a sense of what places look like and feel like, which cannot be conveyed by indicators collected on a wide geographical basis. They argue that a local focus should complement a general, environmental approach, in order to arrive at a more complex and a sensitive analysis of the experience of health and illness.

This line of thinking has been further developed by Sloggett and Joshi (1994) who carried out a detailed analysis of census material in order to investigate the association between levels of social deprivation and premature mortality among residents. They argue that deprivation indices can contribute to identification of areas with relative concentration of disadvantage, but that other factors which are not location specific can affect mortality of individuals in different ways. Thus, disadvantaged individuals living in an area of relative affluence are not necessarily protected by that environment, and conversely, advantaged individuals living in a deprived area do not experience excess risk. The conclusion drawn from this study emphasises that 'area based measures of deprivation are not efficient substitutes. For maximum effectiveness, health policy needs to target people as well as places' (p. 1474).

Sloggett and Joshi's study provides a starting point for the research presented in this paper, and is elaborated in a particular manner. Our study focuses on a geographical area and a general population within that area, whilst placing personal data and biography alongside this broader contextual picture. Our study focuses on older people (65 years and over). Sociological research has highlighted important links between ageing and inequality, most notably though an increase in poverty. Gilbert et al (1991) hypothesise that increasing poverty in older people may be the direct result from retirement. It may also be correlated with stages of the life cycle, for example, the older one becomes the chances of living alone increases, and in general, single people are less well off than married people. Furthermore, Gilbert et al point to the cohort effect of the contemporary older population originating from an era where occupational pension arrangements were less beneficial than at present. This is consistent with Victor's (1991) argument that variations in income levels throughout life are reflected in access to resources from occupational pensions. Thus, the income of older people mirrors the differentials and inequalities established during working life, which persist and are exacerbated in old age.

We will argue that assessing health needs of a specific population - in this case elderly people - can be done through the application of a health status measure, thus providing an understanding of the patterning of ill-health, and revealing any possible inequalities. This can then be augmented with a more in-depth understanding of the personal experience of ill-health in order to gain insight into the possible variations within the target population. Furthermore, in terms of health policy assessing the use of health services, and the perceived appropriateness and effectiveness at the individual level offers the opportunity to respond more to the needs of a general population whilst simultaneously creating flexibility in personal targeting. This paper does not elaborate in any detail on policy formulation, but focuses on the methodological aspects of combining area based research and qualitative, biographical research. One of the key questions to be explored is whether qualitative, in-depth interviews with a small number of respondents within specific localities can provide the detailed and sensitive analysis that MacIntyre et al and Sloggett and Joshi advocate, if we are to improve understanding of the interaction between places and people living within them.

In order to explore this question the study presented here will be discussed in the following manner: first, the study design is outlined and the geographical area will be described. Second, selected results from the population based survey, using the Short Form-36 are presented. Third, biographical case studies will be discussed, before drawing the two parts of the study together.

Design and Methods

The assessment of need has increasingly become an important area in health and social policy in developed countries, and a growing body of literature is concerned with assessing needs at the level of populations (Pickin and St. Leger, 1993), client groups (Cassel, 1994) or individuals (McWalter et al, 1994). Within the scope of this article we have no opportunity to elaborate on the concept of need itself and refer to other authors who discuss the theoretical and philosophical basis of understanding need (Doyal and Gough, 1991; Hewitt, 1993; Bradshaw, 1994). However, we draw upon this body of work by combining a health status measure which is capable of describing health needs of large populations with in-depth individual interviews focusing on cultural, social and psychological experiences of health and illness.

The first part of the study consists of primary data from a survey of the general population of elderly people living in the area of Redbridge and Waltham Forest, an outer London area. Redbridge can be broken down into three localities (North, South and West) as can Waltham Forest (North, Central and South) based on localities defined in the Annual Report of Redbridge and Waltham Forest Health Authority (1993). 1991 Census data and local opinion for these localities suggests that the south of the area, Waltham Forest Central and South (in particular) and Redbridge South are the most deprived areas. For example, census figures show that Waltham Forest Central and South both have 42% of their 65 years and over population in households with a limiting long-term illness (standardized by age and sex), whilst Redbridge North and South have 38%, Waltham Forest North 37% and Redbridge West 35%. Standardizing for age and sex gives the district of Waltham Forest a just above average rate of long- term illness compared to the rest of England and Redbridge just below (Forrest and Gordon, 1995) over all age groups. Figure 1 gives a clear indication of the greater deprivation of Waltham Forest Central and South within the area for the age group 65 and over using Census data for the area.

Figure 1: Properties of pensioner households in the separate localities (from 1991 Census)

The purpose of the survey was to construct a needs profile of the elderly population based on their self-perceived health status. The survey used a validated health survey tool, namely the Short Form-36 (SF- 36) (Ware et al, 1993), and a simple supplementary questionnaire consisting of a short (11 questions) form with questions covering further information about characteristics that may affect respondents' answers to the SF-36 and to validate socio-economic background data obtained from the 1991 Census, such as who (if anyone) they live with, whether they are happy with the area in which they reside and recent use of health services. To simplify the questionnaire, and as the main emphasis is on the SF-36, these were mainly YES-NO questions.

The SF-36 is a profile measure with thirty-six questions encompassing eight variables:

There is also one unscaled question relating to change in respondents health over the last twelve months. Each of the question scores on each variable can be summed and transformed to a scale from 0 to 100 (from worst health to best health) for each variable. UK norms have been obtained from surveys by Brazier et al (1992) and Jenkinson et al (1993) but norms for people over 65 have, as yet, not been obtained, although it has recently been validated for use with the elderly (Lyons et al, 1994).

Internal consistency (the extent to which items within each variable are correlated with each other) of the SF-36 was achieved in both British surveys. They also found no reason to question the validity or test-retest reliability of the SF-36 from their analyses. The SF-36 was found to be quick, Brazier et al (1992) reported completion in five minutes, and acceptable to those interviewed. A number of criticisms have been reported in the literature concerning the use of the SF-36 as a self-contained questionnaire. A Scottish study (Garrat et al, 1993) proposed that the SF-36 should be part of a more comprehensive portfolio of measures in order to assess various aspects of patient outcome. Specifically in relation to the elderly population modifications to the questionnaire itself have been proposed so that there is a closer 'fit' between the questions asked and the actual life experience of older people (Lyons et al,1994) and we have made a similar amendment in moving the emphasis from work to regular daily activities for the questions on role limitation as proposed in Hayes et al (1995).

A systematic random sample was used to draw the names and addresses of approximately 750 elderly people aged sixty- five and over from the Redbridge and Waltham Forest Family Health Service Authority (FHSA) register. This should be the most comprehensive and accurate sampling frame available at the time. The sample size accounted for approximately 1.1% of the elderly population and was considered statistically viable. It also was affordable in terms of the time and resources available. The two questionnaires (coded to maintain confidentiality) together with a covering letter explaining the purpose of the survey and a stamped addressed return envelope were sent to the eventual sample of 753. A reminder was sent after about four weeks.

While a survey focuses on enumerative induction, establishing patterns of characteristics in the research population, this does not wholly answer the question as why people experience their health in a particular way. In order to understand the dynamics of the survey findings at the individual level we develop an argument which has similarly been developed by Hill et al (1996) outlining the distinction between what the SF-36 scores represent, namely functional ability, and what people say about their own health, that is expressing their feelings about their daily lives and their general mood and outlook. In their research they note important differences between how people evaluate their own health and what the quantitative data reveal. They conclude that 'the SF-36 is not helpful in assessing the outcome of some health care services for older people' (p. 96). While our study assesses need instead of outcome, we contend that Hill and colleagues' warning about the limitations of the SF-36 is similarly relevant and, equally, we have placed the SF-36 in a wider methodological context by employing both an additional questionnaire and in-depth interviews. By comparing the actual SF- 36 scores of selected individuals with what they say about their health experiences as related to their social and physical environment we can advance the issue highlighted by MacIntyre et al (1993) concerning the need to accumulate empirical information about people's experiences in different areas.

In order to achieve analytical induction (Znaniecki, 1934; Hammersley and Atkinson, 1983) from a small number of cases, a sample of eleven individuals was drawn from those people who scored at the extreme negative health end of the SF-36. The purpose of this selection was to explore with this particular sample of respondents the narrative of their ill-health, the context in which they experienced it, taking into account the social and psychological aspects of their life, and their perceptions of service delivery. The assumption was that people who have the lowest scores on the SF-36 should, theoretically, present the most extreme cases of need. Potentially, this restricts the analysis in that other types of need may not be identified and adequately described. However, because of limited time it was felt that the group that appears to be most needy (in terms of their SF-36 scores) should be examined first.

The concept of need used in this study can be considered to encompass both met and unmet need in terms of functional status (which takes into account the limitations described by Hill et al (1996)). Through in-depth interviews, using these key themes as guidance, the findings from the SF-36 can be compared with the findings from the qualitative study. In particular, it will be possible to explore any discrepancies between the quantitative scores and the experiential interpretations as they relate to the social and physical environment.

Out of the eleven people from the six different localities six individuals agreed to be followed up. While we do not claim statistical representativeness of these six cases, the issues raised through the exploratory interviews could be considered salient for at least those people who scored at the lowest end of the SF-36. In this study the device of personal biographical interviews has been employed.

The respondents were contacted by letter, and if they were willing to be interviewed they sent back a reply slip and a suggested date for a home-based interview. If the respondent was willing to participate an interview was arranged. The researchers visited at the appointed time, explained the purpose of the study and the interview in particular, and requested permission for the interview to be tape recorded. In all cases this permission was granted. The interviews followed a thematic format, elaborating on the SF-36 categories, and discussing service usage. The interviews normally lasted around an hour.

SF-36 Results and Discussion

The response rate after correcting for those who had died, moved or were not known at the address contacted, was 65%.

The average age of respondents was 74.7 years (standard deviation 6.81), and the response figures were similar to the overall age structure of Redbridge and Waltham Forest. 7% of respondents did not state their age. Of those who did, 56% were between 65-74, 34% between 75-84 and 10% 85 years and over, similar to the age structure of the area from Census data.

The lowest score in the SF-36 is zero. As a very general guide, scores from 0 to 10 relate to very severe problems in the relevant scale; scores from 10 to 40 indicate a degree of problems; while high scores of over 80 suggest little or no problems (100) on that dimension. As expected, looking across all the dimensions of the SF-36, the mean scores found for the elderly population in this study were lower than for general populations found in other surveys (Brazier et al, 1992; Jenkinson et al, 1993).

Whilst the primary aim of the survey and the main theory behind the selection of the sample size was to get an overall view of physical ill-health amongst elderly people, and its effect on their everyday life, the survey also allowed comparisons on the basis of sex, age and locality.

An important question is to what extent scores from a health profile tool like the SF-36 reflect the different localities. Analysis of variance tests were conducted for each SF-36 scale using age as a covariate and sex and locality as independent factors. Age was highly significant on each scale (p < 0.001), sex was important for PF and BP (p < 0.05) but locality was not a significant factor. Separating results of the SF-36 by locality, therefore, did not give a clear picture of the differences between the localities although Table 1 shows to some extent the generally lower scores for the south, particularly Waltham Forest South and Redbridge South on the SF-36 scales, standardized by age. Comparison of Figure 1 and Table 1 indicates that the greater deprivation of Waltham Forest South and the lesser deprivation of Redbridge North, Redbridge West and Waltham Forest North are reflected in their SF-36 scores. Waltham Forest Central, by contrast, did not reflect its poor Census results with comparatively low SF-36 scores whilst Redbridge South tended to be comparatively lower amongst the six localities than its Census data would suggest. On the general health scale, 40% of respondents from Waltham Forest South scored below the lower quartile score (bottom 25%) for the region as a whole as opposed to just 16% from the more affluent Waltham Forest North. On all scales, Waltham Forest South and Redbridge South had over a quarter of their respondents in the lower quartile for the region.

Table 1: Means for the localities standardized by age

Redbridge North5855486575575173
Redbridge South5449395465555271
Redbridge West5759446776595573
Waltham Forest North5957577275595372
Waltham Forest Central5755505670575269
Waltham Forest South4953356065564668

The demographic questionnaire did confirm the Census results on the socio-economic variables. For example, more people lived alone in the more deprived areas. It, also showed that less people were happy about the area they lived in Waltham Forest Central and South and Redbridge South although at least three-quarters in each of those localities were happy with their local area. Shortage of space prevents further discussion of the relationship between the SF- 36 scale scores and the demographic questionnaire results. However, the links between these and locality are apparent. Previous work has revealed links between deprivation and illness (Carstairs, 1995; Eachus et al, 1996), however, little work seems to have been done on the link between scores on health profile tools such as the SF-36 and deprivation. Results from this study are not totally clear. Further research needs to be performed in this area.

As seen earlier, age and sex appeared a more important factor to SF-36 scores than locality. There was a significant difference in all scale scores except BP and MH between males 65-74 and males 75 and over. PF, RP and RE were both highly significant (p < 0.01), the remainder not so highly (p < 0.05). For females, all scale scores were highly significant (p < 0.01) between the age groups except for RE and MH which were not significant at the 5% level. Results suggested that scores for females drop further after the age of 75 than males. Males' scores tended to fall more dramatically after the age of 80.

For the age group 65 to 74 years old, males were found to be only significantly different from females for RE (just significant at the 5% level) and MH (p < 0.01). However, for the 75 and over group, males and females are significantly different on PF and BP (p < 0.01) and SF (p < 0.05). Results for the different sex/age groups are given in table 2.

Table 2: Means (standard deviations) for different categories of age and sex

< td>71.0
65-7475+65- 7475+
(2 1.3)
Role - Physical58.3
Role -

The general health (GH) score appears to fall as age increases. This can be broken down as follows: PF, SF and BP are lower in the older people and also in women; RP and VT falls with age whilst RE appears to fall with age in males. Only MH seems to remain relatively stable over the age groups and between both sexes. In none of the categories, did it appear that locality was a more important factor than age or sex.

An important issue is that need has to be considered within a context relevant to the respondent population. Thus, need cannot be evaluated in absolute terms, but has to be compared within a reference framework. For the current study, this framework is defined as the population of Redbridge and Waltham Forest aged 65 years and over, and the needs of age groups, females and males, can then be assessed against the needs of this population. This comparative analysis has been carried out as follows: Figure 2 presents the standard scores, which sets the mean score for the general population of 65 year olds and over to zero. The standard score is calculated by taking each scale mean score and subtracting the general population mean score and dividing by the general population standard deviation. The general mean population score is therefore the horizontal line at 0 and this allows comparison over all the SF-36 scales; above the line showing standard scores above the general population mean, below is under the general population scores. Figure 2 clearly shows the younger of the elderly scoring better (above the zero line) and males scoring better than the females. Females over 75, in particular, are shown to score badly compared to the general population of 65 year olds and over.

GH = General Health; PF = Physical Functioning; RP = Role - Physical;
RE = Role - Emotional; SF = Social Funtioning; BP = Bodily Pain; VT = Vitality; MH = Mental Health

Figure 2: Standard scores for male and female age groups

The overall picture emerging from the SF-36 survey can be summarized as follows:

In order to understand the dynamics of these findings at the individual level we will now turn to a discussion of the case histories.

The Results and Analysis of the Qualitative Interviews

The complex interplay of different definitions of age - chronological, physiological and social age (Arber, 1994) - has methodological implications, in particular when we want to explore how ageing is experienced possibly in different ways in different areas. McKinlay (1992) offers a methodological approach which takes cognizance of this level of diversity. His point of departure is that the group of elderly people is highly heterogeneous, and surveys obscure important differences between subgroups and/or individuals. Therefore, he proposes a revision of traditional methodologies towards a reliance on varied approaches, including personal interviews, use of multi-dimensional instruments, collection of physiological and other data within people's homes, proxy interviewing and self-reports. He argues that choices of separate or combined methods improves the understanding of the heterogeneity of the population. This argument together with the one put forward by Hill et al with respect to the SF-36 strengthens our case for combining the quantitative approach with a qualitative one.

From the six in-depth interviews we have selected three case histories which highlight a number of issues relevant to a better understanding of the SF-36 results and which were common to the other three cases. The case histories can be understood as illness-narratives which have widely been used in sociological research (eg Bury, 1982; Gerhardt, 1996). Narratives have been interpreted as accounts which allow interviewees to construct their own life history (or part thereof). In the process of the interview they can make sense of their illness and its effect on their life within the context of their social and cultural environment. In our study, the accounts are partially 'structured' through the interview focus itself, namely the discussion of the nature and impact of ill-health around the SF-36 dimensions. Yet, at the same time, considerable freedom is accorded to respondents to construct their own narrative.

The scores for all six cases in the ethnographic sample are summarized in Table 3, and provide a comparison with the three selected cases:

Table 3: Characteristics of the ethnographic interview sample

CodeAgeSex General
Role - PhysicalRole -

Note: respondent 5 was included specifically because of having undergone two hip replacements.

Each case history draws on a within- case analysis through a sequence of steps: the description of events which link to make a 'story', the construction of a map of themes, connecting these themes to construct a deeper story which is both variable and process-oriented (Miles and Huberman, 1994). The case histories are presented in turn, and the connecting themes highlighted.

Mrs. A. (Case 3)

This widowed lady, aged 87, lives on her own in a warden controlled flat in a middle/low income area. She has been under some pressure to exchange her flat for a single person one, but she has successfully resisted the move. She has modest financial means, made up of her husband's occupational pension and her own state pension.

Mrs. A. has had a number of health problems during her life, including curvature of the spine, breast cancer (mastectomy in 1973), a number of fractures, high blood pressure, thrombosis, poor eyesight, respiratory illnesses, arthritis and trouble with her feet. She is often in pain especially when carrying things and walking, and has used painkillers intermittently. She has recently been treated by a physiotherapist after she broke her arm, and has four-monthly chiropody treatments.

Mrs. A.'s score on the PF scale was near the bottom quartile reflecting her high level of co-morbidity. The adverse affect of physical problems is clearly reflected in her zero score on the RP scale. Mrs. A. compares her current physical status with how she used to be, namely as someone who was very sport-oriented with swimming, walking and orienteering as her key leisure pursuits. She was a leader in the Girl Guides and a part-time physical education and swimming instructor. Because of her own high standards of physical activity she often feels frustrated by the limitations that her body now places upon her. Her zero score on the RE scale is, however, surprizing given that she demonstrates a lot of will-power:

There are things I can't do, but there's a lot of things I make myself do.

On the MH scale, Mrs. A. demonstrated no problems at all, which was consistent with the interview in which she emphasized the importance of religion to her psychological well-being. Her considerable level of activity in the church was made possible by a combination of social support, people offering her lifts, and her use of public transport. She also discussed the important social network her church provided for her, and the respect that she obviously commanded within that context gave her considerable motivation and drive:

Well, of course, my social contacts are at the church mostly, and here they have a bingo and all that.
(Qs: so you have a lot of friends through the church?)
Well, shall I tell you that I had 156 cards and letters last Christmas. Does that give you an idea?

The contrast between Mrs. A.'s geographical community and her community of interest, the church, is marked: she moved with her husband into the locality because of the availability of warden controlled accommodation, but never established relationships beyond a few within her block of flats. Her main focus is on the social support provided by the church (which is situated outside her locality), and on her other source of support, namely her daughter who has just retired as a nursing sister, and visits her four times a week to help with shopping and washing.

Mrs. A. lives in one of the least deprived wards, but it would be difficult to disentangle whether her SF-36 scores are primarily influenced by her environment or by the specific networks of which she is part. Listening to her own accounts it appears that she considers her well-being directly related to her inner resources, which in turn are fed through her religious congregation. When comparing Mrs. A. with her reference group, namely women aged over 75 years, she experiences many of the key problems highlighted within this group, in particular her physical problems, pain and associated effects on her role functioning. However, her inner resources appear to be uncharacteristically strong and based upon her strong religious feelings. She also appears to be a resilient person, who tends to cope with her difficulties rather than ask for help, and she takes a philosophical approach to ageing:

I can't expect to have perfect health at 87 ... I'm wearing out, that's all there's to it.

Regarding the match between need and service input this case illustrates the level of unmet need, precisely because Mrs. A. appears to be undemanding. Mrs. A. considered an assumption made by the Social Services assessor that her daughter would help her presumptuous, but she felt unable to press her case because she was unsure about the severity of her need. The relatively small service input that she really found important was chiropody which was only delivered once every four months, because of restricted availability, but hardly in line with her actual need.

Mr. B. (Case 4)

Mr. B. lives in one of the medium deprived wards in the area. His life is marked considerably by his World War II experience when he was in a concentration camp 'for no reason whatsoever, except that I am a Jew', and he was sent to England for safety while all his family perished in Germany. He married an English woman, and had a daughter. He has been widowed for almost fifteen years, and has never adjusted to losing his wife.

Mr. B.'s health is poor and he has problems with his heart, lungs, arthritis, diabetes, stomach ulcers, prostrate problems and poor eyesight. He is on many different types of medication. He is often in pain, and walks outside the house with a stick, but manages inside without. Because of his limited mobility Mr. B. moved to a bungalow two years ago, and he can manage on his own with the assistance of a home help once a week, and visits from his daughter two or three times a week. Mr. B. only goes out to shop, and takes the bus to town and then walks back.

Considering the extent of co- morbidity Mr. B.'s score on the GH scale (45) is consistent with his reported symptoms. The effects on his scores on RP and RE are great, and he himself relates this to his more general outlook on life, which centres around his feelings of loss:

I've got no future, what future have I got? I'm almost 83. I didn't think I'd live that long. My wife died when she was 67. I should have died instead of her.

Many widowers experience a contraction in their social life when their partner dies, and Mr. B. stated very clearly that most of his social contacts were through his wife and her family. As a result he now only has a small circle of people, of whom the most important person is his daughter. She has a busy career, but lives sufficiently close to visit her father regularly and to run errands for him.

Mr. B. scored zero on the VT scale, which fitted in with his feelings of pain, tiredness and loss. He expressed his diminished interest in activities in general, rather than personal terms:

You don't want what you used to want when you were young. So one day goes by like the others.

Placed against the background of his personal tragedies and the fact that widowhood continues to be painful to him, this general sentiment has personal poignancy, and explains his passive attitude. However, this does not mean that he is not in need of health and social care. Mr. B. seriously calls into question the appropriateness of GP services, but as a patient of a single-handed practice he feels that he has little choice:

I don't go to the doctor. I used to go every two months. I tried to get tablets to cover me for two months, but the doctor has changed, everything changes. I can't understand them, not because I don't hear so well, but because they don't speak proper English, you know. ... one is from Greece ... The service is not what it should be, but I suppose that's how it is today.

It has been noted that elderly people are less likely to express their dissatisfaction (Blaxter, 1995) and tend to continue using services despite their failure to address people's needs. Mr. B. is no exception to this trend, but the fact that he does delay seeing the doctor can have an adverse effect on his health and well-being.

Mr. B. expresses the sentiment that he likes the area and feels at home. At the same time he does not relate to his environment, mainly as a result of his diminishing social contacts and the selective use of public transport which he only takes for shopping, but not to socialize. His perception of his own health appears to be directly influenced by his physical illness and his feelings of grief at losing his wife. In view of these two overriding factors the beneficial effects of residing in an area he likes, and which suffers a medium degree of deprivation, do not counterbalance the negative experiences.

Mrs. C. (Case 5)

Mrs. C. is a 65 year old widow, living in one of the least deprived wards in the area. She lives alone in a flat, but has regular contact with her daughter, sisters and friends. Her social networks appear to provide a lot of support. She is reasonably well off and has taken out private health insurance after her husband died. This was important in the last five years when she had to undergo two hip replacements. The first one caused considerable discomfort, but the second one was a success. However, as she had to walk on crutches for six weeks it left her with a tendon injury in her shoulder. She also suffers from 'dry eye' and has difficulties in passing urine. In the light of her earlier problems with walking, which have largely been relieved through the hip replacements, Mrs. C. defines her general health as relatively good:

[The hip replacement] has made a complete difference to my life. I mean, I couldn't walk, I couldn't drive ... I could walk round the flat, but I couldn't go anywhere unless I was taken.

She has adjusted her activities, and in particular, has given up playing golf because she cannot bend down. She does not do a lot of cleaning around the house, and her daughter helps her, or sometimes she gets a professional company in.

Mrs. C. has been an active person, and still feels that her diminished mobility has an effect on her self image:

I think mobility, I mean it's like so many things, you don't realize the importance of it until you haven't got it ... it does have a tremendous emotional effect, lack of mobility. You feel old for a start. There was virtually nothing that a forty year old could do that I couldn't do, if I were working or whatever I was doing. But if you can't get about, and having to use a stick has a great emotional effect, not only from your point of view, but anybody looking at you sees you differently...

This feeling of being different was compounded by her widowhood, which she found difficult to cope with emotionally. The improvement since her second hip operation has helped her regain confidence, and this is borne out in her RE score. She has taken up much of the social activities which are not dependent on her mobility such as:

...a little speaking, writing, meetings, community work, and all those things I am able to do because wherever I need to go, people pick me up.

Her VT score, however, is not consistent with her RE and SF scores, as she is in the lower quartile. This can perhaps be explained by her anticipation of old age and the way in which she is prepared:

Well I do try not to think about [the future]. One of the problems of the autumnal years, if you like, is that things can only get worse. You know that it's not getting better, only worse. I hope for five, ten years, but you know it's got to get worse...

In contrast with the above two respondents, who compare themselves with their previous self, or a peer group, Mrs. C.'s comparison with her future self tends to create a gloomy picture. It is possible that this negatively influences her current feelings of vitality.

Apart from the episodes of acute care, Mrs. C. has limited contact with other health services. She is reasonably content with her GP, but feels that he tends to underplay her ailments:

He's very much of the opinion that, well, you're 65. Be thankful that you've got no blood pressure, no heart problems ... you know, live with it. Which, I suppose from the medical point of view is scientifically correct. But at the same time, one knows that there are things that can be done.

While Mrs. C. did not elaborate upon 'the things that can be done' she was obviously not sure what her GP would consider legitimate medical problems. This poses the question as to whether the patient's own perspective is sufficiently validated, and thus the experience of illness can be underestimated.

Mrs. C. has suffered from more extreme health problems than other people within her reference group, but she has sufficient financial resources and social support to make her life as comfortable as possible, and thus her personal circumstances are more important than environment or age. Mrs. A. defies some of the rules around social age and the expectations associated with that in that she continues to assume great responsibility in church work, and keeps up her 'hobby of sick visiting'. She also appears to overcome some of the problems related to physiological ageing through her resilient attitude. Mr. B.'s life is dominated by his widowhood, and the resulting emotional state, coupled with his poor physical health determine his pattern of needs. His limited social support and material circumstances compound his isolation and loneliness.

Specifically focusing on the link between area of residence and the perceptions of health and health need, the qualitative interviews appear to throw up a heterogeneity of experiences that are not directly attributable to locale. All three respondents were selected in the basis of high levels of (co-)morbidity and thus should be considered as extreme cases within their sub- population of the ward. Yet, their accounts of their health experience and health need hardly refer to the area of residence. The explanation in the case of Mrs. A. and Mrs. C. is that they draw on social networks outside their locality and thus the health enhancing impact of their community of interest supersedes that of their geographical community. In the case of Mr. B. his social and physical world is continually contracting as a result of his physical and psychological ill-health. As a result, his environment becomes largely irrelevant. Mrs. C's experiences of health need and service response are perhaps most consistent with that of others in her environment in that she knows how to access both NHS and private provision, and has the means to do so.


McKinlay's (1992) point that the experiences of elderly people are very varied and require a range of methodological approaches to capture them, has been borne out in this study, and the in-depth interviews have provided both depth and context for the findings of the SF-36. A similar study design has been advocated by Farquhar (1995) who researched older people's perceptions of quality of life, and the combination of survey data with open-ended questionnaires and in-depth interviews enabled a complex understanding of how elderly people perceived their health, and the factors influencing it.

The use of the SF-36 for understanding the needs of people over 65 years of age appears to be promising in that it can detect problems on a number of important dimensions, focusing on functional status (physical, social, and role limitations), well-being (mental health, vitality and pain) and offers an overall evaluation of perceived health in terms of general health and health change over time. The limitations of the SF-36, in common with many health status tools, is that it only provides a broad patterning and cannot explain the subjective differences between different individuals' health and related experiences.

On the other hand, qualitative interviews examine the context within which perceptions are formed, but drawing general conclusions from a small number of cases is difficult. This study has attempted to throw light on a number of issues which possibly play an important role in determining individuals' interpretation of their health, so that these issues can be taken into account when looking at the needs of subgroups within the elderly population. We have subdivided these groups in terms of area, age and sex, but the interviews have shown that even these variables do not consistently influence health. While we are not able to extrapolate at an individual level from the qualitative material, the combination of the SF-36 and the in-depth case studies assists in identifying key variables which circumscribe the SF-36 scores. As in Farquhar's study, the interviews highlight the importance of social contacts and support, financial circumstances, psychological well- being (or more generally, someone's outlook on life), the definition of a reference point, environmental factors, alongside physical functioning and general health. Our findings differ from Farquhar (1995) who appears to detect more geographical homogeneity in her sample, while we build on the argument by Slogget and Joshi (1994) that both places and people have to be equally considered in order to account for the enormous variation in needs of the elderly population.

The research on inequalities in health based upon indicators of deprivation relating to area, unemployment, ethnicity and so on remains extremely important. But, integrating a narrative approach allows for a complementary understanding of the processual and contextual aspects, and for people's narratives of their own lives to gain in importance so that the experiences of ill- health and inequality augment the quantitative studies. Unless health policy can address the needs of general populations, while maintaining a degree of flexibility in accommodating individual needs, a certain proportion of people will continue to experience inequalities in health and access to health services.


ARBER, S. (1994) Gender, Health and Ageing, Plenary address, BSA Medical Sociology Group 26th Annual Conference, York, 23 Sept., 1994.

BLAXTER, M. (1995) Consumer Issues within the NHS, Consumers and Research in the NHS. Leeds: NHS Executive.

BRADSHAW, J. (1994) 'The Conceptualization and Measurement of Need: A Social Policy Perspective' in J. Popay & G. Williams (editors) Researching the People's Health. London: Routledge.

BRAZIER, J., HARPER, R. & JONES, N. (1992) 'Validating the SF-36 Health Survey Questionnaire: New Outcome Measure for Primary Care', British Medical Journal, vol. 305, pp.160 - 164.

BURY, M. (1982) 'Chronic Illness as Biographical Disruption', Sociology of Health and Illness, vol. 4, pp. 167 - 182.

CARSTAIRS, V. (1995) 'Deprivation Indices: Their Interpretation and use in Relation to Health', Journal of Epidemiology and Community Health, vol. 49 (supplement 2), pp. S3 - S8.

CASSEL, C. (1994) 'Researching the Health Needs of Elderly People', British Medical Journal, vol. 308, pp.1655 - 1656.

COX, B. D. (1987) The Health and Lifestyle Survey. Cambridge: The Health Promotion Trust.

COX, B., Huppert, F. & Whichelow, M. (1993) The Health and Lifestyle Survey: Seven Years On. Aldershot: Dartmouth.

DOYAL, L. & GOUGH, I. (1991) 'A Theory of Human Need'. London: MacMillan.

EACHUS, J., WILLIAMS, M., CHAN, P., SMITH, G., GRAINGE, M., DONOVAN, J. & FRANKEL, S. (1996) 'Deprivation and Cause Specific Morbidity: Evidence from the Somerset and Avon Survey of Health', British Medical Journal, vol. 312, pp. 287 - 292.

FARQUHAR, M. (1995) 'Elderly People's Defintitions of Quality of Life', Social Science and Medicine, vol. 41, no. 10, pp. 1439 - 1446.

FEINSTEIN, J. (1993) 'The Relationship between Socioeconomic Status and Health: A Review of the Literature', Millbank Quarterly, vol. 71, pp. 279 - 322.

FORREST, R. & GORDON, D. (1995) People and Places: A 1991 Census Atlas of England. Bristol: SAUS Publications.

GARRAT, A., RUTA, D., IBOALA, M. (1993) 'The SF-36 Health Survey Questionnaire: An Outcome Measure Suitable for Routine use within the NHS?', British Medical Journal, vol. 306, pp. 1440 - 1444.

GERHARDT, U. (1996) 'Narratives of Normality: End- Stage Renal Failure Patients' Experience of their Transplant Options' in S. Williams. & M. Calnan (editors) Modern Medicine: Lay Perspectives and Experiences. London: UCL Press.

GILBERT, N., DALE, A., ARBER, S., EVANDROU, M. & LARCZKO, F. (1991) 'Resources in Old Age: Ageing and the Life Course' in Jefferys, M. (editor) Growing Old in the Twentieth Century. London: Routledge.

HAMMERSLEY, M. & ATKINSON, P. (1983) Ethnography: Principles in Practice. London: Tavistock.

HAYES, V., MORRIS, J., WOLFE, C. & MORGAN, M. (1995) 'The SF-36 Health Survey Questionnaire: Is it Suitable for use with Older Adults?', Age and Ageing, vol. 24, pp. 120 - 125.

HEWITT, M. (1993) 'Social Movement and Social Need: Problems with Postmodern Political Theory', Critical Social Policy, vol. 37, pp. 52 - 74.

HILL, S., HARRIES, S. & POPAY, J. (1996) 'Is the Short Form-36 (SF-36) Suitable for Routine Health Outcome Assessment in Health Care for Older People? Evidence from Preliminary Work in Community Based Health Services in England', Journal of Epidemiology and Community Health, vol. 50, pp. 94 - 98.

ILLSLEY, R. (1990) 'Comparative Review of Sources, Methodology and Knowledge', Social Science and Medicine, vol. 31, no. 3, pp. 229 - 236.

ILLSLEY, R. & BAKER, D. (1991) 'Contextual Variations in the Meaning of Health Inequality', Social Science and Medicine, vol. 32, no. 4, pp. 359 - 365.

JENKINSON, C., COULTER, A. & WRIGHT, L. (1993) 'Short Form-36 (SF-36) Health Survey Questionnaire: Normative Data for Adults of Working Age', British Medical Journal, vol. 306, pp. 1437 - 1440.

LYONS, R., PERRY, H. & LITTLEPAGE, B. (1994) 'Evidence for the Validity of the Short Form-36 Questionnaire (SF-36) in an Elderly Population', Age and Ageing, vol. 23, pp. 182 - 184.

MacINTYRE, A., MacIVER, S. & SOOMANS, A. (1993) 'Area, Class and Health: Should we be Focusing on Places or People?' Journal of Social Policy, vol. 22, no. 2, pp. 213 - 235.

McKINLAY, J. (1992) 'Advantages and Limitations of the Survey Approach: Understanding Older People' in J. Daley, I. McDonald and E. Willis (editors) Researching Health Care: Designs, Dilemmas, Disciplines. London: Routledge.

McWALTER, G., TONER, H., CORSER, A. (1994) 'Needs and Needs Assessment: Their Components and Definitions with Reference to Dementia', Health and Social Care, vol. 2, pp. 213 - 219.

MILES, M. & HUBERMAN, A. (1994) Qualitative Data Analysis: An Expanded Source Book. Sage: Thousand Oaks.

PICKIN, C. & St. LEGER, S. (1993) Assessing Health Need Using The Life Cycle Framework. Buckingham: Open University Press.

REDBRIDGE AND WALTHAM FOREST HEALTH AUTHORITY (1993) A Picture of Health 1993 - The Annual Report of the Director of Public Health.

SLOGGETT, A. & JOSHI, H. (1994) 'Higher Mortality in Deprived Areas: Community or Personal Disadvantage?', British Medical Journal, vol. 309, pp.1470 - 1474.

TOWNSEND, P., PHILLIMORE, P. & BEATTIE, A. (1989) Health and Deprivation. London: Routledge.

VICTOR, C. (1991) 'Income Inequality in Later Life' in Jeffreys, M. (editor) Growing Old in the Twentieth Century. London: Routledge.

WARE, J., SNOW, K., KOSINSKI, M. & GANDEK, B. (1993) SF-36 Health Survey: Manual and Interpretation Guide. Boston: Nimrod Press.

ZNANIECKI, F. (1934) The Method of Sociology. New York: Farrars and Rhinehart.

Copyright Sociological Research Online, 1997