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

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

Perceived
Health
...........................................................................................
How is your health in general? A qualitative
study 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 presents
researchers with a summary of an individuals general state of health. A qualitative study was initiated to
find out which particular aspects are included in health self-assessments; which aspects do people
consider when answering the question How is your health in general?. Subgroup differences were
studied with respect to gender, age, health status and health assessment. Methods: Qualitative study
with 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, physical
problems, medical treatment, age-related complaints, prognosis, bodily mechanics, and resilience).
However, when assessing their health, participants also include aspects that go beyond the physical
dimension of health. In total, 80 percent of the participantswhether or not in addition to physical
aspectsreferred to other health dimensions. Besides physical aspects, participants considered the
extent to which they are able to perform (functional dimension 228%), the extent to which they
adapted to, or their attitude towards an existing illness (coping dimension 228%), and simply the way
they feel (wellbeing dimension 220%). In this study, health behaviour or lifestyle factors (behavioural
dimension 23%) proved to be relatively unimportant in health selfassessments. Conclusions Selfassessed health proved to be a multidimensional concept. For most part, subgroup differences in selfassessed health could be attributed to experience with ill health: being relatively inexperienced with
health problems versus having a history of health problems.
Keywords: qualitative study, selfassessed health, stratified sample, subgroup differences

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

he single question How is your health in general? is a crude

and simple measure that has been widely used, as it presents
researchers with a summary of an individuals general state of
health. It is presumed that in self-assessed health numerous
aspects of health are combined within the perceptual framework
of the individual respondent.1 3 This measure proved to be a
powerful predictor for mortality; poor self-assessed health
increases 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 which
particular aspects are included in health self-assessments. In
quantitative studies the relationship between a priori defined
health measures and self-assessed health has been analysed. In
these studies, however, a significant proportion of variance in
self-assessed health remains unexplained. This suggests that
when assessing their health, participants may include health
aspects that have not been routinely included in quantitative
analyses. Therefore, some researchers have used a qualitative
approach to identify the remaining and unknown aspects of
self-assessed health. Briefly summarizing, selfassessed health
seems mainly to be associated with physical health problems,
functional capacities, health behaviour, and psychological

.......................................................
1 Department of Public Health, Erasmus University Rotterdam, The
Netherlands
2 Department of Medical Psychology and Psychotherapy, Erasmus
University Rotterdam, The Netherlands
Correspondence: I.M.A. Joung, Department of Public Health, Erasmus
MC, University Medical Center Rotterdam, P.O. Box 1738, 3000 DR
Rotterdam, The Netherlands, tel. +31 10 4087714, fax +31 10 4089449,
e-mail: i.joung@erasmusmc.nl

aspects.5 9 Additionally, some studies found that aspects such

as 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 these
qualitative studies attempted to include equal numbers of
participants of different sociodemographic backgrounds.5,7 The
other studies included convenience samples predominantly
consisting of women, elderly, highly educated participants6,8,9
or participants with health problems.8 However, health
standards may vary among different subgroups, and probably
depend very much upon gender,6 age5,10 and experience with
health problems.10 Therefore, it is difficult to decide whether the
findings in these studies reflect general health conceptions, or
are determined by the most prevalent subgroup. It would be
relevant to know whether participants from different subgroups
consider entirely different aspects when assessing their health,
but with the exception of Krause and Jays study,5 qualitative
studies on self-assessed health rarely examined subgroup
differences.
We initiated a qualitative study on self-assessed health in a
sample that has been stratified on background characteristics,
health status, and health assessment. The present paper focuses
on the aspects that people consider when answering the
question How is your health in general? Is it very good, good,
fair, sometimes good and sometimes poor, or poor?. We believe
that health assessments follow an individual process of ordering
and weighing different health aspects. Therefore, we asked
participants what went through their minds when answering the
question on self-assessed health. The analysis was guided by the
following research questions: Which aspects do participants
consider when answering the question on self-assessed health?
Do participants with different background characteristics (age

Qualitative study on self-assessed health

and gender), and participants with different health status (with

and without current chronic conditions) consider the same or
different aspects when assessing their health? Do participants
with good and less-than-good self-assessed health consider the
same or different aspects when assessing their health?

201

Table 1 Distribution of stratification variables in study

population
Stratification variables

Categories

Gender

Women

20

Men

20

Younger (40 2 )

14

Older (60+)

26

Low education

19

High education

21

No current illness

20

Copd or back complaints

20

Gooda

26

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

Data and methods

Study population
Our study population consists of participants of the GLOBE
study, a longitudinal study designed to describe and explain
sociodemographic inequalities in health in the Netherlands.
Design and objective of the GLOBEstudy have been described in
detail elsewhere.11 At baseline in 1991, participants comprised a
cohort of non-institutionalized men and women with Dutch
nationality, 15 74 years of age, who were living in the city of
Eindhoven or surrounding municipalities. In 1997, a subgroup
of respondents to the baseline interview were approached to
participate in a follow-up study. For our qualitative study, we
drew a stratified sample from the respondents to the 1997
follow-up. The interviews took place in 1998.
The variables for stratification have been chosen because of
their supposed relationship with self-assessed health: gender,
age, socioeconomic status, and health status. In order to obtain
maximum contrast, we included men and women, younger than
40 years of age and older than 60 years of age, with the highest
level of education (university degree) and with the lowest level
of education (primary or lower vocational education), with a
chronic illness (COPD/asthma or chronic back complaints) and
without a current illness. Furthermore, we stratified on the most
recent available (i.e. 1997) health assessment and thus included
participants with (very) good, as well as participants with lessthan-good self-assessed health (stratification table is available
on 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 to
various reasons. First, some strata did not exist in the
population from which we drew our study sample. Second,
the number of possible participants that fitted a particular
profile (i.e. stratum) could be very low. When these participants
all refused to participate in our study, there were no other
eligible participants we could approach. Third, some participants changed their health assessment during the 1998 semistructured interview compared to the followup data (1997) on
which we based our initial selection of respondents. All in all, we
were able to select participants for 74% of the existing strata.
From May till December 1998, we approached 63 people by
mail and telephone. Fourteen persons were unwilling to
participate in the study, we were unable to get into contact
with six persons, and three persons were unavailable during the
study period, although willing to participate. Thus, we
interviewed 40 participants, a response of 63%. The distribution
of the different stratification variables can be seen in table 1.
Semi-structured interview
All participants were interviewed in their homes by the principal
investigator (JS). The semi-structured interviews, lasting
approximately 35 minutes, were tape recorded and transcribed
verbatim. Following a brief introduction the interviewees were
presented 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 particular
response.

.................................................. .
.................................................. .
Socio-economic status

.................................................. .
.................................................. .
Health status

.................................................. .
.................................................. .
Self-assessed health
(during interview)

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

14

a: Includes category very good (n 1).

Interview analysis
We started with analysing the verbatim text of the interviews. In
each interview, we condensed the answers given to the singleitem measure on self-assessed health and the reasons for this
health assessment. Parts of the text representing the same theme
were summarised with a single phrase, hereby paraphrasing the
participant. In this way, each interview could be condensed into
personal themes. Next, we categorized the personal themes of all
participants into a smaller number of recurrent themes, which
we will refer to as health aspects. Finally, on categorization of
these health aspects, five conceptually meaningful health
dimensions emerged (see Appendix 1 for a flow chart of the
coding process). For development of the overall categorization
scheme, and for the data analysis that followed, QSR NUD*IST
software,12 were used.
To ensure reliability in coding and analysing the interviews
four researchers (JS, JB, IJ and HB) independently read and
coded eight of the interviews. The results were compared and
discussed to come to a reliable method for analysing the
interviews. Next, the principal investigator (JS) read and coded
all interviews, and designed the final categorization scheme.
Finally, one of the other researchers (IJ) independently applied
the categorization scheme (on the level of health dimensions) to
eight of the interviews. We then calculated Cohens Kappa, a
measure of interrater reliability, and the level of agreement was
shown to be good (k 0.69).13
This paper presents the overall frequency distribution of the
different dimensions and health aspects, as well as the
distribution of health dimensions by gender, age, health status,
and health assessment. Chi-square analyses are used to examine
whether referring to a particular dimension varies significantly
for 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 with
which the different health dimensions and health aspects were
mentioned are shown in table 2. In Appendix 2 the description
of the health dimensions and health aspects are given and
illustrated with quotations. (1) We considered physical
references, i.e. any reference to disease, illness, medical
treatments, or other bodily-oriented theme to be an aspect
of the physical dimension. (2) Any reference to general

202

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

Table 2 Frequency of health dimensions and health aspects

Health dimensions
Health 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)

functional abilities or limitations we considered to be an aspect

of the functional dimension. (3) We considered any reference to
a positive attitude towards a current illness or having adapted to
its limitations to be an aspect of the coping dimension. (4) Any
reference to feeling fit or energetic (not feeling tired) or to
feelings 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 mental
health were included in the wellbeing dimension. (5) The
behavioural dimension refers to any theme referring to (health)
behaviour.
Within each of the overall health dimensions, we have tried to
maintain the subtle nuances observed in the interviews by
distinghuishing different health aspects. For example, within the
physical dimension, we included two related health aspects:
bodily mechanics and robustness (Appendix 2). Both aspects
refer to some kind of susceptibility, in which the former refers to
the more intrinsic failing bodily mechanics, the latter refers to
the more extrinsic robustness i.e. resistance to external agents.
In cases where nuances were quite subtle (i.e. statements in
which respondents seemed to refer to two or more different
health aspects) the context of the whole interview was used to
guide decisions on where to classify a statement. However, in the
subgroup analyses only the classification in health dimensions
was used.
The number of dimensions participants referred to ranged
from one to three health dimensions. Almost half of the
participants (47%) mentioned aspects from only one dimension, half of the participants (50%) mentioned aspects from two
dimension, and one participant mentioned aspects from three

health dimensions. In total, 40 participants made 62 references

to health dimension, thus on average participants mentioned
1.55 health dimensions.
Differences with respect to background characteristics
and health status
Some differences between participants with different background characteristics (gender and age) can be observed
(table 3). With regard to gender it can be seen that men do
refer to the functional dimension more often than women, 40
versus 15% (not statistically significant), though no differences could be observed with respect to the functional aspects
they mention. No gender differences can be observed in the
frequency of physical aspects, aspects of wellbeing and aspects
of coping.
However, in our study group clear age differences can be
observed. Participants in the 60+ age group refer to the
physical dimension (92%, p , 0.01) and functional dimension (35%, not statistically significant) almost twice as much
as 40- participants. Older participants with a chronic illness
or a history of illness mention aspects such as prognosis of
illness or illness-related functional disability more frequently
than do younger participants. Aspects such as age-related
complaints or age-related functional abilities are only
mentioned by older participants, as these aspects do not
apply to the young. Another significantly age-related dimension is wellbeing ( p , 0.001); feeling fit, feeling good, and
body/mind equilibrium are aspects mentioned almost
exclusively by the young. Half of the younger participants
mentions aspects of wellbeing, whereas only one elderly
participant mentions that he based his health assessment on
feeling fit. With respect to the coping dimension, the age
difference is less marked and not statistically significant. Still,
almost onethird of the older participants versus onefifth of
the younger mentioned aspects of coping.
Participants with and without a current illness differ notably
on two dimensions. First, more than half (55%) of the
participants with a chronic illness mentions coping with a
chronic illness, and, logically, none of the participants with no
current illness mentioned it ( p , 0.001). Second, wellbeing is
considered more frequently, although not significantly, by
participants with no current illness (30%) compared to
chonically ill participants (10%). The functional dimension is
mentioned almost equally frequent by participants with no
current illness (30%) and chronically ill participants (25%).
Although only the former mention functional aspects with a
positive connotation: being able to do almost anything, whether
or not in relation to (a relatively high) age. Participants with and
participants without a chronic illness refer to disability or
impaired mobility due to a chronic illness or a history of disease.
Nevertheless, for participants with and without a current illness
self-assessed health is predominantly associated with the
physical dimension. In both groups almost 80% of the
participants refer to the presence or absence of physical
problems.
The final column in table 3 shows that men, elderly, and
chronically ill participants refer to more health dimensions
than women, younger participants, and those with no current
illness.
Differences between participants with good and lessthan-good health assessment
We will refer to participants describing their health as either very
good or good as being in good health, and to participants
describing their health as either fair, sometimes good and
sometimes poor, or poor as being in less-than-good health. Table
3 shows that the functional dimension is far more important for a
less-than-good health assessment (43%) than for a good health

Qualitative study on self-assessed health

203

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

Health dimensions

Category (n)

Physical

Functional

Coping

(%)

Mean no of

(%)

(%)

Wellbeing

Behaviour

(%)

dimensions

(%)

Gender

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Women (20)

15

(75)

(15)

(30)

(20)

(5)

1.5

t-test
n.s.a

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Men (20)

16

(80)

(40)

(25)

(20)

(0)

1.7

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

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
40 2 (14)

(50)

(14)

(21)

(50)

(0)

1.4

t-test

60+ (26)

24

(92)

(35)

(31)

(4)

(4)

1.7

p ,0.10

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Health status

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
No current illness (20)

15

(75)

(30)

(0)

(30)

(5)

1.4

t-test

Chronically ill (20)

16

(80)

(25)

11

(55)

(10)

(0)

1.7

p ,0.10

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Health assessment

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Goodb (26)

19

(73)

(19)

(23)

(27)

(0)

1.4

(83)

(33)

(17)

(17)

(17)

1.7

Anova

Sometimes poor (5)

(80)

(40)

(60)

(0)

(0)

1.8

p ,0.05

Poor (3)

(100)

(67)

(33)

(0)

(0)

2.0

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Fair (6)

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
c

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

assessment (19%). The gradient from good to poor self-assessed

health is very clear, although not statistically significant. When
functional aspects are mentioned by participants in good health,
it is always with a positive undertone. In addition to positive
functional aspects participants in less-than-good health refer to
disability and impaired mobility due to disease or illness. The
coping dimension is mentioned more frequently, though not
significantly, by participants in less-than-good health (36%) than
participants in good health (23%). Remarkably, only participants
in less-than-good health compare their own health with that of
other people who are worse off. In contrast, aspects of
wellbeingsuch as feeling fit or feeling goodare mentioned
predominantly by participants in good health (27%), only one
participant with less-than-good health mentions an aspect of
wellbeing. Clearly, for good as well as for less-than-good selfassessed health the physical dimension is very important. Still,
slightly more participants in less-than-good health (86%) than
participants in good health (73%) explain their health assessment
in physical terms (not statistically significant). Participants in
good health mention the absence of physical problems, only
experiencing minor illnesses or age-related symptoms, and a
good prognosis. Being in less-than-good health is also associated
with the absence of physical problems or only experiencing agerelated symptoms. However, participants in less-than-good
health also refer to the presence of physical problems. Of those
in less-than-good health, particularly participants in poor health
mention 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 less
favourable health assessment refer to more health dimensions
than participants with the most favourable health assessment. A
clear gradient can be observed, from an average of 1.4 health
dimensions for participants with (very) good selfassessed health
up to an average of 2.0 health dimensions for participants with
poor self-assessed health.

Discussion
Summary of the findings
The physical dimension of health has, traditionally, been viewed
as being the core of self-assessed health, and in our study too
this dimension proved to be a central factor in health selfassessments. Almost 80% of the participants referred to one or
more physical aspects. Nevertheless, when assessing their health
participants also include aspects that go beyond the physical
dimension of health: 80% of the participantswhether or not
in addition to physical aspectsreferred to one or more of the
other health dimensions. Besides physical aspects participants
considered the extent to which they are able to perform
(functional dimension), the extent to which they adapted to, or
their attitude towards an existing illness (coping dimension),
and simply the way they feel (wellbeing dimension). Health
behaviours proved to be relatively unimportant in health selfassessments. All in all, we may well conclude that self-assessed
health is not just a physical but a multidimensional concept.
Methodological issues
When interpreting the results of the present study, some
methodological issues should be kept in mind. First, since most
qualitative studies apply an inductive procedure to analyse the
interviews, our study differs from the other studies on selfassessed health both with respect to the terminology used and
the final categorization of these health aspects. Although not all
studies describe the contents of the final categories/dimensions
in detail, at first glance it seems as if applying our final
categorization scheme to the data in other studies would yield
different results (table 4). For instance, Krause and Jay (5)
categorized references to general energy level as health
problems which in our study would have been categorized as
wellbeing. Different researchers thus apply a different terminology, but table 4 also shows that, in general, qualitative studies

204

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

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 absence
of health problems
and illnesses

Medical/health
conditions

Presence or absence
of disease

Medical conditions,
symptoms, prognosis

Physical functioning

Physical symptoms

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

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Health as an experience

Psychological, emotional health

Experienced symptoms,
illnesses

Age-related complaints

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
General physical
condition

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

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Reproductive, sensory
functions

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
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 other
people

Able to transcend
health 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

Attitudinal,
behavioural

Health as an action

Health risk behaviours

Lifestyle

Lifestyle, health
behaviour

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Behaviour

Health behaviour

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Positive/ negative
behaviour

Health behaviour

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Undefined

Externally focused

Social relationships

External validation,
social support,
external causes

Family relations

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

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

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Luck, faith
8

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

on self-assessed health are quite similar with respect to the

health aspects that have been drawn from the interviews.
Second, some studies only included those aspects in the
analysis which participants mentioned first (single-reference
studies), other studies included all aspects which participants
mentioned (multiple-reference studies). Due to both the
multiple-reference / single-reference disparity and the differential categorization of health aspects over these dimensions it is
quite difficult to compare studies with respect to the average
number of health dimensions referred to by participants (e.g. an
average of 1.55 dimensions in our study, 1.39 dimensions in a
singlereferences study by Krause and Jay5 and 1.19 dimensions
in a multiple-reference study by Borawski-Clark.6
Third, even in our small-scale study we were able to identify
some statistically significant subgroup differences. In this small
study population, it required a difference of over 30% points to

become statistically significant. This does not imply that the

remaining non-significant subgroup differences of 20 to 25% we
identified should be discarded as irrelevant, as these may very
well be real differences. When these findings were to be repeated
in a larger study population, these subgroup differences would
be statistically significant. Therefore, we included these smaller
and non-significant subgroup differences in our interpretation
of the findings regarding subgroup differences.
Categorization scheme
As noted earlier the physical health dimension was very
dominant. On the other hand, virtually no reference was
made to mental health. Only one responder did mention
mental health, stating that for her own health assessment a
balance in physical and mental health was important.

Qualitative study on self-assessed health

Although several psychological mechanisms were mentioned

(e.g. categorized within the the coping dimension) or positive
health was stressed (e.g. feeling good in the wellbeing
dimension), none of the respondents made reference to
mental health as such, nor to specific mental diseases or
complaints, such as depression or anxiety. This is all the more
remarkable since mental health problems are among the most
prevalent diseases in the Netherlands and among the diseases
with the largest consequences for quality of life.14 However,
this finding is in agreement with the findings from other
qualitative studies.5 7,9 Probably, mental health is not
consciously taken into account by responders when assessing
their health.
Differences with respect to background characteristics
and health status
We found that men refer to functional aspects more
frequently than women, although this result is only marginally
significant. In Western societies men are normally the
breadwinner and thus responsible for the main source of
income. This may be the reason that men, more than women,
have incorporated the functional definition of health as being
able to perform the necessary duties.15,16 We also observed
clear and some significant age-differences in our studygroup.
Participants in the 60+ age group referred to physical and
functional aspects almost twice as much as younger
participants. In contrast, half of the younger participants
mentioned aspects of wellbeing, whereas this aspect is
mentioned only incidentally by elderly participants. Although
the distribution of participants with and without a current
illness is equal in both age groups, elderly participants more
frequently mention a history of illness. Elderly participants
probably incorporate these prior episodes of (physical or
functional) ill-health in their health assessments. Furthermore,
we found some differences between participants with and
without a current illness. Aspects of coping are typically
mentioned by participants with a chronic illness. On the other
hand, aspects of wellbeing are typical aspects of participants
with no current illness. Some (predominantly younger)
participants are relatively inexperienced with (coping with)
physical, functional or age-related health problems. Consequently, these participants do not incorporate these health
dimensions in their health assessments, but simply rely on the
way they feel. Other (predominantly elderly) participants are
more experienced with episodes of ill-health. Yet, for these
participants it is not so much the presence of (physical,
functional or agerelated) health problems but the extent to
which they are capable of coping with these problems which
determines their eventual health assessment. The importance
of experience with health problems and the ability to cope
with them is also reflected in the finding that elderly and
chronically ill participants include more health dimensions in
their health assessments than do younger participants and
those with no current illness.
Differences between participants with good and lessthan-good health assessment
There are some differences between participants in good and
participants in less-than-good health, though not statistically
significant. Again, these differences may be the result of some
participants having experienced less health problems than
others. For participants in good health two lines of reasoning
can be distinguished. Participants with no current illness or
other health problems reason: I am not bothered by any
physical or functional health problem, I am feeling good,
participants with (a history of) chronic illness or other (e.g. agerelated) health complaints reason: I am not bothered by
physical or functional complaints, I cope with them. Partici-

205

pants in less-than-good health seem to experience more physical

and functional health problems than participants in better
healthas reflected in the larger number of health dimensions
they refer towhich they also present as being more severe. On
the basis of these interviews we cannot determine whether the
participants in less-than-good health truly suffer from more
severe problems than participants in better health, or that for
some reason these participants are less capable of coping with
health problems.
The role of coping
Besides prior or current experience with physical or functional
health problems, coping with these problems seems to be
important for ones health assessment. It is inherent to our
coding process that only explicit statements referring to
adapting to illness, attitude towards illness, or comparison
were considered to be referring to aspects of coping. However, if
we look more close at the data, we find that there are other,
more implicit, references which could be considered as a way of
coping with health complaints, i.e. referrence to age-related
complaints or functioning. It seems as if participants who
consider age-related physical complaints or functional decline
to be normal, are less bothered by them. And although not all
participants mentioning age-related (normal) complaints or
age-related functioning assessed their health as good, it may be
just the reason why they did not assess their health more
poorly.17

Conclusions
We have shown that that self-assessed health is a multidimensional concept. Over the years several qualitative studies
on self-assessed health have produced comparable results, even
though these studies differed with respect to the subgroups they
included and the methodology they applied. The consistency of
the findings suggests that we have actually taken a step nearer to
identifying which particular aspects are involved in health
assessments.

Acknowledgements
The authors would like to thank Dr Ir E. J. de Min for providing
the software for calculating kappa coefficients, and Dr H. van de
Mheen for participation during the early stages of the research
project. We would also like to thank Ms K. Gribling for her
careful translation of the excerpts from the interviews. The
GLOBE-study is supported by the Dutch Ministry of Public
Health, Welfare and Sports, and the Netherlands Health
Research and Development Council (ZON).

Key points
In this qualitative study we studied which particular
aspects are included in self-assessed health.
Self-assessed health proved to be a multidimensional
concept, including primarily physical, functional,
coping and wellbeing aspects.
Health behaviour or lifestyle factors proved to be
relatively unimportant in health self-assessments.
Subgroup differences in self-assessed health could
primarily be attributed to prior experience with illhealth.
The consistancy of these findings with other qualitative
studies suggest that we have identified the key
dimensions of self-assessed health.

206

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

References
1

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

Emotion 1993;17:205 24.

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

the effects of physical and psychological symptoms. Psychol Med 1982;12:
371 8.

Tissue T. Another look at self-rated health among the elderly. J Gerontol 1972;
27:91 4.

Idler EL, Benyamini Y. Self-rated health and mortality: a review of twentyseven community studies. J Health Soc Behav 1997;38:21 37.

Krause NM, Jay GM. What do global self-rated health items measure? Med
Care 1994;32:930 42.

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:S157 70.

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

respondents. Scand J Soc Med 1998;25:145 53.

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

and pessimism. Res Aging 1999;21:440 57.

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

qualitative and quantitative approach. Res Aging 1999;21:458 76.

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;38:299 308.
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;8:29 38.
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;21:1125 44.

Appendix 1
Flow chart describing the phases in the qualitative
analysis of the interviews

Qualitative study on self-assessed health

Appendix 2
Description of the health aspects are given illustrated
with quotations
Health dimensions
Health aspects
Physical: This dimension refers to the general working of
ones body
Chronic illness: presence of a chronic illness or a history of
chronic illness
Well, I guess you could say that my health is reasonably okay,
only theres no getting away from the fact that Im, uh, thirty, forty
percent asthmatic. Thats what Ive got, so to speak. Man, 60+,
high ses, copd/asthma, fair
Physical problems: reference to physical complaints, not
directly related to any chronic illness, such as never being ill,
never needing to stay at home due to illness, or only
experiencing minor illnesses
Uh, no problems, no headaches, no stomach aches, no menstrual
pains 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 () I
mean, well, if you dont need to see the doctor a lot, and you dont
have a whole lot of complaints () Healthy? Yes, all of us, were
healthy. At least, my husband never has to visit the doctorknock
on woodup to now, so, well. () Never been in hospital for
anything, well, only to have a baby, and thats rather a healthy
reason, wouldnt you say. Woman, 40-, low ses, no current illness,
good
Age-related (normal) complaints: reference to physical
complaints which are considered to be expected, i.e. normal,
considering ones age.
Id say Im fine. Yes. Of course theres always some little thing
going wrong here and there, but all pretty much to be expected.
My arm was giving me problems and the doctor gave me a few
shots, I mean, well, it was painful, and after eighty years its not
a surprise my joints werent working as smoothly as when I was
twenty. But actually Im doing fine. Man, 60+, low ses,
copd/asthma, fair
Prognosis of illness: reference to the course, or prognosis of a
chronic illness
Well I dont know whether you read the previous questionnaires?
Oh, well two years ago I was operated on for breast cancer, so with
that in mind, Im doing very well (..) Like I said, I may have had
an operation but it was localised and Im fine now. No other
complaints. Woman, 60+, high ses, no current illness, good
Bodily mechanics: reference to failing mechanicsof the
body, as a result of which one may suffer from recurring (minor)
physical complaints
The only thing, which is why I was wavering between very good
and 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+, high
ses, chronic back complaints, very good
Robustness: reference to being illness-prone vs. being more
robust 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 that
strong, that for the rest I feel perfectly healthy, but Im very quick to
notice when Ive been overdoing it. Like when Ive had too much to
drink. 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 ate
something like that Id notice right away. My stomach starts acting
up or something and he has no problems at all. Man, 40 2 , low
ses, no current illness, good

207

Functional: This dimension refers to ones ability to perform

certain activities
Not being impaired: reference to general functional abilities
or limitations. Referring to being able to do the things one wants
to, or needs to do, without any reference to disease, illness or
bodily mechanics
Well, because you can do everything, you do everything. But
very good, no, I mean there are also all the days that things dont
go 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 disability
or impaired mobility, due to chronic illness or disease
It started with my lungs () Yes, (my health) its poor. I
mean, if I could get more air. I mean, right now, and then Im
referring primarily to the past few months, after Ive walked for,
say, 200 metres, I have to stop and, and catch my breath. Take
just yesterday. I wanted to go get a haircut, thats 10 minutes
away by bike. Halfway there I had to turn around and go
home. So Im hoping that specialist is right and that if I use
oxygen when I exert myself, itll help. Man, 60+, low ses,
copd/asthma, poor
Age-related functional abilities: relating general functional
abilities or limitations to ageing, being able to function well for
ones age
Uhh, if a persons healthy, uhhh, he can do anything hes
supposed to be able to do at his age. I mean, look, if youre over
sixty, Im sixty-three, obviously you cant be doing all the crazy
things you did when you were twenty or thirty. Man, 60+, high ses,
no current illness, good
Coping: This dimension refers to the psychological mechanism of dealing with an illness
To adapt to illness: reference to having adapted to the
chronic illness or having learned to live with its limitations
Because health is extremely complicated. I mean, purely on the
physical level, you could check whether every bit of the body is in
good working order. And in my case youd find that there are a
great many bits in my body that dont work well. But if you look at
the complex, the aggregate and the combination etc. etc., how I
function 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 hand
adapting and on the other taking steps to be able to do what you
want to do, only in a different way. Woman, 60+, high ses,
chronic back complaints, sometimes good and sometimes
poor
A positive attitude: reference to maintaining a positive
attitude towards the illness
Its however you look at it, I say. I mean, its not going at all well
to be frank, but I try to take the cheerful view. So, chin up, is what I
always say () Well, theyre not actually very healthy. No, well
they always look on the bright side, you see. Yes. And, I mean take
someone who has a bug or something else, whatever, that can make
you feel really ill, that person knows thisll be over in a couple of
days, a few weeks, and that holds for a lot of things. And thats
what I mean by always looking on the bright side. And even when
its 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 the
health of peers; comparing ones own health with the health of
people who are worse off (downward comparison)
But there are always worse things, arent there, and thats some
consolation. I was just in hospital and I saw a person come out who
was bent over nearly double, what a hump! His nose close to
scraping 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 to
aspects of mental health
Feeling fit: reference to feelings of fitness or energy, referring
to feeling fit, energetic and not tired

208

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

Yes, I feel good, Im never tired and uh especially during the past
few years, sure. () Yes, physically healthy? I guess, if youre not
tired () I feel fit, not tired, so I feel healthy. Woman, 40 2 , low
ses, no current illness, good
Feeling good: reference to general feelings without any
objective justification, simply referring to feeling good
Yes, I feel good, I feel absolutely great. For me, health is feeling
good. And I do. Thats how simple it is. () Oh, thats, I guess, not
feeling bad. Man, 40 2 , high ses, chronic back complaints,
good
Body/mind equilibrium: reference to the (im)balance of
physical and mental problems

If youre ill and out of sorts, you can forget it, you just feel
rotten. If you have a psychological problem you feel just as
rotten even though physically, theres nothing wrong. But youre
not completely healthy if youve got a problem with either. ()
Healthy is when you have no infections of any kind. I guess
thats part of it. And that theres no blackness messing up your
mind.() 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
BEHAVIOUR
Eating healthy food: mentioning eating well (all from our
own garden) and not eating sweets