|
|
Publications and Products
The Economics of Diabetes Mellitus:
An Annotated Bibliography
COSTS
OF DIABETES
Direct
Medical Costs
Inpatient
Care
139
TITLE:
Amputations in the Surgical Budget. Solomon, C.; van Rij, A.M.; Barnett,
R.; Packer, S.G.; Lewis-Barned, N.J. New Zealand Medical Journal.
107(973): 78-80. March 9, 1994.
OBJECTIVE:
To describe the extent and distribution of inpatient costs for nontraumatic
lower limb amputations and to identify areas for cost-saving strategies.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Length of hospital stay was the largest determinant of cost. The need
for long preoperative management and postoperative rehabilitation hospital
time might be achieved more economically in less costly settings. Most
amputations among patients with diabetes were of the minor type; among
patients without diabetes, most were major.
RECOMMENDATION:
Greater use of outpatient facilities and providing increased resources
for screening, preventive education, podiatry, and early intervention
may offer both limb- and cost-saving opportunities. Free flow and integration
of clinical and cost information could provide the physician with a more
adequate basis for best use of resources.
ABSTRACT:
Investigators reviewed data on 134 hospital admissions for nontraumatic
lower limb amputations from July 1989 to April 1992 at Otago (New Zealand)
Health Board hospitals in order to assess potential cost-saving options.
Detailed information on costs, clinical activity, and length of hospital
stay was available for analysis. Statistical analysis used the Mann-Whitney
U test and the chi-square test. Total in-hospital cost was $1.09 million
(New Zealand dollars), with an annual cost of $388,000. The mean cost
for general surgical admission was $11,342 (median $21,367), which was
significantly higher (p < 0.001) than that for orthopedic patients,
whose mean cost was $2,318 (median $6,277). General surgical patients,
compared with orthopedic patients, required major amputation (above and
below the knee) more frequently (46.5 percent versus 10.4 percent) and
had concomitant diabetes more frequently (36.0 percent versus 4.2 percent).
Of amputations among patients with diabetes, 73.9 percent were minor (forefoot
and toe), versus 29.0 percent for nondiabetic patients (p < 0.005).
For all admissions, ward costs accounted for 55.6 percent of in-hospital
costs per admission, making length of stay the most important determinant
of admission cost. The long preoperative stay spent to avoid amputation
might be decreased by greater use of outpatient facilities for investigation
and initial management. The longest portion of hospital stay is for accommodating
major amputees until they are fully mobile using an artificial limb; use
of less costly facilities outside acute hospital wards may achieve the
same ends more economically. 3 figures, 19 references.
140
TITLE:
Amputations in the Surgical Budget (letter). Simmons, D.; Thomson, C.;
Scott, D. New Zealand Medical Journal. 107 (978): 208-209. May
25, 1994.
OBJECTIVE:
To report cost findings about patients with diabetes who underwent amputation.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Costs of about $21,000 for the year in which amputation took place were
very close to costs reported in a similar study.
RECOMMENDATION:
With education on foot care, daily self-inspection of the feet, and effective
monitoring and early intervention, at least 50 percent of amputations
should be preventable.
ABSTRACT:
The authors determined the mean 1-year cost of $20,881 (in 1987 dollars)
for treating 20 patients with diabetes during the year of their amputation(s).
This figure is close to the figure of $21,439 for a more recent study
in Dunedin, New Zealand. The authors point out that using diabetes codes
to identify patients with diabetes has been shown to miss 45 percent of
these patients. In the present study, many surgical patients with diabetes
remained undiagnosed throughout their admissions. The authors found in
their own study that 40 percent of patients with diabetes had not had
their feet inspected by their diabetes care attendant in the previous
12 months and that 57 percent had poor nail or skin care. The authors
suggest that a team approach makes early intervention more likely and
state that this would involve close liaison between the diabetes nurse,
physician, podiatrist, and preferably a surgical team with skills in vascular
surgery and a special interest in the diabetic foot. 6 references.
141
TITLE:
The Burden of Diabetes: The Cost of Diabetes Hospitalizations in Wisconsin,
1994. Ford, E.J.; Remington, P.L. Wisconsin Medical Journal. 95(3):
168-169. March 1996.
OBJECTIVE:
To determine for Wisconsin in 1994 the cost of all hospitalizations in
which diabetes was listed as a diagnosis; to analyze these hospitalizations
and their costs.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Societal.
CONCLUSION:
Diabetes cost Wisconsin residents $545 million, or $150 per person, in
1994.
RECOMMENDATION:
Efforts should be made to prevent or delay diabetes and to reduce its
complications. Interventions that reduce diabetes-related health care
costs will benefit all citizens.
ABSTRACT:
The state of Wisconsin maintains a database of information on all hospitalizations,
including costs. For 1994, hospital discharge data revealed 53,791 admissions
in Wisconsin for which diabetes was listed as a primary or secondary diagnosis.
Total costs were $545 million; costs of individual admissions ranged from
$150 to $1.5 million (median, $6,280; mean, $10,120). For 8 percent (4,234)
of admissions, diabetes was listed as the principal diagnosis; the cost
of these admissions was $42 million. For the remaining 92 percent, diabetes
was listed as a secondary diagnosis. Among these events, costs by principal
diagnosis were $208 million for circulatory diseases (38 percent of total
costs); $42 million for digestive diseases (8 percent); $35 million for
respiratory diseases (6 percent); $32 million for musculoskeletal conditions
(6 percent); $30 million for complications of medical care (5 percent);
$28 million for neoplasms (5 percent); $21 million for injury and poisoning
(4 percent); $16 million for genitourinary conditions (3 percent); and
$81 million for all other causes (15 percent). By age, costs were $213
million for persons under age 65 (39 percent), $176 million for those
aged 65-74 (32 percent), and $156 million for persons aged 75 and over
(29 percent). Medicare was billed for 68 percent of the costs; other government-supported
reimbursement, including Medicaid, brought the total supported by taxpayers
to over 75 percent. Private insurance accounted for 23 percent. 1 figure,
1 table, 7 references.
142
TITLE:
Clinical Features and Health-Care Costs of Diabetic Nephropathy. Narins,
B.E.; Narins, R.G. Diabetes Care. 11 (10): 833-839. November/December
1988.
OBJECTIVE:
To review the natural history, treatment, and costs of treating nephropathy
associated with diabetes.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Review of studies.
Perspective: Health care system.
CONCLUSION:
Diabetes-associated nephropathy is rapidly becoming the largest cause
of renal replacement therapy.
RECOMMENDATION:
Further study of the pathogenesis and treatment of diabetes and diabetic
nephropathy is needed.
ABSTRACT:
The authors review the incidence, course, prognosis, and treatment of
diabetes-related nephropathy. Of patients with type 1 diabetes, 40 to
50 percent develop clinically evident renal disease; two-thirds of those
who do (30 percent overall) require replacement therapy for end-stage
renal disease (ESRD). From 1973 to 1980, the percentage of ESRD attributable
to diabetes tripled. ESRD from diabetes is disproportionally common among
blacks (versus whites). Microalbuminuria and macroalbuminuria develop
5 to 10 years and 10 to 15 years, respectively, after initial diagnosis
of diabetes. Glomerular filtration rate and serum creatinine generally
remain constant during the first 10 to 15 years of insulin dependence.
ESRD occurs approximately 5 years after persistent proteinuria and azotemia
develop. Among patients with type 1 diabetes-related nephropathy, almost
two-thirds die from renal failure, 25 to 30 percent from cardiovascular
complications, and 5 to 15 percent from other disorders; patients with
type 2 diabetes die more frequently from cardiovascular complications.
After 5 years, survival is about 75 percent for patients who receive kidney
transplants from living related donors, versus 40 to 50 percent for patients
who receive other ESRD therapy. In 1982, average hospital days for patients
with diabetes receiving dialysis were 37 percent greater than for ESRD
patients without diabetes. In the first post-transplantation year, patients
with diabetes averaged 45 percent more days than those without diabetes.
Respective annual per capita costs for all patients with ESRD and for
patients with diabetic ESRD were $23,833 and $35,616 for hemodialysis,
$23,076 and $36,585 for peritoneal dialysis, and $32,075 and $43,010 for
kidney transplantation. 6 figures, 30 references.
143
TITLE:
The Cost of Hospitalization for the Late Complications of Diabetes in
the United States. Jacobs, J.; Sena, M.; Fox, N. Diabetic Medicine.
8 (Symposium): S23S29. 1991.
OBJECTIVE:
To calculate for U.S. patients with diabetes the risk of hospitalization
for late complications (e.g., cardiovascular and kidney disorders).
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Prevalence-based costs.
Perspective: Health care system.
CONCLUSION:
Late complications of diabetes accounted for 2 percent of the total hospital
admissions in the United States in 1987. The total cost of treating these
complications was estimated at $5.091 billion (cardiovascular, 74 percent;
renal diseases, 10 percent; neuropathy, 3.6 percent; ophthalmic disorders,
1.5 percent; and unspecified diseases, 10 percent).
RECOMMENDATION:
None.
ABSTRACT:
The cost of health care in the United States is usually attributed to
a disease according to the primary diagnosis on the patient's medical
record. This method underestimates the cost of hospital care for patients
with diabetes, which often contributes to a variety of other diseases
or complications. The authors used the 1987 National Hospital Discharge
Survey to determine hospitalization rates for patients with diabetes and
a matched control group; excess hospitalizations and associated costs
were attributed to diabetes. Only those hospitalizations with a diagnosis-related
group that matched a late complication were included in the analysis.
Average cost per day for complications was obtained from the Pracon Med
PROs Audit, a Medicare database. For neuropathy, the highest hospitalization
rate among persons with diabetes occurred in the 45- to 54-year age group,
6.74 per 1.000; the highest rate for controls (1.80) was in the group
75 years of age and over. Patients with diabetes aged 45 or younger were
46 times as likely to be hospitalized due to neuropathy as those in the
control group (95 percent CI, 45.0 to 47.4). Those with diabetes were
21.8 times as likely to be admitted for skin ulcers/gangrene (95 percent
CI, 21.6 to 22.0), 15 times as likely for peripheral vascular disease,
10 times as likely for congestive heart failure, and almost 10 times as
likely for atherosclerosis. The risk of cerebrovascular accident and heart
disease was 6 to 10 times greater for diabetic patients than for controls.
Hospitalization from renal complications was much more common at younger
ages for those with diabetes. Late complications of diabetes resulted
in nearly 7 million hospital days; inpatient hospitalization costs were
estimated to be $5.091 billion. 1 figure, 5 tables, 16 references.
144
TITLE:
The Costs of Diabetes-Related Lower Extremity Amputations in the Netherlands.
van Houtum, W.H.; Lavery, L.A.; Harkless, L.B. Diabetic Medicine.
12(9): 777781. September 1995.
OBJECTIVE:
To identify for 1992 the duration of hospitalization for diabetesrelated
lower extremity amputations and their associated costs in The Netherlands.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Days in the hospital for lower extremity amputations in the population
with diabetes totaled 65,778 (41.8 days per hospitalization); mean cost
per hospitalization was £ 10,531.
RECOMMENDATION:
The direct cost of prevention strategies for high-risk patients with diabetes
should be analyzed to determine the financial impact of the multidisciplinary
clinics that provide these preventive services.
ABSTRACT:
The authors used 1992 data from the Dutch organization SIG Health Care
Information to identify all lower extremity amputations in The Netherlands.
Using average costs associated with such amputations obtained from the
National Health Tariffs Authority, Netherlands, the authors were able
to determine the average cost of hospitalizations. Total direct costs
included those associated with hospital stay and the average procedure-specific
costs (for surgeon and anesthetist fees and the operating room) for the
specific level of amputation. Patient-specific charges were not available.
In 1992, there were 1,810 diabetesrelated lower extremity amputations
involving 1,575 hospitalizations. The mean hospital stay was 41.8 days,
with a mean cost of £ 10,531. The total direct cost associated with
hospitalization and surgery for diabetesrelated amputations was £
16.59 million, and more than 65,000 hospital days were used. Hospital
days for amputation among persons without diabetes totaled almost 46,000.
Hospital stays in this group were significantly shorter (mean: 31.8 days,
p < 0.001) than in persons with diabetes. When outcomes were adjusted
for age, more multiple amputations were performed in persons with diabetes
(13.6 percent) than in those without diabetes (6.6 percent, p < 0.001).
Indirect costs and costs of continued medical care at rehabilitation centers
and other medical facilities were not taken into account in this study,
and, thus, the actual costs associated with lower extremity amputations
in the population with diabetes were even higher than this study reported.
Studies have shown that lower extremity amputations are at least partly
preventable with multidisciplinary treatment programs. 1 figure, 4 tables,
19 references.
145
TITLE:
Counting the Cost of Diabetic Hospital Admissions From a Multi-Ethnic
Population in Trinidad. Gulliford, M.C.; Ariyanayagam-Baksh, S.M.; Bickram,
L.; Picou, D.; Mahabir, D. Diabetic Medicine. 12(12): 1077-1085.
December 1995.
OBJECTIVE:
To measure the impact of diabetes on a major hospital in northern Trinidad
and to evaluate the relationship between ethnicity and morbidity from
diabetes.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Costs of admission for diabetes in the hospital wards surveyed accounted
for at least 8 percent of total expenditures at the hospital and 23 percent
of bed occupancy.
RECOMMENDATION:
Investment in better-quality community care for patients with diabetes
would be economically advantageous for countries such as Trinidad and
Tobago.
ABSTRACT:
Admission records at Port of Spain Hospital in Trinidad were reviewed
for all patients with diabetes admitted to any of seven medical wards,
five general surgical wards, or two ophthalmology wards over a 26-week
period beginning in October 1993. During the study period, 1,722 (13.6
percent) of a total of 12,673 admissions involved patients (n = 1,447)
with diabetes. One hundred seventy-eight of these patients had more than
one admission; 1,269 had a single admission. Admission rates increased
with age and were approximately 40 percent higher in the population of
Indian descent than in those of African origin. Diabetes prevalence in
the population had a primary influence on admission rates, with level
of admissions and morbidity also influenced by accessibility to hospital
services and quality of care in the community. Conditions frequently present
on admission included disorders of blood glucose control, foot diseases,
renal impairment, cardiac failure, angina, and myocardial infarction and
stroke. The fatality rate among admissions was 8.9 percent (154 deaths),
and the mean length of stay was 4 days. The presence of renal impairment
on admission was associated with a fatality rate of 21 percent. The annual
number of bed days occupied by persons with diabetes was 26,659, which
represented 23 percent of all bed days on the study wards. Annual financial
cost for all patients admitted with diabetes was TT$10.7 million, including
TT$3.1 million for patients admitted with foot problems, TT$2.5 million
for those with glucose control problems, and TT$5.1 million for those
with other diagnoses. The mean cost of 1 patient admission with diabetes
was TT$3,096. 1 figure, 5 tables, 28 references.
146
TITLE:
Diabetes - Inpatient Utilisation, Costs and Data Validity: Dunedin 1985-9.
Phillips, D.E.; Mann, J.I. New Zealand Medical Journal. 105(939):
313-315. August 1992.
OBJECTIVE:
To describe the impact of diabetes on public hospital inpatient services
in an urban New Zealand 450-bed university teaching hospital from 1985
through 1989.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Admission, bed utilization rates, and associated costs for diabetes rose
by factors of 3.2, 3.8, and 2.8, respectively, during the study period.
Most utilization increases were accounted for by admissions in which diabetes
was a subsidiary diagnosis. At $16,000 (New Zealand dollars) per patient,
admissions for diabetic peripheral vascular disease were the most costly.
RECOMMENDATION:
Because diabetes consumes many hospital resources in this region, especially
for the growing elderly population, alternative options for providing
care are needed. Data audits must be performed or restricted to well-defined
groups (e.g., diabetes as principal diagnosis only) to achieve an accurate
analysis.
ABSTRACT:
Hospital discharge data were collected to describe trends in utilization
and costs of admissions for diabetes and associated conditions in an urban
teaching hospital in New Zealand from 1985 through 1989. A validation
study was performed to assess the impact of underreporting or nonreporting
and misclassification of diabetes as a discharge diagnosis. Admissions
for diabetes-related conditions represented 5 percent of all hospital
inpatient costs in 1989, but 45 percent of admissions where diabetes should
have been a subsidiary diagnosis were omitted from the discharge data.
During the study period, an average 4.6 percent of total hospital bed
utilization was for diabetes admissions, 1.9 percent as the principal,
and 2.7 percent as the subsidiary diagnosis. Length of stay averaged 12.3
days for admissions with diabetes as the principal diagnosis and 15.4
days for diabetes as a subsidiary diagnosis. The highest daily cost was
for diabetic eye disease (90 percent of these costs were for operative
treatment of cataracts, 10 per-cent for treatment of retinopathy), and
the highest total and mean per-patient costs were for peripheral vascular
disease. Annual bed utilization where diabetes was a subsidiary diagnosis
rose from 1,580 to 5,972 days during the study period, but utilization
days where diabetes was the principal diagnosis declined from 2,036 to
1,860. Macrovascular disease accounted for most admissions and bed utilization
when diabetes was the subsidiary diagnosis. Patients with diabetes are
surviving to an age where vascular disease becomes common. Admission rates
averaged 3/1,000 for the 0 to 44 age group and 60/1,000 for the $ 75 age
group; bed utilization rates were similar, and costs for the two groups
rose from $2,400/1,000 to $181,900/1,000. Increased admissions and bed
utilization rates are likely due to an increase in available facilities
and changes in treatment practices; the validation study suggests that
changes in recording practice by doctors is unlikely to be a major factor.
The accuracy of aggregate costs depends on the accuracy of the discharge
data, and diabetes has been significantly underreported. 4 tables, 23
references.
147
TITLE:
Diabetes-Related Hospitalization and Hospital Utilization. Aubert, R.E.;
Geiss, L.S.; Ballard, D.J.; Cocanougher, B.; Herman, W.H. In: Diabetes
in America. 2nd edition. National Diabetes Data Group, ed. National
Institute of Diabetes and Digestive and Kidney Diseases. NIH Publication
No. 95-1468. 1995: 553-569.
OBJECTIVE:
To describe rates and trends for hospitalization of persons with diabetes
using data from the National Hospital Discharge Survey (NHDS) and the
National Health Interview Survey (NHIS).
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Prevalence-based costs.
Perspective: Health care system.
CONCLUSION:
Persons with diabetes have higher rates of hospitalization than persons
without diabetes.
RECOMMENDATION:
None.
ABSTRACT:
The authors assessed rates and trends of hospitalization for persons with
diabetes between 1980 and 1990. Per the NHDS, in 1990, diabetes was listed
as a primary or secondary diagnosis for 2.8 million hospitalizations (24.5
million hospital days); the proportion of such hospitalizations for which
diabetes was listed as the primary diagnosis declined from 29 percent
in 1980 to about 15 percent in 1990. In the latter year, for discharges
that listed diabetes, only diseases of the circulatory system exceeded
diabetes as a primary diagnosis. Per the 1989 NHIS, among adults with
diabetes, 8.3 percent reported multiple hospital admissions and 15.5 percent
reported a single admission. In the population aged 18 years and over,
persons with diabetes were 3 times more likely than those without diabetes
to report hospitalization in the previous year. Women with diabetes were
6 to 18 percent more likely to report being hospitalized than their male
counterparts. Reported single and multiple hospitalization rates were
proportionally higher with more diabetes-related complications. In 1990,
per the NHDS, 89 percent of admissions with diabetes as a listed diagnosis
were of patients aged 45 years or older. Hospitalizations that listed
diabetic ketoacidosis increased 27 percent from 1980 to 1990 (age adjusted);
hospitalizations for lower extremity amputations related to diabetes rose
from about 36,000 to about 54,000 during the period. From 1980 to 1990,
the average length of stay when diabetes was the primary diagnosis decreased
from 10.5 to 7.8 days. The American Diabetes Association estimated costs
associated with hospital care for diabetes to be $37.2 billion in 1992;
an estimate by Rubin et al. (1994) put the figure at $55 billion. 19 figures,
8 tables, 32 references.
148
TITLE:
Diagnosis Related Groups, Resource Utilization, Age, and Outcome for Hospitalized
Nephrology Patients. Muñoz, E.; Thies, H.; Maesaka, J.K.; Angus,
G.; Goldstein, J.; Wise, L. American Journal of Kidney Diseases.
11(6): 481-488. June 1988.
OBJECTIVE:
To examine hospital costs and outcome by age of patients for all nephrology
diagnosis-related groups at an academic medical center outside New York
City.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Nephrology patients with diabetes generated significantly higher hospital
costs than nephrology patients without diabetes due to greater resource
consumption. Nephrology patients with diabetes generated a mean net loss
per admission of $2,353 relative to diagnosis-related group (DRG) payments.
RECOMMENDATION:
Reimbursement of nephrology patients with diabetes should be made more
equitable under the DRG system.
ABSTRACT:
The authors analyzed hospital cost and outcome for 784 adult and pediatric
nephrology admissions over a 2-year period at an academic medical center.
Seventy-eight of these patients had diabetes mellitus. Hospital costs
(excluding physician fees) were calculated by using per-patient charge
data and converting charges to costs via the cost-to-charge ratio data
used by the hospital. These costs were then compared with DRG reimbursement
for all patients using the DRG payment methodology and DRG case-mix index.
The 784 nephrology admissions generated approximately $5 million in hospital
costs. Patients with diabetes generated significantly higher mean hospital
costs, $8,893, than those without diabetes, $6,153 (p < 0.03). For
nephrology patients overall, those in age groups 55 years and over would
have generated losses relative to DRG payment, peaking at a mean loss
of $5,343 for patients aged 85 and over. The deficit was due primarily
to more expensive care provided to patients 65 and over. Reimbursement
under the DRG system should be made more equitable for patients with diabetes.
5 figures, 8 tables, 7 references.
149
TITLE:
The First Two Years of Type I Diabetes in Children: Length of the Initial
Hospital Stay Affects Costs but not Effectiveness of Care. Simell, T.;
Simell, O.; Sintonen, H. Diabetic Medicine. 10(9): 855-862. November
1993.
OBJECTIVE:
To compare 1-week and 4-week initial hospital stays for children with
type 1 diabetes in terms of cost and outcomes during a 2-year follow-up.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Prospective trial.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Total costs for children who had a 4-week initial hospital stay were 1.6-fold
higher during the 2-year follow-up, but outcomes for the two groups were
similar.
RECOMMENDATION:
The main findings from this analysis can be applied to diabetes care centers
in Finland and other countries.
ABSTRACT:
Sixty-one newly diagnosed children with type 1 diabetes took part in this
Finnish study, which was based at two Helsinki hospitals: Children's Hospital,
University of Helsinki; and the Avrore City Hospital. Thirty-one children
were assigned to the short-term group (1-week initial hospital stay) and
30 children to the long-term group (4-week initial hospital stay). Metabolic
control, psychosocial adjustment, and direct and indirect costs of care
were compared during the 2-year follow-up. First-month costs accounted
for 74 percent of total costs in the long-term group and 59 percent of
total costs in the short-term group. The major first-month expenses were
for hospital costs (82 percent in the long-term group and 78 percent in
the short-term group). The indirect costs of parents' lost work time,
traveling costs, children's living expenses at home, and housekeeping
and babysitter assistance, although minimal, were substantially higher
during the first month in the long-term group. Other direct costs considered
during the 2-year follow-up, including outpatient visits, insulin treatment,
home monitoring of blood glucose and urine tests, and diabetic diet, did
not differ significantly by group. Overall, total costs were 1.6 times
greater in the long-term group. The researchers found no significant differences
between the two groups in metabolic control, family adjustment to diabetes,
psychosocial measures, or satisfaction with patient education during the
2-year follow-up period. They concluded that shortening the average initial
hospital stay of children newly diagnosed with diabetes from 23 days to
9 days achieved a savings of 36 percent in costs during the first 2 years
of the disease without influencing the metabolic or psychosocial outcome
of care. 3 tables, 34 references.
150
TITLE:
Hospitalization and Expenditures for the Treatment of General Medical
Conditions among the U.S. Diabetic Population in 1991. Ray, N.F.; Thamer,
M.; Taylor, T.; Fehrenbach, S.N.; Ratner, R. Journal of Clinical Endocrinology
and Metabolism. 81 (10): 36713679. October 1996.
OBJECTIVE:
To estimate the risk of hospitalization for general medical conditions
among patients with diabetes aged 45 years and over; to estimate how much
of the cost of these hospitalizations was attributable to diabetes.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Prevalence-based costs.
Perspective: Health care system.
CONCLUSION:
Hospitalizations and expenses among patients with diabetes for conditions
that are neither acute nor chronic complications of diabetes are greater
than hospitalizations and expenses for these conditions among patients
without diabetes.
RECOMMENDATION:
Additional research is needed to identify medical reasons for increased
risk of selected general conditions among patients with diabetes and to
identify appropriate interventions for the care of these problems.
ABSTRACT:
The authors studied hospitalization for general medical conditions in
patients with diabetes aged 45 years and over. Data were derived from
the 1991 National Hospital Discharge Survey and the 1987 National Medical
Expenditure Survey. Expenditures attributable to diabetes were estimated
by multiplying the excess number of inpatient days for persons with diabetes
by the mean per-day inpatient cost. Identification of patients with diabetes
was based on primary or secondary diagnosis codes for diabetes; general
medical conditions were all those not considered to be acute or chronic
complications of diabetes. Costs were inflated to 1991 dollars. Mean expenditures
per inpatient day per patient with diabetes were $1,673 for those 45 to
64 years old and $1,192 for those 65 and over. Middle-aged persons with
diabetes were 60 percent more likely (relative risk, 1.6, 95 percent confidence
interval, 1.2-2.0) to be hospitalized for general medical conditions than
were persons without diabetes; the greatest relative risks were for peritonitis/intestinal
abscess, respiratory failure, liver disease, and male genital disorders.
Among the elderly, there was no significant difference between the groups
in overall risk of hospitalization for general medical conditions; those
with diabetes had elevated risks for liver disease, septicemia, diseases
of pulmonary circulation, and various other problems. Middle-aged and
elderly persons with diabetes were hospitalized longer than those without
diabetes (8.1 versus 6.3 days for Middle-aged and 10.1 versus 8.9 days
for elderly patients). Inpatient expenditures attributable to diabetes
were $4.12 billion (52.3 percent of this total was for elderly persons).
These costs may be understated because they do not account for undiagnosed
diabetes among hospitalized patients. 5 tables, 55 references.
151
TITLE:
Hospitals Adopt Intensive Programs for Diabetic Patients to Avoid High
Inpatient Costs. Conklin, M.S. Health Care Strategic Management.
12(6): 1113. June 1994.
OBJECTIVE:
To review methods used by some managed care organizations and hospitals
to reduce inpatient costs for patients with diabetes.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Patient management.
Methodology: Survey.
Perspective: Health care system.
CONCLUSION:
Successful diabetes management programs have reduced inpatient stays by
1 to 2 days.
RECOMMENDATION:
None.
ABSTRACT:
The author reviews efforts to reduce expensive inpatient care for people
with diabetes. According to one national study, $9,493 was expended per
patient with diabetes in 1992, of which $5,885 (more than 63 percent)
was for inpatient care; corresponding figures for patients without diabetes
were $2,604 and $1,222, respectively. The author describes several programs
that have been implemented to reduce inpatient costs for diabetes. Kaiser
Permanente of Northern California has developed a database for all patients
with diabetes (68,000 in May 1994) that will track costs, compliance,
and complication rates. Palmyra Medical Centers in Georgia markets an
intensive diabetes intervention and prevention plan to employers, which
costs $1,000 per patient with diabetes for the first year and $400 per
year thereafter; the program is based on the findings of the Diabetes
Control and Complications Trial. Control of blood glucose concentrations
is stressed, and the program begins with a 2day intensive education program
for newly diagnosed patients. Hemoglobin A1c is tested every 3 months,
and additional education is provided if concentrations indicate control
problems; blood glucose concentrations have decreased in 99 percent of
patients. Support groups meet monthly. Rose Medical Center in Denver has
reduced the average length of hospitalization by 1 day per patient with
diabetes, for a savings of over $500,000, by implementing intensive case
management at admission. The hospital uses its computer system to identify
patients with diabetes, then involves a member of the diabetes program
in the care of the patient. In Macon, Georgia, a similar program at the
Medical Center of Central Georgia that emphasizes inpatient education
has reduced length of stay by 1.7 days in 3 years. In Louisville, Kentucky,
at Saints Mary and Elizabeth Hospital, nurses from all units meet monthly
with diabetes nurse educators to discuss diabetes management problems;
hospital stays for patients with diabetes have been reduced by 1.92 days
in 2 years. 1 figure.
152
TITLE:
Operations, Total Hospital Stay and Costs of Critical Leg Ischemia. A
Population-Based Longitudinal Outcome Study of 321 Patients. Eneroth,
M.; Apelqvist, J.; Troeng, T.; Persson, B.M. Acta Orthopaedica Scandinavica.
67(5): 459-465. October 1996.
OBJECTIVE:
To estimate costs over time for a population of patients with critical
leg ischemia, including many with diabetes mellitus, who underwent surgery
in Sweden.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Prospective.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Mean total costs were $47,000 per patient. Patients with diabetes mellitus
more often underwent major amputation, had longer hospital stays, and
incurred higher medical costs than those without diabetes.
RECOMMENDATION:
None.
ABSTRACT:
The authors conducted a longitudinal analysis of costs (in 1996 U.S. dollars)
incurred by 321 patients, including 118 with diabetes mellitus, who had
surgery for critical leg ischemia in 1987 or 1988 in Malmohus county,
Sweden. Surgical procedures, hospitalizations, and hospital treatment
costs were assessed from the first procedure until death or for at least
six years postoperatively. Only those admissions directly caused by arterial
occlusive disease in the lower limbs were included. Admissions to rehabilitation
clinics and nursing homes were included, but nursing home stays for those
who lived in a nursing home before the surgery and returned to one after
the procedure were not. The costs of angiography, implants, intensive
care, outpatient care, and orthopedic appliances as well as indirect costs
were not included. The initial operation during the inclusion year was
a reconstructive vascular procedure for 96 patients, a restorative or
other vascular procedure for 111, and a major amputation for 114. Total
hospitalization until follow-up in all patients was 37,638 days, of which
only 44 percent was in surgical departments. Estimated overall cost for
hospital stays was $12.87 million, a mean per-patient cost of $40,103.
Overall cost of surgery was estimated to be $2.25 million, a per-patient
cost of $7,050. Patients with diabetes had a longer mean total hospital
stay (134 versus 108 days, p = 0.009) and a higher estimated mean total
hospital cost ($52,000 versus $44,000, p = 0.01) than patients without
diabetes. Both groups had a mean of three operations. Patients with diabetes
were more likely than those without diabetes to undergo major amputation
as the operation (50 percent versus 27 percent) and more often became
bilateral amputees than patients without diabetes (28 percent versus 12
percent). 3 tables, 33 references.
153
TITLE:
Patterns of Hospitalization in a Pediatric Diabetes Clinic in Sydney.
Sutton, D.L.; Greenacre, P.; Howard, N.J.; Cowell, C.T.; Silink, M. Diabetes
Research and Clinical Practice. 7(4): 271-276. November 6, 1989.
OBJECTIVE:
To ascertain the number of children with diabetes admitted to Children's
Hospital in Sydney between 1985 and 1987 and to analyze their admissions.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Patient management.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
For the 351 admissions of 241 children with type 1 diabetes admitted during
the period, the median stay was 10 days (12 for newly diagnosed children,
7 for others). Costs for newly diagnosed admissions were projected at
$Aust 3,658.00; of other admissions, $Aust 2,681.55.
RECOMMENDATION:
Patients' use of existing education and dietary services should be maximized,
and additional services should be developed.
ABSTRACT:
Medical files were obtained retrospectively on all patients with type
1 diabetes admitted to Children's Hospital from 1985 to 1987. Information
was collected on sex, age, number of bed days per admission, and reasons
for admission. The direct cost of a bed day and an admission for diabetes
in this hospital were calculated using a small sample. Two hundred forty-one
children with type 1 diabetes had 351 admissions over the 3 years; 105
stays were for newly diagnosed children. Of these 105 children, 14 (13.3
percent) had additional admissions during this period. Of the 246 admissions
for previously diagnosed children, stabilization of high blood glucose
was a cause for 145 (59.0 percent); other leading causes were a medical
reason (32.5 percent) and education (25.6 percent). The cost of a bed
day for a child with diabetes was $Aust 295.00. 3 figures, 1 table, 12
references.
154
TITLE:
Precipitants of Hospitalization in Insulin-Dependent Diabetes Mellitus
(IDDM): A Statewide Perspective. Fishbein, H.A. Diabetes Care.
8 (Supplement 1): 61-S64. September-October 1985.
OBJECTIVE:
To describe the epidemiology of precipitants of hospitalization for type
1 diabetes in Rhode Island; to evaluate the effectiveness of outpatient
education in reducing hospitalizations for patients with diabetes; and
to estimate potential cost savings of the education program.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Poor diabetes control and infection accounted for nearly one-half of all
hospital admissions in Rhode Island among patients known to have diabetes.
An outpatient education program significantly reduced the incidence of
acute hospitalizations among participants and resulted in substantial
savings.
RECOMMENDATION:
Other investigators may want to consider a registry system for diabetes.
ABSTRACT:
The author evaluated the epidemiology of precipitants of hospitalization
for type 1 diabetes in Rhode Island and the effectiveness of outpatient
education in reducing hospitalizations. A statewide registry of patients
with type 1 diabetes was used to identify and characterize the epidemiology
of patients admitted to Rhode Island hospitals. Data on precipitating
factors were obtained via a review of medical records and physician interviews.
During the study period (April 1978 through March 1983), there were 1,344
hospital admissions for type 1 diabetes among 887 persons below age 30.
Patients with known diabetes accounted for 1,123 admissions (691 persons).
Poor diabetes control (noncompliance with diet/medication) or infection
accounted for 54 percent and 44 percent of single and multiple admissions,
respectively. Admissions for pregnancy care and for various conditions
(e.g., myocardial infarction, stroke) were the next most frequent precipitants
of hospitalization. An intensive 10-hour outpatient education program
with two follow-up sessions was evaluated in 100 patients; the program
reduced the number of hospitalizations by 51 percent. Based on a direct
cost for hospitalization of $2,400, the potential savings of the educational
intervention for those admitted with known diabetes were estimated at
$674,000 (cost of admission x admissions for poor diabetes control or
infection x 50 percent for educational effect). The cost of the education
program was only about $100 per patient. 2 tables, 1 figure, 10 references.
155
TITLE:
Public Cost and Access to Primary Care for Hyperglycemic Emergencies,
Clark County, Nevada. Wilson, B.; Sharma, A. Journal of Community Health.
20(3): 249-256. June 1995.
OBJECTIVE:
To determine the cost and major causes of hospitalizations for emergency
admissions for diabetic hyperglycemia at a large public hospital in Clark
County, Nevada; to determine the effect of insurance coverage on the availability
of primary care providers, hospitalization costs, and precipitators of
admissions in the study population.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Uninsured patients accounted for 49 percent of hyperglycemic admissions
during the study period. Hospital medications and/or charges were lower
for uninsured patients, who were more likely than insured patients to
be hospitalized because they lacked access to primary care.
RECOMMENDATION:
Public funding or universal insurance coverage and more comprehensive
employer-funded insurance programs to provide primary care access for
diabetes mellitus management should be evaluated as cost-savings measures
in areas with large populations of people with diabetes.
ABSTRACT:
The authors evaluated hospital charges and major causes of hospitalization
for diabetic ketoacidosis and nonketotic hyperosmolar state, the major
acute emergencies of diabetes mellitus, at a large public hospital. Medical
insurance coverage associations with availability of primary care and
hospitalization charges were also examined. Data were derived from a retrospective
search of hospital admissions for diabetic ketoacidosis and nonketotic
hyperosmolar state over a 30-month period from 1989-1991. Hospitalization
and emergency room charges were compiled from medical records review.
Over the study period, 247 admissions for diabetic ketoacidosis and nonketotic
hyperosmolar state were identified; uninsured admissions accounted for
49 percent. Only 6 percent of patients in the uninsured group identified
a primary care provider, as compared with 85 percent of the insured group.
Un-insured patients accounted for a majority of emergency room visits
and 52 percent of emergency room charges. However, average hospital charges
(exclusive of physician fees) were significantly less in the uninsured
group ($4,049) than in the insured group ($7,222). Uninsured patients
were much more likely to have been hospitalized because they lacked access
to medication and much less likely to be hospitalized because of late
complications of diabetes mellitus. These patients had milder acute and
chronic disease processes potentially responsive to appropriate out-patient
management. 1 figure, 3 tables, 6 references.
156
TITLE:
Relative Risk and Economic Consequences of Inpatient Care Among Patients
With Renal Failure. Thamer, M.; Ray, N.F.; Fehrenbach, S.N.; Richard,
C.; Kimmel, P.L. Journal of the American Society of Nephrology.
7(5): 751-762. May 1996.
OBJECTIVE:
To identify the leading causes of hospitalization for people with renal
failure; to compare the risks of hospitalization associated with these
causes and with other chronic progressive diseases; and to quantify total
inpatient days, excess length of stay, and associated costs attributable
to renal failure.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Economic assessment.
Methodology: Prevalence-based cost.
Perspective: Health care system.
CONCLUSION:
Chronic renal disease is associated with significant morbidity and medical
costs.
RECOMMENDATION:
None.
ABSTRACT:
The authors studied inpatient care for persons with chronic renal failure,
including those with end-stage renal disease as well as those who had
not reached that stage, using 1991 data from the Medicare End-Stage Renal
Disease program, the National Health Interview Survey, the U.S. Census,
and the National Hospital Discharge Survey. If chronic renal failure was
the primary discharge diagnosis, all charges were included; a portion
of the charges was included if chronic renal failure was a secondary diagnosis.
In 1991, patients with renal failure had 348,962 hospitalizations, of
which 64 percent were for either chronic renal failure or 1 of 15 comorbid
conditions (e.g., vascular access problems, congestive heart failure).
Patients aged 65 years and older accounted for 57.6 percent of these hospitalizations
but 63.0 percent of the inpatient days. The renal failure population averaged
1.4 hospitalizations in 1991. Compared with persons without renal failure,
the age-adjusted relative risk of hospitalization for persons with renal
failure was 10 for all causes, 201.5 for vascular access problems, 32.4
for congestive heart failure, 31.3 for pulmonary edema/respiratory failure,
23.4 for gastrointestinal hemorrhage, 21.7 for diabetes, 19.2 for sepsis/septicemia,
15.6 for electrolyte disorders, 12.4 for hypertension, 10.4 for anemia,
and 10.3 for myocardial infarction. Length of stay was significantly longer
(mean difference: 0.9 days) than in those without renal failure. Patients
with renal failure consumed 1.5 million hospital days, for a total cost
of $2.2 billion; of this amount, $291 million was directly for chronic
renal failure. Hospitalization for vascular access problems cost $168
million; for congestive heart failure, $142 million. 6 tables, 46 references.
157
TITLE:
Resource Utilization in Treatment of Diabetic Ketoacidosis in Adults.
May, M.E.; Young, C.; King, J. American Journal of the Medical Sciences.
306(5): 287-294. November 1993.
OBJECTIVE:
To determine the specific care practices that correlate with decreased
hospital stay for adult patients with ketoacidosis related to diabetes.
CATEGORY:
Cost of diabetes (direct).
Type of
Study: Patient management.
Methodology: Cost analysis.
Perspective: Health care system.
CONCLUSION:
Optimizing hospital care and reducing the incidence of ketoacidosis in
women would markedly affect health care costs.
RECOMMENDATION:
Future studies of ketoacidosis in adults should focus on criteria for
admission to the intensive care unit (ICU), optimum time to move patients
out of the ICU, and the etiology and prevention of recurrent ketoacidosis
in women.
ABSTRACT:
The authors examined the records of 40 women and 25 men with diabetes
who were hospitalized a total of 92 times with ketoacidosis (serum glucose
greater than 200 mg/dL, positive serum or urine ketones, and evidence
of metabolic acidosis). The cases were analyzed by initial level of nursing
care (ICU, stepdown, general); the three fatalities were excluded from
analysis of the correlates of length of stay or cost. Mean days of hospitalization
and total costs (with standard deviations) were 6.3 (5.3) and $12,286
($12,096) for the ICU group, 4.4 (3.0) and $6,804 ($3,759) for intermediate
care, and 6.0 (6.0) and $7,357 ($6,902) for general care. Patients with
concurrent infection had significantly longer stays than those who were
culture-negative (9.1 " 6.8 days versus 4.5 " 3.2 days). Shorter length
of stay was correlated with care by a diabetologist, shorter interval
from presentation until time to administration of an intermediate- or
long-acting insulin, and time of day of the initial presentation. In-hospital
recurrences of ketoacidosis occurred less frequently in patients managed
by a diabetologist (30 versus 47 percent of cases) and were correlated
with longer hospitalization. Rapid initial correction of hyperglycemia
and acidemia did not correlate with reduced resource usage. Level of care
did not affect outcome; if all cases were treated at the intermediate
level, an overall savings of $2,292 per case would have been realized.
The authors conclude that admission to the ICU is indicated for patients
with a concurrent diagnosis that would require intensive care (e.g., myocardial
infarction) or if the ketoacidosis is grade 4 (the highest grade). They
project resource savings of $60 to $90 million yearly from reducing the
level of diabetic ketoacidosis in females to the level in males. 4 tables,
33 references.
Back
to Economics of Diabetes Mellitus Indexes
Top of page
Historical
Page last modified: December 20, 2005
Content Source: National Center for Chronic Disease Prevention and Health Promotion
Division of Diabetes Translation
|
Diabetes & Me