This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.
Please go to www.ahrq.gov for current information.
Does publicly provided home care substitute for family care?
The median age of the U.S. population has been steadily rising, with the elderly segment of the
population expected to double by the year 2030. These demographic changes promise to generate
an unprecedented increase in the demand for long-term care. Concerns about these trends and the
high cost of institutional care have motivated the search for approaches to meeting the long-term
care needs of the elderly population while maintaining them in the community.
One such approach entails extending public long-term care financing to support formal (paid) care
provided in the home. Financing of formal home care remains limited, however, because of
policymakers' concerns that publicly provided home care will substitute for, rather than
supplement, family care currently provided at no direct cost to the Government. This may not
necessarily be the case, however. Results from a recent study by the Agency for Health Care
Policy and Research suggest that the degree of substitution, defined as overall displacement of
informal caregiving efforts by subsidized formal care, appears to be small.
Liliana E. Pezzin, Ph.D., of AHCPR's Center for Cost and Financing Studies, and her colleagues
Peter Kemper, Ph.D., and James Reschovsky, Ph.D., both formerly with AHCPR and now with
the Center for Studying Health System Change, conducted the study. They used data from a large
social experiment, the National Channeling Demonstration project, to analyze the extent to which
public financing of formal home care would reduce the provision of family care
directly—through its effects on informal care received by those in a given living
arrangement—and indirectly, through its effects on living arrangement changes.
The study showed that providing public home care resulted in only small reductions (4.2 hours
per week) in the overall amount of care provided by informal caregivers. The reduction in the
amount of informal care was dominated by the direct effect on hours of care assuming no change
in living arrangement. In fact, about 70 percent of the estimated reduction in overall hours of care
was due to a decrease in the number of resident informal hours. These results suggest that
policymakers concerned with the potential substitution of subsidized formal for informal care
should focus more on specific measures designed to minimize caregivers' behavioral
responses—for example, through case management guidelines on informal and respite
care—rather than on
discouraging effects due to living arrangement changes.
Although the program did not have a substantial effect on informal caregiving hours, the results
indicated a sizable and statistically significant effect of the intervention on living arrangement
choices of unmarried elderly persons. Channeling's more generous intervention increased the
probability of independent living by 7.1 percentage points relative to those individuals not
receiving the subsidy (control group). This increase was associated with a corresponding
significant reduction of 2.4 percentage points in the probability that unmarried elderly persons
would live with others and a 4.7 percentage point reduction in the probability of living in a nursing
or personal care home.
See "Does publicly provided home care substitute for family care?" by Drs. Pezzin, Kemper, and
Reschovsky, in The Journal of Human Resources 31(3), pp. 650-676, 1996. Reprints
Publication No. 97-R011) are available from the AHCPR Publications
Having close family substantially reduces the risk of nursing
home admission for the elderly
Nursing home care costs were about $3,200 per month for each resident and totaled $70 billion
for the Nation in 1993. Avoiding nursing home admission hinges on ones family support. Elderly
persons who are married have about half the risk of entering a nursing home as those who are not
married. Those with at least one daughter or sibling reduce their chance of entering a nursing
home by about one-fourth, according to a recent study supported in part by the Agency for Health
Care Policy and Research.
On the other hand, sons appear to reduce the risk of admission for mothers but not for fathers.
This may be due in part to the fact that sons typically do not supply care directly to a frail parent
but instead offer financial assistance or help with bureaucratic tasks. This approach may be less
palatable to elderly fathers than mothers, explains Vicki A. Freedman, Ph.D., formerly of AHCPR
and now with RAND.
Dr. Freedman used the New Haven Established Populations for Epidemiologic Studies of the
Elderly (EPESE) database, which has collected information since 1982 on nursing home stays by
community-dwelling elderly persons, to examine the relationship between family structure and the
risk of first nursing home admission. By 1989, 22 percent of the elderly at risk for their first
nursing home admission had entered a nursing home, another 33 percent had died with no
institutional experience, and 45 percent were still alive and had not entered a nursing home. Even
after controlling for differences in health, demographic, and economic characteristics of the
elderly, having a spouse, daughter, or sibling significantly reduced the risk of nursing home entry.
Being married reduced the risk of entering a nursing home by 41 percent, having at least one
daughter reduced the risk by 27 percent, and having at least one living sibling reduced the risk by
Other factors also affected the risk of nursing home entry. Persons who were limited in their
ability to carry out activities such as dressing and eating, who considered themselves in poor
health, and those who had poor cognitive functioning had a markedly increased risk of nursing
home admission. Whites had nearly twice the risk of nursing home admission as minorities.
However, sex and income had small and nonsignificant effects on the risk of nursing home
Details are in "Family structure and the risk of nursing home admission," by Dr. Freedman, in the Journal of Gerontology: Social Sciences 51B(2), S61-S69, 1996. Reprints (AHCPR
No. 97-R010) are available from the AHCPR Publications
Return to Contents
Women's and Family Health
Rural physicians use fewer health resources than urban
doctors to care for low-risk obstetric patients
Compared with urban family physicians (FPs) and obstetricians (OBs), rural FPs and OBs use
fewer health care resources, such as tests and procedures, when treating low-risk pregnant
women, according to a study supported in part by the Agency for Health Care Policy and
Research (HS06166). The major differences are less use by rural physicians of costly epidural
anesthesia, shorter hospital stays, and fewer tests for sexually transmitted diseases (STDs), which
have been suggested but not recommended by the American Academy of Obstetricians and
On the other hand, rural physicians tend to use most resources at the same rates, including the
same number of clinically important procedures, such as cesarean sections, and have similar birth
outcomes compared with urban physicians. This indicates similar quality of obstetric care in rural
and urban areas, according to researchers at the WAMI Rural Health Research Center at the
University of Washington, Seattle.
The researchers studied health resource use by a stratified random sample of rural and urban OBs
and FPs in Washington State, using medical record data of their low-risk obstetric patients. Both
rural OBs and FPs were significantly more likely than their urban counterparts to use local and
pudendal anesthesia and less likely to use epidural anesthesia during vaginal deliveries. For
cesarean deliveries for both specialties, rural physicians on average provided far fewer epidurals
and more spinal and general anesthesia than urban physicians.
The researchers point out that epidural anesthesia is probably less available in rural hospitals
because fewer anesthesia personnel are available. Fewer prenatal tests for STDs by rural
physicians may be associated with inadequate training, heavy workload, or that they have more
intimate knowledge of their patients and the area.
Urban OBs used 10.4 percent more overall, 21 percent more prenatal, and 8 percent more
intrapartum health resources in their care of women than did rural OBs. Urban FPs used 6 percent
more total and 10 percent more intrapartum resources than rural FPs, but prenatal care differences
were insignificant. Most of the urban-rural differences in prenatal care for OBs were accounted
for by resource use connected with visits to the hospital, although there were small significant
differences for consults and referrals, blood and urinalysis tests, and hematology work.
For more information, see "Rural and urban differences in physician resource use for low-risk
obstetrics," by L. Gary Hart, Ph.D., Sharon A. Dobie, M.C.P., M.D., Laura-Mae Baldwin, M.D.,
M.P.H., and others, in the October 1996 issue of HSR: Health Services Research 31(4),
Delays in beginning prenatal care are unrelated to the number
of available obstetricians
Women living in areas with only a few obstetric providers are no more apt to delay prenatal care
until after the fourth month of pregnancy than women in areas with numerous obstetricians,
according to a study supported in part by the Agency for Health Care Policy and Research
(HS07321). Apparently, obstetrician availability does not affect when women decide to begin
prenatal care, concludes Joyce M. Piper, Dr.P.H.
Dr. Piper and her colleagues at Vanderbilt University retrospectively analyzed linked birth/death
certificates in the state of Tennessee from 1989 through 1991. The researchers found that of
209,171 births, 13 percent of mothers had their first prenatal care visit after the fourth month of
pregnancy. Women who lived in counties with less than five obstetricians per 10,000 women of
childbearing age had slightly increased odds of delayed onset of prenatal care. However, women
who lived in counties with no obstetricians had less risk of delaying the start of prenatal care than
women in counties with four obstetricians per 10,000 women of childbearing age (odds ratio
[OR] of 1.18 vs. 1.29; 1 is equal odds). Also, outcomes of births in high and low physician density
areas were about the same.
On the other hand, several maternal factors were associated with more than a doubling of the
odds that a woman would delay starting prenatal care: age less than 16 years (OR of 3.72), lack
of a high school education (OR of 3.16), three or more previous pregnancies (OR of 2.83), or
being unmarried (OR of 2.34). After controlling for these maternal characteristics, the study still
showed no consistent relationship between geographic access to obstetric services and delayed
prenatal care. These findings suggest that efforts to change the distribution of obstetric providers
will be beneficial only if they are undertaken in the context of programs specifically designed to
improve the quality of prenatal care, conclude the researchers.
Details are in "Obstetric care provider density and delayed onset of prenatal care," by Dr. Piper,
Edward F. Mitchel, Jr., M.S., and Wayne A. Ray, Ph.D., in Medical Care 34(10), pp.
Poor children are less likely than other children to receive
needed mental health services
It has been estimated in previous studies that 12 to 20 percent of children in the United States
suffer from mental health problems, but only 1 to 5 percent of them receive mental health care.
Now, a recent study by the Agency for Health Care Policy and Research has shown that among
children who experience mental health problems, poor children and adolescents are less likely than
more well-to-do young people to obtain needed mental health services.
The study found that children with poor mental health in high-income families were more than
three times as likely to visit a doctor for mental health problems than similar children in
low-income families. The higher the family income, the greater the number of visits and greater
likelihood of seeing a mental health specialist. These findings point to poverty as a significant
barrier to obtaining needed mental health care services, according to Peter J. Cunningham, Ph.D.,
formerly of AHCPR and now with the Center for Studying Health System Change, and Marc P.
Freiman, Ph.D., of AHCPR's Center for Cost and Financing Studies.
The researchers used data from the 1987 National Medical Expenditure Survey, a national survey
of health care use by the U.S. civilian noninstitutionalized population, to examine use of mental
health services and family income. They found that about 5 percent of children ages 6 to 17 used
some type of health care (mostly outpatient) related to a mental health problem in 1987. Just over
half of children receiving ambulatory care for a mental health problem in 1987 saw a mental health
specialist. Children with poor mental health whose family income was $5,000 were estimated to
have a 12 percent probability of using mental health services compared with a 37 percent
probability for similarly disturbed children whose family income was $75,000. Also, among those
with some ambulatory use, the less affluent children had an estimated 41 percent chance of seeing
a specialist compared with a 64 percent chance for more affluent children.
A mother's use of mental health care also strongly influenced whether or not her children
obtained such care. It is likely that this strong association reflects family problems and
propensities, since children's emotional and behavioral problems are frequently related to family
circumstances, conclude the researchers.
See "Determinants of ambulatory mental health services use for school-age children and
adolescents," by Drs. Cunningham and Freiman, in the October 1996 HSR: Health Services
Research, pp. 409-427. Reprints (AHCPR Publication No. 97-R033) are available from the
AHCPR Publications Clearinghouse.
Return to Contents
Minority women with abnormal screening mammograms have
later followup tests than white women
Breast cancer is the most common cancer diagnosed among women of all ethnic groups. But even
though breast cancer deaths have decreased among white women in recent years, this has not
been the case among minority women, despite similar increased rates of mammography screening.
Less rigorous followup testing after an abnormal mammogram may explain why minority women
have later-stage breast cancer at diagnosis and higher mortality rates than white women,
according to a recent study.
The study found that minority women with abnormal screening mammograms had followup tests
(another mammogram, ultrasound, fine-needle aspiration, or biopsy) on average about a week
later than white women with abnormal mammograms. This suggests that there still may be
barriers related to race/ethnicity that affect timely performance of these followup tests, note
researchers from the University of California, San Francisco (UCSF).
Their study, which was supported in part by the Agency for Health Care Policy and Research
(HS07373), included 317 women (aged 33 to 85 years) who had an abnormal screening
mammogram between July 1993 and May 1994, and was based on records abstracted from the
UCSF Mobile Mammography Screening Program. Even after accounting for patient
characteristics such as presence of a palpable breast mass, family history of breast cancer, and
median income, minority women took longer than white women to complete a diagnostic workup
(a median of 19 vs. 12 days). This small difference had no impact on breast cancer stage at
diagnosis or on mortality, but it occurred even when women's physicians were immediately
notified of abnormal results. It is possible that larger and potentially more important racial/ethnic
differences would be observed in clinical settings where reporting and followup procedures are
not so complete.
It is not clear whether the delays in workup were caused by the patients or due to doctors or
systems of care. For example, previous studies have shown that black and Hispanic women
typically have longer delays from the time they notice breast symptoms to the time they first
consult a doctor. Also, timely communication may be more difficult for some minority women for
example, if their primary language is not English or if they receive most of their care through
overburdened community and public health clinics. Finally, minority women may have poorer
access to care, that is, they may be more apt to have public or no insurance, no regular care site or
provider, and/or difficulty maneuvering within a system.
Details are in "Racial differences in timeliness of follow-up after abnormal screening
mammography," by Sophia W. Chang, M.D., M.P.H., Karla Kerlikowske, M.D., Anna
Napoles-Springer, M.P.H., and others, in the October 1, 1996, issue of Cancer 78(7), pp.
South Central Los Angeles residents are much less likely to
undergo surgery than other county residents
South Central Los Angeles (SCLA) has six times more blacks and half the median income of the
rest of Los Angeles County. Not only do SCLA residents have much higher death rates from
ischemic heart disease and cerebrovascular disease, but they also have five times fewer physicians
and fewer hospital beds (1 per 2,000 vs. 1 per 280 persons) than other county residents. SCLA
residents receive significantly fewer revascularization and other potentially life-saving procedures
than other county residents, perhaps because of factors other than poverty that are associated with
the SCLA community, conclude University of California, Los Angeles, researchers led by David
M. Carlisle, M.D., Ph.D.
In a study supported in part by the Agency for Health Care Policy and Research, the investigators
analyzed California hospital discharge data to assess use of eight surgical procedures by Los
Angeles County residents over a 3-year period (1986 through 1988): coronary angioplasty,
coronary bypass surgery, carotid endarterectomy, cesarean section, hysterectomy, transurethral
prostatectomy (TURP), appendectomy, and mastectomy. Results showed that, without
controlling for other factors, the use of four procedures was significantly lower in SCLA. For
every 10,000 residents in other parts of the county, 15 angioplasties were performed compared
with 9 for every 10,000 SCLA residents. Significant differences also were found for coronary
bypass surgery (18 vs. 11 per 10,000 residents), carotid endarterectomy (7 vs. 3 per 10,000
residents), and cesarean section (274 vs. 212 per 1,000 deliveries).
After adjusting for differences in patient health status, availability of health resources (physician
distribution and distance to the nearest hospital), and income, SCLA remained negatively
associated with the use of all procedures studied, except for mastectomy and TURP. Even after
controlling for SCLA's poverty, residence in SCLA remained a significant negative predictor of
use for six of the eight procedures. One possible explanation, which is supported by a growing
body of research, is the greater prevalence in SCLA of historically disenfranchised ethnic groups
(blacks and Latinos), an effect likely to be complex and multifactorial, according to the
Details are in "The effect of race and ethnicity on the use of selected health care procedures: A
comparison of South Central Los Angeles and the remainder of Los Angeles county," by Dr.
Carlisle, Barbara D. Leake, Ph.D., Robert H. Brook, M.D., Sc.D., and Martin F. Shapiro, M.D.,
Ph.D., in the Journal of Health Care for the Poor and Underserved 7(4), pp. 308-322,
Adult-onset diabetes occurs much more often among
minorities than whites
Most minority groups in the United States have higher rates of adult-onset diabetes
(non-insulin-dependent diabetes mellitus, NIDDM) and more complications and death from the
disease than do whites. In fact, diabetes should be treated as a public health problem among
minorities, concludes a study supported by the Agency for Health Care Policy and Research
The heavier burden of NIDDM on minorities has been attributed to factors associated with
"westernization," such as a high-fat, high-calorie, and low-fiber diet and little exercise; obesity,
which is more prevalent among minority groups; and certain genetic factors. Yet the relative
contributions of these and other socioeconomic factors are not well defined, according to a review
of the scientific literature from 1976 to 1994.
The review shows that all minorities, except Alaska Natives, have NIDDM rates two to six times
greater than those of white persons, as well as higher rates of serious complications. For instance,
minority individuals suffer from six times more cases of diabetic nephropathy (kidney disease)
than whites. Moreover, blacks, Hispanics, and American Indians have higher rates of diabetic
retinopathy (damage to the blood vessels of the retina), which can impair vision. Blacks and Pima
Indians who have diabetes also undergo more amputations than white diabetics (1.5 and 4 times
more, respectively). These amputations are sometimes needed to curtail the spread of infections
from ulcers that won't heal due to both peripheral vascular disease and constant reinjury that
occurs because of lack of sensation in the feet. Finally, more minorities die from diabetes-related
complications than whites.
Serious complications are more likely to develop from long-term and less well-controlled
diabetes. Evidence reviewed for this study suggests that the average time from NIDDM onset to
diagnosis was 7 years, with many patients already experiencing complications by the time their
diabetes was diagnosed. In fact, 50 percent of diabetics identified in community surveys do not
know they have diabetes. The researchers recommend diabetic screening of persons from minority
groups, especially if they are overweight or have other risk factors, as well as dissemination of
culturally sensitive diabetes education materials and cookbooks.
Details are in "Non-insulin-dependent diabetes mellitus in minorities in the United States," by
Janette S. Carter, M.D., Jacqueline A. Pugh, M.D., and Ana Monterrosa, M.D., in the August 1,
1996 Annals of Internal Medicine 125(3), pp. 221-232.
Return to Contents
Functional decline in persons with HIV disease is variable and
Persons infected with the human immunodeficiency virus (HIV) that causes AIDS eventually
reach a point when everyday tasks such as dressing and shopping become difficult. As their ability
to function declines, they often must rely on home health care systems oriented toward the
elderly, whose functional status is fairly stable. Yet the functional status of persons with HIV
disease is variable and episodic, with periods of improvement and worsening, and with women
showing the greatest decline in their ability to function over time, according to a study supported
by the Agency for Health Care Policy and Research (HS06339).
Home care programs designed for the elderly may not be flexible enough for persons with HIV
disease, note Stephen Crystal, Ph.D., and Usha Sambamoorthi, Ph.D., of Rutgers University.
They conducted monthly followup interviews with 246 persons enrolled in New Jerseys
Medicaid waiver program for persons with symptomatic HIV disease or AIDS from 1990 to
1991; 82 percent of study participants had AIDS. They collected data on functional status, race
and sex, risk group, and diagnosis at study initiation to develop several models to account for
decline in functional status. The functional status scale rated a persons ability to carry out 16
daily activities, ranging from rather complex tasks, such as taking care of medical paperwork, and
relatively strenuous tasks such as housecleaning, to basic personal care tasks such as bathing,
eating, and using the bathroom.
Results showed that on average, the persons studied were impaired in 5 of the 16 tasks. Although
ability to function declined by about one-third of a task per month, inability to perform a task
during a given month did not necessarily imply a similar inability the next month. Being female
was associated with a significantly higher level of impairment (an additional 1.88 tasks), as was
AIDS diagnosis (an additional 1.35 tasks), but race and risk group were not. These findings
suggest the need for a flexible approach to the authorization and case management of home care
or other long-term care services for patients with HIV disease. Over-servicing these patients
would produce unnecessary costs and client dependence, but under-servicing would place them at
risk, conclude the researchers.
For more details, see "Functional impairment trajectories among persons with HIV disease: A
hierarchical linear models approach," by Drs. Crystal and Sambamoorthi, in the October 1996
HSR: Health Services Research 31(4), pp. 469-488.
San Francisco's most successful HIV prevention programs
emphasize the messenger more than the message
The messenger is more important than the message when trying to prevent the spread of infection
with the human immunodeficiency (HIV) virus among socially ostracized groups. At least that is
the experience of the San Francisco programs that have been most successful in conveying the
HIV prevention message to these typically distrustful groups.
The most successful programs, most of which were community-based, had several elements in
common, according to a study supported by the Agency for Health Care Policy and Research
(HS07610). First, they narrowly defined their target group (audience segmentation) by specific
HIV risk factors, such as drug use, sexual orientation, ethnicity, age, and sex, to the extent that
some groups focused their messages only on black, teenage runaway, male, drug-using, homeless
Second, they used members of the targeted group to convey the message (homophily), usually
face-to-face, to establish a nonjudgmental relationship based on trust between messenger and
target audience. For instance, recovering American Indian drug addicts discussed HIV prevention
with American Indians currently using drugs. Third, prevention strategies were compatible with
the targeted audience's perceived experiences, existing beliefs, and behavior. For instance,
messages for at-risk Hispanics were in Spanish and were sensitive to Hispanic history and culture.
Finally, the most successful programs collaborated with other programs to share resources and
Unlike other public health programs, the emphasis was not on message content but on establishing
a relationship of trust with members of the target audience so that they would be receptive to the
HIV prevention message, explains James W. Dearing, Ph.D., of Michigan State University, who
led the 2-year study. The researchers surveyed the 49 HIV prevention programs in San Francisco
that targeted socially ostracized groups to determine the extent to which these programs used
strategies based on social marketing and diffusion-of-innovation concepts to deliver preventive
health messages. They then interviewed 38 staff leaders who operated the most and least effective
of the 20 most highly targeted programs to identify the most successful strategies.
For more information, see "Social marketing and diffusion-based strategies for communicating
with unique populations: HIV prevention in San Francisco," by Dr. Dearing, Everett M. Rogers,
Ph.D., Gary Meyer, Ph.D., and others, in the Journal of Health Communication 1, pp.
Social, functional, and clinical factors affect hospitalization of
persons with AIDS
Persons with AIDS (PWAs) who are minorities, homeless, have serious AIDS-related infections
or poor neuropsychological status are more apt to be hospitalized than other persons with AIDS.
These factors and other aspects of health-related quality of life (HRQL), such as functional status,
may explain variations in hospital use better than clinical factors alone, concludes a study
supported by the Agency for Health Care Policy and Research (HS06239).
It shows that PWAs were three times more likely to be hospitalized if they were homeless (odds
ratio [OR] of 3.3; 1 is equal odds) or had severe AIDS-related, opportunistic infections, such as
cytomegalovirus-induced pneumonia (OR of 2.7). Minority PWAs were twice as apt to be
hospitalized (OR of 2.0), as were PWAs with poor cognitive functioning (OR of 1.9).
PWAs who were only moderately impaired in their ability to carry out basic daily living tasks were
only slightly more likely than other persons with AIDS to be hospitalized (OR of 1.3). Among
those persons who were hospitalized, total hospital days were influenced by system of care and
neuropsychological status. These findings suggest that factors not listed in the medical chart need
to be assessed along with clinical factors when considering care options for this group of patients,
according to Joel S. Weissman, Ph.D., and Arnold M. Epstein, M.D., of Harvard Medical School,
and their colleagues.
The researchers explored patterns of future hospitalization for more than 300 PWAs treated at
three clinical sites (clinic, private group practice, and health maintenance organization) in Boston
during 1990 and 1991. Using patient interviews, hospital bills, and medical charts, they developed
statistical models associating disease burden, patient sociodemographic characteristics,
health-related quality of life, use of prophylactic drugs, and system of care with the likelihood of
future hospital admissions and total hospital days during a subsequent 4-month period.
Details are in "The influence of health-related quality of life and social characteristics on hospital
use by patients with AIDS in the Boston health study," by Dr. Weissman, Paul D. Cleary, Ph.D.,
George R. Seage III, D.Sc., M.P.H., and others, in Medical Care 34(10), pp. 1037-1054,
Return to Contents
Proceed to Next Article