Family Practice Advance Access originally published online on August 22, 2005
Family Practice 2006 23(1):34-39; doi:10.1093/fampra/cmi064
Dyspnea in elderly family practice patients. Occurrence, severity, quality of life and mortality over an 8-year period
a Department of General Practice, Maastricht University, b Department of Epidemiology, Maastricht University, c Department of Pulmonology, Academic Hospital Maastricht/Maastricht University, d MEMIC; Centre for Data and Information Management, Maastricht University, e Assistant in General Practice, Maastricht University and f Department of Psychiatry and Neuropsychology, Maastricht University
Correspondence to Dr FG van der Horst, Department of General Practice/P. Deb1, Maastricht University, PO Box 616, 6200 MD Maastricht, The Netherlands; Email: frans.vanderhorst{at}hag.unimaas.nl
Received 25 March 2005; Accepted 20 June 2005.
Huijnen L, van der Horst F, van Amelsvoort L, Wesseling G, Lansbergen M, Aarts P, Nicolson NA, Knottnerus JA. Dyspnea in elderly family practice patients. Occurrence, severity, quality of life and mortality over an 8-year period. Family Practice 2006; 23: 34–39.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
Background and objectives. This study examined the prevalence and severity of dyspnea in patients
70 years of age and its impact on the quality of life. We also examined the prognostic significance of dyspnea for mortality. Methods. The cohort study started with a mailed questionnaire, supplemented with an interview. From the population of elderly patients in a family practice health center, a one in five sample (n = 124) was randomly selected, similar in age and sex distribution to those not in the sample. Demographic and other relevant variables were examined for their association with dyspnea. Cox proportional hazards ratio analysis was done with dyspnea (MRC, BDI scores) as independent and mortality as dependent variable.
Results. Baseline data indicated that 23% (MRC) to 37% (BDI) of the patients had moderate to severe dyspnea. Shortness of breath was associated with older age, poor perceived health, more anxiety and depressive symptoms, impaired daily functioning, and lower happiness. Moderate and severe dyspnea measured with BDI and MRC was a significant predictor of death within eight years due to cardiovascular or lung disease. Selective participation did not appear to have biased this outcome.
Conclusions. Dyspnea occurs frequently in the elderly, is associated with poor health, and interferes with daily functioning. Results suggest that dyspnea contributes to mortality. Development and implementation of guidelines would be highly desirable. Early diagnosis is valuable because this provides opportunities to positively influence the patient's functional condition.
Keywords. Dyspnea, health, daily functioning, quality of life, mortality.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
Shortness of breath or dyspnea is a frequently reported complaint in the elderly.1–3 Dyspnea is defined as a more limited or difficult respiration than expected given the current level of activity.2 In the literature, prevalence estimates in the elderly range from 20% to 60%.1,2,4 Data from the registration network of family practices5 in 1995, 1999, and 2002 indicated that dyspnea was noted as a problem in only 1% of patients
70 years of age. According to the guidelines of the RNH the majority of these complaints were transformed into a cardiac, pulmonary, or other diagnostic category. Research literature concerning dyspnea as a symptom in the elderly is scarce, although much has been written about specific diseases that may cause dyspnea, especially COPD and heart failure. More attention to dyspnea, including its early detection, may be important for a variety of reasons: dyspnea is a common complaint1,2,6 with a marked negative influence on daily functioning and quality of life,3,6 acute or severe dyspnea requires prompt and adequate pharmacological intervention,7 and it is an important contributor to mortality.2,3 Early evaluation of dyspnea can have a positive influence on the patient's functional condition, thus promoting and prolonging an active and independent lifestyle.4 To gain insight into the occurrence and severity of dyspnea, as well as its impact on the quality of life, we initiated a cohort study in 1993–1994. The identified cohort was then followed for eight years. We investigated the following questions:
- In patients 70 years of age, what is the prevalence and severity of dyspnea, and to what extent is dyspnea associated with problems in health, daily functioning, and quality of life?
- What is the prognostic significance of dyspnea for mortality over a period of eight years?
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
Setting, participants and procedure
The study was conducted in a family practice health center with 6900 registered patients, including 650 individuals (430 women and 220 men) 70 years of age and older. From this general population of elderly patients, a one in five sample (n = 124) was randomly selected; the age and sex distribution was similar to that of the group
70 years not included in the sample. The family practitioners (GPs) were asked to identify patients meeting criteria for exclusion on medical grounds from study participation. Ten patients were excluded on the basis of the following criteria: residence in hospital or nursing clinic (n = 3), serious diseases like dementia or carcinoma (n = 6), death of partner in the last 6 months (n = 1). The remaining 114 individuals were mailed the first questionnaire on dyspnea symptoms at the end of 1993. In an accompanying letter, patients were asked by their GPs for their informed consent to participate in the second part of the baseline assessment, consisting of a home interview during which a more extensive set of questionnaires was completed. This interview took place approximately three months later (in 1994). After this baseline assessment, the identified cohort was followed by the GPs themselves over the period 1994–2002.
Measures
The presence and experienced severity of dyspnea at baseline were assessed with two different instruments. The initial questionnaire included the Medical Research Council (MRC) Scale,6,8 an easy to comprehend instrument with five questions in a yes/no format concerning recent dyspnea symptoms (Appendix 1). Total scores on the scale range from 0 to 4, with higher scores reflecting more severe dyspnea.
As part of the home visit, the interviewer filled in the Baseline Dyspnea Index (BDI),6,9 based on the patient's responses. The BDI assesses three aspects of dyspnea, as experienced recently: degree of functional impairment, level of activity, and level of effort. Each aspect is rated on a 5-point scale, ranging from 0 (‘extremely’) to 4 (‘not at all’), so that the total score can range from 0 to 12. To avoid confusion in the current analysis, the BDI measure has been transformed so that (as for the MRC) higher scores reflect more severe dyspnea. The BDI was developed because the MRC relates dyspnea only to the level of activity, without considering the associated effort necessary for that activity. Furthermore, the MRC does not assess the degree of functional impairment, an important consequence of dyspnea. The BDI thus provides more information than the MRC; scores on the two instruments are, however, highly correlated (r > 0.70).2
Additional measures of perceived health, daily functioning, and quality of life were completed during the home visit. These included:
- anxiety, assessed with the STAI (State-Trait Anxiety Inventory);10
- depressive symptoms, assessed with the 30-item GDS (Geriatric Depression Scale), a self-report instrument;11
- subjective health (physical and mental), each rated on 5-point scale ranging from very poor to very good;12
- daily functioning, measured with the abbreviated Dutch version of the Sickness Impact Profile (SIP);13
- quality of life, rated by subjects on a 10 cm visual analogue scale (VAS);14 and
- satisfaction and happiness in life, rated on 7-point scales.
- depressive symptoms, assessed with the 30-item GDS (Geriatric Depression Scale), a self-report instrument;11
Follow-up at eight years
Severity of dyspnea at baseline was used as an independent variable to predict mortality over the follow-up period. In January 2002, we therefore determined which patients in the study had died and, in these cases, due to what cause(s). For comparative purposes, we also determined the status of non-respondents and excluded subjects.
Following a death, the attending physician in The Netherlands is required to submit a declaration that the death was due to natural causes (otherwise a coroner is contacted) and a statement describing both the direct cause of death and possible underlying causes. This information is then recorded in the National Death Register, with cause of death coded according to the ICD Mortality List. Established guidelines of the registration network of family practices require participating GPs to keep patient death registration forms for ten years. Mortality data for the practice where the current study was conducted were complete for all study respondents over the follow-up period; of 75 non-respondents and excluded subjects, 5 could no longer be located at the end of the study.
To evaluate possible effects of selective participation in either of the two baseline assessments, we also compared respondents, non-respondents and excluded subjects with respect to the number of active medical problems registered for them in January 1994. Practices in the RNH maintain lists of active medical problems, coded according to the ICPC, for all patients.5 Measures compared were total number of active problems, number of problems in chapter K (circulatory system), number in chapter R (respiratory system), and number in the remaining chapters. The following groups were compared: (1) for the MRC: respondents to the questionnaire (n = 84); (2) for the BDI: respondents to the interview (n = 49), and (3) non-respondents (n = 58) and excluded subjects (n = 17).
Statistical analysis
For the analysis, we used the chi-squared test (where necessary the Fisher Exact test), Spearman correlations, and the t-test for comparison of the number of medical problems. To assess the prognostic significance of dyspnea for mortality, we used Cox proportional hazards analysis (HR) with MRC score as predictor, adjusted for the effects of age and gender. The Cox regression analysis was repeated with BDI score as predictor in the subsample that completed this measure.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
Response rates for questionnaire and interview
Subject inclusion, exclusion, and response rates at each stage of the baseline assessment are summarized in Figure 1. Of the 114 patients (38 men and 76 women), 84 (74%) returned a completed questionnaire. Reasons given for not participating, ascertained by telephone for the remaining 30 patients, were mainly lack of time or interest. Other reasons were a recent operation, recent CVA, and admission to a nursing clinic. Participation was somewhat higher in men (33 out of 38) than in women (51 out of 76) (P = 0.02).
|
The sample of patients who completed the home visit assessment was smaller than the initial questionnaire sample due both to exclusion and to drop-out. The final sample of BDI respondents consisted of 26 men and 23 women. Compared to the total elderly population in this family practice setting, men were somewhat overrepresented. A similar pattern was observed for study drop-outs: of the MRC questionnaire respondents, relatively more men than women completed the BDI three months later (26 of 33 men versus 23 of 51 women; P = 0.002).
Perceived severity of dyspnea
Table 1 shows the distribution of MRC scores by gender and age class. A minority (45%, n = 38) of the respondents had no dyspnea; 32% (n = 27) had mild dyspnea complaints (MRC = 1 or 2), with no significant difference by gender. Moderate to severe symptoms (MRC 3) were reported by 23% (n = 19) of the respondents. Older women (>80 yrs) tended to report moderate to severe dyspnea more frequently than younger women (P = 0.054), but older men did not differ from younger men in prevalence of symptoms in this range. In the subsample 80 yrs, women were more likely than men to have moderate to severe dyspnea (P = 0.06). In the subsample that completed the BDI interview (n = 49), 37% (n = 18) reported moderate to severe dyspnea.
|
Associations between BDI and health, daily functioning and quality of life
Table 2 shows the associations (Spearman correlations) between BDI scores and other variables measured at baseline. Greater shortness of breath in elderly patients was associated with increasing age, worse physical and mental health, more anxiety and depressive symptoms, worse daily functioning, and lower happiness. Dyspnea complaints showed no significant relationship to self-reported life satisfaction or overall quality of life.
|
Differences between respondents, non-respondents and excluded subjects in morbidity and mortality
To gain more insight into possible biases arising from selective participation, we compared respondents with non-respondents and excluded subjects on the number of active medical problems at baseline, as well as on rates and causes of mortality over the eight-year follow-up. Because non-respondent and excluded subjects did not differ in the number of medical problems we combined these categories in the analysis. Patients who completed the MRC questionnaire (n = 84) proved to have more active problems in chapter R than non-respondents (n = 40) (means 0.29 versus 0.18; P = 0.045), but did not differ from non-respondents in frequency of chapter K problems, total problems in the remaining chapters, or total number of active problems. Respondents to the BDI questionnaire (n = 49) had more active problems than non-respondents (n = 35) in chapter K (means 1.41 versus 0.94; P = 0.007) and in chapter R (means 0.43 versus 0.09; P < 0.001), but did not differ in total number of problems in the remaining chapters or in total number of active problems.
Mortality data over the follow-up period are summarized in Table 3. Of the 124 patients in the initial 1-in-5 random sample, only 5 (3 non-respondents and 2 excluded subjects) could not be traced after 8 years to ascertain mortality status. A total of 39 deaths (46% overall mortality) were recorded among the patients who had completed the MRC questionnaire, whereas 25 deaths (51% overall mortality) occurred within the subsample of BDI respondents.
|
Among the traced non-respondents, overall mortality was similar to that observed among respondents. Mortality was much higher (93%) among excluded subjects, for whom information was available, than among the other categories combined (P < 0.001). Percentages of deaths attributed to cardiovascular or lung diseases were similar over the four categories, ranging from 50% to 64%.
Dyspnea as a prognostic factor for mortality
Results of the Cox regression analysis are summarized in Table 4. In terms of total mortality, the MRC score was not a significant predictor (age and gender-adjusted HR: 1.2; 95% CI: 0.94–1.5 per point increase in MRC score). Results obtained with the BDI score as predictor approached statistical significance (HR: 1.2; 95% CI: 0.99–1.4 per point increase in BDI score).
|
For risk of death from cardiovascular or lung disease (combined in the analysis), both MRC and BDI were stronger and statistically significant predictors. For patients scoring higher on the MRC, hazard ratio of death was 1.4 (95% CI: 1.1–1.9 per point increase in MRC score); for the BDI, results were even more pronounced (HR: 1.5; 95% CI: 1.2–2.0 per point increase in BDI score).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
We found a high prevalence of dyspnea among patients
70 years of age, with 55% of participants in the postal survey (MRC). Of those who completed the extended questionnaire (BDI) during the home interview, 71% reported dyspnea symptoms. Similarly, baseline data indicated that 23% of all patients had moderate to severe dyspnea, whereas BDI scores indicated an even higher percentage (37%). Discrepancies between these two figures may reflect differences in the two instruments and the mode of administration (questionnaire versus interview), the interval of approximately three months between the two assessments, and differences in the composition of the sample due to exclusion or drop-out in that interval. Shortness of breath showed a positive association with age. In addition, shortness of breath was associated with poorer physical and mental health, more anxiety and depressive symptoms, impaired daily functioning, and lower happiness.
Shortness of breath measured with BDI and MRC was also a predictor of death within a period of eight years when age and gender were controlled for in the analysis. Especially in the case of death due to cardiovascular or lung disease, the more dyspnea complaints, the greater the mortality risk.
Analysis of the problem lists of all patients involved over the eight-year period revealed that those with active problems in the cardiovascular and respiratory systems were overrepresented among the respondents. Selective participation of patients with more respiratory and/or cardiovascular problems may have resulted in some overestimation of the prevalence of dyspnea complaints in elderly family practice patients; nevertheless, considering the 74% response rate for the MRC dyspnea questionnaire, real prevalence rates are unlikely to be substantially lower. Overrepresentation of patients with more active problems in the cardiovascular and respiratory systems very probably means that study respondents had more known risk factors at baseline. The high percentage of excluded patients who died is understandable, considering that they were excluded due to serious illness, hospitalization, or dementia at baseline. Excluded patients were not, however, more likely than included patients to die of cardiovascular or lung diseases. Given the morbidity and mortality profiles of respondents, non-respondents and excluded subjects, there is no compelling reason to suspect that the outcomes of the study, especially in relation to death prognosis, are biased.
A strength of the current study is the length of the follow-up period. We were able to follow all but a small number of participants and non-participants over an eight-year period, which provided an ample time window for assessing the predictive value of dyspnea symptoms for mortality. Moreover, the design of the study, including the choice of variables, is compatible with the family practice setting and the perspectives of the family practitioner with regard to medical aspects and the daily life of the patients. A limitation is the small sample size. This makes the finding of a significant predictive effect of self-reported dyspnea symptoms on mortality even more striking. Our findings are in line with those of recent studies: dyspnea occurs frequently in the elderly, is associated with poor health, and interferes with daily functioning.3 The current findings indicate that dyspnea is a risk factor for mortality.2 For this reason, clear guidelines for practitioners dealing with symptoms of dyspnea seem highly desirable. We believe that early diagnosis is valuable, not only because of the underlying cardiovascular or pulmonary pathology, but also because the family practitioner may have opportunities to positively influence the patient's functional condition.4,15 In this regard, it is important to note that relatively few people who experience shortness of breath actually visit their family practitioner for this complaint.15 We therefore recommend that, as one of the first steps, family practitioners should add dyspnea as a complaint to the structured inventory of patient problems, even in cases where the medical background for the complaint has been identified (a so-called double registration). This would increase the probability that the practitioner will focus attention not only on the medical background but also on the patient's perceived physical and mental health, functional impairment, and quality of life, thus encouraging interventions that target patient-centred outcomes with respect to dyspnea. This might help to bring the patient's point of view (back) into the centre of clinical practice.16 The short MRC questionnaire may prove useful for GPs in assessing changes in the severity of dyspnea-related symptoms over time.6
Based on the literature as well as the current findings, we believe that integrated shared care systems need to be established for patients with dyspnea complaints. Efforts are underway to establish such a program in Maastricht. The aim of the collaborative program, involving cardiology and pulmonology divisions of the academic hospital, specially trained nurses, and family practitioners, is to develop more effective and efficient diagnostic procedures and supportive interventions for this relatively large subgroup of patients. Ultimately, this could lead not only to early diagnosis and more effective medical interventions, but also to improvements in the daily functioning of patients with dyspnea. Despite our optimism concerning guideline and program development, we realize that guidelines do not always improve clinical practice and that early detection does not automatically improve health outcomes. Further research is therefore necessary to determine whether early detection of dyspnea and treatment of this (secondary) symptom do indeed improve self-reported health, daily functioning, quality of life, and survival.
|
Declaration |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
Funding: the project was conducted without external funding.
Ethical approval: none.
Conflicts of interest: none.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
Medical Research Council Dyspnea Scale6 Grade/Description
- 0. Not troubled with breathlessness except with strenuous exercise.
- 1. Troubled by shortness of breath when hurrying on the level or walking up a slight hill.
- 2. Walks slower than people of the same age on the level because of breathlessness or has to stop for breath walking at own pace on the level.
- 3. Stops for breath after walking about 100 yards or after a few minutes on the level.
- 4. Too breathless to leave the house or breathless when dressing or undressing.
- 1. Troubled by shortness of breath when hurrying on the level or walking up a slight hill.
|
Acknowledgments |
|---|
We are grateful to the patients who filled in the questionnaire and to those who also took part in the home visit. We thank Dr Jan Joosten for his help in updating the literature references.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1References |
|---|
1 Landahl S, Steen B, Svanborg A. Dyspnea in 70-year-old people. Acta Med Scand 1980; 207: 225–230.[ISI][Medline]
2 Tessier JF, Nejjari C, Letenneur L, Filleul ML, Marty MI, Baarberger Gateau P, Dartigues JF. Dyspnea and 8-year mortality among elderly men and women: The PAQUID cohort study. Eur J Epidemiol 2001; 17: 223–229.[CrossRef][Medline]
3 Barberger-Gateau P, Chaslerie A, Dartigues JF, Commenges D, Gagnon M, Salamon R. Health measures correlates in a French elderly community population: the PAQUID study. J Gerontol Soc Sci 1992; 47: S88–S95.
4 Van Grunsven P. Behandeling van acute, ernstige dyspnea bij Astma en COPD in de huisartspraktijk: een literatuuronderzoek. [Treatment of acute serious dyspnea in asthma and COPD in general gractice: a literature review.] Huisarts Wet 1997; 40: 54–62.
5 Metsemakers JFM, Hoppener P, Knottnerus JA, Kocken RJJ, Limonard CBG. Computerized health information in the Netherlands: a registration network of family practices. Br J Gen Pract 1992; 42: 102–106.[ISI][Medline]
6 Mahler DA, Wells CK. Evaluation of clinical methods for rating dyspnea. Chest 1988; 93: 580–586.[Medline]
7 Silvestri GA, Mahler DA. Evaluation of dyspnea in the elderly patient. Clin Chest Med 1993; 14: 393–404.[Medline]
8 Mahler DA, Gyatt GH, Jones PW. Clinical measurement of dyspnea. In Mahler DA (ed.) Dyspnea. New York: Marcel Dekker; 1997, 149–198.
9 Mahler DA, Weinberg DH, Wells CK, Feinstein AR. The measurement of dyspnea. Contents, interobserver agreement, and physiologic correlates of two new clinical indexes. Chest 1984; 85: 751–758.
10 Spielberger CD, Gorsuch RL, Lushene PR, Vagg RR, Jacobs AG. STAI, Manual for the State-Trait-Anxiety Inventory (Self-Evaluation Questionnaire). Palo Alto: Consulting Psychologists Press; 1983.
11 Chattat R, Ellenal, Curcinotta D, Savorani G, Mucciarelli G. A study on the validity of different short versions of the Geriatric Depression Scale. Arch Gerontol Geriatr 2001; suppl 7: 81–86.
12 Ware JE, Sherbourne CD. The MOS 36-item Short Form Health Survey (SF-36). Med Care 1992; 30: 473–83.[ISI][Medline]
13 De Bruin AF, Diederiks JPM, de Witte IP, Stevens FC, Philipsen H. The development of a short generic version of the Sickness Impact Profile. J Clin Epidemiol 1994; 47: 407–418.[CrossRef][ISI][Medline]
14 De Boer AG, van Lanschot JJ, Stalmeier PF, van Sandick JW, Hulscher JB, Haes JC, Sprangers M. Is a single-item visual analogue scale as valid, reliable and response as multi-item scales in measuring quality of life. Qual Life Res 2004; 13: 311–320.[CrossRef][ISI][Medline]
15 Thoonen BPA, van Weel C. Kortademigheid. [Shortness of breath.] Huisarts Wet 2002; 45: 414–419.
16 Sullivan M. The new subjective medicine: taking the patient's point of view on health care and health. Soc Sci Med 2003; 56: 1595–1604.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||