Skip to main content

Full text of "DTIC ADA228133: Age and Functional Health Status"

See other formats


AD-A228 133 



AGE AND FUNCTIONAL HEALTH STATUS 


Eugene Nelson, Ron D. Hays, Sharon Arnold, 
Kent Kwoh, Cathy Sherbourne 


June 1989 



V 




Health Status 


Acknowledgments 


Thanks are due to Kim Wong, Nelie Gill, and Patti Sue Thompson for 
secretarial support. 



Accession For f 

NTIS 

DTIC 

Unann 

Justi 

GRAAI 

rAB □ 
ounced □ 
ricfltlon 



Distribution/ 

Availability Codes 

Dist 

r 1 

Avail 

Spec 

and/or 

lal 


\ 







Health Status 

9 


Age and Functional Health Status 


Abstract 


The relationship between age and functional health status is examined in 
two cross-sectional studies: (1) a random household sample of 2008 
adults and (2) a sample of 1227 ambulatory patients in northern New 
England. Results reveal diminished and more variable physical 
functioning, role functioning,, and perceived health in older age groups, 
particularly those groups aged 50 and older. Emotional functioning, 
however, tended to be better in older age groups. The association 
between physical and emotional functioning was strongest in older age 
groups,. However, decrements in functional health status occurred 
selectively among older individuals with many of the elderly scoring as 
well as the average young adult on the measures of physical, role and 


emotional functioning. vLa t . 


„ ij. (' » Q, U, >' <L 












Health Status 
3 


AGE. AND FUNCTIONAL HEALTH STATUS 


"The functional declines that typically accompany advancing 
age are often dramatic and depressing" (New York Times, June 
10, 1986). 

As evidenced in the quotation above, conventional wisdom says that 
functional health--that is, physical, mental and role performance of 
daily activities--decreases as peopie get older. But professionals 

v 

concerned with the study of aging are often struck not so much by the 
decline in functional health with age, but. with the increased variation 
in function in those with advanced age. Some notable individuals and 
particular cultural groups seem to enjoy superior functioning at very 
old ages. This raises a series of questions. What is the actual 
relationship between functioning and age? How many people of advanced 
age actually function as well or better than young adults? Why do some 
people continue to function well despite the ravages of disease and the 
onslaught of age? Is there a mind-body connection at work so that those 
with better emotional function have better physical function and vice 
, versa? 

Literature Review 


When one turns to the literature to explore what is known about the 
relationship between functioning and age, many studies exist that focus 
on the relationship between functioning' and disease (e.g., Meltzner, 





Health Status 
4 


Carman & House, 1983; Rice & Cugliani, 1980) or functional limitations 
iii the elderly. (e.g., Branch & Jette, 1-981; Jette & Branch, 1981; Katz, 
Branch, Branson, Papsidero, Beck, & Greer, 1983). The former studies 
tend to suggest, not surprisingly, that functioning and disease are . 
related whereas the latter show that functional problems are common in 
older age groups. However, the literature is very sparse with respect 
to a basic question: What is the relationship between functioning and 
age across a wide range of ages? Only a few studies, all restricted to 
patient populations, have addressed this issue (e.g., Nelson et al., 
1983; Parkerson, Gehlbach, Wagner, James, & Clapp, 1981). In this 
section, we briefly review the literature on the age-health 
relationship. 

Functioning has been defined in diverse ways. Some investigators 
have used a global measure of disability whereas others have used more 
specific clinical and physiological measures of the functioning of organ 
systems. Few studies used comprehensive measures of functioning which 
include physical, role and emotional functioning. We found only two 
studies which employed assorted functioning measures across the age 
spectrum, and their generalizability is limited because they were 
conducted in selected patient populations (Nelson et al., 1983; 

Parkerson et al., 1981). 

Measures of functioning differ from clinical measures of health 
status such as morbidity because they reflect the impact of disease on 
day-to-day life. Functional status of the elderly has been especially 
well-studied with many investigations including persons aged 60 and 
older (Branch & Jette, 1981; Fillenbaum, 1985; Jette & Branch, 1981; 


Health Status 
5 


Katz et al., 1983). Rarely has research examined functional 
status across a wide range of ages. The studies conducted to date 
suggest that physical functioning declines with age (Cape & Henschke, 
1980; Chirikos & Nestel, 1985; Fillenbaum, 1985; Jette & Branch, 1981; 
Katz et al., 1983), and social disability, defined as unmet need for 

social services, increases with age (Branch & Jette, 1981). Parkerson 

1. 

et al. (1981) assessed the impact of age on functional status using the 
Duke-UNC Health Profile on a sample of 395 patients 18 and older in a 
family medicine clinic. Negative relationships between age and physical 
and social functioning were found. A significant positive relationship 
between age and symptoms was noted as well. Nelson et al. (1983) 
reported a negative correlation between physical- functioning and age. 
Similarly, Feller (1983). found an increase in the proportion of persons 
requiring help with daily living activities as a function of age. None 
of these studies examined the interaction between different aspects of 
functioning. 

The relationship between emotional functioning and age has been 
explored in several studies. A nonlinear relationship between emotional 
functioning and age has been suggested, but its exact form is unknown 
(Feinson, 1985). Parkerson et al. (1981) reported a negative 
association between age and emotional functioning. However, in a study 
of 758 outpatients aged 18-99, Cassileth et al. (1984) found that 
emotional functioning increased with age. Feinson (1985) reviewed the 
literature on the relationship between age and emotional functioning and 
noted that of the 31 studies reviewed, ten provided no evidence of 
decreased emotional function with age, eight indicated better emotional 




Health Status 

6 

function with increased age, and three ^concluded that emotional function 

V — 

decreases with age. Koenig (1986, rp .384) .concluded that there, is 
"greater support for a decrease in frequency of mental disorders among 
older persons and an increased ability to cope with major life changes 
when compared with younger age .groups." 

Studies using global, subjective measures of perceived health 
reveal a tendency for health perceptions to be inversely related to age, 
with some exceptions. Consistent with conventional wisdom, Halpert and 
Zimmerman 01986) found in their study of 148 rural elderly that persons 
aged 60 to 74 were more likely to evaluate their health- as excellent or 
good than were persons 75 and older. Similarly, the U.S. Bureau of the 
Census Survey, "Americans Assess Their Health 1978" (U.S. Department of 
Health and Human Services, 1983), showed increasing proportions of 
persons who assessed their health as fair or poor as age increased, up 
until age 80. After age 80, however, a decrease in the proportion of 
persons rating their health as fair or poor was observed. Some studies 
have, found a larger proportion of persons 75 and older than those 65-74 
rating their health as excellent or good (Ferraro, 1980). Interestingly, 
when 660 Illinois adults (18 and above) were asked to assess their 
health compared to others their age, only those 61 and older rated their 
health as better than their peers (Cockerham, Sharp, & Wilcox, 1983). 

Measurement of Functioning Across a Wide Range of Ages 

The measurement of health and functioning has improved dramatically 
in recent years. Applications of standard scaling techniques have 
generated highly refined multi-item scales for measuring physical, 


Health Status 
7 


emotional and role functioning, and current health perceptions. The 
advantages of multi-item health scales over single-item measures are 
well documented (Davies & Ware, 1981; Manning, Newhouse, &Ware, 1982). 
Single-item measures of health frequently are not precise enough to 
achieve satisfactory statistical power for hypothesis testing (Ware & 
Karmos, 1976; Manning et al., 1982). There is a number of standard 
multi-item health measures available (e.g., Bergner, Bobbitt, Carter, & 
Gilson, 1981; Ware, 1984), but these instruments tend to be too long for 
many applications. A short-form measure of health has recently been 
developed and evaluated. It was derived from self-report instruments 
that have been used extensively (c.f. Brook, Ware, Davies-Avery et aL, 
1979). This short-form instrument is comprehensive (assesses perceived 
health and physical, emotional and role functioning), reliable and 
valid, yet it consists of only 17 items (Stewart, Hays, & Ware, 1988). 
Short-form instruments, such as this one, provide comprehensive 
information on individual functioning in a cost-effective manner, 
without excessive respondent burden. 

In general population studies only a small proportion of the 
respondents may have chronic disease or disability. Reliance on the 
measurement of only a narrow range of functioning, representing the most 
severe effects of disease, will provide little information for the 
majority of the sample. Comprehensive assessment of health status 
across a wide range of ages allows for the identification of multiple 
aspects of health which may show differential relationships with age and 


disease. 






Health Status 

8 

The purpose of this paper is to explore the following questions 
which prior research has left largely unanswered. 

• How do physical, emotional and role functioning differ 
between age groups? 

• How many older adults report functioning equal to or 
better than that of the average young adult? 

• How do perceived health and energy level differ between 
age groups? 

• What is the association between physical functioning and 
emotional functioning? 

The strengths of this study include use of a short, standardized 
instrument in a general population with a wide range of ages, a 
comprehensive assessment of health status, and the ability to study 
age-health relations in a patient sample. 

Method 

Our analysis is based on information gathered in samples drawn from 
two different populations. These two cross-sectional, descriptive 
studies are described next. 

Subiects 

Samples were drawn from two populations: (1) a random sample of 
adults living in private households in the United States; and (2) a 



Health Status 
9 


consecutive series of patients visiting primary care practices in 
northern New England. Louis Harris and Associates (Harris & Associates, 
1984) conducted the former study of 2008 adults in 1984 and the 
Dartmouth Primary Care Cooperative Information Project (Nelson et al. 
1981; Nelson & Green, 1984) performed the latter research on 1227 
ambulatory patients in 1981. The average age of the household sample 
was 36; the average age of the patient sample was 47. Fifty-six 
percent of the household sample and 67% of the patient sample were 
females. A total of 84% and 77% of the respondents reported that they 
were high school graduates in the household and patient samples, 
respectively. 

Data Collection 

The household sample consists of adults 18 years of age and older 
who were interviewed by telephone from August through October, 1984. 

Half of those sampled represent households enrolled in health 
maintenance organizations (HMOs) and half represent those in the 
fee-for-service (FFS) system. The FFS sample was identified using the 
random-digit-dialing method, based on an unclustered sampling frame, and 
stratified by region and by the Census Bureau's "size of place" 
designations. The same procedure was used for the HMO sample; 
households were first screened to identify those falling into known HMO 
areas, and sampling from 195 Standard Metropolitan Statistical Areas 
known to include HMOs yielded additional HMO enrollees. The methods 
used are documented in detail elsewhere (Montgomery & Paranjpe, 1985). 
The telephone interviews included measures of health status, source of 






Health Status 
10 

medical care, satisfaction with health care, and demographic variables. 
The patient sample was selected from a series of adults visiting 27 
predominantly rural primary care medical practices. Patients completed 
self-administered questionnaires which assessed health status and 
demographic variables. The questionnaires were filled out while 
patients were waiting to see their physician and took approximately 10 
minutes to complete. Approximately 90% of the patients who were asked 
to participate in the study did so. 

Measures of Health 

The health status measures that were used to assess physical, 
emotional and role functioning in the two samples were short-form 
adaptations of longer measures developed at the RAND Corporation for 
the Health Insurance Experiment and Medical Outcomes Study (Stewart & 
Ware, forthcoming; Stewart, Ware, & Brook, 1981; Veit & Ware, 1983; Ware, 
1976, 1984). Items were selected for the short-form measures based on 
the criterion that the selected set best represents or captures the 
information contained in the constructs measured by the long-form 
scales. Table 1 presents details about the three functional health 
measures--physical, emotional and role functioning--which were used 
in both samples, a measure of perceived health used in the household 
sample, and a measure of energy level used in the patient sample. 

In this study we examine data from both samples, although the data 
were not originally collected for this purpose. Because of the 
differences in the measures and the methods used to gather the data 









Health Status 
11 


Table 1 


Description of Health Measures Used In Household And Patient Samples 


Name of 
Scale 


No. of 


Items Reliability Description of Content 


Physical Function 

Household 6 

Patient 7 


Emotional Function 
Household 
Patient 


Role Function 
Household 
Patient 


.88 (.55) 
.80 (.36) 


.82 (.48) 
.93 (.62) 


.76 (.61) 
.88 (.71) 


Vigorous activities; 
Bending, lifting, 
stooping. Moderate 
activities; walk uphill, 
walk one block. Eating 
and dressing. 

Nervous person; Felt calm 
and peaceful; felt 
downhearted; a happy 
person; down in the dumps. 


Unable to work at a job, around 
the house, or go to school 
because of health; unable to 
do certain kinds of work, 
housework, or schoolwork because 
of health. 


8 







Health Status 
12 


Table 1 continued 


Name of No. of 

Scale Items Reliability Description of Content 


Perceived Health 

Household 4 .83 (.55) Somewhat ill; Healthy as 

anybody; Feeling bad; 
Health is excellent. 

Energy Level 

Patient 4 .85 (.59) How much energy; Felt 

tired; Feeling sluggish; 
waking up fresh 


Internal consistency reliability was estimated using Crombach's alpha 
coefficient. The estimated reliability for a single item is given 
in parentheses. 

Note . Alpha reliability was also computed within five age groups: 

18-24, 25-34, 35-49, 50-64, and 65 and over. Alpha ranged across age 
groups from .73 to .85 for physical function in the household sample and 
from .70 to .78 in the patient sample. Alpha ranged from .76 to .84 for 
emotional function in the household sample and .92 to .94 in the patient 
sample. Alpha ranged from .62 to .80 for role function in the patient 





Health Status 
13 


Table 1 continued 

sample and from .85 to .92 in the patient sample. Alpha ranged from .75 
to .84 for perceived health in the household sample. Alpha ranged 
from .82 to .89 for energy level in the patient sample. 








Health Status 
14 


(telephone versus self-administration), we do not compare the samples 
directly in this study. 

Analysis Plan 

Internal consistency reliabilities (Cronbach, 1951) of the health 
measures were computed within five different age groups (18-24, 25-34, 
35-49, 50-64, 65+). In addition to this traditional test of item 
convergent validity, item discriminate validity across scales was 
evaluated. Discriminant validity is supported if an item correlates 
significantly higher with its hypothesized scale than it does with 
other scales. For example, items measuring physical functioning are 
expected to correlate higher with the physical functioning scale 
(corrected for the item being evaluated) than they correlate with scales 
measuring mental health or role functioning. After establishing that 
the items adequately represented the hypothesized scales, we summed 
appropriate items together to form derived health status scales. Scale 
scores were then transformed to a 0-100 scale for each measure, with 
higher scores representing better health. 

The relations between age and different indicators of health status 
were assessed for males and females separately. Average scores on the 
health status variables for each of five age groups (18-24, 25-34, 

35-49, 50-64, 65+) were compared using one-way ANOVAs (n's in different 
age groups were 243, 624, 592, 353, and 194 in the household sample, 
respectively, and 150, 303, 218, 241, and 311 in the patient sample, 
respectively). Thus, age group served as an independent variable in 
analyses of health status differences. We also calculated the 






Health Status 
15 


percentage of respondents in each age group that scored equal to or 
better than the average health score of the youngest age group (i.e., 
18-24). Finally, the correlation between physical functioning and 
emotional functioning was evaluated by age group in both samples. 

Results 

Reliability and Validity of Health Status Measures 

The internal consistency of the multi-item measures was excellent as 
evidenced by alpha reliability coefficients ranging from 0.76 to 0.93 
for the overall samples (see Table 1). Reliability was also 
satisfactory in each of the different age groups; alpha internal 
consistency reliability was 0.62 or higher in all age groups. Thus, the 
multi-item scales exceeded the minimum standard for reliability of 0.50 
needed for group comparisons (Helmstadter, 1964). In addition, item 
discriminant validity for these measures was strongly supported. In 
both samples, 80% or more of the item correlations with hypothesized 
scales were significantly higher than correlations with other scales. 
Thus, items tended to correlate higher with the. scales they were 
designed to represent than they did with other scales. 

Physical, Emotional and Role Functioning By Age Group 

Table 2 and Table 3 provide descriptive statistics for measures of 
physical, emotional and role functioning by age group. Figures 1-6 
display trends in measures of functional status by age and sex for the 
household and patient samples. Physical functioning was found to differ 






Health Status 
16 


Table 2 


Distribution of Health Scores By Age Group in Household Sample 


Standard 

Scale _ Age Group Mean Deviation Range 

Physical Functioning 


18-24 

94.49 

15.02 

20-100 

25-34 

95.77 

12.42 

0-100 

35-49 

92.87 

18.21 

0-100 

50-64 

83.23 

26.62 

0-100 

65+ 

76.91 

29.55 

0-100 


F motional Functioning 


18-24 

75.08 

16.57 

12-100 

25-34 

77.03 

16.41 

0-100 

35-49 

78.08 

15.43 

12-100 

50-64 

79.42 

16.3 c 

4-100 

65+ 

81.70 

16.36. 

0-100 


Role Functioning 


18-24 

95.88 

17.11- 

0-100 

25-34 

96.39 

15.99 

0-100 

35-49 

93.24 

23.01 

0-100 

50-64 

84.14 

33.97 

0-100 

65+ 

74.74 

39.83 

0-100 


18-24 

85, 

.38 

17. 

,39 

25-100 

25-34 

87 

.14 

17, 

.68 

6-100 


Perceived Health 



Health Status 

17 


TabTe 2 continued 


Standard 


Scale 

. - Age Group 

. Mean 

Deviation 


35-49 

83.52 

21.58 


50-64 

72.56 

28.99 


65+ 

71.99 

27.40 


Range 

0-100 

0-100 

0-100 





Health Status 
18 


Table 3 


Distribution of Health Scores By Age Group in Patient Sample 


Standard 

Seale _ Age Group _ Mean Deviation Range 

Physical Functioning 


18-24 

96.80 

9.49 

43-100 

25-34 

96.27 

10.68 

36-100 

35-49 

91.12 

16.52 

14-100 

50-64 

83.06 

20.73 

0-100 

65+ 

69.01 

24.67 

0-100 


Emotional Functioning 


18-24 

68.33 

20.78 

10-100 

25-34 

69.87 

18.64 

8-100 

35-49 

64.58 

22.12 

2-100 

50-64 

71.53 

18.74 

5-100 

65+ 

74.27 

19.61 

2-100 


Role Functioning 


18-24 

93.78 

21.98 

0-100 

25-34 

90.40 

25.88 

0-100 

35-49 

81.10 

34.31 

0-100 

50-64 

76.86 

38.76 

0-100 

65+ 

60.57 

42.11 

0-100 


Energy Fatigue 


18-24 

62, 

,56 

19 

.95 

10-100 

25-34 

63, 

,81 

17 

.41 

10-100 



Health Status 
19 


Table 3 continued 


Scale 

Age Grout) 

Mean 

Standard 

Deviation 

Range 


35-49 

56.35 

22.06 

5-100 


50-64 

60.07 

21.22 

0-100 


65+ 

58.37 

21.23 

0-100 


r 



























Figure 4: Emotional Functioning by Age for Females 




+1 


li 

CD 

■ 

O 

H 

iO 

#s, 

[HI 


O bo o 

«> a .s ^ 

|g 

a s 


( □ = Household sample ; ■ = patient sample ; 





























Health Status 
26 


significantly by age in both samples (household males: F=18.67, pC.Ol; 
household females; F-30.35, pC.Ol; patient males: F-25.93, pC.Ol; 
patient -females: F=95.18, p<.01). However, the difference between 
groups was not significant (Duncan's multiple range test) in three of 
the four subsamples (i.e., household sample males and females; patient 
sample males) until after age 49. The lower' level of physical 
performance with increasing age was more apparent in the patient sample 
than for the household sample. Males tended to score better on physical 
functioning than females for each age group but the difference between 
the sexes was smaller for the three younger age groups and larger for 
the two older groups. Males scored significantly higher than females in 
the patient sample for every group except the 25-34 age group. 

Emotional functioning also differed significantly by age in both 
samples (household males: F=2.73, p<.05; household females: F=4.74, 
pC.Ol; patient males: F=»4.64, pC.Ol; patient females: F=*4.73, pC.Ol), 
but, unlike physical functioning, emotional functioning tended to be 
better in the older age groups. In three of the four subpopulations the 
oldest age group scored significantly better than the youngest age group 
(see Figures 3-4). The age group trends in emotional functioning varied 
by subsample. For example, the trend was curvilinear (linear, 
quadratic, cubic, and quartic trends were statistically significant) for 
females in the patient sample, with women in the middle age group 
scoring significantly worse than women in the younger and older age 
groups, whereas the trend for women in the household sample was 
monotonic with a gradual improvement in emotional functioning with older 
age. Males had significantly higher emotional functioning scores than 


l 



i 

i 


i 



Health Status 
27 

females for the 25-34 age group in the household sample and in the older 
age groups (35-49, 50-64, 65+) in the patient sample. 

Significant differences in role functioning by age group were also 
observed (household males: F=11.54, pC.Ol; household females: F=24.84, 
p<.01; patient males: F=10.11, pC.Ol; patient females: F=28.11, p<.01). 

Role functioning age trends paralleled those observed for physical 
functioning, but with a more pronounced decrement observed for the older 
age groups, especially for the patient samples (see Figures 5-6). For 
example, the difference between role functioning scores for males in the 
youngest and oldest age groups was -16 for the household sample, but was 
-30 for the patient sample, with roughly half of the difference 
occurring between the 50-64 and 65+ age groups. Males scored 
significantly better on role functioning than females for the 25-34 and 
50-64 age groups in the household sample. 

Comparison of Older Adults With the Average Young Adult 

Because levels of physical, emotional, and role functioning represent a 
continuum and absolute standards to define a "poor" level of functioning 
versus a "good" level of functioning are nonexistent, the clinical 
interpretation of differences in the levels of functioning may be 
difficult to make. One method to determine the clinical significance of 
a given level of functioning is to assume that the average young adult 
enjoys relatively "good" health and to compare the percentage of 
individuals in each age group who have equivalent or better levels of 
functioning. 

The findings presented above showed that physical and role 


Health Status 
28 

functioning scores were lower for older persons, on average. However,, 
Table 4 demonstrates that a majority of elderly in the community sample 
scored equal to or better than the mean score for people who were 40 
years younger than them. Fifty-one percent (51%) of individuals aged 
65+ had physical functioning scores equal to or better than the mean 
score of the 18-24 age group. Furthermore, 69% of the elderly had role 
functioning scores and 79% had emotional functioning scores that matched 
or exceeded the mean score of the youngest age group. Thus, only a 
subset of the aged displayed poorer functional status than younger 
respondents in the sample. 

The patient population had smaller proportions of elderly scoring 
equal to or better than the mean score of the youngest age group than 
did the household sample (compare Table 4 with Table 5). This is 
particularly apparent in the area of physical functioning where only 17% 
of the age 65 and older group had scores that matched or exceeded the 
mean score for the youngest age group in the patient sample. Even in 
this sample, however, a substantial proportion of older people had 
emotional (70%) and role functioning (47%) scores equal to or better 
than the mean value registered in the 18-24 group. 

Perceived Health and Energy Level 

Perceived health for the household sample exhibited a "rectangular" 
trend with age (see Figure 7). The distribution of perceived health was 
relatively flat for the three younger age groups, then it dropped to a 
lower level and remained at this lower level. In contrast to the sex 
differential favoring males on physical and emotional functioning in 


Health Status 
29 


Table 4 

Percentage of People in Older Age Groups Who Score Equal to or Better 
Than Mean of Youngest Age Group on Selected Measures of Health Status 
In Household Sample (^=2008) 


Mean Percent Scoring Equal To Or 

Health Status in 18-24 Better Than 18-24 Mean 


Measure 

Age Group 

18-24 

25-34 

35-49 

50-64 

65+ 

Physical Function 

95 

85% 

86% 

82% 

61% 

51% 

Emotional Function 

75 

65% 

69% 

69% 

74% 

79% 

Role Function 

96 

94% 

95% 

91% 

80% 

69% 

Perceived Health 

85 

63% 

68% 

65% 

48% 

42% 

(N) 


(243) 

(624) 

(592) 

(353) 

(194) 

Note: Scores were 

transformed to 

a 0 to 100 

distribution and 

rounded 

to 


the nearest whole number. The large percentage of persons scoring above 
the mean is due to very skewed distributions, with the mode (i.e., most 
frequently occurring score) exceeding the mean. 




Table 5 


Health Status 
30 


Percentage of People in Older Age Groups Who Score Equal to or Better 
Than Mean of Youngest Age Group on Selected Measures of Health Status 
in Patient Sample (N=1227’) 


Mean Percent Scoring Equal To Or 

Health Status in 18-24 Better Than 18-24 Mean 


Measure 

Age Group 

18-24 

25-34 

35-49 

50-64 

65+ 

Physical Function 

97 

82% 

81% 

62% 

37% 

17% 

Emotional Function 

68 

60% 

61% 

48% 

61% 

70% 

Role Function 

94 

91% 

86% 

72% 

71% 

47% 

Energy Level 

63 

57% 

57% 

43% 

49% 

49% 

(N) 


(150) 

(303) 

(218) 

(241) 

(311) 


Note: Scores were transformed to a 0 to 100 distribution and rounded to 
the nearest whole number. The large percentage of persons scoring above 
the mean is due to very skewed distributions, with the mode (i.e., most 
frequently occurring score) exceeding the mean. 






























Health Status 
33 


Table 6 

Correlations Between Physical Functioning And Emotional Functionin g 


Age Group 



18-24 

25-34 

35-49 

50-64 

65+ 

Household Sample 






CORR 

.18 

.21 

.31 

.43 

.35 

SE 

.06 

.04 

.04 

.05 

.07 

(N) 

(243) 

(624) 

(590) 

(353) 

(194) 

Patient Sample 






CORR 

.13 

.16 

.28 

.30 

.34 

SE 

.08 

.06 

.07 

.07 

.06 

(N) 

(150) 

(302) 

(215) 

(239) 

(309) 

Note. CORR<=Pearson's 

product-moment 

correlation; SE = 

Standard 

error of 


correlation, N=number of cases. 







Health Status 
34 

this sample, both sexes had similar scores on Perceived Health. 

Energy level was lower for male patients as a function of age, with 
the largest difference being between the two younger age groups and the 
35-49 age group (see Figure 8). For female patients, energy level was 
constant, across age groups except for a sharp drop in the 35-49 age 
group. Male patients had significantly more energy than females in all 
age groups except for the oldest, where the scores of males and females 
were similar. 

Relation Between Physical and Emotional Functioning 

Finally, we were interested in looking at the association between 
physical and emotional functioning by age. Although previous research 
has shown moderate correlations between these two types of functional 
health, we were primarily interested in changes in the association as a 
function of age group. Results appear in Table 6. All correlations are 
statistically significant, ranging from 0.18 to 0.43 in the household 
sample and 0.13 to 0.34 in the patient sample. The correlations between 
physical and emotional functioning tended to increase with age. In the 
household sample, the correlation increased from 0.18 for the youngest 
age group to 0.43 for the 50-64 age group, and then dipped to 0.35 in 
the oldest age group. In the patient sample, the correlation increased 
throughout the age span, ranging from 0.13 to 0.34. 

Discussion 

Results show that perceived health, energy level, and physical, 
role and emotional functioning vary in different age groups. Perceived 










Health Status 
35 

health and physical and role functioning tended to be diminished in the 
older compared to the younger age groups. These decrements were 
particularly evident in the older age groups (i.e., 50-64 and 65+) and 
are consistent with conventional wisdom concerning physiologic changes 
in physical abilities with advancing age. Nevertheless, many older 
adults continue to enjoy levels of physical and role functioning 
equivalent to that of young adults. More than half of the older adults 
(i.e., age 50 and above) in the household sample had physical and role 
functioning scores that were equal to or better than the average scores 
for young adults (i.e., aged 18-24). In the patient sample, a notably 
smaller proportion of older adults had physical and role functioning 
scores that were equal to or better than the average for young adults. 

In contrast to the findings for perceived health and physical and 
role functioning, emotional functioning, on average, tends to be better 
in the older than in the younger age groups. Almost 8 of 10 older 
adults in the household sample and 7 out of 10 older adults in the 
patient sample reported equal or better emotional health than the 
average young adult. Relatively good emotional functioning 
may be associated with greater life satisfaction among older adults, 
greater acceptance of their life situation, or the acquisition of skills 
over time that allow better adaptation to life (Koenig, 1986). The 
relationship between age and energy level interacted with gender such 
that energy level declined with older age for males, but energy level 
was lowest for females in the 35-49 age group. 

The correlations between physical and emotional functioning for 
different age groups are intriguing. The magnitude of these 








Health Status 
36 


correlations for the two youngest age groups (18-24, 25-34) is similar 
to that reported for a sample of 1209 Health Insurance Experiment 
participants whose average age was 34.3 years (Ware, Davies-Avery, & 
Brook, 1980). The increasing strength of the relationship between 
physical and emotional functioning with age suggests that age-related 
decrements in physical functioning may have increasing effects on 
emotional well-being. In fact, larger associations between different 
areas of functioning with increasing age has been called the 
vulnerability hypothesis and confirmed in previous research (Fillenbaum, 
1977-78; Youmans & Yarrow, 1971). 

On the whole, the results reported here parallel those found by 
previous investigators. However, direct comparisons are not possible 
because the measures of health used in previous studies differ from 
those used here. Feller (1983) reported on data from a civilian 
noninstitutionalized population from the 1979 National Health Interview 
Survey. The results showed an association of poorer levels of health 
with increasing age, with women reporting poorer health than men at all 
ages. Similar results were reported by Branch and Jette (1981) in their 
research from the Framingham Disability Study. They reported 
significant age and sex differences, particularly in the performance of 
physical activities. Although the extent of disability in each age 
group varied, depending on the indicator of physical health, all three 
indicators showed decreased physical health with advancing age. 

Cassileth et al. (1984), in a study of outpatients in a university 
health clinic, showed that emotional health was better in older 
individuals. This trend was consistent among patients with six 








Health Status 
37 




different diagnoses (arthritis, depression, diabetes, cancer, renal 
disease, and dermatologic disorder). 

The similarity of the results observed in the two different samples 
(a large representative sample of households and a large sample of 
primary care patients) increases our confidence in the findings of the 
present study. However, the limitations of the study should be noted. 
The cross-sectional design allows us to speculate about age group 
differences in functioning, but it does not permit us to detect 
longitudinal variations in functioning that may occur within the same 
individual. Ontogenetic changes are not separable from generational 
differences in this study (Schaie, 1981). A longitudinal study of a 
cohort of individuals would be necessary to illustrate the effects of 
aging on functioning free of generational effects. 

Another limitation of this study is the lack of adjustment for 
differential mortality by age group. Results may therefore be biased 
due to the exclusion of patients from the older age groups, because 
individuals with poorer functioning may have greater risk of mortality 
than individuals with better functioning. If so, individuals with poor 
functioning would have been disproportionately excluded from the older 
age groups because of their higher mortality rates. The older age 
groups in this study may represent survivors who, not surprisingly, 
function better than non-survivors. 

Examination of the association of age with clinical factors such 
as the presence of a chronic disease was not possible because data of 
this type was not collected. Finer gradations of age groupings may have 
been more informative, but that information was not available to us in 



Health Status 
38 


the household sample. These age groupings allow comparisons of 
functioning between younger, middle-aged, and older adults. Our data 
suggest that the rate of decline in physical functioning is worst among 
older adults. An important group to examine more closely in future 
studies is the older group, which should be divided into finer age 
divisions to explore more specific age-related relationships. Data from 
other studies (Feller, 1983; Jette & Branch, 1981; Katz, 1983) suggest 
that the incremental decline in function with age is greatest in those 
over 75. Analysis of associations of functioning with socioeconomic 
status may also have been enlightening, but this information was 
unavailable. Katz et al. (1983) found that active life expectancy was 
longer in the non-poor compared to the poor group. Future studies are 
needed to further examine the relation between socioeconomic status and 
functional health. 

The picture of older adults which emerges from our results stands 
in sharp contrast to the sterotypes of the aged which characterize them 
as having "nonproductive, impaired, incapable, useless status with loss 
of virtually everything that contributes to personal capacity, 
performance, roles, and status of individuals in the world" (Eisdorfer, 
1983, p. 198). Although stereotypes such as the above are clearly 
unsupported by data in this study, emphasis on this point is critical in 
light of the fact that public policy decisions concerning people of 
advanced age can be influenced by such inaccurate beliefs. A model of 
aging which encompasses a broad definition of health that includes not 
only the biologic dimension, but also the physical, emotional and social 
functioning dimensions may show that, although certain aspects of health 









Health Status 

39 


decline with age, others may actually improve. Such a multidimensional 
model of health allows for a more favorable view of the large proportion 
of the elderly who function well, and encourages the identification of 
subgroups of individuals at risk for declines in function. Strategies 
can then be developed to remedy or compensate for limitations which may 
occur in these select individuals. 

The development and acceptance of a broad and functionally oriented 
model of aging requires that accurate data on physical, emotional and 
social functioning be collected longitudinally on people of various 
ages, taking into account important sociodemographic and clinical 
factors. This data can be used to construct the most accurate models of 
aging. Such models can be used to inform policymakers and to help 
forecast the future health of our aging population. 








Health Status 
AO 


References 

Bergner, M. , R. A. Bobbitt, W. B. Carter, and B. S. Gilson, "The 

sickness impact profile: Development and final revision of a health 
status measures," Medical Care, Vol. 19, 1981, pp. 787-805. 

Branch, L. G. , & A. M. Jette, "The Framingham disability study: 1. 

Social disability among the aging," American Journal of Public Health, 
Vol. 71, 1981, pp. 1202-1210. 

Cape, R. D. T., & P. J. Henschke, "Perspective of health in old age," 

Journal of the American Geriatrics Society, Vol. 28, 1980, pp. 295-299. 

Cassileth, B. R., E. J. Lusk, T. B. Strouse, D. S. Miller, L. L. Brown, 

P. A. Cross, and A. N. Tenaglin, "Psychosocial status in chronic 
illness," New England Journal of Medicine, Vol. 311, 1984, pp. 506-511. 

Chirikos, T. N., and G. Nestel, "Longitudinal analysis of functional 

disabilities in men," Journal of Gerontology, Vol. 40, 1985, pp. 426-433. 

Cockerham, W. C., K. Sharp, J. A. Wilcox, "Aging and Perceived 

Health Status," Journal of Gerontology, Vol. 38, 1983, pp. 349-355. 

Cronbach, L. J., "Coefficient alpha and the internal structure 
of tests," Psychometrika, Vol. 16, 1951, pp. 297-334. 

Davies, A. R., and J. E. Ware, "Measuring health perceptions in 

the Health Insurance Experiment," R-2711-HHS, Santa Monica, CA: The 
RAND Corporation, 1981. 

Eisdorfer, C., "Conceptual models of aging: The challenge of a new 
frontier," American Psychologist, Vol. 38, 1983, pp. 197-202. 

Feinson, M. C., "Aging and mental health," Research on Aging, 


Vol. 7, 1985, pp. 155-174. 





Health Status 
41 

Feller, B. A., "Americans needing help to function at home," Advance 
data, Vol. 92, 1983, pp. 1-12. 

Ferraro, K. F., "Self-ratings of health among the old and the old-old," 
Journal of Health and Social Behavior, Vol. 21, 1980, pp. 377-383. 

Fillenbaum G. G., "Screening the elderly," Journal of the American 
Geriatric Society, Vol. 33, 1985, pp. 698-705. 

Halpert, B. P., and M. K. Zimmerman, "The health status of the "old-old" 
A reconsideration," Social Science and Medicine , Vol. 22, 1986, pp. 
893-899. 

Harris & Associates, Inc., A report card on health maintenance 
organizations, 1980-1984. 

Helmstadter, G. C., Principles of psychological measurement. New York: 
Appleton-Century-Crofts, 1964. 

Jette, A. M., and L. G. Branch, "The Framingham disability study: II. 
Physical disability among the aging," American Journal of Public 
Health, Vol. 71, No. 11, 1981, pp. 1211-1216. 

Katz, S. "Assessing self-maintenance: Activities of daily living, 
mobility, and instrumental activities of daily living," Journal of 
American Geriatric Society, Vol. 31, 1983, pp. 721-727. 

Katz, S., L. G. Branch, M. H. Branson, J. A, Papsidero, J. C. Beck, and 
D. S. Greer, "Active life expectancy," New England Journal of 
Medicine, Vol. 309, 1983, pp. 1218-1224. 

Katz S., A. B. Ford, R. W. Moskowitz, B. A. Jackson, and M. W. Jaffe, 
"Studies of illness in the aged: The index of ADL: A standardized 
measure of biological and psychosocial function," Journal of 
the American Medical Association, Vol. 185, 1963, pp. 914-919. 






Health Status 
42 

Koenig, H., "Depression and dysphoria among the elderly: Dispelling a 
myth," The Journal of Family Practice, Vol. 23, 1986, pp. 383-385. 

Linn, B. S., and M. W. Linn, “Objective and self-assessed health in the 
old and very old," Social Science and Medicine, Vol. 14A, 1980, 
pp. 311-315. 

Manning, W. G., J. P. Newhouse, and J. E. Ware, "The status of health 

in demand estimation; beyond excellent, good, fair, poor," in V. R. Fuchs 
(ed.), Economic aspects of health, 1982, pp. 143-184, Chicago: 

University of Chicago Press. 

Meltzner, H. L., W. J. Carman, and J. House, "Health practices, risk 
factors, and chronic disease in Tecumseh," Preventive Medicine, 

Vol. 12, 1983, pp. 491-507. 

Montgomery, E. A., and A. V. Paranje, A report card on HMOs 1980-8A, 

Menlo Park, California: Henry J. Kaiser Family Foundation, 1985. 

Nelson, E. C., and L. A. Green, "The evaluation of medical practice 

network computer systems: Lessons from two regional projects," Journal 
of Family Practice, Vol. 19, 1984, pp. 59-65. 

Nelson, E. C., J. W. Kirk, B. W. Bise, R. J. Chapman, F. A. Hale, P. L. 
Stamps, and J. H. Wasson, "The cooperative information project: 

Part 1: A sentinal practice network for service and research in primary 
care," Journal of Family Practice, Vol. 13, 1981, pp. 641-649. 

Nelson E., B. Conger, R. Douglass, D. Gephart, J. Kirk, R. Page, A. Clark, 

K. Johnson, K. Stone, J. Wasson, and M. Zubkoff, "Functional health 
status levels of primary care patients," Journal of the American 
Medical Association, Vol. 249, 1983, pp. 3331-3338. 

Parkerson, G. R., S. H. Gehlbach, E. H. Wagner, S. A. James, and N. E. 









Health Status 
43 

Clapp, "The Duke-UNC health profile: An adult health status 
instrument for primary care," Medical Care, Vol. 19, 1981, pp. 806-828. 

Rice, D. P., and A. S. Cugliani, "Health status of American women," 
tfomen and Health, Vol. 5, 1980, pp. 5-22. 

Schaie, K. W. , "Psychological changes from mid-life to early old age: 
Implications for the maintenance of mental health. American Journal 
of Orthopsychiatry, Vol. 51, 1981, pp. 199-218. 

Stewart, A. L., R. D. Hays, and J. E. Ware, "The MOS Short Form General 
Health Survey: Reliability and validity in a patient population, 

Medical Care, Vol. 26, 1988, pp. 724-735. 

Stewart A. L., and J. E. Ware, Measuring functional status and 
well-being: The Medical Outcomes Study approach, forthcoming, 

Stewart, A. L., J. E. Ware, and R. H. Brook, "Advances in the measurement 
of functional status: Construction of aggregate indexes," Medical Care, 
Vol. 19, 1981, pp. 473-488. 

U.S. Department of Health and Human Services, Public Health Service, 

Americans assess their health: United States, 1978. DHHS Publication 
No. (PHS) 83-1570, 1983. 

Veit, C. T., and J. E. Ware, "The structure of psychological distress 
and well-being in general populations," Journal of Consulting 
and Clinical Psychology, Vol. 51, 1983, pp. 730-742. 

Ware, J. E., "Scales for measuring general health perceptions. 

Health Services Research, Vol. 11, 1976, pp. 396-415. 

Ware, J. E. "Methodological considerations in the selection of health 
status assessment procedures," in N. K. Wenger, M. E. Mattson, 

C. D. Furberg, and J. Elinson (eds.), Assessment of quality of life in 



Health Status 
44 

clinical trials of cardiovascular therapies, 1984, pp.87-117, New York 
LeJacq publishing. 

Ware, J. E., A. Davies-Avery, and R. H. Brook, Conceptualization and 
measurement of health for adults in the Health Insurance Study: 

Vol. VI, Analyses of relationships among health status measures , 

Santa Monica: The RAND Corporation, R-1987/6-HEW, 1980. 

Ware, J. E., and A. H. Karmos, Development and validation of scales 
to measure perceived health and patient role propensity: Volume II 
of a final report. Springfield, VA: National Technical Information 
Services, NTIS Publ. No. PB288-331, 1976. 

Ware, J. E., C. D. Sherbourne, A. R. Davies, and A. L. Stewart, 

"A short-form general health survey," submitted for publication.