Trends in the Prevalence and Comorbidities of Diabetes Mellitus in Nursing Home Residents in the United States: 1995-2004

Article (PDF Available)inJournal of the American Geriatrics Society 58(4):724-30 · April 2010with48 Reads
DOI: 10.1111/j.1532-5415.2010.02786.x · Source: PubMed
Abstract
To estimate trends in the prevalence and comorbidities of diabetes mellitus (DM) in U.S. nursing homes from 1995 to 2004. SAS callable SUDAAN was used to adjust for the complex sample design and assess changes in prevalence of DM and comorbidities during the study period in the National Nursing Home Surveys. Trends were assessed using weighted least squares linear regression. Multiple logistic regressions were used to calculate predictive margins. A continuing series of two-stage, cross-sectional probability national sampling surveys. Residents aged 55 and older: 1995 (n=7,722), 1997 (n=7,717), 1999 (n=7,809), and 2004 (n=12,786). DM and its comorbidities identified using a standard set of diagnosis codes. The estimated crude prevalence of DM increased from 16.9% in 1995 to 26.4% in 2004 in male nursing home residents and from 16.1% to 22.2% in female residents (all P<.05). Male and female residents aged 85 and older and those with high functional impairment showed a significant increasing trend in DM (all P<.05). In people with DM, multivariate-adjusted prevalence of cardiovascular disease increased from 59.6% to 75.4% for men and from 68.1% to 78.7% for women (all P<.05). Prevalence of most other comorbidities did not increase significantly. The burden of DM in residents of U.S. nursing homes has increased since 1995. This could be due to increasing DM prevalence in the general population or to changes in the population that nursing homes serve. Nursing home care practices may need to change to meet residents' changing needs.
Trends in the Prevalence and Comorbidities of Diabetes Mellitus in
Nursing Home Residents in the United States: 1995–2004
Xinzhi Zhang, MD, PhD,
Frederic H. Decker, PhD,
w
Huabin Luo, PhD,
z
Linda S. Geiss, MA,
William S. Pearson, PhD,
§
Jinan B. Saaddine, MD, MPH,
Edward W. Gregg, PhD,
and
Ann Albright, PhD, RD
OBJECTIVES: To estimate trends in the prevalence and
comorbidities of diabetes mellitus (DM) in U.S. nursing
homes from 1995 to 2004.
DESIGN: SAS callable SUDAAN was used to adjust for the
complex sample design and assess changes in prevalence of
DM and comorbidities during the study period in the Na-
tional Nursing Home Surveys. Trends were assessed using
weighted least squares linear regression. Multiple logistic
regressions were used to calculate predictive margins.
SETTING: A continuing series of two-stage, cross-sec-
tional probability national sampling surveys.
PARTICIPANTS: Residents aged 55 and older: 1995 (n 5
7,722), 1997 (n 5 7,717), 1999 (n 5 7,809), and 2004 (n 5
12,786).
MEASUREMENTS: DM and its comorbidities identified
using a standard set of diagnosis codes.
RESULTS: The estimated crude prevalence of DM in-
creased from 16.9% in 1995 to 26.4% in 2004 in male
nursing home residents and from 16.1% to 22.2% in female
residents (all Po.05). Male and female residents aged 85 and
older and those with high functional impairment showed a
significant increasing trend in DM (all Po.05). In people
with DM, multivariate-adjusted prevalence of cardiovascu-
lar disease increased from 59.6% to 75.4% for men and
from 68.1% to 78.7% for women (all Po.05). Prevalence of
most other comorbidities did not increase significantly.
CONCLUSION: The burden of DM in residents of U.S.
nursing homes has increased since 1995. This could be due
to increasing DM prevalence in the general population or to
changes in the population that nursing homes serve. Nurs-
ing home care practices may need to change to meet res-
idents’ changing needs. J Am Geriatr Soc 58:724–730,
2010.
Key words: diabetes mellitus; nursing home; elderly;
comorbidities
D
iabetes mellitus (DM) is the sixth leading cause of
death in the United States.
1
The attributable economic
effect of DM exceeded $174 billion in 2007.
2
DM preva-
lence has increased substantially in the United States, with
the largest absolute increases observed in those aged 65 and
older.
3
In 2007, 23.6 million people, or 7.8% of the U.S.
population (23.1% of those aged 60), had DM (diag-
nosed or not).
4
In the Medicare population, the annual in-
cidence of DM increased 23%, and prevalence increased
62% between 1994/95 and 2003/04.
5
Moreover, in addi-
tion to microvascular and macrovascular complications,
DM in older adults is associated with cognitive disorders,
physical disability, falls and fractures, and other geriatric
syndromes.
6–10
The increasing burden of DM in older
adults raises the question about its effects in long-term care
settings such as nursing homes.
The increasing life expectancy and the potentially es-
calating demand from the country’s baby boomer popula-
tion could pose future challenges for nursing homes.
Although growing postacute care in nursing homes and in-
creasing nursing home alternatives (e.g., assisted living and
home health care) may have altered nursing home us-
age,
11,12
the long-term care component of the nursing home
has not markedly changed.
13
Research on the 1970s and 1980s indicated an increas-
ing prevalence of DM and its comorbidities in the nursing
home population.
14
Little is known about trends in the
prevalence of DM in nursing home residents since the 1990s.
A recent report shows that, of nursing home residents aged
Address correspondence to Xinzhi Zhang, Division of Diabetes Translation,
National Center for Chronic Disease Prevention and Health Promotion,
Centers for Disease Control and Prevention, 4770 Buford Hwy, N.E. (K-10),
Atlanta, GA 30341. E-mail: XZhang4@cdc.gov
DOI: 10.1111/j.1532-5415.2010.02786.x
From the
Divisions of Diabetes Translation and
§
Adult and Community
Health, National Center for Chronic Disease Prevention and Health
Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia;
w
Long-Term Care Statistics Branch, Division of Health Care Statistics,
National Center for Health Statistics, Centers for Disease Control and
Prevention, Hyattsville, Maryland; and
z
Mount Olive College, Mount Olive,
North Carolina.
JAGS 58:724–730, 2010
r 2010, Copyright the Authors
Journal compilation r 2010, The American Geriatrics Society
0002-8614/10/$15.00
65 and older, approximately one-quarter had DM in 2004.
15
Other research has also provided estimates showing that
nursing home residents with DM in the 21st century have
more comorbidities, such as cardiovascular diseases, visual
problems, and kidney failure, than residents without
DM,
16,17
although it is not known whether these point es-
timates of recent prevalence represent a substantial increase
in DM and comorbidities, a continuation of a stable prev-
alence over the years, or even a decrease associated perhaps
with the increase in postacute care and the changing role of
nursing homes. No study has formally analyzed national
trends in the prevalence of DM and its comorbidities in
nursing home residents for the 1990s and beyond.
Linear trends in the prevalence of DM and comorbid-
ities in nursing home residents were investigated using data
from all National Nursing Home Surveys (NNHS) con-
ducted between 1995 and 2004. Each year’s prevalence,
overall and stratified according to demographics, were de-
termined, and trends were tested for. Adjusted prevalence,
accounting for age, sex, race and ethnicity, length of stay
(LOS), and dependence in activities of daily living (ADLs)
were also studied, enabling secular trends to be distin-
guished from variation caused by a changing population.
These findings may help understand changes in nursing
home DM and help policy makers better prepare for the
future.
METHODS
Data Source
The NNHS is a two-stage, cross-sectional probability sam-
pling survey of U.S. nursing homes.
18–20
The series of
NNHS have collected facility information and data on
current residents. All nursing homes included in this survey
had at least three beds and were certified (by Medicare
or Medicaid) or had a state license to operate as a nursing
home. Data from nursing home residents aged 55 and
older in 4 years of NNHS data were examined: 1995
(n 5 7,722), 1997 (n 5 7,717), 1999 (n 5 7,809), and 2004
(n 5 12,786).
In each NNHS, a sample of nursing homes was selected
from a data set (universe file) listing all U.S. nursing homes.
Each universe file was constructed primarily from provider
data for the period from the Centers for Medicare and
Medicaid Services (CMS) and state licensing lists. In each
participating nursing home, NNHS interviewers drew a
random sample of current residents at the time of the in-
terview. Up to six current residents were sampled at each
nursing home in the 1995, 1997, and 1999 surveys and 12
in 2004. Detailed information on each resident was col-
lected from staff members at the nursing home, typically
registered nurses, who answered questions by referring
to the residents’ medical records and other documentation.
No residents were directly interviewed. Data collected
included residents’ demographic characteristics, date of
admission, current functional status, and admission and
current diagnoses.
Measuring DM and Comorbidities
DM and its commonly associated comorbidities were iden-
tified using a standard set of diagnosis codes (International
Classification of Diseases, Ninth Revision, Clinical Modi-
fication (ICD-9-CM)), the method that the American Dia-
betes Association (ADA) uses.
21
DM and its comorbidities
were identified from up to six admission diagnoses and up
to six current diagnoses in every year, except in 2004, when
only the primary admission diagnosis was available, along
with six current diagnoses. Nursing home residents were
identified as having DM if the first three digits of the ad-
mission or current diagnosis ICD-9 code, or both, was 250.
Outcomes commonly associated with, but not unique to,
DM were divided into seven categories: neurological dis-
eases, peripheral vascular diseases, cardiovascular diseases,
renal diseases, endocrine and metabolic diseases, ophthal-
mic diseases, and other chronic conditions, as previously
used by ADA.
21
Covariates
Residents’ cha racteristics included in the an alysis were
age, sex, race or e thnicity, LOS, and ADLs. Age was cat-
egorized as 55 to 64, 65 to 74, 75 to 84, and 85 and older.
Race and ethnicity was c ategoriz ed as non-Hispanic white,
non-Hispanic black, Hispanic, and other. Following pre-
vious research,
22
LOSwascategorizedaslessthan30
days, 30 to 89 da ys, 90 to 365 days, and more than 365
days. Similar to a previous study,
23
ADL dependence was
categorized according to the deg ree of de pendenc e in the
four ADLs that were available in all years of data (bathing,
eating, dressing, and toileting): lo wer dependence (0–1
ADLs) a nd higher dependence (2–4 ADLs). Transfe rring
was e xcluded beca use question s across years of the NHHS
were not sufficie ntly compa rable; in particular, questions
on transferring were skipped for residents using wheel-
chairs in the NNHS before 2004 ( 40% of the samples in
1995 through 1999), whe reas in the 2004 NNHS , the
transferring ability of all residents was captured, including
those in whee lchairs not requiring tran sferring assista nce.
For the 1995, 1997, and 1999 NNHS, reside nts were
classified as depend ent in an ADL if the interv iewed nurs-
ing home staff ’s response was affirmative on questions,
‘Does [the resident] currently receive any assistance in
[name of ADL]?’ The 2004 NNHS que stions ca me from
the current CMS-mandated Minimum Data Set (MDS)
resident assessment form
24
and asked a bout a resident’s
level of self-performance in an ADL, with more response
categories in dicating levels of functioning: independe nt,
supervision, limited assistanc e, extensive assistance, total
dependence, or activity did not oc cur during 7 days. For
eating, dressing, a nd toileting, the definition of supervision
includes some provision of physical assistance, wherea s
for bathing, supervision is only oversight a s define d in the
MDS resident assessment.
24
Therefore for eating, d ressing,
and toileting, res idents in the 2004 NNHS were classified
as dep endent in these respective ADLs for responses of
supervision through activity d id not occ ur during 7 days.
For bathing, responses of limited assistance through ac-
tivity did not occur during 7 d ays were classified as need-
ing bathing assistance.
Socioeconomic characteristics, such as education status
and income level, were not available and therefore were not
included in the analysis.
DIABETES TRENDS IN NURSING HOMES 725JAGS APRIL 2010–VOL. 58, NO. 4
Statistical Analyses
SAS version 9.1 (SAS Institute, Cary, NC) with SUDAAN
version 9.0 (Research Triangle Institute, Research Triangle
Park, NC) was used to adjust for the complex sample de-
sign. T tests were used to assess changes in the prevalence of
DM and its comorbidities during the study period. Esti-
mated population numbers and percentages were calculated
for men and women separately for each resident character-
istic (age, race and ethnicity, LOS, and ADL dependence)
according to survey year. The prevalence of DM for men
and women according to each resident characteristic also
was calculated. Estimates were identified as unreliable if the
relative standard error was greater than 30%. Statistical
difference was set at Po.05. The linear trend in the esti-
mates of the prevalence of DM and comorbidities within
each resident characteristic (e.g., male residents) was as-
sessed by applying weighted least squares regression. The
predictor was the variable indicating the year, and the es-
timated prevalence in each year (the values of the dependent
variable) was weighted according to the inverse of the es-
timate’s variance.
Crude and adjusted prevalences of DM and its comor-
bidities were computed. Multiple logistic regressions were
used to calculate predictive margins and the standard error
of estimates for diagnosed DM and its comorbidities, ad-
justing for residents’ characteristics. Predictive margins are
a type of direct standardization in which the predicted val-
ues from the logistic regression model are averaged over the
covariate distribution of the population. The Taylor linear-
ization method was used for variance estimation.
RESULTS
Resident Characteristics
During the study period, the total estimated resident pop-
ulation in nursing homes aged 55 and older decreased from
approximately 1,489,000 to 1,410,000. Approximately
three-quarters of the current residents in nursing homes
were female. Female residents were usually older and more
likely to be non-Hispanic white than were male residents.
The mean age of male nursing home residents decreased
from 79.5 in 1995 to 78.4 in 2004 (P 5 .001), whereas the
mean age of female residents did not significantly change
(from 84.2 to 84.1; P 5 .74). In male residents, the propor-
tion aged 55 to 64 years increased from 8.3% to 12.8%
during the study period (Po.001), whereas in female res-
idents, the proportion increased from 3.0% to 4.2%
(P 5 .001). The proportion of residents with a LOS shorter
than 30 days increased from 10.8% in 1995 to 13.6% in
2004 of male residents (P 5 .007) and from 6.6% to 9.3%
of female residents (Po.001).
Prevalence of DM
From 1995 to 2004, the estimated number of nursing home
residents with DM increased from approximately 242,000
to 329,000, an overall increase in prevalence of current
residents from 16.3% to 23.4% (P for trend .01; average
change 0.8 percentage points per year). The estimated
prevalence of DM increased from 16.9% to 26.4% of male
nursing home residents (P for trend .009; average change
1.1 percentage points per year) and from 16.1% to 22.2%
of female residents (P for trend .03; average change 0.7
percentage points per year) (Table 1). The crude prevalence
difference between male and female residents was not sig-
nificant at the .05 level in 1995, but it became a significant
difference in 1997, and it remained significant (3–4 per-
centage points) thereafter.
During the study period, DM prevalence increased for
most categories of age, race and ethnicity, LOS, and ADL
dependence (Table 1). Of the oldest old residents (aged
85), the estimated prevalence of DM increased from
13.1% in 1995 to 19.5% in 2004 in male residents, for
whom there was a significant linear trend (P 5 .02), and the
estimated prevalence increased from 11.3% to 16.1% in
female residents (P 5 .04). Linear trends also were signifi-
cant for men aged 65 to 74 (P 5 .047) and 75 to 84
(P 5.02). Linear trends were significant for non-Hispanic
white male residents (P 5 .005), for non-Hispanic white
female residents (P 5 .049), for non-Hispanic black female
residents (P 5 .004), and for Hispanic female residents
(P 5.01). Also, linear trends were significant for male res-
idents with a LOS less than 30 days (P 5 .007), from 30 to
89 days (P 5 .045), and longer than 365 days (P 5 .01) and
for female residents with a LOS from 90 to 365 days
(P 5.002) and longer than 365 days (P 5 .049). Of male and
female nursing home residents, those with high ADL de-
pendence had a significant linear trend (P 5 .006 and .03,
respectively).
Prevalence of Comorbidities
Table 2 lists the prevalence of comorbidities of male and
female nursing home residents with and without DM.
Cardiovascular diseases were the major DM-related co-
morbidities. In 2004, the estimated prevalence of cardio-
vascular diseases was significantly higher in nursing home
residents who had DM than it was in those without DM
(male, 74.1% vs 67.8%; female, 78.9% vs 68.3%). The
estimated prevalence of renal symptoms also was higher for
those with DM than it was for those without DM (male,
19.1% vs 13.7%; female, 14.7% vs 10.3%). From 1995 to
2004, there was an increase in the prevalence of cardiovas-
cular diseases in nursing home residents regardless of DM
status, although the prevalence was greater for residents
with DM. The rate of change was similar for residents with
and without DM. At the same time, peripheral vascular
disease significantly decreased in female residents regardless
of DM status.
Multivariate-Adjusted Prevalence
During the study period, the multivariate-adjusted preva-
lence of DM increased from 15.4% to 23.5% (linear trend,
P 5.001) for male nursing home residents and increased
from 16.6% to 23.3% (P 5.03) for female residents (Table 3).
Moreover, the multivariate-adjusted prevalence of
cardiovascular diseases increased in residents with DM,
from 59.6% to 75.4% for men (P 5 .04) and from 68.1%
to 78.7% for women (P 5 .01). Multivariate-adjusted
prevalence of most other comorbidities did not change
significantly.
726 ZHANG ET AL. APRIL 2010–VOL. 58, NO. 4 JAGS
DISCUSSION
Data from the 1970s and 1980s suggested that DM was
beginning to increase in nursing homes.
14
The current study
found that, in nursing home older residents, the prevalence
of DM increased from 16.3% to 23.4% during a 10-year
period between 1995 and 2004. It found the prevalence of
DM in nursing home residents in 2004 to be nearly 25%,
which is consistent with the findings of a previous study.
15
These findings highlight a continuing and substantial
growth of DM in the nursing home population from the
1990s. This growth remained even after adjusting for res-
ident characteristics. The results of the current study are
similar to those of a study of the Medicare population that
found that the prevalence of DM increased from 15.3% in
1994 to 24.8% in 2003.
5
Just as other conditions, such as
obesity, have increased in the nursing home populations,
25
so has DM.
These findings are consistent with a previous report
that suggests that older Americans are reducing their use of
nursing home care.
26
Although the estimated number of
older residents in nursing homes decreased approximately
5% from 1995 to 2004, the estimated number of residents
with DM increased approximately 36%. In 1997, the cri-
teria for diagnosing DM changed from a fasting glucose
level of 140 mg/dL to a level of 126 mg/dL.
27
After this
change, some individuals who previously would not have
been diagnosed with DM might receive this diagnosis,
which may account for part of the greater DM prevalence.
Additionally, there may be an increasing tendency to assign
a DM diagnostic code, owing to greater awareness on the
part of healthcare providers and nursing home personnel.
The prospective payment system (PPS), based upon Re-
source Utilization Groups, was implemented in U.S. nursing
homes in 1998, and its effect on better identifying DM for
reimbursement is not clear. In any case, using the DM di-
agnostic code (ICD-9-CM code 250) from administrative
records to identify individuals with DM has been done
previously
5,15
and has been validated.
28,29
It also seems
unlikely that such a significant increase in DM prevalence in
the nursing home is due only to changes in definitions and in
diagnostic practices.
Microvascular and macrovascular complications of
DM have been well documented.
6–10
Recent studies have
found that heart and circulatory comorbidities were most
common in residents with DM.
16,17
Similarly, the findings
of the current study suggest that cardiovascular disease re-
mains the most common DM-related comorbidity. The
Table 1. Trends in Prevalence of Diabetes Mellitus in Nursing Home Residents Aged 55 and Older, According to Sex,
United States, 1995–2004
Characteristic
Men
P-Value
Women
P-Value
% (SE)
% (SE)
1995 1997 1999 2004 1995 1997 1999 2004
n 5 2,041 n 5 2,052 n 5 2,126 n 5 3,496 n 5 5,681 n 5 5,665 n 5 5,683 n 5 9,290
Total 16.9 (0.9) 20.0 (1.0) 20.7 (0.9) 26.4 (0.9) .009 16.1 (0.5) 16.7 (0.5) 17.4 (0.5) 22.2 (0.5) .03
Age
55–64 14.0 (2.7) 25.6 (3.5) 19.5 (2.9) 32.5 (2.8) .11 17.1 (2.9) 30.0 (3.6) 24.9 (3.0) 32.9 (2.6) .15
65–74 21.1 (2.0) 25.3 (2.2) 27.7 (2.3) 30.7 (2.1) .047 26.1 (1.9) 31.3 (2.0) 28.7 (2.0) 36.2 (1.9) .10
75–84 18.8 (1.5) 21.8 (1.6) 21.9 (1.5) 27.8 (1.4) .02 20.6 (1.0) 17.7 (0.9) 21.7 (1.0) 27.0 (1.0) .14
85 13.1 (1.3) 13.3 (1.3) 15.8 (1.5) 19.5 (1.4) .02 11.3 (0.6) 12.4 (0.6) 12.3 (0.6) 16.1 (0.6) .04
Race or ethnicity
Non-Hispanic white 16.1 (1.0) 19.0 (1.1) 20.1 (1.1) 25.0 (0.9) .005 15.1 (0.5) 15.4 (0.5) 15.8 (0.6) 19.8 (0.6) .049
Non-Hispanic black 21.3 (2.6) 27.4 (3.0) 25.1 (2.9) 31.6 (2.8) .10 26.3 (2.3) 28.3 (2.2) 29.8 (2.2) 35.8 (1.9) .004
Hispanic 19.3 (4.4) 30.1 (5.9) 22.1 (4.5) 32.6 (4.2) .19 16.8 (3.8) 22.2 (3.7) 27.7 (3.9) 35.4 (3.8) .01
Other 17.2 (3.2) 14.4 (2.9) 18.6 (3.7) 27.3 (5.0) .18 16.6 (1.6) 16.4 (1.7) 17.4 (1.8) 36.2 (4.2) .24
Length of stay, days
o 30 18.5 (2.7) 19.9 (2.6) 23.4 (2.9) 28.5 (2.4) .007 16.8 (2.1) 19.8 (2.0) 21.4 (2.1) 23.3 (1.6) .05
30–o90 24.3 (3.1) 25.1 (3.0) 25.2 (3.0) 29.7 (3.2) .045 20.3 (1.9) 19.9 (1.9) 18.0 (1.8) 25.2 (1.8) .28
90–365 15.7 (1.6) 21.4 (1.8) 21.1 (1.8) 24.7 (1.6) .09 17.0 (1.1) 18.3 (1.1) 19.4 (1.1) 23.0 (1.1) .002
4365 15.6 (1.1) 18.4 (1.3) 19.1 (1.3) 25.9 (1.3) .01 15.0 (0.6) 15.1 (0.6) 16.0 (0.7) 21.3 (0.7) .049
Activity of daily living dependence
0–1 14.7 (2.0) 19.8 (2.4) 21.2 (2.9) 25.0 (2.7) .05 16.8 (1.6) 16.8 (1.5) 17.1 (1.7) 22.9 (1.8) .08
2–4 17.4 (0.9) 19.8 (1.0) 20.8 (1.0) 26.6 (0.9) .006 16.0 (0.5) 16.6 (0.5) 17.3 (0.6) 22.2 (0.6) .03
Note: Data source is the National Nursing Home Survey from 1995 through 2004.
P-value 5 significant level for linear trend, 1995–2004.
Data were not available for all residents for all characteristics.
Weighted percentage.
SE 5 standard error.
DIABETES TRENDS IN NURSING HOMES 727JAGS APRIL 2010–VOL. 58, NO. 4
findings also show cardiovascular disease to be the most
frequent condition of nursing home residents without DM,
but the prevalence was greater for those with DM. It
was found that prevalences of renal symptoms were also
significantly higher in residents with DM than in those
without.
In addition to crude rates, the adjusted prevalence of
conditions of nursing home residents over multiple years
was considered. The results indicate that the growth in
nursing homes of the prevalence of DM and its most com-
mon comorbidity, cardiovascular disease, is not simply a
matter of changes in the demographics and usage of nursing
home residents. After adjusting for resident characteristics,
it was found that the adjusted prevalence of DM and car-
diovascular diseases were generally similar to the crude
prevalence, although in contrast to the crude difference, the
adjusted prevalence of and the increase in DM in men and
women were nearly identical. The adjusted prevalence ac-
counts for the apparent differences in nursing home demo-
graphics and usage according to sex (e.g., standardizes for
the greater proportion of men aged 55–64 and with a LOS
o30 days).
The incidence of microvascular complications may
have declined in recent decades owing to improvements in
the clinical management of people with DM.
30
A compar-
ison of four groups of patients with DM diagnosed between
1965 and 1984 showed that the cumulative incidences of
diabetic nephropathy and proliferative retinopathy were
lower in the later DM-onset groups, a decrease attributed to
better clinical management.
31
Better care management may
have reduced some complications associated with DM in
recent years in the general population, but this may not be
associated with a reduction in complications in nursing
home residents with DM. The findings, although covering a
more-recent and shorter time period, found no improve-
ments in ophthalmic and several other comorbidities over
time in nursing home residents with DM. Future care in
nursing homes may not only require practices preventing
complications in residents with DM but also, it seems, re-
quire practices to manage an increasing complexity of the
clinical conditions of persons with DM who may have ad-
ditional physical and cognitive conditions.
This study had several limitations. The NNHS did not
collect information on socioeconomic characteristics of
residents such as education and income. The set of ques-
tions on ADL dependence in 2004 had more response cat-
egories to differentiate dependence levels than did the
questions used from 1995 through 1999. This difference in
ADL questions warrants caution in the interpretation of the
meaning of the comparisons of trends in ADL dependence
between 2004 and other years. Nonetheless, the 2004 ag-
gregated ADL dependence measure was a useful control for
ADL dependence in estimating the adjusted prevalence of
DM and associated comorbidities. Lack of admission di-
agnoses in 2004 might negatively bias the estimate of the
growth in prevalence of DM. Finally, because the NNHS is
a cross-sectional survey, DM outcomes could not be linked
to DM clinical management in the facility. Future research
Table 2. Trends in Prevalence of Comorbidities in Nursing Home Residents Aged 55 and Older with and without
Diabetes Mellitus, According to Sex, United States, 1995–2004
Comorbidity
Men
P-Value
Women
P-Value
% (SE)
% (SE)
1995 1997 1999 2004 1995 1997 1999 2004
With diabetes mellitus n 5 351 n 5 408 n 5 438 n 5 911 n 5 895 n 5 957 n 5 997 n 5 2,054
Neurological disease 29.6 (2.6) 31.6 (2.4) 29.2 (2.3) 28.6 (1.8) .32 28.2 (1.5) 33.0 (1.6) 27.2 (1.5) 23.6 (1.1) .20
Peripheral vascular disease 21.6 (2.3) 16.7 (2.0) 19.3 (2.1) 17.0 (1.4) .36 15.4 (1.3) 14.8 (1.2) 14.6 (1.1) 12.8 (0.9) .01
Cardiovascular disease 59.6 (2.8) 66.8 (2.8) 70.0 (2.4) 74.1 (1.7) .06 68.0 (1.6) 69.2 (1.6) 73.5 (1.5) 78.9 (1.1) .009
Renal disease 14.4 (1.9) 13.5 (1.7) 17.1 (1.9) 19.1 (1.7) .08 15.6 (1.3) 14.6 (1.2) 14.1 (1.2) 14.7 (0.9) .60
Endocrine and metabolic disease 1.3 (0.6)
w
1.9 (0.7)
w
1.3 (0.6)
w
10.4 (1.2) .25 1.4 (0.4) 1.3 (0.4)
w
3.0 (0.6) 8.5 (0.7) .07
Ophthalmic disease 10.3 (1.7) 8.2 (1.5) 10.1 (1.5) 9.0 (1.3) .78 12.4 (1.2) 9.8 (1.0) 9.3 (1.0) 10.0 (0.8) .56
Other condition 8.7 (1.5) 4.9 (1.1) 4.8 (1.0) 2.6 (0.6) .07 4.8 (0.8) 5.3 (0.7) 4.5 (0.7) 2.2 (0.4) .04
Without diabetes mellitus n 5 1,690 n 5 1,644 n 5 1,688 n 5 2,585 n 5 4,786 n 5 4,708 n 5 4,686 n 5 7,236
Neurological disease 26.1 (1.1) 26.1 (1.2) 23.4 (1.1) 27.8 (1.2) .66 20.9 (0.6) 19.9 (0.6) 19.1 (0.6) 20.1 (0.6) .67
Peripheral vascular disease 14.7 (0.9) 15.0 (0.9) 11.5 (0.8) 12.4 (0.8) .36 13.6 (0.5) 13.1 (0.5) 11.6 (0.5) 10.7 (0.5) .048
Cardiovascular disease 55.0 (1.3) 54.9 (1.3) 56.8 (1.3) 67.8 (1.2) .047 57.8 (0.8) 58.4 (0.8) 59.4 (0.8) 68.3 (0.7) .04
Renal disease 11.6 (0.8) 10.7 (0.8) 12.8 (0.9) 13.7 (0.8) .14 12.8 (0.5) 11.2 (0.5) 10.9 (0.5) 10.3 (0.4) .12
Endocrine and metabolic disease 0.8 (0.2) 1.1 (0.3) 2.1 (0.4) 6.8 (0.6) .07 1.1 (0.2) 1.1 (0.2) 1.7 (0.2) 5.5 (0.3) .11
Ophthalmic disease 10.2 (0.7) 9.2 (0.7) 8.3 (0.7) 9.3 (0.7) .64 12.0 (0.5) 11.1 (0.5) 11.0 (0.5) 11.0 (0.5) .30
Other condition 1.9 (0.3) 1.3 (0.3) 2.3 (0.4) 2.6 (0.4) .30 2.5 (0.2) 1.8 (0.2) 2.0 (0.2) 2.4 (0.2) .82
Note: Data source is the National Nursing Home Survey from 1995 through 2004.
P-value 5 significant level for linear trend, 1995–2004.
Weighted percentage.
w
Estimated prevalence has a relative standard error430% and is considered to be statistically unreliable.
SE 5 standard error.
728 ZHANG ET AL. APRIL 2010–VOL. 58, NO. 4 JAGS
is needed to examine the extent of appropriate DM care and
management of this population.
CONCLUSION
This study provides detailed information on linear trends in
the prevalence of DM and its comorbidities in a population
not usually captured in disease surveillance studies. The
findings have important public health implications because
increasing DM prevalence in the nursing home population
may place greater challenges on nursing homes. The prev-
alence of DM has increased in long- and short-stay resi-
dents. This increase may be due to a greater prevalence of
DM, in general, in the older population, as well as the
changing role of nursing homes within the long-term care
system. As the population ages, and life expectancy in-
creases, coupled with the increasing prevalence of obesity,
more individuals with DM may enter nursing homes when
inpatient chronic care is required. More studies seem war-
ranted on the care practices and resources needed to ensure
high-quality care for current and future populations of
nursing home residents with DM.
ACKNOWLEDGMENTS
We thank the National Center for Health Statistics for pro-
viding the data in conducting this study and the National
Center for Chronic Disease Prevention and Health Promo-
tion for administrative support and effort.
The findings and conclusions in this paper are those of
the authors and do not necessarily represent the official
position of the Centers for Disease Control and Prevention.
Conflict of Interest: The editor in chief has reviewed the
conflict of interest checklist provided by the authors and has
determined that the authors have no financial or any other
kind of personal conflicts with this paper.
Author Contributions: Zhang, Decker, Luo, Geiss,
Pearson, Saaddine, Gregg, and Albright: study concept and
design, interpretation of data, and revision of manuscript.
Zhang, Decker, and Luo: acquisition of subjects and data
and analysis. Zhang and Decker: preparation of main
manuscript draft.
Sponsor’s Role: None.
REFERENCES
1. Minin
˜
o AM, Heron MP, Murphy SL et al. Deaths: Final Data for 2004. Na-
tional vital statistics reports; vol 55 no 19. Hyattsville, MD: National Center
for Health Statistics [on-line]. Available at http://www.cdc.gov/nchs/data/nvsr/
nvsr55/nvsr55_19.pdf, 2007 Accessed November 29, 2007.
2. American Diabetes Association. Economic costs of diabetes in the U.S. in
2007. Diabetes Care 2008;31:596–615.
3. Cowie CC, Rust KF, Ford ES et al. Full accounting of diabetes and pre-diabetes
in the U.S. population in 1988–1994 and 2005–2006. Diabetes Care
2009;32:287–294.
4. Centers for Disease Control and Prevention. National diabetes fact sheet:
general information and national estimates on diabetes in the United States,
2007. Atlanta, GA: U.S. Department of Health and Human Services, Centers
for Disease Control and Prevention [on-line]. Available at http://apps.nccd.
cdc.gov/DDTSTRS/FactSheet.aspx, 2008 Accessed June 26, 2008.
5. Sloan FA, Bethel MA, Ruiz D Jr. et al. The growing burden of diabetes mellitus
in the U.S. elderly population. Arch Intern Med 2008;168:192–199; discussion
199.
6. Bertoni AG, Krop JS, Anderson GF et al. Diabetes-related morbidity and
mortality in a national sample of U.S. elders. Diabetes Care 2002;25:471–475.
7. Fox CS, Coady S, Sorlie PD et al. Increasing cardiovascular disease burden due
to diabetes mellitus: The Framingham Heart Study. Circulation 2007;115:
1544–1550.
8. Gregg EW, Engelgau MM, Narayan V. Complications of diabetes in elderly
people. BMJ 2002;325:916– 917.
Table 3. Multivariate-Adjusted Prevalence of Diabetes Mellitus (DM) and Comorbidities in Nursing Home Residents
Aged 55 and Older, According to Sex, United States, 1995–2004
Characteristic
Men
P-Value
Women
P-Value
Predictive Margin, %
(SE) Predictive Margin, %
(SE)
1995 1997 1999 2004 1995 1997 1999 2004
DM
w
15.4 (0.8) 17.5 (0.9) 18.8 (0.9) 23.5 (0.8) .001 16.6 (0.5) 17.4 (0.6) 18.0 (0.6) 23.3 (0.6) .03
Residents with DM
z
Neurological disease 28.7 (2.5) 29.7 (2.3) 28.6 (2.2) 28.6 (1.8) .52 28.3 (1.5) 33.4 (1.6) 27.7 (1.5) 23.5 (1.1) .20
Peripheral vascular disease 22.4 (2.4) 16.5 (2.1) 18.9 (2.1) 16.4 (1.4) .31 15.0 (1.2) 15.0 (1.2) 15.0 (1.2) 13.0 (0.9) .07
Cardiovascular disease 59.6 (2.8) 67.6 (2.9) 69.9 (2.4) 75.4 (1.7) .04 68.1 (1.6) 68.7 (1.6) 73.6 (1.8) 78.7 (1.2) .01
Renal disease 13.7 (1.8) 12.6 (1.6) 16.8 (1.9) 18.0 (1.7) .12 16.2 (1.3) 15.1 (1.3) 14.0 (1.2) 15.1 (1.0) .67
Endocrine and metabolic disease 2.6 (0.9)
§
6.6 (1.4) 1.2 (0.6)
§
9.8 (1.1) .46 2.8 (0.6) 6.2 (0.8) 3.0 (0.6) 8.7 (0.7) .24
Ophthalmic disease 10.5 (1.8) 8.6 (1.6) 11.0 (1.6) 9.4 (1.3) .80 12.5 (1.2) 9.6 (1.0) 9.0 (1.0) 9.6 (0.8) .47
Other condition 8.3 (1.5) 4.7 (1.1) 4.5 (1.0) 2.6 (0.6) .09 4.9 (0.8) 5.5 (0.8) 4.7 (0.8) 2.3 (0.4) .04
Note: Data source is the National Nursing Home Survey from 1995 through 2004.
P-value 5 significant level for linear trend, 1995–2004. Multiple logistic regression models included age, race and ethnicity, length of stay, and activity of daily
living dependence as independent variables in calculating adjusted prevalence.
Weighted adjusted percentage.
w
The sample sizes in the analysis of adjusted prevalence of DM in all residents for 1995, 1997, 1999, and 2004 were 7,662, 7,643, 7,695, and 12,679,
respectively.
z
The sample sizes in the analysis of adjusted prevalence of comorbidities in residents with DM for 1995, 1997, 1999, and 2004 were 1,235, 1,344, 1,413, and
2,938, respectively.
§
Estimated prevalence has a relative standard error 430% and is considered to be statistically unreliable.
SE 5 standard error.
DIABETES TRENDS IN NURSING HOMES 729JAGS APRIL 2010–VOL. 58, NO. 4
9. Gregg EW, Gu Q, Williams D et al. Prevalence of lower extremity diseases
associated with normal glucose levels, impaired fasting glucose, and diabetes
among U.S. adults aged 40 or older. Diabetes Res Clin Pract 2007;77:485–488.
10. Gregg EW, Yaffe K, Cauley JA et al. Is diabetes associated with cognitive
impairment and cognitive decline among older women? Study of Osteoporotic
Fractures Research Group. Arch Intern Med 2000;160:174–180.
11. McCormick JC, Chulis GS. Growth in residential alternatives to nursing
homes: 2001. Health Care Financ Rev 2003;24:143–150.
12. Shaughnessy PW, Kramer AM. The increased needs of patients in nursing homes
and patients receiving home health care. N Engl J Med 1990;322:21–27.
13. Decker FH. Nursing Homes, 1977–99: What Has Changed, What Has Not?
Hyattsville, MD: National Center for Health Statistics, 2005.
14. Mayfield J, Deb P, Potter D. Diabetes and long-term care. In: Harris M, Cowie
C, Stern M, editors. Diabetes in America, 2nd Ed. Bethesda, MD: National
Institutes of Health, National Institute of Diabetes and Digestive and Kidney
Diseases, 1995, pp. 571–590.
15. Resnick HE, Heineman J, Stone R et al. Diabetes in U.S. nursing homes, 2004.
Diabetes Care 2008;31:287–288.
16. Duffy RE, Mattson BJ, Zack M. Comorbidities among Ohio’s nursing home
residents with diabetes. J Am Med Dir Assoc 2005;6:383–389.
17. Travis SS, Buchanan RJ, Wang S et al. Analyses of nursing home residents with
diabetes at admission. J Am Med Dir Assoc 2004;5:320–327.
18. Gabrel C, Jones A. The National Nursing Home Survey: 1995 summary. Vital
Health Stat 2000;13:1–83.
19. Gabrel CS, Jones A. The National Nursing Home Survey: 1997 summary. Vital
Health Stat 2000;13:1–121.
20. Jones A. The National Nursing Home Survey: 1999 summary. Vital Health
Stat 2002;13:1–116.
21. Hogan P, Dall T, Nikolov P. Economic costs of diabetes in the US in 2002.
Diabetes Care 2003;26:917– 932.
22. Decker FH. The relationship of nursing staff to the hospitalization of nursing
home residents. Res Nurs Health 2008;31:238–251.
23. McNabney MK, Wolff JL, Semanick LM et al. Care needs of higher-func-
tioning nursing home residents. J Am Med Dir Assoc 2007;8:409–412.
24. Centers for Medicare and Medicaid Services. Minimum Data Set (MDS)
Nursing Home Resident Assessment. Baltimore, MD: U.S. Department of
Health and Human Services [on-line]. Available at http://www.cms.hhs.gov/
NursingHomeQualityInits/ Accessed March 16, 2009.
25. Lapane KL, Resnik L. Obesity in nursing homes: An escalating problem. J Am
Geriatr Soc 2005;53:1386–1391.
26. Bishop CE. Where are the missing elders? The decline in nursing home use,
1985 and 1995. Health Aff (Millwood) 1999;18:146–155.
27. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus.
Report of the expert committee on the diagnosis and classification of diabetes
mellitus. Diabetes Care 1997;20:1183–1197.
28. Hux JE, Ivis F, Flintoft V et al. Diabetes in Ontario: Determination of prev-
alence and incidence using a validated administrative data algorithm. Diabetes
Care 2002;25:512–516.
29. Saydah SH, Geiss LS, Tierney E et al. Review of the performance of methods to
identify diabetes cases among vital statistics, administrative, and survey data.
Ann Epidemiol 2004;14:507– 516.
30. Chaturvedi N. The burden of diabetes and its complications: Trends and
implications for intervention. Diabetes Res Clin Pract 2007;76(Suppl 1):
S3–S12.
31. Hovind P, Tarnow L, Rossing K et al. Decreasing incidence of severe diabetic
microangiopathy in type 1 diabetes. Diabetes Care 2003;26:1258–1264.
730 ZHANG ET AL. APRIL 2010–VOL. 58, NO. 4 JAGS
    • "Although there are differences in the physical and cognitive effects of the different types of dementias, all are progressive, involve increasing physical and mental deterioration and lead to a person with dementia to become increasingly dependent. Diabetes mellitus is seen in 10–25 % of older people [5, 6], and in nursing homes, up to 27 % of residents may have diabetes789. As with dementia, the prevalence of type 2 diabetes is increasing globally [10, 11] and there is evidence to suggest there is a link between cognitive dysfunction and type 2 diabetes [1, 12, 13]. "
    [Show abstract] [Hide abstract] ABSTRACT: Background Worldwide, the prevalences of diabetes and dementia are both increasing, particularly in older people. Rates of diabetes in people with dementia are between 13 and 20 %. Diabetes management and diabetic self-care may be adversely affected by the presence of dementia. There is a need to know what interventions work best in the management of diabetes in people living with dementia (PLWD) in different settings and at different stages of the dementia trajectory. The overall aim is to develop an explanatory account or programme theory about ‘what works’ in the management of diabetes in people in what context and to identify promising interventions that merit further evaluation. Methods/design This study uses a realist approach including studies on the management of diabetes in older people, medication management, diabetes-related self-care, workforce issues and assessment and treatment. We will use an iterative, stakeholder driven, four-stage approach. Phase 1: development of initial programme theory/ies through a first scoping of the literature and consultation with key stakeholder groups (user/patient representatives, dementia-care providers, clinicians, diabetes and dementia researchers and diabetes specialists). Phase 2: systematic searches of the evidence to test and develop the theories identified in phase 1. Phase 3: validation of programme theory/ies with a purposive sample of participants from phase 1. Phase 4: actionable recommendations for the management of diabetes in PLWD. Discussion A realist synthesis of the evidence will provide a theoretical framework (i.e. an explanation of how interventions work, for whom, in what context and why) for practice and future research work that articulates the barriers and facilitators to effective management of diabetes in people with dementia. By providing possible explanations for the way in which interventions are thought to work and how change is achieved, it will demonstrate how to tailor an intervention to the setting and patient group. The propositions arising from the review will also inform the design of future intervention studies. Systematic review registration PROSPERO registration number CRD42015020625.
    Full-text · Article · Dec 2016
    • "Thus, a growing proportion of the population is affected by multimorbidity as populations age [9], particularly in countries with demographic patterns like the United Kingdom [10]. Previous studies111213 most commonly reported the following disease groups as likely to concur: cardiovascular diseases, diabetes mellitus, chronic kidney disease, chronic musculoskeletal disorders, chronic lung disorders, and mental ill health (particularly dementia and depression). There is also a greater burden of multimorbidity at younger ages (younger than 65 years) in deprived areas [7]. "
    [Show abstract] [Hide abstract] ABSTRACT: Patients with multiple conditions have complex needs and are increasing in number as populations age. This multimorbidity is one of the greatest challenges facing health care. Having more than 1 condition generates (1) interactions between pathologies, (2) duplication of tests, (3) difficulties in adhering to often conflicting clinical practice guidelines, (4) obstacles in the continuity of care, (5) confusing self-management information, and (6) medication errors. In this context, clinical decision support (CDS) systems need to be able to handle realistic complexity and minimize iatrogenic risks. The aim of this review was to identify to what extent CDS is adopted in multimorbidity. This review followed PRISMA guidance and adopted a multidisciplinary approach. Scopus and PubMed searches were performed by combining terms from 3 different thesauri containing synonyms for (1) multimorbidity and comorbidity, (2) polypharmacy, and (3) CDS. The relevant articles were identified by examining the titles and abstracts. The full text of selected/relevant articles was analyzed in-depth. For articles appropriate for this review, data were collected on clinical tasks, diseases, decision maker, methods, data input context, user interface considerations, and evaluation of effectiveness. A total of 50 articles were selected for the full in-depth analysis and 20 studies were included in the final review. Medication (n=10) and clinical guidance (n=8) were the predominant clinical tasks. Four studies focused on merging concurrent clinical practice guidelines. A total of 17 articles reported their CDS systems were knowledge-based. Most articles reviewed considered patients' clinical records (n=19), clinical practice guidelines (n=12), and clinicians' knowledge (n=10) as contextual input data. The most frequent diseases mentioned were cardiovascular (n=9) and diabetes mellitus (n=5). In all, 12 articles mentioned generalist doctor(s) as the decision maker(s). For articles reviewed, there were no studies referring to the active involvement of the patient in the decision-making process or to patient self-management. None of the articles reviewed adopted mobile technologies. There were no rigorous evaluations of usability or effectiveness of the CDS systems reported. This review shows that multimorbidity is underinvestigated in the informatics of supporting clinical decisions. CDS interventions that systematize clinical practice guidelines without considering the interactions of different conditions and care processes may lead to unhelpful or harmful clinical actions. To improve patient safety in multimorbidity, there is a need for more evidence about how both conditions and care processes interact. The data needed to build this evidence base exist in many electronic health record systems and are underused.
    Full-text · Article · Jan 2015
    • "The majority of nursing home patients receive multiple drug therapy and drug-related problems (DRPs) are common [14] . Patients with diabetes have a higher burden of comorbidities compared to patients without diabetes [10,15], further complicating management of care. Hypoglycemic episodes occur frequently, due to both an overly intensive drug regime [7,11,16] and concurrent diseases [17,18]. "
    [Show abstract] [Hide abstract] ABSTRACT: Aims Determine prevalence of diabetes, and describe use of blood glucose lowering (BGL) drugs and glycemic control in Norwegian nursing homes. Methods In this cross-sectional study we collected details of BGL drugs, capillary blood glucose measurements (CBGM) in the last four weeks and HbA1c measurements in the last 12 months from the medical records of patients with diabetes, within a population of 742 long-term care patients from 19 randomly selected nursing homes in Western Norway. Descriptive statistics were applied, and Pearson's chi-squared (P≤0.05) or non-overlapping 95% confidence intervals were interpreted as significant effects. Results 116 patients (16%) had diabetes, 100 of these gave informed consent and medical data were available. BGL treatment was a s follows 1) insulin only (32%), 2) insulin and oral antidiabetics (OADs) (15%), 3) OADs only (27%) and 4) no drugs (26%). Patients with cognitive impairment were less likely to receive medical treatment (P = 0.04). CBGM and HbA1c measurements were performed for 73% and 77% of patients, respectively. Mean HbA1c was 7.3% (57 mmol/mol), 46% of patients had an HbA1c <7.0% (53 mmol/mol), and CBGM consistent with risk of hypoglycemia was found for 60% of these patients. Conclusions Prevalence of diabetes and BGL treatment in Norwegian nursing homes is comparable to other European countries. Although special care seems to be taken when choosing treatment for patients with cognitive impairment, there are signs of overtreatment in the population as a whole. The strict glycemic control unveiled may negatively affect these frail patients’ quality of life and increase the risk of early death.
    Full-text · Article · Jul 2014
Show more