Upload
barbara-ek-klein
View
222
Download
1
Embed Size (px)
Citation preview
Frailty, morbidity and survival
Barbara E.K. Klein *, Ronald Klein, Michael D. Knudtson,Kristine E. Lee
Department of Ophthalmology and Visual Sciences, University of Wisconsin Medical School,
610 North Walnut Street, 405 WARF, Madison, WI 53726-2336, USA
Received 20 February 2004; received in revised form 13 January 2005; accepted 14 January 2005
Available online 16 March 2005
Abstract
Frailty, as a reflection of decreased physical reserve rather than disability, is assessed by
various functional tests rather than by specific disease burden. We investigated association of
measures of frailty to disease outcomes and survival in a population-based study of Midwestern
adults. The markers of frailty we evaluated were: time to walk a measured course (gait-time),
handgrip strength, peak respiratory flow rate, ability to stand from a sitting position without using
arms, and best corrected visual acuity. A history of cardiovascular disease, cancer, and hyperten-
sion were obtained. Data were collected at the third examination (1998–2000) of the Beaver Dam
Eye Study cohort (n = 2962). Follow-up for mortality occurred up to 412 years after the 1998–2000
examinations. Markers of frailty were significantly associated with age. Values in the highest
quartile (slowest) of gait-time, lowest quartile of peak expiratory flow rate, lowest quartile of
handgrip strength, inability to stand from sitting in one try (those not in a wheelchair), and visual
impairment were combined in an index to denote a general description of frailty. The range of the
index was 0 (no frailty) to 5 (maximum frailty). Greater frailty was significantly associated with
cardiovascular disease and hypertension. Frailty was associated with poorer survival over an
interval of 412 years after adjusting for age, sex, hypertension, diabetes, and cardiovascular disease.
Greater frailty was associated with greater likelihood of concurrent medical conditions and with
decreased survival.
# 2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: Frailty of elderly; Mortality; Morbidity
www.elsevier.com/locate/archger
Archives of Gerontology and Geriatrics 41 (2005) 141–149
* Corresponding author. Tel.: +1 608 263 0276; fax: +1 608 263 0279.
E-mail address: [email protected] (Barbara E.K. Klein).
0167-4943/$ – see front matter # 2005 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.archger.2005.01.002
1. Introduction
Frailty results from declines in multiple systems (Lipsitz and Goldberger, 1992; Bortz,
1993; Campbell and Buchner, 1997; Hamerman, 1999). It differs from disability or
morbidity in that it is meant to describe a general decrease in functional status (Chin et al.,
1999; Fried et al., 2001). Markers of frailty include low lung function (Sharp et al., 1997),
decreased mobility, slower gait-time, age-related decreases in balance and muscle strength,
and poorer visual function (Kahn et al., 1977; Dana et al., 1990; Tielsch et al., 1990; Klein
et al., 1999, 2003; Hyman et al., 2001; Wang et al., 2001). These signs may be accompanied
by a reduced ability to rebound from challenges, leading to increased mortality (Klein
et al., 1995; Sharp et al., 1997; Anstey et al., 2001; Fried et al., 2001; Hennis et al., 2001;
Wang et al., 2001) and morbidity (Tinetti et al., 1986, 1995; Nevitt et al., 1989; Klein et al.,
1998). In addition, it has been postulated that this apparently multifaceted syndrome may
be characterized as a phenotype which leads to the possibility of exploring environmental
and genetic antecedents to it (Fried et al., 2001).
It is the purpose of this report to describe the associations of morbidity and survival to an
index of frailty in the Beaver Dam Eye Study cohort.
2. Methods
All persons of 43–86 years of age living in Beaver Dam Wisconsin were invited to
participate in a study evaluation from 1988 to 1990 (Campbell and Palit, 1988; Klein et al.,
1991). Tenets of the Declaration of Helsinki were followed. Signed consent was obtained
from each participant at each examination. Institutional review board approval was obtained
yearly. All subjects who were eligible for participation in the baseline examination (Klein
et al., 1991) were invited for a second examination 5 years later and for a third examination 5
years after that (10-year follow-up). The differences between participants and nonparti-
cipants at the baseline (Klein et al., 1991), 5-year (Klein et al., 1996), and 10-year follow-up
(Klein et al., 2001) examinations have been previously published. Those who were alive but
not participating at later visits were older and had poorer visual acuity. Those who died were
older, were more likely to be men, had poorer visual acuity, and were more likely to have
diabetes (Klein et al., 1991, 1996, 2001). For purposes of this inquiry, we use measurements at
the 10-year examination of the cohort because some of the measurements were not taken at
previous examinations. Only pertinent parts of the examination are described.
Distance visual acuity was measured according to a modification of the Early Treatment
Diabetic Retinopathy Study (ETDRS) protocol (Klein et al., 1996) for each eye and was
denoted as best-corrected visual acuity. Results were given as the number of letters read, as
well as Snellen equivalents ranging from 20/10 to no light perception. We used the Snellen
equivalents in tables.
Preliminary analyses were performed for each eye. There were no systematic
differences between the eyes. We used data for responses in the better eye based on the
assumption that performance is a function of vision in the better eye. However, there were
no systematic differences in the associations of visual function in either eye or the better
eye with the measures of frailty.
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149142
Ambulatory participants were instructed to walk a measured course at their usual pace.
The time was recorded (gait-time) (Guralnik et al., 1994). The participants were then
seated in a standard chair (seat 19.5 in. from the floor) which was against the wall. The
participants who felt that it was safe for them to stand up without help were asked to do so
without using their arms. If unable to rise without using their arms, the participants were
instructed to stand up using their arms. The number of attempts to rise was recorded. The
method (with or without arms) was recorded. The peak expiratory flow rate was measured
using the mini-Wright meter (Klein and Klein, 1999). The participants were asked to stand
and were instructed to take as deep a breath as possible and to blow as hard and fast as s/he
was able to. This was repeated two more times, and the best value (greatest flow rate) was
used in the analysis. Dynamometry was performed in each hand two times. The mean of
two measures for the dominant hand was used in these analyses (Evans, 1995).
Study participants were asked whether a doctor ever told them that they had had a heart
attack, angina, a stroke or cancer (non-skin cancer). A positive response to any of these
questions was used in the analyses of morbidity. Hypertension was defined as mean systolic
blood pressure of 160 mmHg or mean diastolic blood pressure of 95 mmHg, measured
according to protocol (Klein et al., 1999), or taking medications to lower blood pressure.
Mortality was monitored throughout the study by yearly telephone calls to all study
participants or a contact person and by reviewing obituaries.
A frailty index combining poorer function for each characteristic was devised according
to the following scheme: highest quartile of gait-time (�3.37 s in women, �3.19 s in men);
lowest quartile of peak expiratory flow rate (�290 l/min for women, �440 l/min for men);
lowest quartile for hand grip strength for the dominant hand (�18.5 kg for women,
�34.5 kg for men); not being able to stand from a sitting position in one try (without use of
arms) and visual impairment (best-corrected visual acuity of 20/40 or poorer in the better
eye). Equal weight was given to each measure. Frailty was further categorized into four
levels: none (none of the characteristics), mild (1–2 characteristics), moderate (3
characteristics), and severe (4–5 characteristics). These four categories of severity are used
in the analyses.
Analytical techniques including computation of means, standard deviations, chi-square
tests, Spearman and Pearson correlation coefficients, survival analysis, and regression
models were performed using version 8.1 of SAS (Cary, NC). For the cross-sectional
analyses of comorbid conditions, logistic regression was used. Age was adjusted as a
continuous variable. Similarly, the frailty level was entered linearly. A test for non-linearity
of the frailty level was performed, and it was determined that the relationship was linear.
Both Kaplan–Meier and Cox proportional hazard models were used in the mortality
analysis. In this analysis, models were first adjusted for age (continuous) and sex. To
determine if the relationship was independent from other comorbidities, we further
adjusted for hypertension, diabetes, and history of cardiovascular disease.
3. Results
There were 2515 of the 2962 participants for whom we had data for all frailty markers.
Those without all measurements tended to be older, female, had poor peak expiratory flow
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149 143
rate, had poor grip strength, were less able to do the chair stand, had poor visual acuity, and
resided in a nursing home.
Values for gait-time, grip strength in the dominant hand, and peak expiratory flow rate
differed by age and gender (Fig. 1). For the chair stand, there was decreased ability to rise in
one attempt with age, but differences between women and men were not significant
( p = 0.80). Visual impairment increased with age.
Table 1 describes the frailty index levels for women and men by age. In both women and
men, youngest persons were unlikely to be frail. Level of frailty increased with age
( p < 0.001). Women were significantly less frail than men ( p = 0.02).
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149144
Fig. 1. Frailty measures by age. (*) Anyone not in a wheelchair who could not rise from a chair in one try without
using their arms.
We computed the age-adjusted odds of having cardiovascular disease (self-reported
angina, myocardial infarction or stroke), hypertension, and cancer (excluding skin cancers)
for a level increase in frailty level (Table 2). In both men and women, a one level increase in
frailty was associated with about a 35% increase in odds of having cardiovascular disease
and about a 20% increase in odds for hypertension. A 15–20% increase in odds of having
cancer (excluding skin cancer) was not significant.
We performed survival analyses to determine whether the frailty index predicted
mortality over the relatively short interval of follow-up (3.5–4.5 years). We confirmed that
292 persons had died since they were seen, but only 149 of these persons had measurements
on all five frailty markers. We confined the survival analyses to the latter 2515 persons with
complete information of who 149 died. Because of the relatively small number of deaths
considering the number of levels of frailty, and because the hazard ratios were the same for
women and men, we combined the data from the sexes for the survival analyses. Greater
frailty was significantly associated with decreased survival (Fig. 2). The age- and sex-
adjusted hazard ratio for an increase in frailty level using the four categories as described in
Section 2 was 1.69 (95% confidence interval [CI]; 1.38, 2.08). When we further adjusted
for hypertension, diabetes, and cardiovascular disease, the hazard ratio was 1.56 (95% CI;
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149 145
Table 1
Frailty index levela by age and sex
Age (years) Women (percent) Men (percent)
N None Mild Moderate Severe py N None Mild Moderate Severe py
0 1–2 3 4–5 0 1–2 3 4–5
53–64 581 81.2 18.6 0.2 0.0 488 74.2 24.6 1.0 0.2
65–74 455 53.9 41.1 3.7 1.3 361 49.9 44.6 4.4 1.1
75–84 328 24.4 53.7 15.6 6.4 217 21.7 56.2 15.7 6.5
85+ 61 6.6 42.6 23.0 27.9 <0.001 24 0.0 41.7 33.3 25.0 <0.001
Age-adjusted p-value women vs. men: 0.02.a Frailty index score combined into four levels (none: 0 frailty markers; mild: 1–2 frailty markers; moderate: 3
frailty markers; severe: 4–5 frailty markers).y Cochran–Mantel–Haenszel test of trend.
Table 2
Age-adjusted odds ratios for cardiovascular disease, hypertension, and cancer associated with the frailty index
level
Outcome Women Men
No. of events
/total no.
ORa (95% CI) p No. of events
/total no.
ORa (95% CI) p
Cardiovascular disease 153/1424 1.43 (1.13, 1.82) 0.003 231/1088 1.33 (1.06, 1.67) 0.01
Hypertension 697/1424 1.22 (1.02, 1.46) 0.03 488/1089 1.22 (1.00, 1.49) 0.05
Cancer (excluding skin) 162/1423 1.21 (0.95, 1.54) 0.13 123/1087 1.17 (0.89, 1.55) 0.27
a OR: age-adjusted odds ratio representing a one-level increase in frailty (for four levels, see footnote of
Table 1).
1.27, 1.92). In an attempt to compensate for morbidities that might be associated with the
frailty score and with early mortality, we then excluded deaths within 1 year of the
examination (n = 25). Frailty was still strongly associated with survival with the age- and
sex-adjusted hazard ratio for an increase in frailty level being 1.79 (95% CI; 1.43, 2.23).
4. Discussion
Men were more likely to be frail than women in our population. The reverse has been
found in the Cardiovascular Health Study (CHS) (Fried et al., 2001). This may have
resulted from different sex-specific cutoffs for defining characteristics of frailty. Our
definition of frailty, which allows for differences in values for the markers between the
sexes, results in a similar relationship of the index to death for the sexes. This is in keeping
with the notion of frailty being a measure of vulnerability that is not much influenced
by sex.
The concept of frailty implies a condition lowering ‘‘resistance’’ to all or many diseases.
In our study, we found that frailty was associated with cardiovascular disease and
hypertension. We were unable to accurately assess whether other conditions might be
related to frailty because few people had reported other conditions in Beaver Dam. During
the relatively short interval between our measuring the frailty markers and the end of our
subsequent mortality follow-up, there were too few deaths to determine whether mortality
attributable to many specific causes, aside of cardiovascular disease, increased with
increased frailty level. We plan to re-evaluate this with further follow-up of the cohort. We
have taken a conservative approach to testing the usefulness of our model of frailty and
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149146
Fig. 2. Survival curves stratified by levels of frailty.
‘survival by confining these analyses to the 149 deaths among 2515 participants for whom
we had all measurements prior to their death. By not including the additional 143 deaths,
we may be underestimating the information about mortality that may be extant in the
markers we did measure in these persons. We note that many subjects for whom we did not
have all measures were nursing home residents. The age-adjusted hazard ratio for mortality
for those in nursing homes was 4.66 (95% CI; 3.44, 6.32) (data not previously shown). Our
continued follow-up of this cohort will permit us to determine whether there is increased
mortality for those who were characterized as frail for any particular as well as for all
markers.
While visual acuity is not always included as a marker of frailty (Lipsitz and
Goldberger, 1992; Bortz, 1993; Campbell and Buchner, 1997; Chin et al., 1999;
Hamerman, 1999; Fried et al., 2001), it has been associated with increased mortality and
morbidity (Tinetti et al., 1986, 1995; Nevitt et al., 1989; Klein et al., 1995, 1998; Anstey
et al., 2001; Hennis et al., 2001; Wang et al., 2001). Including visual acuity as a marker of
frailty in its own right also increases the range of the frailty index. It may be that with
longer follow-up for incident morbidities and for survival, visual acuity will improve the
informativeness of the index of frailty.
We used visual function rather than specific age-related eye diseases because our
concept of frailty, as in the CHS (Fried et al., 2001), is that frailty differs from specific
disabilities and is best reflected in function. Limiting our analyses to those without specific
eye diseases would be akin to eliminating data from those persons with cardiovascular
disease from other markers of frailty.
The concept of frailty is used here to denote diminished functional abilities. It can also
be construed to mean diminished life span or early mortality (Yashin et al., 1999; Begun
et al., 2000; Kohler and Kohler, 2000). It may be that using a frailty index as a quantitative
genetic trait (phenotype) would enhance the investigations of genetic and environmental
determinants of longevity.
Acknowledgement
This research is supported by National Institutes of Health grants EY06594 and
EY13438.
References
Anstey, K.J., Luszcz, M.A., Giles, L.C., Andrews, G.R., 2001. Demographic, health, cognitive, and sensory
variables as predictors of mortality in very old adults. Psychol. Aging 16, 3–11.
Begun, A., Desjardins, B., Iachine, I., Yashin, A., 2000. Multivariate frailty model with a major gene:
application to genealogical data. In: Hasman, A., Blobel, B., Dudeck, J., Engelbrecht, R., Gell, G.,
Prokosch, H. (Eds.), Medical Infobahn for Europe—Proceedings of MIE 2000 and GMDS 2000. IOS
Press, Amsterdam, pp. 412–416.
Bortz II, W.M., 1993. The physics of frailty. J. Am. Geriatr. Soc. 41, 1004–1008.
Campbell, A.J., Buchner, D.M., 1997. Unstable disability and the fluctuations of frailty. Age Aging 26, 315–
318.
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149 147
Campbell, J.A., Palit, C.D., 1988. Total digit dialing for a small area census by phone. In: Proceedings of the
Survey Research Methods Section, American Statistical Association, Alexandria, VA, pp. 549–551.
Chin, A., Paw, M.J., Dekker, J.M., Feskens, E.J., Schouten, E.G., Kromhout, D., 1999. How to select a frail elderly
population? A comparison of three working definitions. J. Clin. Epidemiol. 52, 1015–1021.
Dana, M.R., Tielsch, J.M., Enger, C., Joyce, E., Santoli, J.M., Taylor, H.R., 1990. Visual impairment in a rural
Appalachian community, prevalence and causes. J. Am. Med. Assoc. 264, 2400–2405.
Evans, W.J., 1995. Exercise, nutrition and aging. Clin. Geriatr. Med. 11, 725–734.
Fried, L.P., Tangen, C.M., Walston, J., Newman, A.B., Hirsch, C., Gottdiener, J., Seeman, T., Tracy, R., Kop, W.J.,
Burke, G., McBurnie, M.A., for the Cardiovascular Health Study Collaborative Research Group, 2001. Frailty
in older adults: evidence for a phenotype. J. Gerontol. 56, M146–M156.
Guralnik, J.M., Simonsick, E.M., Ferrucci, L., Glynn, R.J., Berkman, L.F., Blazer, D.G., Scherr, P.A.,
Wallace, R.B., 1994. A short physical performance battery assessing lower extremity function: associa-
tion with self-reported disability and prediction of mortality and nursing home admission. J. Gerontol. 49,
M85–M94.
Hamerman, D., 1999. Toward an understanding on frailty. Ann. Intern. Med. 130, 945–950.
Hennis, A., Wu, S.Y., Li, X., Nemesure, B., Leske, M.C., Barbados Eye Study Group, 2001. Lens opacities and
mortality: the Barbados Eye Studies. Ophthalmology 108, 498–504.
Hyman, L., Wu, S.Y., Connel, A.M., Schachat, A., Nemesure, B., Hennis, A., Leske, M.C., 2001. Prevalence and
causes of visual impairment in the Barbados Eye Study. Ophthalmology 108, 1751–1756.
Kahn, H.A., Leibowitz, H.M., Ganley, J.P., Kini, M.M., Colton, T., Nickerson, R.S., Dawber, T.R., 1977. The
Framingham Eye Study. I. Outline and major prevalence findings. Am. J. Epidemiol. 106, 17–32.
Klein, R., Klein, B.E., 1999. Beaver Dam Eye Study III. Manual of Operations. U.S. Department of Commerce,
Springfield, VA NTIS Accession No. PB99-137861.
Klein, R., Klein, B.E., Linton, K.L., De Mets, D.L., 1991. The Beaver Dam Eye Study: visual acuity.
Ophthalmology 98, 1310–1315.
Klein, R., Klein, B.E., Moss, S.E., 1995. Age-related eye disease and survival. The Beaver Dam Eye Study. Arch.
Ophthalmol. 113, 333–339.
Klein, R., Klein, B.E., Lee, K.E., 1996. Changes in visual acuity in a population. The Beaver Dam Eye Study.
Ophthalmology 103, 1169–1178.
Klein, B.E., Klein, R., Lee, K.E., Cruickshanks, K.J., 1998. Performance-based and self-assessed measures of
visual function as related to history of falls, hip fractures, and measured gait time. The Beaver Dam Eye Study.
Ophthalmology 105, 160–164.
Klein, B.E., Klein, R., Lee, K.E., Cruickshanks, K.J., 1999. Associations of performance-based and self-reported
measures of visual function. The Beaver Dam Eye Study. Ophthalmic Epidemiol. 6, 49–60.
Klein, R., Klein, B.E., Lee, K.E., Cruickshanks, K.J., Chappell, R.J., 2001. Changes in visual acuity in a
population over a 10-year period: the Beaver Dam Eye Study. Ophthalmology 108, 1757–1766.
Klein, B.E., Klein, R., Knudtson, M.D., Lee, K.E., 2003. Relationship of measures of frailty to visual function. The
Beaver Dam Eye Study. Trans. Am. Ophthalmol. Soc. 101, 191–199.
Kohler, H.P., Kohler, I., 2000. Frailty modelling for adult and old age mortality: the application of a modified
DeMoivre hazard function to sex differentials in mortality. Demogr. Res. 3 (8), 1–32.
Lipsitz, L.A., Goldberger, A.L., 1992. Loss of complexity and aging. Potential applications of fractals and chaos
theory to senescence. J. Am. Med. Assoc. 267, 1806–1809.
Nevitt, M.C., Cummings, S.R., Kidd, S., Black, D., 1989. Risk factors for recurrent nonsyncopal falls. A
prospective study. J. Am. Med. Assoc. 261, 2663–2668.
Sharp, D.S., Burchfiel, C.M., Curb, J.D., Rodriguez, B.L., Enright, P.L., 1997. The synergy of low lung function
and low body mass index predicting all-cause mortality among older Japanese-American men. J. Am. Geriatr.
Soc. 45, 1464–1471.Tielsch, J.M., Sommer, A., Witt, K., Katz, J., Royall, R.M., 1990. Blindness and visual impairment in an American
urban population. The Baltimore Eye Survey. Arch. Ophthalmol. 108, 286–290.
Tinetti, M.E., Williams, T.F., Mayewski, R., 1986. Fall risk index for elderly patients based on number of chronic
disabilities. Am. J. Med. 80, 429–434.
Tinetti, M.E., Inouye, S.K., Gill, T.M., Doucette, J.T., 1995. Shared risk factors for falls, incontinence and
functional dependence. Unifying the approach to geriatric syndromes. J. Am. Med. Assoc. 273, 1348–
1353.
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149148
Wang, J.J., Mitchell, P., Simpson, J.M., Cumming, R.G., Smith, W., 2001. Visual impairment, age-related cataract,
and mortality. Arch. Ophthalmol. 119, 1186–1190.
Yashin, A.I., De Benedictis, G., Vaupel, J.W., Tan, Q., Andreev, K.F., Iachine, I.A., Bonafe, M., DeLuca, M.,
Valensin, S., Carotenuto, L., Franceschi, C., 1999. Genes, demography, and life span: the contribution of
demographic data in genetic studies on aging and longevity. Am. J. Hum. Genet. 65, 1178–1193.
B.E.K. Klein et al. / Archives of Gerontology and Geriatrics 41 (2005) 141–149 149