| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2000;31:2610.)
© 2000 American Heart Association, Inc.
Original Contributions |
Clinical Meaning of the Stroke-Adapted Sickness Impact Profile–30 and the Sickness Impact Profile–136
From the Departments of Social Medicine (A. van S., G.A.M. van den B.), Clinical Epidemiology and Biostatistics (R.J. de H.), and Clinical Informatics (M.L.), Academic Medical Center, University of Amsterdam; Netherlands Heart Foundation (M.L.); and Department for Health Services Research, National Institute of Public Health and the Environment, Bilthoven (G.A.M. van den B.), Netherlands.
Correspondence to A. van Straten, PhD, Trimbos Institute, PO Box 725, 3500 AS Utrecht, Netherlands. E-mail astraten{at}trimbos.nl
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose—Handicap or health-related quality of life (HRQL) measures are seldom used in stroke trials, although the importance of these measures has been stressed frequently. We studied the clinical meaning of the Stroke-Adapted Sickness Impact Profile–30 (SA-SIP30) and the original SIP136 for use in stroke research.
Methods—We included 418 patients who had had a stroke 6 months earlier. We studied the associations between the SA-SIP30 and SIP136 scores versus other frequently used outcome measures from the International Classification of Impairments, Disabilities, and Handicaps (ICIDH) (Barthel Index, Rankin Scale) and the HRQL model (health perception items, Euroqol). To interpret the continuous SA-SIP30 and SIP136 scores, we used receiver operating characteristic curve analysis with the aforementioned measures as external criteria.
Results—The psychosocial dimension scores of both SIP versions remained largely unexplained. The physical dimension and total scores of both SIP versions were mainly associated with the disability measures derived from the ICIDH model, as well as with the physical HRQL domains. Most patients with an SA-SIP30 total score >33 or an SIP136 total score >22 had poor health profiles. There were no major differences between the SA-SIP30 and the SIP136, although the SA-SIP30 scores were less skewed toward the healthier outcomes than the SIP136.
Conclusions—Our study showed that (1) both SIP total scores primarily represent aspects of physical functioning and not HRQL; (2) both SIP versions provide more clinical information than the frequently used disability measures; and (3) the SA-SIP30 should be preferred over the SIP136.
Key Words: health • psychometrics • quality of life • stroke • validity
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Stroke may have a devastating impact on patients’ lives. In stroke research, this impact should be accurately assessed. Since many patients die after a stroke, patient outcomes should always be expressed in terms of mortality. The health of the survivors can be hierarchically described according to the International Classification of Impairments, Disabilities, and Handicaps (ICIDH)1 or in terms of Health-Related Quality of Life (HRQL). The ICIDH classification defines impairments as disturbances at the level of the organ (eg, paresis, eye deviation, or dysarthria) and disabilities as the consequences of the impairments on a personal level. In general, disabilities are defined as limitations in (Instrumental) Activities of Daily Living [(I)ADL]. Handicap is defined as a disadvantage for a given individual, resulting from impairments or disabilities, that limits or prevents the fulfillment of a role that is normal for that individual. Although no formal HRQL definition exists, there is consensus that it should encompass physical, psychological, and social aspects of health.2 3 Moreover, HRQL is usually considered a personal and subjective evaluation of a patient’s health status. HRQL can be seen as an umbrella concept that embodies measures of impairment, disability, and handicap.4 5
In general, disabilities, handicap, and HRQL are the most meaningful for patients. Nevertheless, a recent review on stroke outcome measures, which included all acute stroke trials performed between 1955 and 1995, showed that death was recorded in 76%, impairment in 76%, disability in 42%, and handicap or HRQL in only 2% of the trials.6 There was a (statistically nonsignificant) trend for more trials to record disability over time. One of the reasons that HRQL is seldom assessed may be the lack of a comprehensive yet feasible instrument.
Quality of life measures can be differentiated into generic scales and disease-specific scales. Generic scales are not specifically developed for a specific target population and may be suitable for use with many patient populations. An advantage of generic scales is that they allow comparison of HRQL results across patient populations. Generic scales, however, are not always sufficiently focused on the specific problems of any given patient population. For this reason, disease-specific HRQL scales have been developed, which are often more sensitive to the HRQL issues that are particularly relevant to a specific patient group. At present, no stroke-specific HRQL scales exist.
Currently, the Sickness Impact Profile (SIP) is frequently used as a
generic HRQL measure in stroke research.7 It has been
demonstrated that the SIP is reliable and valid in many patient
populations.3 8 Its validity has also been demonstrated in
stroke patients, although data on the reliability in this patient group
are limited.9 10 11 However, the SIP has a major
disadvantage: its length. In stroke populations, it usually takes 
30
minutes to complete the 136 items. Therefore, we designed a
Stroke-Adapted Version of the Sickness Impact Profile, the
SA-SIP30.12 The 12 subscales and the 136 items of the
original SIP were reduced to 8 subscales with 30 items for the
SA-SIP30. We demonstrated that the considerable reduction of items is
associated with a relatively small loss of clinical information. The
SA-SIP30 scores could explain almost 90% of the variation in scores of
the original SIP. Furthermore, the reliability and the validity of the
SA-SIP30 were comparable to the psychometric properties of the original
SIP 136-item version.
Although the SA-SIP30 addresses the feasibility problem of assessing HRQL in stroke patients, 2 other issues remain. First, since the original SIP (and consequently the SA-SIP30) measures observable behavior instead of more subjective health perceptions, both SIP versions can be regarded as a measure of disability and not as a measure of HRQL.13 14 The second issue concerns the clinical interpretation of continuous scale scores. Since clinical trials often make use of dichotomous end points (eg, poor and good health outcome), more knowledge is needed regarding the clinical meaning of continuous scale scores of the SIP. Therefore, in this study we determined the clinical meaning of the SA-SIP30 and SIP136. First, we compared the SA-SIP30 and SIP136 with other frequently used stroke outcome measures derived from both the ICIDH model and the HRQL model. Second, we interpreted the continuous SA-SIP30 and SIP136 scale scores.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients
The study comprised 418 patients who had had a stroke 6 months earlier. They were the survivors of an original cohort of 760 consecutively admitted stroke patients. All 760 patients were admitted within 1 week after stroke onset to 1 of 23 randomly selected Dutch hospitals. Patients were traced by administrative records shortly after hospital discharge. Patients were considered to have had a stroke if there was a focal neurological deficit of sudden onset that lasted at least 24 hours with no known alternative to a vascular cause. Both first and recurrent strokes were included. Patients with a transient ischemic attack, a subarachnoid hemorrhage, or nonstroke pathology were excluded. Six months after stroke, all surviving patients were contacted for an interview at their place of residence.
Of all 760 patients, 258 died within the first 6 months after stroke. Of the 502 survivors, 17 patients refused to participate in the 6-month follow-up interview; 31 were interviewed by telephone because they declined a home visit, and SIP data were not collected. Finally, for 36 of the remaining 454 patients, either SIP data were not collected during the follow-up interview (because of the length of the interview) or the obtained SIP data were insufficient for analyses (arbitrarily defined as >10% of all SIP136 items missing or >50% of items missing on 1 subscale). Of the 418 studied patients, 99 were not communicative because of severe speech, language, or cognitive disorders. These patients were rated by a proxy respondent, primarily the partner.
The original SIP version consists of 136 dichotomous items that are grouped into 12 subscales: Body Care and Movement, Mobility, Ambulation, Social Interaction, Emotional Behavior, Alertness Behavior, Communication, Household Management, Sleep and Rest, Recreation and Pastimes, Eating, and Work. An aggregate score can be obtained for the total SIP136, as well as for each subscale individually. Additionally, 3 subscale scores can be aggregated into a physical dimension (Body Care and Movement, Mobility, Ambulation) and 4 into a psychosocial dimension score (Social Interaction, Emotional Behavior, Alertness Behavior, Communication). By convention, scores are presented as a percentage of maximal dysfunction, ranging from 0% to 100%. Therefore, higher scores indicate less desirable health outcomes. Since the majority of the patients (77%) did not work before their stroke occurred, the subscale Work was removed from all analyses.
The SA-SIP30 consists of 8 subscales: Body Care and Movement, Mobility, Ambulation, Social Interaction, Emotional Behavior, Alertness Behavior, Communication, and Household Management. The scoring of items, subscales, dimensions, and total score is the same as for the original SIP136 version. The scores are also presented as a percentage of maximal dysfunction, ranging from 0% to 100%.
Comparison of the SA-SIP30 and SIP136 With Other ICIDH and
HRQL Measures
We first determined the association between the various SIP
subscale scores on the one side and the total SIP scores and the
dimension SIP scores on the other side by multiple linear regression
analyses (forward selection procedure). Thereafter, we studied
the associations between the SIP and the measures of disability, global
functioning, health perceptions, and HRQL. Disabilities and global
functioning were assessed with the Barthel Index and the modified
Rankin Scale, respectively.15 16 17 The health perceptions
were based on 2 single items: "How would you rate your present
health?" and "How satisfied are you with your present life?"
Both questions were rated on a 5-point scale (from [very]
healthy/satisfied to moderately healthy/satisfied to [very]
unhealthy/dissatisfied). HRQL was assessed with the Euroqol,
encompassing 5 items: mobility, self-care, main activity,
pain/discomfort, and mood (anxiety and/or
depression).18 19 Each item was rated on a 3-point scale:
no problems, some problems, very severe problems. Furthermore, on the
basis of the 5 Euroqol dimensions, we computed a single index
score.20 This index score incorporates patients’ values
for the different health states. The associations were examined by
correlation analyses. We used Pearson’s correlation
coefficients since they were almost equivalent to the
nonparametric Spearman’s correlation coefficients.
Univariate associations were expressed as the percentage of
variance of the SIP scores that could be explained by the score on the
other outcomes.
Interpretation of Continuous SIP Scores
To interpret SIP continuous scores, we used receiver operating
characteristic (ROC) curve analysis. The ROC curve shows the
ability of the SIP to detect poor and good health outcomes by using an
external criterion. The curve depicts the true-positive rate
(sensitivity) and false-positive rate (1 minus specificity). The area
under the ROC curve represents the probability that a random
pair of patients will be correctly classified as having a good or a
poor outcome with the SIP. A value of 0.50 is obtained if the SIP does
not perform better than chance, and a value of 1.0 indicates perfect
accuracy. For this analysis, we dichotomized the criterion
scores for disability, global functioning, health perceptions, and HRQL
as follows: ADL dependent (Barthel Index <20); unable to live
independently (Rankin classification 3 to 5); feeling moderately to
(very) unhealthy (item score 3 to 5); feeling moderately to (very)
dissatisfied (item score 3 to 5); and poor HRQL (score 2 to 3 for each
item of the Euroqol; score below median [<0.34] for the single index
score).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subject Characteristics and Scores
The mean age of the 418 patients was 70 years (SD 13 years), and 55% were men. The type of stroke was unspecified in 24 of the 418 patients (6%). Of the remaining 394 patients, 48 (12%) suffered intracerebral hemorrhages and 346 patients (88%) suffered ischemic strokes; 57 of these were infratentorial strokes, 79 lacunar infarctions, and 210 (sub)cortical infarctions.
The mean SA-SIP30 total score for all patients was 32.0 (SD 20.7), and
the mean SIP136 score was 22.5 (SD 14.2). The distributions of both
total scores were skewed toward the healthier outcomes but to a greater
extent for the original SIP136 than for the SA-SIP30 (median
scores, 29.8 and 20.8 for the SA-SIP30 and SIP136, respectively)
(Figure
). This skewed distribution could
also be observed for the physical dimensions and especially for the
psychosocial dimensions (median physical dimension scores, 35.6 and
20.0 for the SA-SIP30 and the SIP136, respectively; median psychosocial
dimension scores, 20.0 and 15.4 for the SA-SIP30 and SIP136,
respectively).
|
Comparison of the SA-SIP30 and SIP136 With Other ICIDH and
HRQL Measures
Linear stepwise regression analyses showed that the SIP136
and SA-SIP30 physical dimension scores were primarily explained by
scores on the subscale Body Care and Movement, whereas the psychosocial
dimension scores were mainly explained by the subscale Social
Interaction (Table 1). The total scores
of both SIP versions were more strongly associated with the physical
subscales than with the psychosocial subscales.
|
The total scores of the 2 SIP versions, as well as the physical
dimension scores, were substantially associated with other physical
functioning measures: disabilities, global functioning, and (I)ADL
domains of the Euroqol (Table 2).
Furthermore, SIP scores were substantially correlated with the Euroqol
index score (the valuation of the Euroqol health states). No clear
associations could be demonstrated between both SIP versions and the
patients’ health perceptions or with the Euroqol dimensions of
pain/discomfort and mood. The psychosocial dimensions of the SIP
versions were partially correlated with physical functioning, health
ratings, and mood (Table 2).
|
Interpretation of Continuous SIP Scores
With the ROC curves, we identified SIP cutoff scores for poor
health outcomes. In general, patients with an SA-SIP30 total score of
>33 or an SIP136 total score of >22 were ADL disabled; unable to live
independently; experienced at least some problems in mobility,
self-care, and in performing their main activity; and valued their HRQL
as poor (Table 3). The same profile was
observed in patients with an SA-SIP30 physical dimension score of >40
or an SIP136 physical dimension score of >23. In regard to the SIP
psychosocial dimension scores, we could not demonstrate clear cutoff
scores for poor health outcomes.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We studied the clinical meaning of the SA-SIP30 and the original SIP136 by comparing them with other frequently used stroke outcome measures from the ICIDH model and the HRQL model. Furthermore, we also tried to determine which SIP scores might be interpreted as a poor health outcome.
In our study the psychosocial dimension scores of both SIP versions remained largely unexplained. The physical dimension scores of both SIP versions, however, were substantially associated with the disability measures derived from the ICIDH model, the physical domains of the Euroqol, and the Euroqol single index score. The results of the SIP total scores closely resemble those of the physical dimension scores. This is not surprising because the total scores of both SIP versions have a strong correlation with the SIP physical dimension scores and less of a correlation with the psychosocial scores.
Most patients with an SA-SIP30 physical dimension score >40 or an SIP136 physical dimension score >23 were ADL disabled, were unable to live independently, and experienced at least some problems with mobility and self-care. The same profile could be demonstrated for patients with an SA-SIP30 total score >33 or an SIP136 total score >22. No cutoff values for poor psychosocial functioning could be demonstrated.
Comparison of the SA-SIP30 and SIP136 With Other ICIDH and
HRQL Measures
In most stroke research, measures of impairment are inadequate for
describing health outcomes of surviving patients.21 The
Barthel Index has been suggested, and accepted, as the standard measure
of disability.22 23 The Rankin Scale, originally
presented as a handicap measure, is currently considered a
global measure of disability.5 16 However, limitations in
the higher levels of physical functioning are not fullycovered
by these measures.24 25 26 The Barthel Index merely focuses
on very basic daily activities, such as transfers and dressing. The
Rankin Scale provides a crude rating score only.
As stated in the introduction, it is generally acknowledged that HRQL measures should encompass physical, emotional, and social aspects of functioning. Although the SA-SIP30 and SIP136 aim to measure all 3 aspects, the psychosocial dimension scores could not be explained by other ICIDH measures or by any of the HRQL dimensions. Since we did not include measures of social functioning as an external criterion, we do not know to what extent the SA-SIP30 and SIP136 total scores measure social aspects of functioning. Further research is needed to reveal what is measured with the psychosocial dimension scores. Interestingly, the total scores of both SIP versions were mainly explained by the physical dimension of the SIP and to a lesser extent by the psychosocial dimension.
HRQL should also be a subjective evaluation of a patient’s health status. In this report we demonstrated that the SA-SIP30 and SIP136 total scores were barely associated with other psychological HRQL domains or with health perceptions as measured with the health and satisfaction rating. However, the total scores of both SIP versions were associated with the subjective evaluation of patients’ health states (single Euroqol index score). We conclude that both SIP total scores are largely based on physical disabilities, but apparently these physical disabilities are of major importance in explaining the patient’s valuation of HRQL. Since the SA-SIP30 and SIP136 describe the (physical) functioning of the patients in more detail than the Barthel Index and the Rankin Scale, we recommend the (additional) use of the short version of the SIP in stroke outcome research.
Interpretation of Continuous SIP Scores
To estimate an effect size in a clinical trial, many stroke
researchers prefer a dichotomous primary end point (eg, poor versus
good health outcome). This approach gives the investigator the
opportunity to compare 2 percentages of response rates, using
statistics such as relative risk, relative or absolute risk reduction,
or number needed to treat. We demonstrated which cutoff scores might be
used for the SA-SIP30 and SIP136. Therefore, by using the SA-SIP30 or
SIP136 it is possible not only to demonstrate a statistical difference
in the risks for poor outcome between 2 groups but also to designate
the involved area of functioning.
SA-SIP30 Versus SIP136
There were 2 differences between the results of the SA-SIP30 and
those of the original version of the SIP. The first is concerned with
the association between the dimension scores and the subscale scores.
The SIP136 dimension scores could be largely explained by 1 subscale
each (Body Care and Movement, Social Interaction). The SA-SIP30
dimension scores, however, were substantially related to all relevant
subscales. Second, the score distributions of the SA-SIP30 scores were
less skewed than the SIP136 scores. The skewed distribution of the
SIP136 toward good health outcomes implies that the long version of the
SIP is less able to discriminate between patients in relatively good
functional health. The skewed distribution of the SIP136 scores was
also demonstrated in other studies.10 27 28 Therefore, we
conclude that the feasible SA-SIP30 is to be preferred over the
original version of the SIP.
In summary, our analyses showed that (1) the original generic SIP and the stroke-adapted 30-item version of the SIP mainly measure aspects of disability instead of HRQL; (2) both SIP versions provide more clinical information than the frequently used disability measures; and (3) in stroke outcome research, the SA-SIP30 should be preferred over the SIP136.
|
Acknowledgments |
|---|
This study was supported by the Netherlands Organization of Scientific Research (NWO/KWAZO 900-571-032). Additional support was provided by the National Committee on Investigative Medicine of the Health Insurance Executive Board of the Netherlands (OG91-037) and the Netherlands Heart Foundation (NHS 40.004).
Received January 24, 2000; revision received June 26, 2000; accepted June 26, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
- World Health Organization. International Classification of Impairments, Disabilities and Handicaps: A Manual of Classification Relating to the Consequences of Disease. Geneva, Switzerland: WHO; 1980.
- Aaronson NK. Quality of life: what is it? how should it be measured? Oncology. 1988;2:69–76.[Medline] [Order article via Infotrieve]
- Gill TM, Feinstein AR. A critical appraisal of the quality-of-life measurements. JAMA. 1994;272:619–626.[Abstract]
- Ebrahim S. Clinical and public health perspectives and applications of health-related quality of life measurement. Soc Sci Med. 1995;41:1383–1394.
- Van den Bos GAM, Triemstra AHM. Quality of life as an instrument for need assessment and outcome assessment of health care in chronic patients. Qual Health Care. 1999;8:247–252.[Medline] [Order article via Infotrieve]
-
Roberts L, Counsell C. Assessment of clinical outcomes
in acute stroke trials. Stroke. 1998;29:986–991.
[Abstract/Free Full Text] - Bergner M, Bobbit RA, Carter WB, Gilson BS. The Sickness Impact Profile: development and final revision of a health status measure. Med Care. 1981;19:787–805.[Medline] [Order article via Infotrieve]
- Anderson RT, Aaronson NK, Wilkin D. Critical review of the international assessment of health related quality of life. Qual Life Res. 1993;2:369–395.[Medline] [Order article via Infotrieve]
- Schuling J, Greidanus J, Meyboom-De Jong B. Measuring functional status of stroke patients with the Sickness Impact Profile. Disabil Rehabil. 1993;15:19–23.[Medline] [Order article via Infotrieve]
- Nydevik I, Hulter-Asberg K. Sickness impact after stroke: a 3-year follow-up. Scand J Prim Health Care. 1992;10:284–289.[Medline] [Order article via Infotrieve]
-
De Haan R, Limburg M, van der Meulen J, Jacobs H,
Aaronson N. Quality of life after stroke: impact of stroke type and
lesion location. Stroke. 1995;26:402–408.
[Abstract/Free Full Text] -
Van Straten A, de Haan RJ, Limburg M, Schuling J,
Bossuyt PM, van den Bos GAM. A stroke-adapted 30-item version of the
Sickness Impact Profile to assess quality of life (SA-SIP30).
Stroke. 1997;28:2155–2161.
[Abstract/Free Full Text] - Essink-Bot M-L. Health Status as a Measure of Outcome of Disease and Treatment [thesis]. Rotterdam, Netherlands: Humanitas; 1995.
- Jurkovich G, Mock C, MacKenzie E, Burgess A, Cushing B, deLateur B, McAndrew M, Morris J, Swiontkowski M. The Sickness Impact Profile as a tool to evaluate functional outcome in trauma patients. J Trauma. 1995;39:625–631.[Medline] [Order article via Infotrieve]
- Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J. 1965;14:61–65.[Medline] [Order article via Infotrieve]
-
Van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJA,
van Gijn J. Interobserver agreement for the assessment of handicap
in stroke patients. Stroke. 1988;19:604–607.
[Abstract/Free Full Text] -
De Haan R, Limburg M, Bossuyt P, van der Meulen J,
Aaronson N. The clinical meaning of Rankin "handicap" grades.
Stroke. 1995;26:2027–2030.
[Abstract/Free Full Text] -
Dorman PJ, Waddell FM, Slattery J, Dennis MS,
Sandercock PAG. Is the Euroqol a valid measure of health-related
quality of life after stroke? Stroke. 1997;28:1876–1882.
[Abstract/Free Full Text] - The Euroqol Group. Euroqol: a new facility for the measurement of health related quality of life. Health Policy. 1990;16:199–208.[Medline] [Order article via Infotrieve]
- Dolan P. Modeling valuations for Euroqol health states. Med Care. 1997;35:1095–1108.[Medline] [Order article via Infotrieve]
-
De Haan R, Horn J, Limburg M, van der Meulen J, Bossuyt
P. A comparison of five stroke scales with measures of disability,
handicap, and quality of life. Stroke. 1993;24:1178–1181.
[Abstract/Free Full Text] - Wade DT, Collin C. The Barthel ADL Index: a standard measure of physical disability? Int Disabil Stud. 1988;10:64–67.[Medline] [Order article via Infotrieve]
-
Kwakkel G, Wagenaar RC, Kollen BJ, Lankhorst GJ.
Predicting disability in stroke: a critical review of the literature.
Age Ageing. 1996;25:479–489.
[Abstract/Free Full Text] - Anderson C. Baseline measures and outcome predictions. Neuroepidemiology. 1994;13:283–289.[Medline] [Order article via Infotrieve]
-
Greveson GC, Gray CS, French JM, James OFW. Long-term
outcome for patients and carers following hospital admission for
stroke. Age Ageing. 1991;20:337–344.
[Abstract/Free Full Text] -
Duncan PW, Samsa GP, Weinberger M, Goldstein LB, Bonito
A, Witter DM, Enarson C, Mtachar D. Health status of individuals with
mild stroke. Stroke. 1997;28:740–745.
[Abstract/Free Full Text] - Stucki G, Liang MH, Philips C, Katz JN. The Short Form-36 is preferable to the SIP as a generic health status measure in patients undergoing elective total hip arthroplasty. Arthritis Care Res. 1995;8:174–181.[Medline] [Order article via Infotrieve]
- Hochstenbach J, Donders R, Mulder T, Limbeek J van, Schoonderwaldt H. Long-term outcome after stroke: a disability orientated approach. Int J Rehabil Res. 1996;19:189–200.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  C. M. Boyd, C. O. Weiss, J. Halter, K. C. Han, W. B. Ershler, and L. P. Fried Framework for Evaluating Disease Severity Measures in Older Adults With Comorbidity J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2007; 62(3): 286 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Foerch, K R Kessler, D A Steckel, H Steinmetz, and M Sitzer Survival and quality of life outcome after mechanical ventilation in elderly stroke patients J. Neurol. Neurosurg. Psychiatry, July 1, 2004; 75(7): 988 - 993. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||