public health

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Perceived Health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

How is your health in general? A qualitativestudy on self-assessed health

J.G. Simon1, J.B. De Boer2, I.M.A. Joung1, H. Bosma1, J.P. Mackenbach1

Background: The single-item measure on self-assessed health has been widely used, as it presentsresearchers with a summary of an individuals general state of health. A qualitative study was initiated tofind out which particular aspects are included in health self-assessments; which aspects do peopleconsider when answering the question How is your health in general?. Subgroup differences werestudied with respect to gender, age, health status and health assessment. Methods: Qualitative studywith stratification by background characteristic, health status and health assessment (n 40). Results:Almost 80% of the participants referred to one or more physical aspects (chronic illness, physicalproblems, medical treatment, age-related complaints, prognosis, bodily mechanics, and resilience).However, when assessing their health, participants also include aspects that go beyond the physicaldimension of health. In total, 80 percent of the participantswhether or not in addition to physicalaspectsreferred to other health dimensions. Besides physical aspects, participants considered theextent to which they are able to perform (functional dimension 228%), the extent to which theyadapted to, or their attitude towards an existing illness (coping dimension 228%), and simply the waythey feel (wellbeing dimension 220%). In this study, health behaviour or lifestyle factors (behaviouraldimension 23%) proved to be relatively unimportant in health selfassessments. Conclusions Self-assessed health proved to be a multidimensional concept. For most part, subgroup differences in self-assessed health could be attributed to experience with ill health: being relatively inexperienced withhealth problems versus having a history of health problems.

Keywords: qualitative study, selfassessed health, stratified sample, subgroup differences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The single question How is your health in general? is a crudeand simple measure that has been widely used, as it presentsresearchers with a summary of an individuals general state ofhealth. It is presumed that in self-assessed health numerousaspects of health are combined within the perceptual frameworkof the individual respondent.1 3 This measure proved to be apowerful predictor for mortality; poor self-assessed healthincreases the mortality risk, even when other (more objective)indicators of health status have been controlled for.4

Many studies have been conducted to find out whichparticular aspects are included in health self-assessments. Inquantitative studies the relationship between a priori definedhealth measures and self-assessed health has been analysed. Inthese studies, however, a significant proportion of variance inself-assessed health remains unexplained. This suggests thatwhen assessing their health, participants may include healthaspects that have not been routinely included in quantitativeanalyses. Therefore, some researchers have used a qualitativeapproach to identify the remaining and unknown aspects ofself-assessed health. Briefly summarizing, selfassessed healthseems mainly to be associated with physical health problems,functional capacities, health behaviour, and psychological

aspects.5 9 Additionally, some studies found that aspects suchas health comparison,5 health transcendence, externally focused,non-reflective,6 social role activities, and social relationships9

were included in health self-assessments. Only two of thesequalitative studies attempted to include equal numbers ofparticipants of different sociodemographic backgrounds.5,7 Theother studies included convenience samples predominantlyconsisting of women, elderly, highly educated participants6,8,9

or participants with health problems.8 However, healthstandards may vary among different subgroups, and probablydepend very much upon gender,6 age5,10 and experience withhealth problems.10 Therefore, it is difficult to decide whether thefindings in these studies reflect general health conceptions, orare determined by the most prevalent subgroup. It would berelevant to know whether participants from different subgroupsconsider entirely different aspects when assessing their health,but with the exception of Krause and Jays study,5 qualitativestudies on self-assessed health rarely examined subgroupdifferences.

We initiated a qualitative study on self-assessed health in asample that has been stratified on background characteristics,health status, and health assessment. The present paper focuseson the aspects that people consider when answering thequestion How is your health in general? Is it very good, good,fair, sometimes good and sometimes poor, or poor?. We believethat health assessments follow an individual process of orderingand weighing different health aspects. Therefore, we askedparticipants what went through their minds when answering thequestion on self-assessed health. The analysis was guided by thefollowing research questions: Which aspects do participantsconsider when answering the question on self-assessed health?Do participants with different background characteristics (age

Correspondence: I.M.A. Joung, Department of Public Health, ErasmusMC, University Medical Center Rotterdam, P.O. Box 1738, 3000 DRRotterdam, The Netherlands, tel. +31 10 4087714, fax +31 10 4089449,e-mail: [email protected]

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Department of Public Health, Erasmus University Rotterdam, The

Netherlands2 Department of Medical Psychology and Psychotherapy, Erasmus

University Rotterdam, The Netherlands

European Journal of Public Health, Vol. 15, No. 2, 200208

q The Author 2005. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.

doi:10.1093/eurpub/cki102

and gender), and participants with different health status (withand without current chronic conditions) consider the same ordifferent aspects when assessing their health? Do participantswith good and less-than-good self-assessed health consider thesame or different aspects when assessing their health?

Data and methods

Study population

Our study population consists of participants of the GLOBEstudy, a longitudinal study designed to describe and explainsociodemographic inequalities in health in the Netherlands.Design and objective of the GLOBEstudy have been described indetail elsewhere.11 At baseline in 1991, participants comprised acohort of non-institutionalized men and women with Dutchnationality, 1574 years of age, who were living in the city ofEindhoven or surrounding municipalities. In 1997, a subgroupof respondents to the baseline interview were approached toparticipate in a follow-up study. For our qualitative study, wedrew a stratified sample from the respondents to the 1997follow-up. The interviews took place in 1998.

The variables for stratification have been chosen because oftheir supposed relationship with self-assessed health: gender,age, socioeconomic status, and health status. In order to obtainmaximum contrast, we included men and women, younger than40 years of age and older than 60 years of age, with the highestlevel of education (university degree) and with the lowest levelof education (primary or lower vocational education), with achronic illness (COPD/asthma or chronic back complaints) andwithout a current illness. Furthermore, we stratified on the mostrecent available (i.e. 1997) health assessment and thus includedparticipants with (very) good, as well as participants with less-than-good self-assessed health (stratification table is availableon request).

Non-response and changes in health assessments

In each stratum, participants were randomly selected. It was,however, not possible to select participants in all strata, due tovarious reasons. First, some strata did not exist in thepopulation from which we drew our study sample. Second,the number of possible participants that fitted a particularprofile (i.e. stratum) could be very low. When these participantsall refused to participate in our study, there were no othereligible participants we could approach. Third, some partici-pants changed their health assessment during the 1998 semi-structured interview compared to the followup data (1997) onwhich we based our initial selection of respondents. All in all, wewere able to select participants for 74% of the existing strata.

From May till December 1998, we approached 63 people bymail and telephone. Fourteen persons were unwilling toparticipate in the study, we were unable to get into contactwith six persons, and three persons were unavailable during thestudy period, although willing to participate. Thus, weinterviewed 40 participants, a response of 63%. The distributionof the different stratification variables can be seen in table 1.

Semi-structured interview

All participants were interviewed in their homes by the principalinvestigator (JS). The semi-structured interviews, lastingapproximately 35 minutes, were tape recorded and transcribedverbatim. Following a brief introduction the interviewees werepresented with the core question How is your health in general?Is it very good, good, fair, sometimes good and sometimes poor,or poor?, and were then asked to explain their particularresponse.

Interview analysis

We started with analysing the verbatim text of the interviews. Ineach interview, we condensed the answers given to the single-item measure on self-assessed health and the reasons for thishealth assessment. Parts of the text representing the same themewere summarised with a single phrase, hereby paraphrasing theparticipant. In this way, each interview could be condensed intopersonal themes. Next, we categorized the personal themes of allparticipants into a smaller number of recurrent themes, whichwe will refer to as health aspects. Finally, on categorization ofthese health aspects, five conceptually meaningful healthdimensions emerged (see Appendix 1 for a flow chart of thecoding process). For development of the overall categorizationscheme, and for the data analysis that followed, QSR NUD*ISTsoftware,12 were used.

To ensure reliability in coding and analysing the interviewsfour researchers (JS, JB, IJ and HB) independently read andcoded eight of the interviews. The results were compared anddiscussed to come to a reliable method for analysing theinterviews. Next, the principal investigator (JS) read and codedall interviews, and designed the final categorization scheme.Finally, one of the other researchers (IJ) independently appliedthe categorization scheme (on the level of health dimensions) toeight of the interviews. We then calculated Cohens Kappa, ameasure of interrater reliability, and the level of agreement wasshown to be good (k 0.69).13

This paper presents the overall frequency distribution of thedifferent dimensions and health aspects, as well as thedistribution of health dimensions by gender, age, health status,and health assessment. Chi-square analyses are used to examinewhether referring to a particular dimension varies significantlyfor different subgroups.

Results

Which health aspects are taken into consideration?

The final categorization scheme consists of 17 health aspects,categorized into five health dimensions. The frequencies withwhich the different health dimensions and health aspects werementioned are shown in table 2. In Appendix 2 the descriptionof the health dimensions and health aspects are given andillustrated with quotations. (1) We considered physicalreferences, i.e. any reference to disease, illness, medicaltreatments, or other bodily-oriented theme to be an aspectof the physical dimension. (2) Any reference to general

Table 1 Distribution of stratification variables in studypopulation

Stratification variables Categories N

Gender Women 20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Men 20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Age Younger (40 2 ) 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Older (60+) 26. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Socio-economic status Low education 19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

High education 21. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Health status No current illness 20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Copd or back complaints 20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Self-assessed health(during interview)

Gooda 26

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Less-than-good 14

a: Includes category very good (n 1).

Qualitative study on self-assessed health 201

functional abilities or limitations we considered to be an aspectof the functional dimension. (3) We considered any reference toa positive attitude towards a current illness or having adapted toits limitations to be an aspect of the coping dimension. (4) Anyreference to feeling fit or energetic (not feeling tired) or tofeelings without any further justification (simply feeling good),were considered to be an aspect of the wellbeing dimension.Also, references to the (im)balance between physical and mentalhealth were included in the wellbeing dimension. (5) Thebehavioural dimension refers to any theme referring to (health)behaviour.

Within each of the overall health dimensions, we have tried tomaintain the subtle nuances observed in the interviews bydistinghuishing different health aspects. For example, within thephysical dimension, we included two related health aspects:bodily mechanics and robustness (Appendix 2). Both aspectsrefer to some kind of susceptibility, in which the former refers tothe more intrinsic failing bodily mechanics, the latter refers tothe more extrinsic robustness i.e. resistance to external agents.In cases where nuances were quite subtle (i.e. statements inwhich respondents seemed to refer to two or more differenthealth aspects) the context of the whole interview was used toguide decisions on where to classify a statement. However, in thesubgroup analyses only the classification in health dimensionswas used.

The number of dimensions participants referred to rangedfrom one to three health dimensions. Almost half of theparticipants (47%) mentioned aspects from only one dimen-sion, half of the participants (50%) mentioned aspects from twodimension, and one participant mentioned aspects from three

health dimensions. In total, 40 participants made 62 referencesto health dimension, thus on average participants mentioned1.55 health dimensions.

Differences with respect to background characteristics

and health status

Some differences between participants with different back-ground characteristics (gender and age) can be observed(table 3). With regard to gender it can be seen that men dorefer to the functional dimension more often than women, 40versus 15% (not statistically significant), though no differ-ences could be observed with respect to the functional aspectsthey mention. No gender differences can be observed in thefrequency of physical aspects, aspects of wellbeing and aspectsof coping.

However, in our study group clear age differences can beobserved. Participants in the 60+ age group refer to thephysical dimension (92%, p , 0.01) and functional dimen-sion (35%, not statistically significant) almost twice as muchas 40- participants. Older participants with a chronic illnessor a history of illness mention aspects such as prognosis ofillness or illness-related functional disability more frequentlythan do younger participants. Aspects such as age-relatedcomplaints or age-related functional abilities are onlymentioned by older participants, as these aspects do notapply to the young. Another significantly age-related dimen-sion is wellbeing ( p , 0.001); feeling fit, feeling good, andbody/mind equilibrium are aspects mentioned almostexclusively by the young. Half of the younger participantsmentions aspects of wellbeing, whereas only one elderlyparticipant mentions that he based his health assessment onfeeling fit. With respect to the coping dimension, the agedifference is less marked and not statistically significant. Still,almost onethird of the older participants versus onefifth ofthe younger mentioned aspects of coping.

Participants with and without a current illness differ notablyon two dimensions. First, more than half (55%) of theparticipants with a chronic illness mentions coping with achronic illness, and, logically, none of the participants with nocurrent illness mentioned it ( p , 0.001). Second, wellbeing isconsidered more frequently, although not significantly, byparticipants with no current illness (30%) compared tochonically ill participants (10%). The functional dimension ismentioned almost equally frequent by participants with nocurrent illness (30%) and chronically ill participants (25%).Although only the former mention functional aspects with apositive connotation: being able to do almost anything, whetheror not in relation to (a relatively high) age. Participants with andparticipants without a chronic illness refer to disability orimpaired mobility due to a chronic illness or a history of disease.Nevertheless, for participants with and without a current illnessself-assessed health is predominantly associated with thephysical dimension. In both groups almost 80% of theparticipants refer to the presence or absence of physicalproblems.

The final column in table 3 shows that men, elderly, andchronically ill participants refer to more health dimensionsthan women, younger participants, and those with no currentillness.

Differences between participants with good and less-

than-good health assessment

We will refer to participants describing their health as either verygood or good as being in good health, and to participantsdescribing their health as either fair, sometimes good andsometimes poor, or poor as being in less-than-good health. Table3 shows that the functional dimension is far more important for aless-than-good health assessment (43%) than for a good health

Table 2 Frequency of health dimensions and health aspects

Health dimensionsHealth aspects (n)

N (% of total)

Physical 31 (78%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Chronic illness (15). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Physical problems (11). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Medical treatment (6). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Age-related (normal) complaints (6). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Prognosis of illness (4). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Bodily mechanics (1). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Robustness (1). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Functional 11 (28%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Not being impaired (4). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Illness-related disability (5). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Age-related functional abilities (3). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Coping 11 (28%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .To adapt to illness (5). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .A positive attitude (4). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Social comparison (2). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Wellbeing 8 (20%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Feeling fit (5). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Feeling good (2). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Body/mind equilibrium (1). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Behaviour 1 (3%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Eating healthy food (1)

202 European Journal of Public Health Vol. 15, No. 2, 200208

assessment (19%). The gradient from good to poor self-assessedhealth is very clear, although not statistically significant. Whenfunctional aspects are mentioned by participants in good health,it is always with a positive undertone. In addition to positivefunctional aspects participants in less-than-good health refer todisability and impaired mobility due to disease or illness. Thecoping dimension is mentioned more frequently, though notsignificantly, by participants in less-than-good health (36%) thanparticipants in good health (23%). Remarkably, only participantsin less-than-good health compare their own health with that ofother people who are worse off. In contrast, aspects ofwellbeingsuch as feeling fit or feeling goodare mentionedpredominantly by participants in good health (27%), only oneparticipant with less-than-good health mentions an aspect ofwellbeing. Clearly, for good as well as for less-than-good self-assessed health the physical dimension is very important. Still,slightly more participants in less-than-good health (86%) thanparticipants in good health (73%) explain their health assessmentin physical terms (not statistically significant). Participants ingood health mention the absence of physical problems, onlyexperiencing minor illnesses or age-related symptoms, and agood prognosis. Being in less-than-good health is also associatedwith the absence of physical problems or only experiencing age-related symptoms. However, participants in less-than-goodhealth also refer to the presence of physical problems. Of thosein less-than-good health, particularly participants in poor healthmention the severity of their chronic illness and a poor prognosis:their illness has deteriorated.

The final column of table 3 shows that participants with a lessfavourable health assessment refer to more health dimensionsthan participants with the most favourable health assessment. Aclear gradient can be observed, from an average of 1.4 healthdimensions for participants with (very) good selfassessed healthup to an average of 2.0 health dimensions for participants withpoor self-assessed health.

Discussion

Summary of the findings

The physical dimension of health has, traditionally, been viewedas being the core of self-assessed health, and in our study toothis dimension proved to be a central factor in health self-assessments. Almost 80% of the participants referred to one ormore physical aspects. Nevertheless, when assessing their healthparticipants also include aspects that go beyond the physicaldimension of health: 80% of the participantswhether or notin addition to physical aspectsreferred to one or more of theother health dimensions. Besides physical aspects participantsconsidered the extent to which they are able to perform(functional dimension), the extent to which they adapted to, ortheir attitude towards an existing illness (coping dimension),and simply the way they feel (wellbeing dimension). Healthbehaviours proved to be relatively unimportant in health self-assessments. All in all, we may well conclude that self-assessedhealth is not just a physical but a multidimensional concept.

Methodological issues

When interpreting the results of the present study, somemethodological issues should be kept in mind. First, since mostqualitative studies apply an inductive procedure to analyse theinterviews, our study differs from the other studies on self-assessed health both with respect to the terminology used andthe final categorization of these health aspects. Although not allstudies describe the contents of the final categories/dimensionsin detail, at first glance it seems as if applying our finalcategorization scheme to the data in other studies would yielddifferent results (table 4). For instance, Krause and Jay (5)categorized references to general energy level as healthproblems which in our study would have been categorized aswellbeing. Different researchers thus apply a different termi-nology, but table 4 also shows that, in general, qualitative studies

Table 3 Frequency of different health dimensions, by gender, age, health status and health assessment

Subgroup Health dimensions Mean no of

Category (n) Physical Functional Coping Wellbeing Behaviour dimensions

N (%) N (%) N (%) N (%) N (%)

Gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Women (20) 15 (75) 3 (15) 6 (30) 4 (20) 1 (5) 1.5 t-testn.s.a

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Men (20) 16 (80) 8 (40) 5 (25) 4 (20) 0 (0) 1.7

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

40 2 (14) 7 (50) 2 (14) 3 (21) 7 (50) 0 (0) 1.4 t-test. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

60+ (26) 24 (92) 9 (35) 8 (31) 1 (4) 1 (4) 1.7 p ,0.10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Health status. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

No current illness (20) 15 (75) 6 (30) 0 (0) 6 (30) 1 (5) 1.4 t-test. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Chronically ill (20) 16 (80) 5 (25) 11 (55) 2 (10) 0 (0) 1.7 p ,0.10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Health assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Goodb (26) 19 (73) 5 (19) 6 (23) 7 (27) 0 (0) 1.4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Fair (6) 5 (83) 2 (33) 1 (17) 1 (17) 1 (17) 1.7 Anova. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Sometimes poorc (5) 4 (80) 2 (40) 3 (60) 0 (0) 0 (0) 1.8 p ,0.05. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Poor (3) 3 (100) 2 (67) 1 (33) 0 (0) 0 (0) 2.0

a: n.s. Not significantb: Includes category Very good (n 1)c: In full: Sometimes good and sometimes poor


on self-assessed health are quite similar with respect to thehealth aspects that have been drawn from the interviews.

Second, some studies only included those aspects in theanalysis which participants mentioned first (single-referencestudies), other studies included all aspects which participantsmentioned (multiple-reference studies). Due to both themultiple-reference / single-reference disparity and the differen-tial categorization of health aspects over these dimensions it isquite difficult to compare studies with respect to the averagenumber of health dimensions referred to by participants (e.g. anaverage of 1.55 dimensions in our study, 1.39 dimensions in asinglereferences study by Krause and Jay5 and 1.19 dimensionsin a multiple-reference study by Borawski-Clark.6

Third, even in our small-scale study we were able to identifysome statistically significant subgroup differences. In this smallstudy population, it required a difference of over 30% points to

become statistically significant. This does not imply that theremaining non-significant subgroup differences of 20 to 25% weidentified should be discarded as irrelevant, as these may verywell be real differences. When these findings were to be repeatedin a larger study population, these subgroup differences wouldbe statistically significant. Therefore, we included these smallerand non-significant subgroup differences in our interpretationof the findings regarding subgroup differences.

Categorization scheme

As noted earlier the physical health dimension was verydominant. On the other hand, virtually no reference wasmade to mental health. Only one responder did mentionmental health, stating that for her own health assessment abalance in physical and mental health was important.

Table 4 Overview of the main dimensions (in italics) of five qualitative studies on selfassessed health

This paper Krause and Jay (5) Borawski-Clark et al. (6) Manderbacka (7) Idler et al. (9)

Physical Health problems Physical health Absence of ill-health Physical health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Presence or absenceof health problemsand illnesses

Medical/healthconditions

Presence or absenceof disease

Medical conditions,symptoms, prognosis

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Physical functioning Physical symptoms Health as an experience Psychological, emotional health

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..General physicalcondition

Experienced symptoms,illnesses

Age-related complaints

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Other

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Reproductive, sensoryfunctions

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Functional Physical functioning Physical health Health as a function Physical functioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Physical functioning,mobility

Functional capacities Functional restrictions Daily activities

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Social role activities

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Social responsibilities

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Coping Health comparisons Health transcendence Health as an action Psychological, emotional health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Comparing to otherpeople

Able to transcendhealth problems

Strength, coping Attitude

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Attitudinal, behavioural Social relationships

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Psychological Social comparison

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Wellbeing Mental health Non-reflective Health as an experience Physical health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Psychological wellbeing Feeling good Feeling good Energy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Physical functioning Health as an action Psychological, emotional health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

Energy level Fitness, vitality,equilibrium

Positive emotions, happiness

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Behaviour Health behaviour Attitudinal,

behaviouralHealth as an action Health risk behaviours

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Positive/ negativebehaviour

Lifestyle Lifestyle, healthbehaviour

Health behaviour

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Undefined Externally focused Social relationships. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

External validation,social support,external causes

Family relations

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Psychological, emotional health

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Luck, faith

Note Van Doorns paper8 provided little information on the exact contents of the dimensions that were distinguished; thisstudy is not included in the overview.


Although several psychological mechanisms were mentioned(e.g. categorized within the the coping dimension) or positivehealth was stressed (e.g. feeling good in the wellbeingdimension), none of the respondents made reference tomental health as such, nor to specific mental diseases orcomplaints, such as depression or anxiety. This is all the moreremarkable since mental health problems are among the mostprevalent diseases in the Netherlands and among the diseaseswith the largest consequences for quality of life.14 However,this finding is in agreement with the findings from otherqualitative studies.5 7,9 Probably, mental health is notconsciously taken into account by responders when assessingtheir health.

Differences with respect to background characteristics

and health status

We found that men refer to functional aspects morefrequently than women, although this result is only marginallysignificant. In Western societies men are normally thebreadwinner and thus responsible for the main source ofincome. This may be the reason that men, more than women,have incorporated the functional definition of health as beingable to perform the necessary duties.15,16 We also observedclear and some significant age-differences in our studygroup.Participants in the 60+ age group referred to physical andfunctional aspects almost twice as much as youngerparticipants. In contrast, half of the younger participantsmentioned aspects of wellbeing, whereas this aspect ismentioned only incidentally by elderly participants. Althoughthe distribution of participants with and without a currentillness is equal in both age groups, elderly participants morefrequently mention a history of illness. Elderly participantsprobably incorporate these prior episodes of (physical orfunctional) ill-health in their health assessments. Furthermore,we found some differences between participants with andwithout a current illness. Aspects of coping are typicallymentioned by participants with a chronic illness. On the otherhand, aspects of wellbeing are typical aspects of participantswith no current illness. Some (predominantly younger)participants are relatively inexperienced with (coping with)physical, functional or age-related health problems. Conse-quently, these participants do not incorporate these healthdimensions in their health assessments, but simply rely on theway they feel. Other (predominantly elderly) participants aremore experienced with episodes of ill-health. Yet, for theseparticipants it is not so much the presence of (physical,functional or agerelated) health problems but the extent towhich they are capable of coping with these problems whichdetermines their eventual health assessment. The importanceof experience with health problems and the ability to copewith them is also reflected in the finding that elderly andchronically ill participants include more health dimensions intheir health assessments than do younger participants andthose with no current illness.

Differences between participants with good and less-

than-good health assessment

There are some differences between participants in good andparticipants in less-than-good health, though not statisticallysignificant. Again, these differences may be the result of someparticipants having experienced less health problems thanothers. For participants in good health two lines of reasoningcan be distinguished. Participants with no current illness orother health problems reason: I am not bothered by anyphysical or functional health problem, I am feeling good,participants with (a history of) chronic illness or other (e.g. age-related) health complaints reason: I am not bothered byphysical or functional complaints, I cope with them. Partici-

pants in less-than-good health seem to experience more physicaland functional health problems than participants in betterhealthas reflected in the larger number of health dimensionsthey refer towhich they also present as being more severe. Onthe basis of these interviews we cannot determine whether theparticipants in less-than-good health truly suffer from moresevere problems than participants in better health, or that forsome reason these participants are less capable of coping withhealth problems.

The role of coping

Besides prior or current experience with physical or functionalhealth problems, coping with these problems seems to beimportant for ones health assessment. It is inherent to ourcoding process that only explicit statements referring toadapting to illness, attitude towards illness, or comparisonwere considered to be referring to aspects of coping. However, ifwe look more close at the data, we find that there are other,more implicit, references which could be considered as a way ofcoping with health complaints, i.e. referrence to age-relatedcomplaints or functioning. It seems as if participants whoconsider age-related physical complaints or functional declineto be normal, are less bothered by them. And although not allparticipants mentioning age-related (normal) complaints orage-related functioning assessed their health as good, it may bejust the reason why they did not assess their health morepoorly.17

Conclusions

We have shown that that self-assessed health is a multi-dimensional concept. Over the years several qualitative studieson self-assessed health have produced comparable results, eventhough these studies differed with respect to the subgroups theyincluded and the methodology they applied. The consistency ofthe findings suggests that we have actually taken a step nearer toidentifying which particular aspects are involved in healthassessments.

Acknowledgements

The authors would like to thank Dr Ir E. J. de Min for providingthe software for calculating kappa coefficients, and Dr H. van deMheen for participation during the early stages of the researchproject. We would also like to thank Ms K. Gribling for hercareful translation of the excerpts from the interviews. TheGLOBE-study is supported by the Dutch Ministry of PublicHealth, Welfare and Sports, and the Netherlands HealthResearch and Development Council (ZON).

Key points

In this qualitative study we studied which particularaspects are included in self-assessed health.

Self-assessed health proved to be a multidimensionalconcept, including primarily physical, functional,coping and wellbeing aspects.

Health behaviour or lifestyle factors proved to berelatively unimportant in health self-assessments.

Subgroup differences in self-assessed health couldprimarily be attributed to prior experience with ill-health.

The consistancy of these findings with other qualitativestudies suggest that we have identified the keydimensions of self-assessed health.


References

1 Idler EL. Perceptions of pain and perceptions of health. Motivation and

Emotion 1993;17:20524.

2 Murray J, Dunn G, Tarnopolsky A. Self-assessment of health: an exploration of

the effects of physical and psychological symptoms. Psychol Med 1982;12:

3718.

3 Tissue T. Another look at self-rated health among the elderly. J Gerontol 1972;

27:914.

4 Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-

seven community studies. J Health Soc Behav 1997;38:2137.

5 Krause NM, Jay GM. What do global self-rated health items measure? Med

Care 1994;32:93042.

6 Borawski-Clark EA, Kinney JM, Kahana E. The meaning of older adults

health appraisals: congruence with health status and as determinants of

mortality. J Gerontol 1996;51b:S15770.

7 Manderbacka K. Examining what self-rated health is understood to mean by

respondents. Scand J Soc Med 1998;25:14553.

8 Van Doorn C. A qualitative approach to studying health optimism, realism,

and pessimism. Res Aging 1999;21:44057.

9 Idler EL, Hudson SV, Leventhal H. The meanings of self-ratings of health. A

qualitative and quantitative approach. Res Aging 1999;21:45876.

10 Blaxter M. Health & Lifestyles. London: Routledge, 1990.

11 Mackenbach JP, Van de Mheen H, Stronks K. A prospective cohort study

investigating the explanation of socio-economic inequalities in health in the

Netherlands. Soc Sci Med 1994;38:299308.

12 QSR. NUD * IST. In. N4 ed. Melbourne: Qualitative Solutions and Research,

1997.

13 Fleiss J. Statistical methods for rates and proportions. New York: John Wiley &

Sons, 1973.

14 Van Oers JAM, Editor. Health on Course? The 2002 Dutch Public Health Status

and Forecasts Report. Houten: Bohn Stafleu Van Loghum, 2002.

15 Twaddle AC. The concept of health status. Soc Sci Med 1974;8:2938.

16 Seedhouse D. Health: the foundations for achievement. Chichester: John Wiley

& Sons, 1986.

17 Suls J, Marco CA, Tobin S. The role of temporal comparison, social

comparison, and direct appraisal in the elderlys self-evaluations of health.

J Appl Soc Psychol 1991;21:112544.

Appendix 1

Flow chart describing the phases in the qualitative

analysis of the interviews


Appendix 2

Description of the health aspects are given illustrated

with quotations

Health dimensionsHealth aspectsPhysical: This dimension refers to the general working of

ones bodyChronic illness: presence of a chronic illness or a history of

chronic illnessWell, I guess you could say that my health is reasonably okay,

only theres no getting away from the fact that Im, uh, thirty, fortypercent asthmatic. Thats what Ive got, so to speak. Man, 60+,high ses, copd/asthma, fair

Physical problems: reference to physical complaints, notdirectly related to any chronic illness, such as never being ill,never needing to stay at home due to illness, or onlyexperiencing minor illnesses

Uh, no problems, no headaches, no stomach aches, no menstrualpains like I used to get. Woman, 60+, high ses, no current illness,good

Medical treatment: (not) being under medical treatment, or(not) being prescribed medication

I never see the doctor, so, uh, sure, Im in good shape () Imean, well, if you dont need to see the doctor a lot, and you donthave a whole lot of complaints () Healthy? Yes, all of us, werehealthy. At least, my husband never has to visit the doctorknockon woodup to now, so, well. () Never been in hospital foranything, well, only to have a baby, and thats rather a healthyreason, wouldnt you say. Woman, 40-, low ses, no current illness,good

Age-related (normal) complaints: reference to physicalcomplaints which are considered to be expected, i.e. normal,considering ones age.

Id say Im fine. Yes. Of course theres always some little thinggoing wrong here and there, but all pretty much to be expected.My arm was giving me problems and the doctor gave me a fewshots, I mean, well, it was painful, and after eighty years its nota surprise my joints werent working as smoothly as when I wastwenty. But actually Im doing fine. Man, 60+, low ses,copd/asthma, fair

Prognosis of illness: reference to the course, or prognosis of achronic illness

Well I dont know whether you read the previous questionnaires?Oh, well two years ago I was operated on for breast cancer, so withthat in mind, Im doing very well (..) Like I said, I may have hadan operation but it was localised and Im fine now. No othercomplaints. Woman, 60+, high ses, no current illness, good

Bodily mechanics: reference to failing mechanicsof thebody, as a result of which one may suffer from recurring (minor)physical complaints

The only thing, which is why I was wavering between very goodand good, uh, mechanically Im not in great condition. Right now,for example, Ive got a stiff neck, but Ive always got a backache.And, uh, thats because well, its just not strong. Man, 60+, highses, chronic back complaints, very good

Robustness: reference to being illness-prone vs. being morerobust to illness

I guess it all has to do with constitution, how strong your body is,you know. What I notice in my case is that thats not all thatstrong, that for the rest I feel perfectly healthy, but Im very quick tonotice when Ive been overdoing it. Like when Ive had too much todrink. Or forget stuff. That S. I was talking about just now, well,hes a good example. He can eat, say, halfdone chicken legs. If I atesomething like that Id notice right away. My stomach starts actingup or something and he has no problems at all. Man, 40 2 , lowses, no current illness, good

Functional: This dimension refers to ones ability to performcertain activities

Not being impaired: reference to general functional abilitiesor limitations. Referring to being able to do the things one wantsto, or needs to do, without any reference to disease, illness orbodily mechanics

Well, because you can do everything, you do everything. Butvery good, no, I mean there are also all the days that things dontgo very well, so I guess good is a happy medium. Woman, 40 2 ,low ses, no current illness, good

Illness-related disability: reference to some kind of disabilityor impaired mobility, due to chronic illness or disease

It started with my lungs () Yes, (my health) its poor. Imean, if I could get more air. I mean, right now, and then Imreferring primarily to the past few months, after Ive walked for,say, 200 metres, I have to stop and, and catch my breath. Takejust yesterday. I wanted to go get a haircut, thats 10 minutesaway by bike. Halfway there I had to turn around and gohome. So Im hoping that specialist is right and that if I useoxygen when I exert myself, itll help. Man, 60+, low ses,copd/asthma, poor

Age-related functional abilities: relating general functionalabilities or limitations to ageing, being able to function well forones age

Uhh, if a persons healthy, uhhh, he can do anything hessupposed to be able to do at his age. I mean, look, if youre oversixty, Im sixty-three, obviously you cant be doing all the crazythings you did when you were twenty or thirty. Man, 60+, high ses,no current illness, good

Coping: This dimension refers to the psychological mechan-ism of dealing with an illness

To adapt to illness: reference to having adapted to thechronic illness or having learned to live with its limitations

Because health is extremely complicated. I mean, purely on thephysical level, you could check whether every bit of the body is ingood working order. And in my case youd find that there are agreat many bits in my body that dont work well. But if you look atthe complex, the aggregate and the combination etc. etc., how Ifunction the way I am, well, the answer is good, I would say. ()So to my mind it has a lot to do with uhh, on the one handadapting and on the other taking steps to be able to do what youwant to do, only in a different way. Woman, 60+, high ses,chronic back complaints, sometimes good and sometimespoor

A positive attitude: reference to maintaining a positiveattitude towards the illness

Its however you look at it, I say. I mean, its not going at all wellto be frank, but I try to take the cheerful view. So, chin up, is what Ialways say () Well, theyre not actually very healthy. No, wellthey always look on the bright side, you see. Yes. And, I mean takesomeone who has a bug or something else, whatever, that can makeyou feel really ill, that person knows thisll be over in a couple ofdays, a few weeks, and that holds for a lot of things. And thatswhat I mean by always looking on the bright side. And even whenits like there isnt one, you still always have to find that tiny spark.Woman, 40 2 , low ses, chronic back complaints, good

Social comparison: comparing ones own health with thehealth of peers; comparing ones own health with the health ofpeople who are worse off (downward comparison)

But there are always worse things, arent there, and thats someconsolation. I was just in hospital and I saw a person come out whowas bent over nearly double, what a hump! His nose close toscraping the ground, I mean imagine going through life like that?That would really be awful. Man, 60+, low ses, copd/asthma,poor

Wellbeing: This dimension refers to general feelings and toaspects of mental health

Feeling fit: reference to feelings of fitness or energy, referringto feeling fit, energetic and not tired


Yes, I feel good, Im never tired and uh especially during the pastfew years, sure. () Yes, physically healthy? I guess, if youre nottired () I feel fit, not tired, so I feel healthy. Woman, 40 2 , lowses, no current illness, good

Feeling good: reference to general feelings without anyobjective justification, simply referring to feeling good

Yes, I feel good, I feel absolutely great. For me, health is feelinggood. And I do. Thats how simple it is. () Oh, thats, I guess, notfeeling bad. Man, 40 2 , high ses, chronic back complaints,good

Body/mind equilibrium: reference to the (im)balance ofphysical and mental problems

If youre ill and out of sorts, you can forget it, you just feelrotten. If you have a psychological problem you feel just asrotten even though physically, theres nothing wrong. But yourenot completely healthy if youve got a problem with either. ()Healthy is when you have no infections of any kind. I guessthats part of it. And that theres no blackness messing up yourmind.() I mean, you dont have flu, mentally youre okay.And its like everythings good, Im doing fine. Woman, 40 2 ,high ses, chronic back complaints, fair

BEHAVIOUREating healthy food: mentioning eating well (all from our

own garden) and not eating sweets


How is your health in general? A qualitative study on self-assessed healthData and methodsStudy populationNon-response and changes in health assessmentsSemi-structured interviewInterview analysis

ResultsWhich health aspects are taken into consideration?Differences with respect to background characteristics and health statusDifferences between participants with good and less-than-good health assessment

DiscussionSummary of the findingsMethodological issuesCategorization schemeDifferences with respect to background characteristics and health statusDifferences between participants with good and less-than-good health assessmentThe role of coping

ConclusionsAcknowledgementsReferences&?th;Appendix 1&?th;&?th=9pt;Flow chart describing the phases in the qualitative analysis of the interviews

&?th;Appendix 2&?th;&?th=9pt;Description of the health aspects are given illustrated with quotations

Documents

public health