The disablement process in patients with pulmonary disease.

Health has been defined by the World Health Organization (WHO) as "a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity."[1] This definition suggests a model for conceptualizing health that is multidimensional and goes beyond a definition that focuses strictly on disease. Nagi[2] has proposed a multidimensional framework focusing on the pathway from disease to its broad outcomes at the individual and societal levels. Conceptually, the pathway progresses from pathology (biochemical or physiological dysfunction at the cellular level), to impairment (dysfunction in body systems), to functional limitation (restriction in mental or physical actions at the level of the individual), to disability (problems doing daily actvities within the context of society). This model distiguishes actions, such as walking or standing, from activities that require multiple actions, such as bathing or working.

An expanded model for conceptualizing the progression from disease to disability has been introduced by Verbrugge and Jette[3] (Fig. 1). Furthering the work of Nagi,[2] they described a disablement process that encompasses the functional consequences of pathology and the factors that affect the pattern, rate, and direction of those consequences. Their model characterized factors affecing disability as risk factors, extraindividual factors, and intraindividual factors. Risk factors are attributes of the individual that exist prior to initiation of the disablement process. Extraindividual factors are social and environmental factors imposed on the patient that can decrease, limit, or increase the level of disability. Medication, physical therapy, and job or home modifications are all considered extraindividual factors. Intra-individual factors are the individual's behavior, coping skills, and other accommodations made in response to the disablement process. The pathway is not unidiretional. For example, additional impairments can result from any functional limitations or they can be the consequences of disability. This process can result in a cycle of additional dysfunctions in other areas not initially associated with the original pathology. Because of the variety of influencing factors, this model suggests that similar pathologies will not result in similar functional limitations or levels of disability in all individuals.

The purposes of our study were (1) to describe the disabilities of patients with pulmonary disease as mesured by the Functional Status Questionnaire (FSQ) and (2) to elaborate on the disablement process in patients with pulmonary disease by examining the relationships among pulmonary impairments, functional limitation as measured by the 6-minute walk distance (6MWD), and disability as measured by the FSQ.[4]

Method

Subjects

Data were collected retrospectively from the records of 154 patients enrolled in an outpatient pulmonary rehabilitation program at Boston's Beth Israel Hospital over a 1-year period (1994-1995). All subjects were medically stable and were referred by their physicians with the major complaint of difficulty performing desired activities due to dyspnea. The routine initial physical therapy examination included determination of the 6MWD and completion of the ESQ by the patient. Patients who could not speak English or who could not read or understand the questions did not complete the FSQ. All patients who had completed a 6MWD test and the FSQ, and for whom information on other factors of interest for the study was available, were included. See Table 1 for a description of the subjects.

Table 1.
Patient Characteristics(a) (N=54)

Age (y)
  X                                          59
  SD                                         14
  Range                                      24-86
Gender (%)
  Female                                     64
  Male                                       36
Work status (%)
  Not working due to health                  50
  Working not working due to reasons other
   than health                               50
Disease classification
  Restrictive                                16
  obstructive                                84
% FVC
  X                                          62
  SD                                         19
  Range                                      26-116
% FE[V.sub.1]
  X                                          44
  SD                                         20
  Range                                      10-110
FE[V.sub.1]/FVC
  X                                          57
  SD                                         20
  Range                                      16-112
Comorbidity (%)
  None                                       34
  One system                                 39
  More than one system                       27
 BMI (%)
   [is greater than] 27                      33
   20-27                                     46
   [is less than] 20
 Activity Sa[O.sub.2] (%)
   [is less than] 88%                        23
   [is greater than] 87%                     77

(a) % FVC=percentage of predicted forced vital capacity, %FE[V.sub.1]=percentage of predicted 1-second forced expiratory volume, FE[V.sub.1]/FVC=ratio of 1-second forced expiratory volume to forced vital capacity, BMI=body mass index, Sa[o.sub.2]=activity oxyhemoglobin saturation.

Procedure

Data concerning risk factors and intraindividual factors that potentially might affect the development of functional limitations and disability were collected from physical therapy records. The data included each patien's age, gender, height, weight, and comorbidities. Height and weight were used to calculate the body mass index (BMI), and patients were classified as having a low ([is less than] 20), high ([is greater than] 27), or normal (20-27) BMI. The medical history was used to classify patients as having no comorbidity, comorbidity in one organ system, or comorbidity in more than one organ system.

The following measures of pulmonary impairment were determined: activity oxyhemoglobin saturation (Sa[o.sub.2]), percentage of predicted forced vital capacity (%FVC), percentage of predicted 1-second forced expiratory volume (%FE[V.sub.1]), and 1-second forced expiratory volume (FE[V.sub.1]) to forced vital capacity (FVC) ratio (FE[V.sub.1]/FVC). The Sa[o.sub.2] was measured continuously during a 6MWD test with a Nellcor digital pulse oximeter.(*) The lowest Sa[o.sub.2] during the walk was recorded, and patients were classified has having an Sa[o.sub.2] of less than 88% or greater than 87%. Measurement of Sa[o.sub.2] with a digital pulse oximeter has been shown to correlate well with invasive measures of Sa[o.sub.2] (r=.80)[5] and to have a test-retest coefficient of variation of 2.5% during exercise.[6]

Pulmonary function tests were performed using a Collins Survey Spirometer [dagger] and a standardized protocol[7]; FVC and FE[V.sub.1] were recorded. Only one maneuver was performed unless the patient did not understand or follow instructions on the first maneuver. Almost all of the patients had performed spirometric measurement previously and were familiar with the maneuver. Although learning and effort affect the test results, the mean within-subject coefficient of variation for measures of FE[V.sub.1], has been reported as 10.2%.[8]

Predicted FE[V.sub.1] was calculated separately for men and women using prediction equations derived by Knudson et al.[9] The %FVC and %FE[V.sub.1] were calculated by dividing the actual value by the predicted value for each patient. The FE[V.sub.1]/FVC ratio was also calculated.

Functional limitation was defined by the 6MWD test.[10] All patients entering the pulmonary rehabilitation program completed one 6MWD trial as a routine part of the initial evaluation. Patients were asked to walk as far as possible in ra minutes. A physical therapist followed patients during the walk while monitoring with pulse oximetry. No encouragement was given, and patients were told to choose their own pace, resting and starting again as needed regardless of their Sa[o.sub.2] during the activity. Patients used their usual assistive walking devices. Those patients who routinely used supplemental oxygen used their usual activity flow rates during the walk, with the therapist carrying the oxygen tank. Patients who did not have a prescription for supplemental oxygen completed the 6MWD test without supplmental oxygen. Six minutes was recorded with a stopwatch, and the distance walked (in feet) was determined at the end of the time period. Tests were terminated by the physical therapist only if there were complaints of angina, dizziness, light-headedness, or instability. If patients chose to stop walking prior to the end of 6 minutes, they rested in the corridor until the completion of 6 minutes and the distance was recorded. The test was conducted within a hospital corridor 137.7 m (455 ft) long, and one 6MWD trial was performed by each patient. The 6MWD test has been shown to be a reliable measure in patients with chronic heart or lung disease (intraclass correlation coefficient=.92 for six walks).[11]

Disability was assessed using the FSQ.[4] Patients completed the FSQ in the waiting room at their first visit or at home after their first visit, but prior to the second visit. The FSQ is a self-administered health status questionnaire that is used in an outpatient setting and that is generic, meaning that it was designed to be used for patients regardless of the type of health-related problem being assessed. The FSQ takes approximately 10 to 15 minutes to complete. Patients are queried about how their health has affected a variety of activities over the past month.

The FSQ provides scores for five scales: basic activities of daily living (BADL) (3 items), instrumental activities of daily living (IADL) (6 items) items@, mental health (5 items), social activity (3 items), and quality of interaction (5 items). The patient is asked to categorize his or her work situation, and if the patient is working, a work performance score is calculated (6 items). Additionally, there are 6 items that are not included within the FSQ scales. Examples of questions asked in summary scales are presented in Table 2.

[TABULAR DATA NOT REPRODUCIBLE IN ASCII]

Scales are scored separately, and the scores for each scale range from 0 to 100, with 100 representing the highest level of health. Raw scores can be entered into a computer program that generates a report placing the patient's score for each scale within the context of benchmarks, or "warning zones," which have been established by clinical consensus for determining when scores should prompt concern on the part of the health care provider.[4]

The FSQ scales have been found to be valid and reliable.[4] The internal consistency reliability ranged from .64 to .82 in a previous report.[4] The BADL, IADL, and mental health scores achieved the highest reliability.

Data Analysis

Multivariate analyses were used to determine the relationships between functional limitation and impairments, disability and impairments, and disability and functional limitation. To determine the relationship between functional limitation and impairments, four multivariate analyses were done with 6MWD as the dependent variable and FE[V.sub.1]/FVC ratio, %FVC, %FE[V.sub.1] or activity Sa[o.sub.2] as the independent variable of interest. To determine the relationship between disability and impairments, 24 separate multivariate analyses were carried out with FSQ scales as the dependent measures and FE[V.sub.1]/FVC ratio, %FVC, %FE[V.sub.1], or activity Sa[o.sub.2] as the independent variable of interest. To determine the relationship between disability and functional limitation, six separate multivariate analyses were conducted with FSQ scale scores as the dependent measures and 6MWD as the independent variable of interest. All analyses controlled for the effects of the potential risk factors and intraindividual factors identified from the patients, records. These factors, entered into the analyses as additional independent variables, included age, gender, BMI, and comorbidity. For analyses involving five of the FSQ scales, general linear models were used. For analyses involving the work performance item of the FSQ logistic regression was used because patients were classified as to whether or not they had left work because of health problems. An alpha level of .05 was used in all analyses. All analyses were performed using SAS software. [double dagger]

Results

Figure 2 illustrates the pattern and magnitude of disability of patients with chronic pulmonary disease. Mean FSQ scores were 83.3 (SD=17.9, range=22-100) for BADL, 52.6 (SD=24.4, range=0-100) for IADL, 65.8 (SD=18.5, range=20-100) for mental health, 64.8 (SD=32.8, range=0-100) for social activity, and 77.5 (SD=14.1, range=36.100) for quality of interaction. Mean scores were below the benchmark scores that prompt concern in BADL, IADL, and social activity. The mean 6MWD was 307.2 m (SD=131.7, range=25.9-630.1) (1.008 ft [SD=432, range=85-2-100]). Table 3 summarizes the regression models that described the relationships between pulmonary impairment, functional limitation, and disability,

[TABULAR DATA NOT REPRODUCIBLE IN ASCII]

Relationship of Impairment to Functional Limitation

The %FE[V.sub.1], one measure of impairment, was positively related to the 6MWD; as the %FE[V.sub.1] increased, the distance walked increased. Age and BMI also contributed to the model. Walking distance declined as age increased, and a high BMI was associated with decreased 6MWD (model [R.sup.2]=.21). The %FVC was also associated with 6MWD, with walking distance increasing as the %FVC increased ([R.sup.2]=.21). The FE[V.sub.1]/FVC ratio, another measure of impairment, was also associated with 6MWD. A higher FE[V.sub.1]/FVC ratio was associated with a greater 6MWD. Increasing age was also associated with a lower 6MWD in this model, as were both high and low BMI classifications ([R.sup.2]=.10). Finally, a low Sa[o.sub.2] was associated with a decreased 6MWD. Age was also a factor in this model, with greater age associated with a decreased 6MWD ([R.sup.2]=.10).

Relationship of Impairment to Disability

None of the models designed to analyze the relationship between impairment measures and FSQ scores were significant (model [R.sup.2]=.03-,.08).

Relationship of Functional Limitation to Disability

A common measure of functional limitation, 6MWD, was positively related to BADL, LADL, and social activity. Subjects who walked farther tended to have better ability in BADL (model ([R.sup.2]=.24). The 6MWD was positively related to IADL; subjects walking farther tended to have better IADL ability (model ([R.sup.2]=.35). The 6MWD was positively associated with social activity; subjects who walked farther reported better social function (model ([R.sup.2]=.26). The 6MWD was also related to work status, with subjects who were able to walk farther being less likely to have left work due to health reasons. Age also contributed to this model, with younger patients being less likely to have left work due to health reasons. The 6MWD was not related to the mental health or quality-of-interaction scales of the FSQ.

Discussion

The results of this study support the tenets of the disablement process model (Fig. 3). The model suggests that impairments will be more closely related to functional limitations than to disability and that functional limitations will be related to disability. In addition, the relationship will be affected by factors present in the individual such as age and gender. Our findings suggest that in this group of patients, the measures of impairment (%FE[V.sub.1], %FVC, FE[V.sub.1]/FVC, and Sa[o.sub.2] that are related to functional limitation (6MWD) are not directly related to measures of disability (FSQ scales). In addition, a common measure of functional limitation (6MWD) was moderately associated with several measures of disability (BADL, IADL, social activity, and work status). The models tested also lend support to the idea that there is an effect of risk factors (eg, age) and intraindividual factors (eg, BMI) on the associations between impairments and functional limitation.

Reports of the relationship of pulmonary impairment and walking distance vary widely.[12-14] Alonso et al[12] reported a correlation of .23 between %FE[V.sub.1], and 6MWD. Swinburn et al[13] and McGavin et al[14] found correlations of .406 and 0.13, respectively, between actual FVC and 12-minute walk distance (12MWD); correlations of FE[V.sub.1], with 12MWD were .283 and .17, respectively. Discrepancies may be accounted for by age and gender of subjects, disease severity, and differences in the pulmonary impairment measures examined. In our study, analyses accounted for potentially confounding variables such as age and gender.

We found that pulmonary impairments were not associated with disability (FSQ scale scores). Similarly, Alonso et al[12] also found that neither %FE[V.sub.1] nor arterial oxygen saturation was related to the total score on a generic health status instrument. Mahler et al[15] found that measures of pulmonary function (%FVC and %FE[V.sub.1]) were related to disability. In univariate analyses, %FE[V.sub.1] and %FVC were moderately associated with the physical function, role function, and health perceptions scale of the Medical Outcomes Study 20-Item Short-Form Health Survey (SF-20). In a later study, Mahler and Mackowiak[16] reported similar findings using the 36-Item Short-Form Health Survey (SF-36). The physical function scale, which is identical in the SF-20 and SF-36, includes items similar to those found in the IADL and BADL scales of the FSQ The disagreement between our findings and those of the previous studies[15,16] may be related to the fact that the patients in each previous study were carefully selected and those with almost any other kind of medical problem were excluded. In spite of our controlling for number of comorbidities, it is likely that the results of our analyses examining the relationship between disability and impairments were affected by the fact that 66% of our patients had impairments, other than pulmonary impairments, that could have had an impact on disability. Our findings, however, probably more closely resemble the actual situation of patients with pulmonary disease than the effect reported by Mahler et al[15] or Mahler and Mackowiak.[16] In addition, our analyses were multivariate, controlling for potentially confounding variables such as comorbidity and BMI. Jones et al[17] found that %FE[V.sub.1] and %FVC, but not Sa[o.sub.2], were related to Sickness Impact Profile (SIP) scores in univariate analyses. When 6MWD was added in a multiple regression analysis, there was no effect of pulmonary impairment.

One previous study examining the relationship of a functional limitation (timed walk) to disability measured by a generic instrument supports our findings. Jones et al[17] reported that 6MWD accounted for approximately 40% of variability in total SIP score. Guyatt et al[11] reported moderate relationships (r=.47-.59) between 6MWD and several measures of function designed for use with patients with cardiovascular and pulmonary diseases. Other researchers[18,19] have reported the relationship of exercise impairment (oxygen consumption), but not functional limitation, to disability, and the correlations were low.

Our finding that 6MWD accounts for more of the variability in IADL than in any other disability scale may be due to the fact that the activities assessed by the IADL scale involve a degree of energy expenditure similar to walking. Two of the six items in that scale assess the ability to walk under different circumstances. Recent studies have shown a moderate correlation of 6MWD and oxygen consumption in patients with end-stage lung disease[20] and advanced heart failure.[21]

Although the BADL scale also measures the ability to accomplish physical activities, 6MWD accounted for a smaller degree of variability in this scale than in the IADL scale. The BADL, involving toileting, bathing, and dressing, may require less oxygen consumption than IADL, and often do not require walking. Furthermore, the activities are at a more personal level than IADL and their private nature may create added incentive to complete them independently despite energy costs. Such activities are not discretionary as are some IADL.

Social interaction may very well involve energy expenditure at a level similar to that required for walking. In order to visit friends or attend religious services, individuals need to be mobile in a less-controlled environment than at home, that is, an environment in which stairs, ramps, uneven terrain, and scarcity of restrooms may be encountered.

We found that factors we believed might contribute to the degree of disability, such as comorbidity and BMI, did not add to the variability in disability scales once walking distance was accounted for. As these variables help to explain the 6MWD, they most likely have an indirect effect on disability.

Factors that were not included in our database but that might contribute to the explanation of disability in this population include risk factors and intraindividual and extraindividual factors such as psychological state, dyspnea, socioeconomic level, education level, and medications. Prigatano et al[19] found that along with aerobic capacity, measures of tension and anxiety and education level predicted total SIP score. Other researchers have also reported the effect of psychological state on health status in patients with pulmonary disease. Jones et al[22] found that a measure of depression contributed to the variability in total SIP score in a model that included 6MWD and a measure of dyspnea. Weaver and Narsavage[23] also found that 12MWD and depression were related to functional status. Alonso et al,[12] however, found that of the factors they considered in multivariate analyses, including 6MWD, age, and impairment measures, only dyspnea contributed to a global measure of health provided by the Nottingham Health Profile. Similarly, using multivariate analyses, Mahler and et al[15] and Mahler and Mackowiak[16] found that dyspnea was correlated with the SF-20 and SF-36.

Our study also demonstrates the usefulness of measuring health using a generic instrument that provides a health profile, allowing the practitioner to determine in which areas a patient is having the greatest difficulty. Our patients demonstrated difficulty in performance of IADL and social activity. Previous studies using the SIP to measure health in patients with pulmonary disease have shown results inconsistent with our findings: the psychosocial summary score was worse than the physical health summary score.[22,23] Recreation/pastime when measured as a separate scale of the SIP showed the greatest loss. Some of this loss may be measured in the social activity scale of the FSQ The pattem and magnitude of disability of patients with pulmonary disease demonstrated when health has been measured using the SF-36 are somewhat consistent with those found in our study.[12,16] Scales measuring the physical dimension of health had poorer scores than scales measuring the psychological dimension of health. Social functioning, however, was one of the better scoring scales. Patients in these studies, however, were selectively chosen to exclude patients with comorbidities or, as in the study by Alonso et al,[12] patients using supplemental oxygen. We believe that the patients in our study better represent a cross section of patients with pulmonary diseases seen in most outpatient clinical settings.

Limitations

Our study was observational in nature and was based on analysis of a clinical database. These data were generated for the major purposes of patient evaluation and clinical decision making. The issues related to the use of clinical databases have been clearly oudined by Pryor and Lee[24] and include the use of data to answer questions not determined a priori, missing observations, selection bias, referral bias, and other biases that may limit the validity. For example, patients unable to read English and those incapable of comprehending the questions did not complete the FSQ and therefore were excluded from the sample. Such patients may be very different from those included in the analyses due to cultural or class influences. In addition, some variables, such as comorbidity, were reduced to facilitate analysis and interpretation. Such reduction results in the loss of some information and the increased chance of misclassification. Certainly, further studies are needed to further elucidate the disablement process in patients with pulmonary disease.

Additional limitations relate to the timing of the administration of the tests and the reliability of the 6MWD test. The initial evaluation of patients often spanned more than one visit, and patients may not have completed the FSQ and the 6MWD test within the same visit. A significant change in functional limitation or self-perceived health was not expected, however, because no treatment was given during the interval and the interval was usually no more than 1 week. Although walking distance and reliability of the walking test have been shown to improve with a second trial,[13,14] subjects in our study completed the 6MWD test only one time due to clinical time constraints. This limitation most likely resulted in the 6MWD being lower for our patients than for those in other studies in which multiple trials of a walking test have been performed. As most patients are likely to improve with additional trials and improvements are not likely to differ in any systematic way among patients, the relationships described in this report are not likely to be substantially different from those that might have been obtained had measurements from a second or third trial been used in the analyses. Swinburn et al[13] demonstrated that correlations between walking distance and measures of pulmonary impairment did not change with repeat testing.

Conclusion

This study is the first to our knowledge to examine the relationship between the 6MWD test, a measure of functional limitation, and the FSQ a generic measure of health status, among patients with pulmonary disease. Our results support the use of the disablement process model, showing that common measures of pulmonary impairment do not explain disability as well as does a measure of functional limitation. The risk factors and intraindividual and extraindividual factors that affect this relationship need further examination, as those included in this study did not contribute in any consistent manner to explaining the degree of functional limitation associated with impairments or the degree of disability associated with functional limitation seen in our patients. Patients showed clinically important dysfunction in BADL, IADL, and social activity. These findings strongly suggest that health care providers should consider explicitly measuring all domains of a patient's health. Future research should more fully explore and test the disablement process model and the factors affecting the development of disability in patients with pulmonary disease. Ultimately, both explicit measurement and an improved understanding of the meaning and process of disablement may help health care practitioners to clarify patients, goals and their attainment and to focus treatment through improved decision making.

References

[1] Constitution of the World Health Organization. In: Basic Documents. Geneva, Switzerland: World Health Organization; 1948.

[2] Nagi S. Some conceptual issues in disability and rehabilitation. In. Sussman M, ed. Sociology and Rehabilitation. Washington, DC: American Sociological Association; 1965:100-113.

[3] Verbrugge LM, Jette AM. The disablement process. Soc Sci Med. 1993;38:1-14.

[4] Jette AM, Davies AR, Cleary PD, et al. The Functional Status Questionnaire: reliability and validity when used in primary care. J Gen Intern Med. 1986;1:143-149.

[5] Escourrou PJ, Delaperche MF, Vissueaux A. Reliability of pulse oximetry during exercise in pulmonary patients. Chest. 1990;97: 635-638.

[6] Marciniuk DD, Watts RE, Gallagher CG. Reproducibility of incremental maximal cycle ergometer testing in patients with restrictive lung disease. Thorax. 1993;48:894-898.

[7] American Thoracic Society. Standardization of spirometry: 1987 update. Am Rev Respir Dis. 1987;136:1285-1289.

[8] Noseda A, Carpiaux JP, Prigogine T, Schmerber J. Lung function, maximum and submaximum exercise testing in COPD patients: reproducibility over a long interval. Lung. 1989;167:247-257.

[9] Knudson RJ, Lebowitz MD, Holberg CJ, et al. Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis. 1983;127:725-734.

[10] Butland RJA, Pang J, Gross ER, et al. Two-, six-, and twelve-minute walking tests in respiratory disease. BMJ. 1982;284:1607-1608.

[11] Guyatt GH, Thompson PJ, Berman LB, et al. How should we measure function in patients with chronic heart and lung disease? J Chronic Dis. 1985;38:517-524.

[12] Alonso J, Anto JM, Gonzalez M, et al. Measurement of general health status of non-oxygen dependent chronic obstructive pulmonary disease patients. Med Care. 1992;30(suppl):MS125-MS135.

[13] Swinburn CR, Wakefield JM, Jones PW. Performance, ventilation, and oxygen consumption in three different types of exercise test in patients with chronic obstructive pulmonary disease. Thorax. 1985;40: 581-586.

[14] McGavin CR, Gupta SP, McHardy GJR. Twelve-minute walking test for assessing disability in chronic bronchitis. BMJ. 1976;1:822-823.

[15] Mahler DA, Faryniarz K, Tomlinson D, et al. Inpact of dyspnea and physiological function on general health status in patients with chronic obstructive pulmonary disease. Chest. 1992;102:395-401.

[16] Mahler DA, Mackowiak JI. Evaluation of the short-form 36-item questionnaire to measure health-related quality of life in patients with COPD. Chest. 1995;107:1585-1589.

[17] Jones PW, Quirk FH, Baveystock CM. The St George's respiratory questionnaire. Respir Med. 1991; 85 (suppl B):25-31.

[18] Carlson DJ, Ries AL, Kaplan RM. Prediction of maximum exercise tolerance in patients with COPD. Chest. 1991;100:307-311.

[19] Prigatano GP, Wright EC, Levin D. Quality of life and its predictors in patients with mild hypoxemia and chronic obstructive pulmonary disease. Arch Intern Med. 1984;144:1613-1619.

[20] Cahalin LP, Pappagianopoulos P, Prevost S, et al. The relationship of the 6-minute walk test to maximal oxygen consumption in transplant candidates with end-stage lung disease. Chest. 1995;108:452-459.

[21] Cahalin LP, Mathier MA, Semigran MJ, et al. The six-minute walk test predicts peak oxygen uptake and survival in patients with advanced heart failure. Chest. 1996;110:325-332.

[22] Jones PW, Baveystock CM, Littlejohns P. Relationships between general health measured with the sickness impact profile and respiratory symptoms, physiological measures, and mood in patients with chronic airflow limitation. Am Rev Respir Dis. 1989;140:1538-1543.

[23] Weaver TE, Narsavage GL. Physiological and psychological variables related to functional status in chronic obstructive pulmonary disease. Nurs Res. 1992;41:286-291.

[24] Pryor DB, Lee KL. Methods for the analysis and assessment of clinical databases: the clinician's perspective. Stat Med. 1991;10:617-628.