Family Practice Vol. 18, No. 2, 174-180
© Oxford University Press 2001
Decision-making in the consultation |
Health from the patient's point of view. How does it relate to the physician's judgement?
Research Centre in General Medicine, Borgmästarvillan, Karolinska Hospital Stockholm, and
a Institute for Social Work, University of Stockholm, Sweden.
Undén A-L and Elofsson S. Health from the patient's point of view. How does it relate to the physician's judgement? Family Practice 2001; 18: 174–180.
Received 23 March 2000; Revised 23 August 2000; Accepted 30 October 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
Objectives. The purpose of this study was to evaluate the relationship of self-rated health to a measure of physical status, based on a professional rating of the individual's health from a strictly physical point of view.
Methods. A random selection of 407 people over the age of 20 from the north-western catchment area of greater Stockholm were invited in 1995 to a physical examination, including a self-report questionnaire with questions about self-rated health, lifestyle, psychosocial factors and quality of life. A measure of physical health on a 5-point graded scale was constructed using the information from the records of the physical examination as a base.
Results. Self-rated health and the professional ratings of health coincided in ~60% of the cases. There were a relatively large number of cases where the ratings were contradictory. The correlation between the scales was 0.45. Comparison between the two ratings with respect to association with potential determinants showed that physical factors naturally explained most of the variances in physical health, whereas social and mental well-being and somatic conditions (women) were the most important explanatory variables for self-rated health. Irrespective of whether they had ‘favourable’ or ‘unfavourable’ health, those with ‘poor’ self-rated health also had perceived lower social and mental well-being, less appreciation, more somatic conditions (women) and worse coping abilities (men).
Conclusions. With mental, psychosocial and social problems becoming more pronounced in sickness patterns for primary care patients, self-rated health could be a helpful device, especially when time resources for consultations are short. This measure could also give a more global view of the patient's situation when effectivity and rationality can be a threat to a holistic view of the patient.
Keywords. Physician's rating of health, psychosocial factors, self-rated health.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
A strong focus on saving money within health care, combined with a shift of care from in-patient to out-patient, has led to major changes within primary care in Sweden. The discharging of patients from institutional psychiatric care has also had consequences. There have also been changes within the profile of diseases, such that more and more of those seeking help within primary care have underlying psychosocial problems.1 Similar results are also found in international studies.2,3 This altered disease profile that the GP encounters might suggest that longer consultation times are required.4 In Sweden, insufficient ecomic resources and an actual shortage of physicians have enforced shorter consultation times, which do not always allow the GP sufficient time to obtain a complete picture of the patient's situation. It is thus necessary to find new methods within the existing time and efficiency constraints for acquiring a more complete picture of the patient's situation.
One possibility would be through simple forms of self-rating on the patient's part, which would provide a basis to make the consultation more ‘effective’ and would also allow the GP to concentrate on the central problems in the patient's life situation. Moreover, this is in line with the goal of medical care which should be to achieve a more effective life, and preserve function and well-being.5,6 This also implies that health care outcome has to consider the extent to which results of treatment meet a patient's need and expectations. Nowadays, there is widespread use of different short form measures of health and health-related quality of life in health care settings.7,8 These measures make it practical to look beyond traditional measures of biological functioning to larger issues of functioning and well-being. A very simple health measure which is widely used nowadays is the individual's own evaluation of health, i.e. self-rated health, measured with a single question which prompts the respondent to provide an overall assessment of health. Measurements of self-rated health are also used as common indicators of the health status of the population in many epidemiological studies. A very intriguing finding in the literature is that perceived health is a powerful independent predictor of mortality, even after controlling for clinical health status.9 The general consistency of results in these studies is impressive, particularly in light of the fact that the studies are different with regard to design, population, follow-up period, control for confounders and the way in which self-rated health is assessed. However, the predictive capacity of self-rated health shows a remarkable consistency across all studies. The research into determinants for self-rated health reveals that the global health status measure reflects physical health problems and functional ability as well as social and mental well-being, and the measure provides a succinct way of summarizing the diverse components under the broader health status rubric.9,10
Thus self-rated health seems to agree with the World Health Organization's definition of health as a ‘state of complete physical, mental and social well-being and not merely the absence of disease or infirmity’.11 Although patients are the best source of information regarding the achievement of the goals of medical care described above, their experiences of disease, treatment and global health status do not usually form part of the medical record or of the health care database.12
The purpose of this study was to evaluate the relationship of self-rated health to a measure of physical status, based on a professional rating of the individual's health from a strictly physical point of view. An underlying question was if the patients's own assessment of his health status may provide information that could constitute a basis for identifying the new disease profile which confronts today's health care to an ever-increasing extent.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
Study group
The study group consisted of a random, age-stratified selection of 407 men and women who were invited to a physical examination after having participated in a population-based questionnaire study involving 4200 randomly selected people over the age of 20 in the north-western catchment area of greater Stockholm in 1995. The questionnaire targeted health care utilization, general health and quality of life, and the response rate was 71%.
In the present study, 30 men and 30 women were selected from each 10-year age group between the ages of 25 and 75 years. The age group 20–24 years also had 30 men and 30 women, making a total of 359 persons (one person from the selection list gave the incorrect age). An additional 48 people fairly evenly distributed between 20 and 74 years, who had rated their health as fair or poor, were also invited for a physical examination. Seventy-one percent of the 407 invited agreed to come, making a total sample size of 287 (141 men and 146 women). The mean age was 47.2 years for men and 46.9 years for women.
Procedure
A 30-minute physical examination including blood draw was performed after the subject had fasted for 8 hours. Before seeing the doctor, the subject was asked to fill in a questionnaire concerning health, lifestyle and psychosocial factors. The physical examinations were carried out during the period November 1995–May 1996.
Questionnaire
For health-related symptoms and social, physical and mental well-being, the Gothenburg Quality of Life Instrument (GQLI) was used.13 For health-related symptoms, the respondents were given a list of 30 symptoms and asked to indicate whether they had experienced any of these symptoms during the previous 3 months. The second part of the GQLI comprises ratings of social well-being (home and family, housing, work situation, finances, leisure), physical well-being (vision, hearing, memory, appetite, fitness), mental well-being (patience, mood, self-confidence, energy, sleep) and feeling appreciated (within and outside of the home). The ratings were done on a 7-point graded Lickert scale ranging from ‘very bad’ (1) to ‘excellent, could not be better’ (7). The two ratings concerning appreciation were summed to produce a simple measure of social support. Self-rated health is included in this part of the GQLI and was rated in the same way. Use of health care services was measured as the self-reported number of visits to a physician during the last year. Educational attainment was classified into five levels ranging from mandatory education to university education (1 = secondary, 2 = lower-secondary school, 3 = 2-year upper-secondary school, 4 = 3-year upper-secondary school and 5 = university education). Physical activity during leisure time was estimated on a 4-point scale where 1 = no activity, 2 = moderate activity (cycling, walking) at least 2 hours a week, 3 = moderately heavy activity (swimming, tennis, running, physical training) at least twice a week and 4 = hard training at least three times a week.14 Depression was assessed according to MADRS, a 9-item scale, developed with the goal of providing an instrument which would be sensitive to changes, and practical in clinical settings.15 Coping was measured by a 7-item scale. This scale measures the extent to which people see themselves as being in control of forces that have an important effect on their lives.16 Hypochondriacal attitudes was measured with seven questions.17
Physician's rating of health
To obtain an assessment of physical health, two physicians developed a 5-point graded scale based on the information from the records of the physical examinations conducted. The record provided detailed information on specific health problems, both chronic and acute, medication and blood analysis. The aim was to achieve an easily applicable scale, which could be used to define health from the information received from the record. The scale had to cover the physical status of the patient, ranging from healthy to severely ill, from a strictly physical point of view. Social needs were not to be included. A scale for somatic diseases was used as a basis for the development of the scale.18 Judgements were made in accordance with the incidence and severity of disease according to the following definitions: ‘healthy’, ‘healthy with slight problems’, ‘fair’ ‘rather poor’ and ‘poor’ (see Appendix 1). The inter-rater reliability between the two physicians was good (r = 0.80). In the results presented below, we have used the estimates made by the physician who had the main responsibility for developing the scale.
Statistical methods
Associations between the health ratings and some hypothesized determinants were estimated as product–moment correlations, according to Spearman. Multiple regression analyses were used to determine the independent effects of the hypothesized determinants. Results are presented as explained variance, standardized regression coefficients and significance levels. Differences in coping, social and mental well-being and physical factors between different ‘health groups’ were tested by Student's t-test (two-tailed).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
In Table 1
, the distributions of the scales for self-rated health (a 7-point graded scale ranging from ‘very bad’ to ‘excellent, could not be better’) and physician's rating of health (a 5-point graded scale ranging from ‘healthy’ to ‘severely ill’, rated from a strictly physical point of view) are compared. The distribution characteristics were very much in agreement with each other, both with marked positive skewness. To look at the concordance between the two assessments, the ratings have been grouped in the following way. The physician's ratings ‘healthy’ and ‘healthy with slight problems’ are hypothesized to correspond to ratings 6–7 in self-rated health, ‘fair’ to ratings 4–5 and ‘rather poor’ or ‘poor’ to ratings 1–3. With this grouping, 60% of the cases were rated in the same way. The concordance was slightly higher among women (62%) than men (58%). To study the distribution characteristics in more detail, self-rated health in relation to the physician's ratings of health was calculated (Fig. 1). The results showed that the distribution of self-rated health was relatively similar in the two groups that were rated by the physician as completely ‘healthy’ or ‘healthy with slight problems’. The average values for self-rated health in these groups were 5.8 and 5.7, respectively. About 4% of the individuals in these groups rated their health as <4 on the scale, and ~15% as <5, which means that some of these ‘healthy’ (as defined by the physician) persons regarded their health as less good. In the group ‘fair’, 30% rated their health as excellent or just below excellent, and the mean value of self-rated health was 4.8. In the ‘poor health’ group, no one gave the highest score on self-rated health (excellent) but 20% in this group rated their health as just below excellent. The mean value of self-rated health was 3.8 in this group. The results thus showed that even if there was an agreement between the physician's rating of health and self-rated health, there were a relatively large number of cases where the ratings were contradictory. The correlation between the two ratings was 0.45 (P < 0.001).
|
|
Self-rated health and physician's rating of health were then compared with respect to correlation with some hypothesized determinants based on patient's ratings, including factors representing physical health as well as social and mental well-being. The results showed that most of the factors were significantly correlated to the health ratings (Table 2
), with some exceptions. Social well-being and hypochondriacal attitudes were significantly correlated only to self-rated health, whereas age was correlated only to physician's rating of health. The correlation pattern was also similar for men and women, with the exception of physical activity, which was significantly correlated with self-rated health and physician's rating only among men. There was also a tendency for the correlations to be somewhat stronger throughout for men than for women. To control for confounding effects, regression analyses were performed (Table 3). The results gave a more discriminating structure, indicating that factors representing physical health explained most of the variances in physician's rating of health, whereas psychosocial factors explained the variances in self-rated health to a greater extent. Concerning physician's rating of health, there were significant associations with age, chronic disease, visits to the physician and depression for both men and women. For self-rated health, the patterns differed more clearly between men and women. Among men, the most important explanatory variables were social and mental well-being, and hypochondriacal attitudes. For women, somatic conditions and mental well-being were the most important explanatory variables. The variables explained ~50–70% of the variance in self-rated health as well as in the physician's rating.
|
|
To study whether the relationship of self-rated health to other factors was the same irrespective of physical health status, the physician's ratings of health were dichotomized simply as ‘favourable’ (4–5) or ‘unfavourable’ (1–3), and self-rated health as ‘good’ (5–7) or ‘poor’ (1–4), and were compared with each other (Table 4
). Irrespective of having ‘favourable’ or ‘unfavourable’ health, those with ‘poor’ self-rated health perceived lower social and mental well-being, less appreciation, more somatic conditions (women) and worse coping abilities (men). Among those with ‘unfavourable’ health, there were also significant differences between persons with ‘good’ self-rated health and ‘poor’ self-rated health for depression and hypochondriacal attitudes. Those with ‘favourable’ health (as defined by the physician) and ‘poor’ self-rated health were compared with those with ‘unfavourable’ health and ‘good’ self-rated health (significance analyses are not presented in the table). The first group appeared to have significantly lower social well-being among both men and women. Among women, this group also had significantly lower mental well-being and felt less appreciated than those with ‘unfavourable’ health and ‘good’ self-rated health. These differences could be even more pronounced as the former group is significantly younger.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
In this study, self-rated health has been compared with a measure of physical status, based on a professional rating of the individual's health from a strictly physical point of view. Of special interest is whether the individual's evaluation of health could constitute a base for identifying the new disease profile which confronts today's health care to an ever-increasing extent.
Comparing self-rated health and physician's rating of health with respect to distributions, the scales seemed to run together quite well. When comparing physician's ratings and self-rated health, the results showed that 60% of the self-rated health corresponded to the physician's rating. Self-rated health has been compared with professional ratings of physical health in only a few studies.19–21 However, as different measures and methods were used, the results are not direcly comparable. The professional ratings and those of the participants coincided in ~40–60% of the cases in these studies.
The results showed that even if there was an agreement between the ratings, there were a relatively large number of cases where the ratings were contradictory, which means that some of these ‘healthy’ (as defined by the physician) people regarded their health as less good and some of the ‘unhealthy’ (as defined by the physician) persons regarded their health as good. This implies that comparing only the distributions between self-rated health and physician's rating of health gives a somewhat false-positive picture as the discrepancies are in both directions. This became apparent when looking at the correlation between the two scales, which was modest, r = 0.45.
Regression analysis showed that physical factors primarily explained the variance in physician's rating of health, more pronounced among women, whereas psychosocial factors primarily explained the variance in self-rated health, more pronounced among men. For self-rated health, there were some differences in the pattern between men and women. Social and mental well-being were stronger explanatory variables among men than among women, whereas somatic conditions and mental well-being were the strongest explanatory variables for women. Separate analyses showed that among women, somatic conditions were strongly correlated to social as well as mental well-being. This means that somatic conditions could ‘catch’ the association with social and mental well-being, an indication that women express social and mental problems as somatic or psychosomatic conditions to a greater degree than men. Among men, hypochondriacal attitudes were also associated independently with self-rated health. Low self-rated health could thus, in some cases, be an indication of abnormal worries about health or illness.
Irrespective of physician's rating (dichotomized into ‘favourable’ and ‘unfavourable’ health), people with ‘poor’ self-rated health had lower social and mental well-being compared with those with ‘good’ self-rated health. An interesting finding was that those with ‘unfavourable’ health but with ‘good’ self-rated health in many cases showed better well-being than persons with ‘favourable’ health and ‘poor’ self-rated health.
The results presented suggest that the patient's own assessment of his health may provide a complement to a strictly physical evaluation of health. The accumulated results of studies on self-rated health show that it is a very strong independent predictor for future illness and death, independent of physical health status.9 This knowledge has implications not only for future research but also for clinical and preventive medicine. Seen from a clinical point of view, the results from research on self-rated health ought to remind the physician of the importance of listening carefully to what the patient has to say about his/her health. This is especially true when the patient has a health assessment that differs from that of the physician, which could lead to different consequences. For example, chronically or seriously ill people with low self-rated health could need more support not only in medical terms but also from a social and mental point of view. On the other hand, the chronically or seriously ill who rate their health as good could be supposed to have more of their own resources to deal with the situation by themselves. The results also showed that people with ‘good’ self-rated health had better coping abilities compared with those with ‘poor’ self-rated health.
What then is the practical use of self-rated health in primary health care? Psychosocial health indicators have been proposed as an alternative for the evaluation of primary medical care, on the grounds that such indexes can reflect not only medical aspects of morbidity but also the patient's social functioning and his emotional and psychological well-being.22 One application of this type of measure could be as a means for the patient to communicate his main areas of concern or felt needs to the physician or health care team. The information may also be used both in medical education and in encouraging physicians to adopt a comprehensive view of the patient's problems. In before-and-after studies of medical care, the patient's judgement of his health provides an important component in the overall evaluation of the care provided. With mental, psychosocial and social problems being more pronounced in sickness patterns for primary care patients,23 self-rated health could be a helpful device, especially when time resources for consultations are short. It could also be important as this measure could give a more global view of the patient's situation when effectivity and rationality can be a threat to a holistic view of the patient. These measures make it practical to look beyond traditional measures of biological functioning to larger issues of functioning and well-being. However, further research is needed on the practicability of self-rated health in clinical settings.
|
Appendix 1 Physician's rating of health |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
- No illness, no problems, no treatment (‘healthy’ = 5).
- Patient without problems but under treatment or patient with one or more slight problems without serious underlying causes (‘healthy with slight problems’ = 4).
- Chronically ill or temporarily ill patient with one (or several) disease that does not unduly limit (for age) normal activities (‘fair’ = 3).
- Chronically ill patient with present symptoms that give only limited reduction for age in normal activities, or temporary illness in a healthy individual that gives the same limitations (‘rather poor’ = 2).
- Acute illness or chronic illness with pronounced limitations of normal activities (for age) (‘poor’ = 1).
|
Acknowledgments |
|---|
The authors gratefully acknowledge Johanna Rubin and Stefan Österholm for the work on developing the scale to rate physical health. The study was supported by grants to A-L U from the Swedish Council for planning and co-ordination of research and the Harald and Louise Ekman's research foundation.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1 Physician's rating...References |
|---|
1 Hansson L, Stefansson C-G, Nettelbladt P, Borgquist L, Nordström G. Psykisk ohälsa hos patienter i primärvården. En multicenterstudie. Nord Med 1993; 5: 145–147.
2 Üstün T, Sartorius N. Mental Illness in General Health Care. An International Study. Chichester: Wiley & Sons, 1995.
3 Sartorius N. Psychological Disorders in General Medical Settings. Toronto: Hogrefe & Huber Publishers, 1990.
4
Schire S. Frequent attenders—a review. Fam Pract 1986; 3: 272–275.
5 Ellwood PM. Outcomes management: a technology of patient experience. N Engl J Med 1988; 318: 1549–1556.[ISI][Medline]
6 Schroeder S. Outcome assessment 70 years later: are we ready? N Engl J Med 1987; 316: 160–162.[ISI][Medline]
7 Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992; 30: 473–483.[ISI][Medline]
8 Stewart AL, Hays JA, Ware EJ. The MOS Short-Form General Health Survey: reliability and validity in a patient population. Med Care 1988; 26: 724–735.[ISI][Medline]
9 Bjorner JB, Söndergaard-Kristensen T, Orth-Gomér K, Tibblin G, Sullivan M, Westerholm P. Self-rated Health. A Useful Concept in Research, Prevention and Clinical Medicine. Swedish Council for Planning and Coordination of Research, 1996.
10 Fylkesnes K, Førde OH. Determinants and dimensions involved in self-evaluation of health. Soc Sci Med 1992; 35: 271–279.
11 World Health Organization. World Health Organization Constitution. Basic Documents. Geneva: WHO, 1948.
12 Ware JE. The status of health assessment 1994. Annu Rev Public Health 1995; 16: 327–354.[ISI][Medline]
13 Tibblin G, Tibblin B, Peciva S, Kullman S, Svärdsudd K. The Goteborg quality of life instrument—an assessment of well-being and symptoms among men born in 1913 and 1923. Methods and validity. Scand J Prim Health Care 1990; Suppl 1: 33–38.
14 Grimby G, Wilhelmsen L, Björnstorp P, Saltin B, Tibblin G. Habitual physical activity. Aerobic power and blood lipids. In Pernow B, Saltin B (eds). Muscle Metabolism During Exercise. New York: Plenum Press, 1971.
15
Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979; 134: 382–389.
16 Pearlin LI, Menaghan EG, Lieberman MA, Mullan JT. The stress process. J Health Soc Behav 1981; 22: 337–356.[ISI][Medline]
17 Troein M, Råstam L, Selander S. Health beliefs and heart disease risk among middle-aged Swedish men. Scand J Prim Health Care 1997; 15: 198–202.[ISI][Medline]
18 Bygren L-O. Met and unmet needs for physical and social services. Scand J Soc Med 1974; Suppl 8, Thesis.
19 Friedsam H, Martin H. A comparison of self and physician's health ratings in an older population. J Hum Soc Behav 1963; 4: 179–183.
20 LaRue A, Bank L, Jarvik L, Hetland M. Health in old age: how do physicians' ratings and self-ratings compare? J Gerontol 1979; 34: 687–691.[ISI][Medline]
21 Maddox G, Douglas E. Self-assessement of health: a longitudinal study of elderly subjects. J Health Soc Behav 1973; 14: 87–93.[ISI][Medline]
22
McDowell I, Martini C. Problems and new directions in the evaluation of primary care. Int J Epidemiol 1976; 5: 247.
23 Royal College of General Practitioners. Present State and Future Needs of General Practice. London: Royal College of General Practitioners, 1973.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. V. O'Toole, R. C. Castillo, A. N. Pollak, E. J. MacKenzie, M. J. Bosse, and the LEAP Study Group Determinants of Patient Satisfaction After Severe Lower-Extremity Injuries J. Bone Joint Surg. Am., June 1, 2008; 90(6): 1206 - 1211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Graz Prognostic Ability of Practitioners of Traditional Arabic Medicine: Comparison with Western Methods Through a Relative Patient Progress Scale Evid. Based Complement. Altern. Med., March 13, 2008; (2008) nen022v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Breidablik, E. Meland, and S. Lydersen Self-rated health in adolescence: A multifactorial composite Scand J Public Health, January 1, 2008; 36(1): 12 - 20. [Abstract] [PDF]
|
|
|
|
 |
|
 A. Wikman, S. Marklund, and K. Alexanderson Illness, disease, and sickness absence: an empirical test of differences between concepts of ill health J. Epidemiol. Community Health, June 1, 2005; 59(6): 450 - 454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lekander, S. Elofsson, I.-M. Neve, L.-O. Hansson, and A.-L. Unden Self-rated Health Is Related to Levels of Circulating Cytokines Psychosom Med, July 1, 2004; 66(4): 559 - 563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Okkes, M Veldhuis, and H Lamberts Severity of episodes of care assessed by family physicians and patients: The DUSOI/WONCA as an extension of the International Classification of Primary Care (ICPC) Fam. Pract., August 1, 2002; 19(4): 350 - 356. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||