39
Managed Care for Children:
Effect on Access to Care
and Utilization of Health
Services
Peter G. Szilagyi
Abstract
The rapid trend toward enrolling children in managed care has occurred largely without conclusive evidence about the effects of these arrangements on two important
aspects of care: access and utilization. Although the effect of managed care on these
measures has been studied more widely in the adult population, the results may not
be applicable to children, who have unique health care needs centering around prevention and early treatment of acute illnesses to avoid long-term health problems.
Moreover, several methodological challenges make it difficult to evaluate the impact
of managed care on health care access and utilization in general.
This article reviews what is known about the effect of managed care on access to health
services, as well as utilization of hospital care, emergency department (ED) visits, primary care services, and specialty services for the pediatric population. In each area,
findings from privately insured children and Medicaid enrollees are considered separately. There is little conclusive evidence on the effect of managed care on access to
and utilization of pediatric health services. A recurring theme is that the effect of managed care is dependent on several factors, including whether providers assume financial risk through capitated reimbursements or retain fee-for-service payments; the
comprehensiveness of benefits offered by health plans; and the level of cost sharing
required of families. Among privately insured children, for example, managed care
usually has been associated with higher primary care visit rates, though the benefit of
managed care is reduced when fee-for-service plans cover preventive care and require
minimal or no cost sharing for these services. Among Medicaid recipients, studies suggest that managed care is more likely to be associated with a decrease in preventive
visits when provider payments are capitated. Attempts to decipher effects by health
plan type are made more difficult by the rapid evolution of both managed care and
fee-for-service plans, which often blurs the distinction between these two entities.
Nonetheless, in some areas, managed care does appear to have an identifiable effect
on pediatric health services. For Medicaid recipients, managed care has been associated with decreased emergency department use and decreased access to specialty care
for chronically ill children. As enrollment of children in managed care plans increases,
the need continues for methodologically sound studies evaluating the effect of these
arrangements on the delivery of pediatric health services and on health outcomes.
The Future of Children CHILDREN AND MANAGED HEALTH CARE Vol. 8 • No. 2 – Summer/Fall 1998
Peter G. Szilagyi, M.D.,
M.P.H., is associate
professor of pediatrics,
director of Pediatric
Ambulatory Services,
and chief of the Child
Health Outcomes Research
Division at the University
of Rochester School of
Medicine and Dentistry.
40
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
T
he rapid growth of managed care is one of the most significant
changes affecting the delivery of child health care since the turn of
the century. Children are disproportionately represented in managed care plans because these plans tend to market through employers to
enroll working adults and their families. In addition, most states are rapidly
enrolling Medicaid recipients—primarily women of childbearing age and
children—in managed care plans. Managed care now plays a central role in
the delivery of children’s health care services within both the private and the
public sector.
Despite the explosive growth of managed care in recent years, only a
small body of scientific information is available about the effect of managed
care arrangements on access to and utilization of health care services by children and adolescents in the United States. The purpose of this article is to
summarize this information. In each section, findings about the effects of
managed care for privately insured children will be discussed first, followed
by a discussion of the effects of Medicaid managed care. The final section
and Table A1 at the end of this article will summarize results.
Potential Effect of Managed
Care on Access to and
Utilization of Care for
Children
Unique Characteristics of Child
Health Care
Most children are physically healthy. Their
health care needs include preventive services; acute services for frequent illness and
injuries; management of developmental,
school-related, psychosocial, and emotional
problems; and the occasional use of
specialty, emergency, or inpatient care.
Evaluation of access and utilization should
focus on these types of services.
Preventive care is fundamental to child
health care.1 Children are and should be
high utilizers of preventive services, because
the appropriate receipt of preventive services may reduce adverse health outcomes
later in life.2 Children are also high utilizers
of acute care services for illnesses, injuries,
and minor acute conditions. Prompt and
appropriate treatment for acute conditions
can prevent long-term complications. While
most acute care is provided in primary care
offices, a substantial amount is provided in
emergency departments.3 There is much
debate about the costs and overuse of emergency departments, particularly by impoverished patients. It is important to examine the
performance of managed care systems with
respect to both preventive and acute care for
children.
While most children are physically
healthy, about 3% to 5% suffer from serious,
handicapping chronic conditions,4 and
these children account for a disproportionate share of all child health care expenditures.5,6 There is substantial concern that
managed care systems restrict services for
this population. A related issue is the provision of care for individuals with chronic
developmental, emotional, and psychosocial
problems, which affect up to 20% of all children and adolescents. This is another group
known for its frequent use of health services.4 It is critical to assess the effect of managed care on these vulnerable groups of
children.
Another crucial issue in pediatrics is the
health status and health care of poor children, many of whom are covered by
Medicaid. Poverty is one of the best predictors of mortality, activity limitations, and use
of health care resources,7 and the impact of
Medicaid managed care should be systematically evaluated.
Potential Advantages of
Managed Care for Children
A major theoretical advantage of managed
care over traditional fee-for-service delivery
systems is that managed care plans normally
have more comprehensive information
about their enrolled populations and can
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
more effectively track service use patterns.
For example, while fee-for-service plans
may not even know how many children
they have enrolled, managed care plans
usually have well-developed data systems containing information about children enrolled
in their health plans and the services they
receive, including preventive visits and immunizations. Managed care plans can use these
data to develop strategies aimed at improving
access to care and the quality of services
received by the children and families
enrolled. However, unless automated data
systems in these plans are quite sophisticated,
the information available may capture only
the overall patient encounter, rather than the
specific activities included during a patient’s
visit to his or her health care provider.
Managed care plans theoretically have a
greater incentive than traditional fee-forservice systems to promote preventive care
in order to maximize the role of the primary
care gatekeeper and reduce future health
care expenditures for therapeutic care. This
is particularly important for children, who
may benefit most from preventive services.
The use of preventive health services can be
encouraged using numerous mechanisms,
including lower out-of-pocket costs for
patients; financial or administrative incentives for providers, such as higher reimbursement rates for preventive services or
provider profiles that include the implementation of preventive health services as a
quality indicator; and coordination of preventive care with other private or public
health care providers. Health maintenance
organization (HMO) enrollees traditionally
have had lower out-of-pocket costs for preventive care than fee-for-service patients.
However, this difference may be narrowing
for two reasons. First, health insurance legislation in some states now requires insurance
plans to cover preventive services for children with no cost sharing required.8 Second,
the Vaccines for Children (VFC) program
has restricted patient charges and improved
physician reimbursement for immunizations.9 The Medicaid program does not rely
on copayments and offers fairly broad
coverage, so direct patient costs for preventive services are irrelevant to that population, reducing the advantage of Medicaid
managed care. Moreover, from the point of
view of managed care plan administrators,
the incentive to promote preventive care
may be tempered by the fact that the plan’s
investment in preventive care will result in
cost savings well into the future, when many
enrollees will no longer be plan members
because of the high disenrollment rates and
plan switching that typically occur in the
health care sector. As a result, there may
be substantial variability in the extent to
which managed care plans promote prevention; and external factors, such as Health
Plan Employer Data and Information Set
(HEDIS) guidelines to report on and reward
the delivery of preventive services, may play
an important role.10
Other potential advantages of managed
care in general, and HMOs in particular,
have been listed elsewhere.11–14 For children with chronic illnesses, analyses of the
benefit packages and organizational characteristics of HMOs suggest some potential
advantages resulting from an improved
emphasis on prevention, primary care, and
Managed care plans theoretically have a
greater incentive than traditional fee-forservice systems to promote preventive care
in order to reduce future health care
expenditures.
service coordination.15,16 The potential cost
savings and efficiencies of managed care
plans might be translated into better access
to services, enhanced quality of care for
patients, and lower total expenditures on
health services.
Potential Disadvantages of
Managed Care for Children
The major concern about managed care
arrangements is that the strong incentives to
control costs may limit needed medical
services, particularly for vulnerable groups
whose medical care is either costly or difficult to deliver. For the pediatric population,
these vulnerable groups include impoverished children and those with chronic
medical, mental, or emotional conditions.
Concerns about obtaining appropriate
services and paying for these services are
paramount for most families with chronically ill children.17 Managed care plans,
which attempt to minimize costs by limiting
hospitalizations, referrals, and expensive
41
42
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
therapeutic services, may have greater
incentives to either exclude potential
enrollees with severe chronic illnesses or limit
the breadth of services covered. Because the
majority of children in managed care plans
are healthy, a plan may be able to satisfy the
majority of families while denying needed
care to that subset of children who have the
most costly medical needs. Furthermore, a
high degree of risk sharing by primary care
physicians may encourage them to limit necessary referrals or diagnostic tests for children with complex conditions. Burdensome
authorization requirements and reduced
access or denied referrals to needed specialists have been noted as disadvantages of
HMOs for chronically ill children.15,16
A similar concern exists about managed
care programs for impoverished children,
because these patients, like children with
chronic conditions, often face barriers to
health care, which may increase under the
administrative constraints of gatekeeper
systems; and because of the higher rates of
health problems among poor children.1 By
Burdensome authorization requirements and
reduced access or denied referrals to needed
specialists have been noted as disadvantages
of HMOs for chronically ill children.
limiting services, managed care systems may
compromise the quality of care for this
vulnerable population. Concern has also
been raised about the exclusion of traditional “safety net” providers (such as public
health clinics or neighborhood health
centers) from managed care contracts,
because many disadvantaged and high-risk
children traditionally have received comprehensive services in these settings. Shifting to
a managed care plan may interrupt the
usual source of care for this population,
forcing them to obtain services from
providers less experienced in caring for children with complex needs.18
Methodological Challenges in
Evaluating Managed Care
Systems
Experts have noted methodological difficulties in evaluating the effects of managed care
plans.11,13,14,19 These constraints are a major
reason for the dearth of studies evaluating
managed care effects, and perhaps for the
inconsistent and often contradictory findings. As described in the article by Hughes
and Luft in this journal issue, managed
care plans differ widely in their structure,
financial arrangements, and utilization
management, and it is difficult to make
generalizations about even theoretical
effects on access or utilization of care.
The effect of managed care depends on the
extent of financial incentives and disincentives placed on the providers, and on the
structural and financial arrangements of
the comparison groups. Consequently, it is
possible to observe even greater variability
among different types of managed care
plans than between managed care plans
and fee-for-service arrangements.
Fee-for-service plans are evolving in the
direction of managed care, particularly in
terms of utilization management mechanisms, including preapproval for elective
procedures; the monitoring of patients’
progress following hospital admissions; discharge planning; and case management
services for high-cost patients. Some feefor-service plans are also negotiating discounted charges for services. Perhaps
because of these changes, the rise in health
care costs has been dampened, even in feefor-service systems.19 These changes blur the
distinction between managed care and feefor-service plans, making it more difficult to
isolate the effects of managed care on access,
utilization, and outcomes.
One of the difficulties in studying the
impact of managed care is the problem of
patients themselves choosing managed care
or fee-for-service plans, and physicians electing to join organizations in part based on
whether the practices accept managed care
patients.14 Because of this self-selection of
patients and physicians into different types
of plans, it is difficult to estimate whether
any differences in health care utilization or
outcomes are attributable to financial and
administrative incentives, or whether baseline differences in enrollees’ behavior and
health status or physicians’ practice styles
partially explain these variations. Several
studies have documented that patients
enrolled in HMOs tend to have lower prior
health care expenditures and utilization
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
patterns, particularly for hospitalizations,
than patients who remain in fee-for-service
care.20,21 However, in most studies, the
health status of HMO enrollees does not
appear to be different from the health status
of fee-for-service patients, even though the
HMO enrollees used fewer services prior to
enrollment. Although the reason for this
relationship remains uncertain, it may
reflect the imprecise measures of health
status currently available (resulting in no
detectable difference), while the availability
of more precise prior utilization and cost
data allows smaller differences among
groups to be detected. The exception to this
pattern is that Medicare beneficiaries who
enroll in HMOs tend to be healthier than
their counterparts in traditional fee-forservice Medicare plans.22,23
services were more likely to be enrolled in
the IPAs.28 That study, however, did not
address the total costs of health care, and
evaluated an IPA plan that placed relatively
weak cost-control incentives on physicians.
The lower prior health care utilization of
HMO enrollees is not entirely understood.
One possibility is that HMOs attract patients
who prefer not to use medical services and
are cost-conscious and health-conscious individuals. Another possibility is that patients
who must switch physicians to join HMOs are
more likely to do so if they have weak prior
relationships with their physicians. Similarly,
high utilizers of medical care may have established bonds with their physicians and may
be less likely to switch health plans when
this requires switching physicians. This is an
important point because the newer forms of
managed care, such as independent practice
associations (IPAs) and preferred provider
organizations (PPOs), often do not require
switching physicians at enrollment. Thus, the
selection of lower-risk individuals into managed care plans historically experienced by
HMOs may not occur with these newer managed care arrangements. A study of the types
of individuals who chose to enroll in HMOs,
PPOs, and fee-for-service plans found that
the voluntary enrollment of lower-risk
patients was strongest in HMOs and much
weaker in PPOs, where switching physicians
was less likely to be required.24 Studies evaluating voluntary enrollment into Medicaid
managed care also have found that children
who chose to enroll in HMOs had lower
baseline costs.25–27 In contrast, a New York
study comparing privately insured children
enrolled in fee-for-service plans to those
enrolled in IPA plans that did not require
switching physicians found that children who
were high utilizers of acute care and specialty
Studies evaluating voluntary enrollment into
Medicaid managed care also have found
that children who chose to enroll in HMOs
had lower baseline costs.
The voluntary enrollment of individuals
with varying degrees of risk into different
types of health plans may falsely cause a plan
to look “good” or “bad” based on enrollee
utilization patterns or health outcomes, if
prior use and health status are not considered. It is important to recognize that voluntary enrollment of individuals into different
types of health plans may be due to patient,
provider, or health plan factors, or a combination thereof. Just as patients with a lower
tendency to use services may elect to enroll
in HMOs, so might primary care providers
influence patients’ choices. For example,
providers who serve both Medicaid fee-forservice and Medicaid managed care patients
might encourage families with special needs
children to remain under fee-for-service
coverage. Health plans also can facilitate
enrollment of healthy populations by carefully selecting providers and hospitals, and
through marketing practices.
It is important to note that financial disincentives used by managed care plans,
including patient copayments, do not apply
to the Medicaid population. Findings on the
effects of managed care for the privately
insured population may not apply to
Medicaid recipients, and vice versa. The
majority of studies on managed care among
children have focused on Medicaid beneficiaries and must be interpreted carefully
before extrapolating to the privately insured
population.
Effect of Managed Care on
Access to Health Services
Although access to health care is an intuitive concept and some experts have developed frameworks for defining access,29 it is
43
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
44
difficult to measure the effect of health
care arrangements such as managed care
on access to health services, and information in this area is limited. Some measures
that imply access include having a regular
primary care provider or usual source of
care; the first contact being with a primary
care provider; and a patient’s perception
of, or satisfaction with, the availability of
needed services. Utilization of services,
however, is a reflection of access and
many other factors.
The fact that managed care plans assign
a primary care provider to each enrolled
patient implies some level of potential
access; fee-for-service systems do not usually
engage in this practice. However, the
assignment of a primary care provider does
Children enrolled in HMOs were more likely
to report an unmet health care need and
difficulty in scheduling an appointment,
but reported higher overall visit rates.
not, by itself, ensure access to health services, and the majority of children—even
the uninsured—have a usual source of
health care.30,31
Access to Health Care for
Privately Insured Children
There is a dearth of information about the
effect of managed care systems on access to
care for nonpoor children who are covered
by private insurance, and the information
that is available addresses only some of the
indicators of access that have been identified. For example, one useful barometer of
access to care is parent satisfaction with
access, but only a few studies on this topic
exist. A Commonwealth Fund survey in
three U.S. cities found that HMO and PPO
enrollees were less satisfied with access to services for their children and for themselves
than were enrollees in fee-for-service
plans.32,33 Moreover, low-income families in
managed care plans were less satisfied with
their access to health services than either
low-income families in fee-for-service plans
or higher-income families in managed care
plans. It is possible that the poorer satisfaction ratings observed among lower-income
patients enrolled in managed care reflect
the lack of choice among plans offered to
this group by their employers, rather than
any true managed care effect.34 Nonetheless,
the finding of less satisfaction with access
to health services among managed care
enrollees was supported by results from the
landmark Medical Outcomes Study, which
focused on adults.35,36
The finding of poorer access to care
among disadvantaged groups is supported
by the 1994 Robert Wood Johnson Foundation National Access Survey, which documented that adults and children enrolled
in HMOs were more likely to report an
unmet health care need and difficulty in
scheduling an appointment, but reported
higher overall visit rates, than either PPO
or traditional fee-for-service enrollees.37
When the analysis was stratified by income
level, low-income enrollees in both HMOs
and fee-for-service plans reported poorer
access to care than higher-income enrollees.
These results suggest that, compared to
high-income enrollees, privately insured
low-income persons may experience greater
access barriers within both HMOs and traditional fee-for-service settings.
Data on preventive health visits and
referrals for specialty services—indicators
of both access and utilization—are discussed in the utilization section of this
article. Most studies reviewed suggest that
managed care may be associated with
higher preventive visit rates than fee-forservice arrangements, though any benefits
attributable to managed care are likely to
be reduced as more fee-for-service plans
cover preventive care with minimal or no
patient copayments required.8,28,38 In referrals for specialty care, results from previous
studies have been mixed. However, for
chronically ill children, there is evidence
that although HMOs offer some advantages
for this population, they also limit the
breadth of services available by restricting
the number of visits and the types of services that are covered, as well as the choice
of network physicians.16
Access to Health Care Under
Medicaid Managed Care
Substantially more research has been done
on access to care under managed care systems for poor children and pregnant
women covered by Medicaid. Although his-
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
torically improvements in access to care
were noted with enrollment in traditional
fee-for-service Medicaid,39 compared to
nonpoor children, poor children still lack
adequate access to both primary and
specialty health services in spite of their
Medicaid coverage.31 Thus, it is important
to assess whether children fare better under
Medicaid managed care than under traditional fee-for-service Medicaid. Overall, studies have yielded mixed results, with no
consistent improvement in access to care
resulting from Medicaid managed care.
In Medicaid managed care demonstration projects during the 1980s, the effect of
managed care on indicators of access was
mixed. In Arizona, for example, families
were more likely to report a usual source of
care,40 but less likely to report being able to
obtain sick care for their child, after the
implementation of managed care.41 An
analysis of the Medicaid primary care gatekeeper programs in California and New
Jersey found that in California, where primary care gatekeepers were paid on a fee-forservice basis with a case management fee,
the percentage of all children with at least
one primary care visit did not change substantially after enrollment in managed care
(42% predemonstration versus 38% postdemonstration). In New Jersey, however,
where gatekeepers were paid on a capitated
basis, the percentage of children with a primary care visit decreased (64% predemonstration versus 58% postdemonstration). In
comparison sites that continued fee-forservice Medicaid, the percentage of children with a primary care visit did not change
during the same time period.42 A similar
study in Ohio also found a reduced likelihood of having at least one physician visit for
continuously enrolled children and adults
(mostly women) in managed care.26 These
results suggest that, for Medicaid beneficiaries, the effect of managed care on access to
primary care services may depend on the
payment structure within the Medicaid managed care program, with a reduction in
access to primary care services being more
likely with capitated payments.
Among low-income pregnant women,
studies have found no consistent effect of
managed care on access to prenatal services.
A study in California and Missouri counties
found no overall improvement in prenatal
care when comparing prepaid capitated
plans to fee-for-service plans.43 Another
study examining the effect of a mandatory
Medicaid case management program on
prenatal care found low utilization of prenatal care in both study and comparison feefor-service groups, with no improvement in
the case management group.44 Finally, an
analysis comparing Medicaid recipients in
three managed care plans to those in fee-forservice plans in the state of Washington
found no consistent managed care effect on
the prevalence of late or no prenatal care.45
Under all arrangements, Medicaid enrollees
had lower use of prenatal care than did nonMedicaid enrollees.
Some information is available regarding
patients’ perceptions of access to child
health services under Medicaid managed
care. A telephone survey of adult Medicaid
recipients, many of whom had children
enrolled in Medicaid, found that those in
managed care plans were more likely to have
a usual source of care other than the emergency department; see the same clinician at
Poor children still lack adequate access to
both primary and specialty health services
in spite of their Medicaid coverage.
their usual source; contact their regular
provider or health plan rather than the
emergency department for urgent or afterhours care; and experience shorter appointment waiting times than did fee-for-service
Medicaid recipients.46 Several surveys of
managed care enrollees from the Medicaid
demonstration projects during the 1980s
also documented higher patient satisfaction
with quality of care, availability of care, and
choice of physician under Medicaid
managed care than under fee-for-service
Medicaid.47,48
For chronically ill children covered by
Medicaid, there is some evidence of
decreased access to specialty services associated with managed care;15,16,49,50 this is
discussed in greater detail in the utilization
section of this article.
In summary, there are insufficient information and mixed results regarding the
45
46
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
effect of managed care on access to care for
children. Among privately insured children, for whom baseline access under feefor-service care is relatively high, there is
some evidence of improved access to preventive health services, but possibly
reduced access to specialty services for
chronically ill children under managed
care. In addition, managed care plans may
fare worse on other indicators of access,
such as perceived satisfaction, unmet
needs, and difficulty scheduling appointments, though additional research is
needed in this area.
For children covered by Medicaid, information is available from early Medicaid managed care demonstrations and experience
during the 1980s, but surprisingly little information is available from the recent surge of
state Medicaid populations into managed
care plans. Some studies have reported
better perceptions of access to health care
under Medicaid managed care, as evidenced
by having a usual source of care, improved
continuity of care, shorter waiting times, and
improved patient satisfaction. Other studies,
One recent investigation found that 82%
of HMO enrollees, 61% of POS enrollees,
and 48% of fee-for-service enrollees were
discharged within one day of delivery.
however, have found that managed care had
little effect, or a negative effect, on access to
primary care or office visits for children and
pregnant women, depending on the reimbursement mechanism. There is also some
evidence of decreased access to services for
special needs children in managed care
plans. It is important to note that baseline
levels of access for children covered by traditional fee-for-service Medicaid often have
been quite poor, and switching to Medicaid
managed care has had greater potential to
improve access for Medicaid enrollees than
for privately insured children. More studies
are needed to evaluate changes in access to
care for Medicaid enrollees who have
recently switched from fee-for-service to
managed care plans, and to evaluate the relative impact of capitation versus fee-forservice reimbursement within Medicaid
managed care.
Effect of Managed Care on
Utilization of Health Services
The effects of managed care on utilization of
services for children have been reviewed
previously.19,49,51 The preponderance of studies have involved Medicaid recipients, and
extrapolating findings from Medicaid managed care to the privately insured, nonpoor
child population, for whom the system of
fee-for-service health care is often markedly
different, should be done with caution. Also,
most studies of managed care for children
have involved group- or staff-model HMOs,
with fewer studies assessing IPA or point-ofservice (POS) model health plans, in which
cost-control strategies and provider incentives may be quite different.
Hospital Utilization
Because hospital care accounts for approximately 40% to 50% of total health care
expenditures for children,52 many studies
have evaluated the ability of managed care
plans to affect hospital utilization. For
insured adults and children combined, managed care systems, particularly HMOs, have
reduced hospital admissions and hospital
costs substantially.53 However, findings for
children specifically are more mixed.
For Privately Insured Children
Because hospitalization rates for children are
low and the average length of stay is short,
one would expect managed care to have
smaller effects on hospital use for children
than for adults. Consequently, it would take
large-scale studies to evaluate the effect of
managed care on inpatient use for children.
The RAND Health Insurance Experiment
(HIE), a landmark randomized clinical trial,
allocated families into a staff-model HMO,
Group Health Cooperative of Puget Sound,
or into traditional fee-for-service plans that
required no deductible or copayments (freecare) or various levels of cost sharing. This
study found that the hospitalization rate for
children in the HMO experimental group
was 45% lower than in the free-care plan, but
not significantly different from the fee-forservice plans that required cost sharing.38 It is
unlikely that this reduction in hospitalizations achieved by the HMO in comparison to
the free-care group could be achieved today,
because the HIE was performed 15 years ago,
and since that time hospitalization rates
for children have declined.54 Another study
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
PHOTO OMITTED
comparing health care utilization by children in IPA plans to that of a similar population under fee-for-service care found no
differences in hospital admission rates
between the two groups, but slightly higher
inpatient expenses in the IPA group.
However, the authors noted that the sample
of 1,280 children in this one community was
too small to evaluate hospital use in depth.28
One area that has received attention
recently is postpartum maternal and newborn discharges under managed care plans.
During the past 40 years, the average length
of postpartum hospital stays in the United
States has declined dramatically for mothers
and their newborns, from 7 to 10 days in the
1950s to 1.4 days in 1996. Currently, postpartum lengths of stay in the United States
are among the shortest in the world.55
Others have summarized the literature on
outcomes associated with early discharge
and have cautioned about the potential
medical risks of this practice, particularly
for poor or medically high-risk women.56
Although managed care is believed to have
influenced the trend toward shorter postpartum stays, it is important to note that
hospital lengths of stay are decreasing in all
populations. Studies comparing fee-forservice to managed care plans have found
somewhat mixed results, though most have
reported shorter lengths of stay under
managed care. One recent investigation
found that 82% of HMO enrollees, 61% of
POS enrollees, and 48% of fee-for-service
enrollees were discharged within one day of
delivery.57 These differences varied across
geographic regions, with no differences in
the Northeast. Also, among women discharged within the first postpartum day,
readmissions were higher for HMO enrollees
than for fee-for-service enrollees, although
readmission rates for newborns did not vary
by plan type.
Under Medicaid Managed Care
Substantial research has been conducted on
the effect of Medicaid managed care on hospital utilization, though no consistent managed care effect has been observed across
studies. A review of 25 published and unpublished studies on Medicaid managed care
prior to 1992 found that in most cases there
were reductions or no change in hospital utilization associated with managed care
among adults and children combined.58
Another analysis of mandatory and voluntary enrollment in Medicaid managed care
during the late 1980s in Ohio found no difference in hospital utilization associated
with mandatory enrollment in managed
care. However, in the sites where managed care enrollment was voluntary, hospital
admissions and the number of inpatient
days were reduced for continuously enrolled
women and children; pregnancy and newborn care did not account for these
changes.26 In a controlled clinical trial of a
Medicaid HMO in New York City in the late
1980s, Medicaid families were randomly
assigned to either fee-for-service or HMO
plans, though actual HMO enrollment was
voluntary.25 No differences were observed in
47
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
48
hospitalization rates, hospital resource utilization, or hospital costs between the HMO
and fee-for-service groups. Also, no difference was found in hospital utilization when
comparing the randomly assigned HMO
group to a group that voluntarily chose to
participate in the Medicaid HMO.
In conclusion, findings regarding the
impact of managed care on pediatric hospital utilization are mixed. For children covered by private insurance, there is a paucity
of recent information from rigorous studies,
in spite of a great deal of rhetoric about the
effects of managed care. For women and
newborns, however, there is a growing body
of evidence suggesting that enrollment in
A number of studies have found that
emergency department use is lower under
managed care than under fee-for-service
Medicaid.
managed care plans may be associated
with shortened postpartum hospitalizations,
though the effect of this change on health
and health care service outcomes remains
uncertain. Among children enrolled in
Medicaid, results usually have shown no difference by plan type, or somewhat lower hospitalization rates associated with managed
care enrollment, with little data from recent
evaluations. There is some evidence that
lower hospitalization rates observed in some
Medicaid managed care plans were due to
patients self-selecting into managed care,
and did not reflect any true effect of managed care per se.
With trends leading to reductions in
hospitalization rates and stricter controls on
hospital use across all types of health plans,
the effect of managed care on pediatric hospitalization rates—which were low to begin
with—is likely to be small and difficult to
detect. This finding is in contrast to those of
hospital use among adults, for whom managed care has been shown to reduce both
admission rates and total inpatient days.53
Emergency Department (ED)
Utilization
Because one of the objectives of managed
care is increased reliance on the primary
care provider or “gatekeeper” and reduced
use of discretionary episodic care, one would
expect managed care systems to minimize
ED utilization. Most managed care plans
have patient and/or provider controls that
pertain to ED use. These include patient
copayments, required authorization by the
primary care provider, gatekeeping leading
to denials within the ED, provider profiling,
and financial penalties for providers.
For Privately Insured Children
There are no published reports of the effect
of managed care on ED utilization among
nonpoor children covered by private
insurance specifically; studies have not
differentiated ED use from the use of other
ambulatory services.
Under Medicaid Managed Care
Because poor, urban populations covered by
Medicaid have the highest rate of ED visits
for nonurgent conditions,59 Medicaid managed care plans have focused on reducing
inappropriate ED use. A number of studies
have found that ED use is lower under
managed care than under fee-for-service
Medicaid. A review of Medicaid demonstration projects during the 1980s found that all
nine evaluations reporting ED utilization
documented that use was reduced by
about one-third in managed care plans, as
compared to fee-for-service plans.19 Other
Medicaid managed care programs also have
noted consistent reductions in ED visits.60,61
In addition, an increase in the severity of ED
visits has been observed,62 suggesting that
discretionary or inappropriate ED visits
were avoided in managed care settings. A
randomized clinical trial of a Medicaid
HMO found no differences in ED visits comparing HMO to fee-for-service enrollees.25
However, ED use was measured by parent
diaries in this trial and may have been
recorded inaccurately.
There is a major concern about managed care plans placing controls on pediatric ED visits and consequently reducing
ED use among poor children. A small group
of high-risk, critically ill children enrolled in
Medicaid legitimately require ED care, but
are denied or fail to receive appropriate care
in the ED or elsewhere,63 or are lost to
follow-up64 because of restrictions on utilization practices. A survey of children who were
denied authorization for nonemergency
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
care in a university-based pediatric ED
under Medicaid managed care found no
adverse health outcomes resulting from the
delay in care. However, although the subsequent ED utilization rate of the denied
group was similar to the rate of a comparison group, the subsequent hospitalization
rate in the denied group was higher.63 Since
the preponderance of evidence demonstrates that Medicaid managed care does
reduce ED use, it is important to continue to
monitor the health outcomes of patients
whose care may be adversely affected.
Utilization of Primary Care
Ambulatory Services
Because most managed care plans assign
patients to a primary care provider, and this
typically does not occur among fee-forservice indemnity plans, one may expect a
greater utilization of primary care ambulatory services for children enrolled in
managed care plans. As discussed below,
however, research suggests that any difference in utilization associated with health
plan types is most likely due to differences in
cost sharing and physician reimbursement
mechanisms across health plans.
For Privately Insured Children
For privately insured children, it is important to interpret the effects of managed care
in comparison to the benefit coverage for
the fee-for-service population. For example,
most early studies of HMO enrollment
among adults and children combined documented higher rates of office and preventive
visits than among fee-for-service controls
who lacked coverage for these services.11
Reviews of more recent studies, again for
children and adults combined, typically
found more physician office visits per
enrollee, or no difference in utilization, in
managed care plans, as compared to fee-forservice indemnity plans.53 These findings
held true across various types of managed
care arrangements. The attenuated effect of
managed care in more recent investigations
may be attributed to recent changes in benefit coverage across many fee-for-service
plans, which now typically cover preventive
care, immunizations, and other primary
care services with little or no copayment.
In the RAND HIE, children assigned to
the HMO had 20% more office visits and
40% more preventive visits than did children
assigned to the cost-sharing plans, and 20%
more office visits but no difference in preventive visits compared to children assigned
to free-care.38 Similarly, a study in Rochester,
New York, found significant increases in preventive visits, acute illness visits, chronic illness visits, after-hours visits, weekend visits,
out-of-office laboratory evaluations, and
referrals to medical specialists among IPA
enrollees as compared to children under
fee-for-service care, after controlling for
self-selection.28 At the time of the study, feefor-service control patients typically had
copayments for preventive and many primary care services at levels similar to the
RAND HIE copayment groups.
A more recent study examined previously uninsured or underinsured children
in New York State who enrolled in Child
Health Plus, a state-funded health plan
for poor children who do not qualify for
Medicaid.65 While the overall population of
children experienced increased use of preventive, acute, primary care, and specialty
services after enrollment in Child Health
Plus, these increases did not vary among
Any difference in utilization associated
with health plan types is most likely due to
differences in cost sharing and physician
reimbursement mechanisms across
health plans.
children enrolled in managed care versus
fee-for-service commercial plans. Patient
copayments for ambulatory services were
similar for managed care and fee-for-service
enrollees, most likely accounting for the lack
of an effect attributable to managed care.
The effect of patient cost sharing on differences in ambulatory care utilization has
been documented elsewhere.66 Earlier investigations also have found that differences in
ambulatory use between HMO and feefor-service populations disappeared when
HMO enrollees were compared to control
patients having comparable ambulatory coverage.11 These studies suggest that the historical advantage of managed care may be
diminished with the reduction or elimination of copayments for preventive care services in many fee-for-service plans.
49
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
50
Under Medicaid Managed Care
Studies of the effect of managed care on
ambulatory utilization for Medicaid enrollees
again have found mixed results. The systematic evaluation of Medicaid demonstration
projects found an increase in ambulatory
office visits among adults and children
enrolled in managed care in nearly half of the
programs (11 of 25), though the remainder
showed no managed care effect (5 of 25) or a
decrease (8 of 25) in the visit rate compared
to fee-for-service Medicaid.58 Managed care
programs that paid the primary care gatekeeper on a fee-for-service basis were more
likely to be associated with increases in visits,
Ambulatory costs and the probability of
any ambulatory visit were about 40%
lower for children randomly assigned to
the Medicaid HMO.
while programs that used capitated reimbursements were more likely to experience
decreases or no change in visit rates. Groupand staff-model HMOs were more strongly
associated with a decrease in office visits than
were IPA plans, irrespective of whether the
payment structure of the IPA plan was fee-forservice or capitated. The Medicaid managed
care evaluation in Ohio noted decreases in
office visits for children and adults enrolled
in managed care in most sites.26
The randomized clinical trial of a
Medicaid HMO found that ambulatory costs
and the probability of any ambulatory visit
were about 40% lower for children randomly assigned to the Medicaid HMO than
for children who either chose to remain in a
Medicaid fee-for-service plan or were randomly assigned to the Medicaid fee-forservice plan.25 Patients who switched out of
the Medicaid HMO and patients who were
randomly assigned to the HMO but never
enrolled had even higher ambulatory utilization than the fee-for-service patients,
suggesting that the HMO enrolled and
retained a population of patients that were
low utilizers of ambulatory care. These results
highlight the importance of considering
the self-selection of patients into different
types of health plans when interpreting
differences between managed care and
fee-for-service systems.
Another study examined utilization patterns in a voluntary Medicaid managed
care program, the Children’s Medicaid
Program, in Suffolk County, New York.61
Within the managed care program, one
group of physicians was reimbursed under
a capitated system, while the other group
received fee-for-service payments at marketlevel rates that resembled those of an IPA
plan. This study found that the switch
from traditional fee-for-service Medicaid to
capitated Medicaid managed care was not
associated with a change in office visits, but
that the switch to fee-for-service Medicaid
managed care was associated with a significant increase in office visits. Patients
in both managed care groups, but particularly those in the fee-for-service group,
were more likely than patients in the traditional Medicaid group to be compliant
with recommended guidelines for wellchild care.67
Overall, these findings do not demonstrate a consistent increase or decrease in
childhood ambulatory office visits attributable to managed care, though distinct patterns of utilization by plan type do appear
for different child populations. For privately
insured children, enrollment in managed
care was most likely to increase the use of
ambulatory services when managed care was
compared to a fee-for-service plan that did
not cover ambulatory services or required
patient copayments for these services. Thus,
it is difficult to tell whether this observed difference is due to managed care arrangements per se, or simply the reduced cost
sharing more often experienced among
managed care enrollees.
For poor children covered by Medicaid
and for whom cost sharing is not required,
differences in ambulatory care use by plan
type were usually associated with provider
reimbursement mechanisms. Managed care
arrangements using fee-for-service reimbursement and less restrictive managed
care models were more likely to increase
ambulatory primary care visits, while capitated reimbursements and more restrictive
managed care arrangements, such as staffor group-model HMOs, tended to decrease
or have little effect on these services.
Primary care visit rates by poor children
enrolled in Medicaid usually have been
found to be below recommended guide-
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
lines, whether or not they were enrolled in
managed care.
Utilization of Specialty Services
The effect of managed care arrangements is
of particular concern for children with complex physical or emotional disorders, who
may require expensive specialty services.
Relatively healthy children, too, may
require specialty services for acute conditions at various times throughout their
development. Theoretically, primary care
physicians who assume some degree of
financial risk for the care of patients may
face strong cost disincentives to refer children with complex needs to specialists,
resulting in the inappropriate underutilization of such services.
For Privately Insured Children
One rigorous investigation specifically examined the effect of managed care on specialty
utilization for privately insured children. In
the Child Health Plus program in New York,
the utilization of specialists increased dramatically for the entire group of previously
uninsured children after enrollment in
Child Health Plus, but there was no managed care effect on this overall increase in
specialty care.65 It is likely that the addition
of any health insurance coverage overshadowed any possible effect of managed care in
specialty utilization.
A few other studies have explored referrals to specialty services for privately insured
children under managed care, and results
from these investigations have been mixed.
A 1988 survey of more than 1,000 pediatricians reported fewer referrals to specialists
and a higher likelihood of denied referrals
in managed care than in fee-for-service
plans, with greater perceived barriers in
PPOs than in staff-model HMOs.68 In contrast, two studies of specialty referrals
within IPA plans in Rochester, New York,
found a higher rate of specialty referrals for
children enrolled in IPA plans than for children enrolled in Blue Cross fee-for-service
plans.14,28 It is important to note, however,
that the financial incentives and fee-forservice payments within these IPAs represented relatively weak controls on specialty
referrals.
Among chronically ill children, an analysis of the 1989 National Health Interview
Survey found that although HMOs offered
some advantages, including reduced cost
sharing and improved availability of
mental health, ancillary, and home care services, many HMOs also limited care by
restricting the breadth of specialty services,
the number of visits, and the choice of
providers that were covered.16 A major
determinant of access to and utilization of
specialty services is coverage in the benefit
package, which varies substantially among
managed care plans. Other mechanisms
implemented by managed care organizations to reduce the use of specialty services
include financial, administrative, and educational strategies.
Under Medicaid Managed Care
There is considerably more information
available about the use of specialists from
studies that examined Medicaid managed
care, though few recent studies exist. The
evaluations of the Medicaid demonstrations
consistently found reductions in the probability of visiting a specialist of about 30% to
50% among managed care enrollees as compared to fee-for-service enrollees (except in
Evaluations of the Medicaid demonstrations
consistently found reductions in the
probability of visiting a specialist of about
30% to 50% among managed care enrollees
as compared to fee-for-service enrollees.
Missouri), and typically found that a greater
proportion of visits occurred at the primary
care office.26,58 Similarly, the Children’s
Medicaid Program in Suffolk County, New
York, found reduced non–primary care office
visits in the capitated managed care group,
although the fee-for-service managed care
group and a comparison group both experienced increased specialty visits during the
same time period.61 While lower specialty
utilization among children may reflect
reduced scattering of health care, it is possible that some of the specialty visits averted by
Medicaid managed care may have been
needed for children with complex health
conditions.
For children with special needs, another
way that managed care plans have limited
the use of specialty care is by excluding from
51
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
52
their benefit packages specific services that
are critical for this population.49,50 In addition, Medicaid managed care plans have
been noted to limit network providers and
require prior authorization for out-of-plan
referrals to specialists, thus reducing the use
of necessary services for chronically ill and
disabled children.15,16,49
In summary, little research exists to draw
sound conclusions about the effect of managed care arrangements on the use of specialty services by privately insured children.
The few studies that do exist have mixed
results, and any managed care effect
observed appears to be influenced by the
strength of cost incentives felt by providers
or by children’s baseline health insurance
status. In contrast, for poor children enrolled
in Medicaid, findings of a negative effect of
managed care on the use of specialty services
are stronger, though whether this is attributable to managed care alone or to the degree
Medicaid managed care plans limit network
providers and require prior authorization
for out-of-plan referrals to specialists,
reducing the use of necessary services for
chronically ill and disabled children.
of financial risk within different managed
care plans cannot be drawn out using available data. Moreover, the effect that referrals
to pediatric specialists have on the quality of
health care for children has not been determined. Additional investigations are needed
to evaluate health outcomes related to the
use of specialty services and the process of
specialty referrals to judge whether reductions of referrals in managed care settings
are beneficial or harmful to disadvantaged
children.
Conclusion
The bulk of research on the effect of managed care on access and utilization has
focused on adults. Yet there are unique
characteristics about child health care that
highlight the importance of examining the
effect of managed care on the health of this
population specifically. An example characteristic is the critical role of prevention and
ambulatory care for all children. In addition,
it is particularly important to determine the
effect of managed care on children with
severe chronic physical or mental illnesses,
and on the much larger population of poor
children covered by Medicaid. Managed
care has potential advantages and disadvantages with respect to the delivery of child
health care, and its performance must be
monitored carefully as the number of children enrolled in managed care plans continues to expand.
Studies of access to care for children
have found mixed results for managed care.
For privately insured children, managed
care has had a consistent effect on some
indicators of access to health services.
However, potentially worrisome findings
have been noted with regard to reduced
access to needed specialty care for children
with chronic conditions. For children covered by Medicaid, results from several evaluations in the 1980s found improved access to
primary care services. Other studies, however, have documented inconsistent findings
on indicators of access to care for pregnant
women and children, which have varied by
reimbursement mechanisms and across
individual plans. In addition, there is a
paucity of research from the recent wave of
state Medicaid recipients enrolled in managed care. Concern has been noted about
reduced access to specific specialty services
for Medicaid recipients with chronic conditions. The impact of managed care depends
to a large extent on baseline levels of care
in the comparison indemnity population,
and on changes in benefit packages by sponsors. Access to services usually has been
worse for the Medicaid population than for
privately insured children; thus, the potential positive impact of managed care is
greater for Medicaid recipients. However, as
baseline access to care improves, the potential improvement from managed care is
reduced, and adverse effects may become
more salient.
A number of studies have examined the
effect of managed care on utilization of care
by children. For privately insured children,
there is little recent evidence that managed care reduces hospitalization rates or
inpatient services, except for postpartum
maternal and newborn hospitalizations, for
which managed care has been noted to
reduce lengths of stay. Similarly, for the population covered by Medicaid, there are scant
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
PHOTO OMITTED
data supporting an effect of managed care
on hospitalization rates, particularly after
the effects of patient self-selection are
eliminated. These results are in contrast to
findings for the adult population, for whom
managed care has been noted to reduce
hospitalizations. It is likely that low baseline hospitalization rates for children reflect
a general reluctance to hospitalize children unless absolutely necessary, thereby
minimizing any potential effect of managed care.
While there are virtually no data on the
effect of managed care on emergency
department use for privately insured children, there is relatively good evidence that
managed care has reduced ED visits for
Medicaid recipients, and in many instances
has reduced low-severity ED visits. This trend
concerns many child health experts, because
some evidence suggests there are dangers to
denying pediatric ED care to high-risk children, who may suffer subsequent adverse
health consequences.
Studies of ambulatory utilization have
not found consistent effects of managed
care on office visits, and any managed care
effects observed depend in large part on the
ambulatory coverage for the comparison
group and the payment mechanism of the
managed care group. When compared to
groups having poor coverage for office visits,
some studies have found that privately
insured children enrolled in managed care
plans have higher visit rates. For Medicaid
recipients, managed care plans that pay
providers on a capitated basis are more likely
to decrease or have no effect on office visits,
while those that pay providers on a fee-forservice basis are more likely to increase the
use of ambulatory services. Managed care
has not consistently improved utilization of
preventive services for vulnerable children
in Medicaid, and overall levels of preventive
care remain lower than for privately insured
children.
There is strong evidence for reduced
use of specialty services resulting from managed care, particularly for children enrolled
in Medicaid. In addition, there is some
evidence for reduced coverage of high-cost
specialized services for children with chronic
illnesses enrolled in managed care plans.
Children are particularly vulnerable to
inadequate access to and utilization of
health services. Findings pertaining to adults
often have not been observed for children,
and findings specific to early evaluations of
managed care may not apply to newer managed care arrangements implemented in a
rapidly changing health care environment.
Moreover, there is little evidence of the
appropriateness of health care services used
by children in managed care settings. It is,
therefore, critical during the next decade to
continue to monitor the effect of managed care on access to and utilization of
services by children, as well as to evaluate its
effect on quality of care and child health
outcomes.
53
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
54
Table A1
Summary of Access and Utilization of Health Services for
Children and Families by Type of Health Plan
Health Plan
Comparisona
Indicator
Finding
Access for Privately
Insured
a
MCb vs. FFSc,28,38
Preventive health/office visits
Increased in HMOd if copay
in FFS; otherwise little effect
HMO/PPOe vs. FFS34,35
Satisfaction with access
Lower in HMO/PPO
HMO vs. PPO/FFS37
Unmet health care needs;
difficulty getting appointment;
overall visit rates
Greater in HMO
Summary: Access to preventive health visits may increase under managed care if cost sharing is
required in fee-for-service plans. Managed care generally fares worse on other indicators of access,
including satisfaction, unmet health needs, and difficulty scheduling appointments.
Access Under Medicaid
Managed Care
MC vs. FFS40
Usual source of care
Availability of sick child care
Improved in MC
Worse in MC
MC vs. FFS46
Usual source of care; continuity
of care; regular provider for urgent/
after-hours care; waiting times
Improved in MC
MC vs. FFS42
Any primary care visit
Decreased in MC if capitated
payment; no difference if FFS
payment in MC
MC vs. FFS26
Any physician visit
Decreased in MC
Prenatal visits
No consistent effect
MC vs.
FFS43–45
MC vs. FFS47,48
a
Satisfaction with quality of care;
Higher in MC
availability of care; physician choice
Summary: Numerous indicators of access are better under Medicaid managed care, including having
a usual source of care, continuity of care, shorter waiting times, satisfaction with quality of care, and
provider choice. However, there is some evidence that Medicaid managed care may decrease or have
no impact on availability of primary care visits for children and pregnant women.
Hospital Utilization for
Privately Insured
MC vs. FFS57
HMO vs. FFS38
Postpartum length of stay
Hospitalization rate
Usually shorter in MC
Lower in HMO if no copay in
FFS; otherwise no difference
IPAf vs. FFS28
Hospitalization rate
No difference
a
Summary: Little evidence of any managed care effect for children. Some evidence that managed care
is associated with shorter postpartum hospital stays.
a
See the related endnotes following this article for the complete citation of a publication in which the
relevant study is described and/or discussed.
b
MC = managed care plan, type unspecified
c
FFS = fee-for-service plan
d
HMO = health maintenance organization, type unspecified
e
PPO = preferred provider organization
f
IPA = independent practice association
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
Table A1 (continued)
Summary of Access and Utilization of Health Services for
Children and Families by Type of Health Plan
Health Plan
Comparisona
Indicator
Finding
Hospital Utilization Under
Medicaid Managed Care
a
MC vs. FFS19,47,58
Hospitalization rate
No consistent effect
MC vs. FFS26
Hospitalization rate
Lower if voluntary MC
enrollment; no difference if
mandatory MC enrollment
HMO vs. FFS25
Hospitalization rate; resource
utilization; costs
No difference
Summary: Little evidence of any managed care effect.
Emergency Department
Utilization for Privately
Insured
No Studies
Emergency Department
Utilization Under Medicaid
Managed Care
a
MC vs. FFS19,60–62
Emergency department (ED) visits
Lower in MC
MC vs. FFS62
Severity of ED visits
Higher in MC
HMO vs. FFS25
ED visits
No difference
Summary: Fewer emergency department visits and increased severity of visits associated with
Medicaid managed care.
Primary Care Ambulatory
Utilization for Privately
Insured
HMO vs. FFS11
Office/preventive visits
Increased in HMO
HMO vs. FFS38
Office/preventive visits
Increased in HMO if copay
in FFS; otherwise little effect
IPA vs. FFS28
Preventive, acute, chronic illness,
after-hours, and weekend visits;
lab evaluations; specialty
referrals
Increased in IPA compared to
FFS with copay
MC vs. FFS13,53
Office visits
Increased in MC or no effect
Ambulatory visits
No change
MC vs.
FFS65
a
Summary: Some evidence of increased visits associated with managed care when cost sharing for
these services is required in the comparison fee-for-service plan.
a
See the related endnotes following this article for the complete citation of a publication in which the
relevant study is described and/or discussed.
b
MC = managed care plan, type unspecified
c
FFS = fee-for-service plan
d
HMO = health maintenance organization, type unspecified
e
PPO = preferred provider organization
f
IPA = independent practice association
55
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
56
Table A1 (continued)
Summary of Access and Utilization of Health Services for
Children and Families by Type of Health Plan
Health Plan
Comparisona
Indicator
Finding
Primary Care Ambulatory
Utilization Under Medicaid
Managed Care
MC vs. FFS58,61
Office visits
Increased in MC if FFS
payment; decreased in MC or
no effect if capitated payment
MC vs. FFS26
Office visits
Usually decreased in MC
Office visits; costs
Decreased in HMO
HMO vs.
a
FFS25
Summary: Findings mixed. Managed care most likely to increase probability of visits if reimbursement
method is fee-for-service. Managed care most likely to lower probability of visits or have no effect if
reimbursement method is capitation.
Specialist Utilization for
Privately Insured
a
MC vs. FFS65
Visits to specialists
No change
MC vs. FFS68
Specialty referrals
Denied referrals
Lower in MC
Higher in MC
IPA vs. FFS14,28
Specialty referrals
Higher in IPA
HMO vs. FFS16
Coverage for breadth of services;
Restricted in HMO
number of visits; choice of providers
HMO vs. FFS16
Availability of mental health,
ancillary, home health services
for children with special needs
Better in HMO
Summary: No consistent difference in referrals and service use. Some evidence for restricted
coverage under managed care for chronically ill children.
Specialist Utilization
Under Medicaid
Managed Care
MC vs. FFS26,42,58
Visits to specialists
Lower in MC
MC vs. FFS61
Non–primary care office visits
Lower in MC if capitated
payment
MC vs. FFS15,16,49,50
Coverage for breadth of services;
limit network providers;
require preauthorization
for out-of-plan specialty
referrals
Restricted in HMO
a
Summary: Evidence for decreased use of specialty services and restricted coverage under
managed care.
a
See the related endnotes following this article for the complete citation of a publication in which the
relevant study is described and/or discussed.
b
MC = managed care plan, type unspecified
c
FFS = fee-for-service plan
d
HMO = health maintenance organization, type unspecified
e
PPO = preferred provider organization
f
IPA = independent practice association
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
1. Szilagyi, P.G., and Schor, E.L. The health of children. Health Services Research (1998), in
press.
2. Starfield, B. The effectiveness of medical care: Validating clinical wisdom. Baltimore, MD: Johns
Hopkins University Press, 1985.
3. Halfon, N., Newacheck, P.W., Wood, D.L., and St. Peter, R.F. Routine emergency department
use for sick care by children in the United States. Pediatrics (July 1996) 98:28–34.
4. Newacheck, P.W., and Taylor, W.R. Childhood chronic illness: Prevalence, severity, and
impact. American Journal of Public Health (1992) 82:364–71.
5. Newacheck, P.W., and McManus, P. Financing health care for disabled children. Pediatrics
(March 1988) 81:385–94.
6. Lewit, E.M., and Monheit, A.C. Expenditures on health care for children and pregnant
women. The Future of Children (Winter 1992) 2,2:95–114.
7. Newacheck, P.W. Poverty and childhood chronic illness. Archives of Pediatrics and Adolescent
Medicine (1994) 148:1143–49.
8. For example, on December 13, 1993, the New York State Assembly amended the insurance
law to cover preventive and primary care services to dependent children and adolescents.
9. Centers for Disease Control and Prevention. Reported vaccine-preventable diseases: United
States, 1993, and the Childhood Immunization Initiative. Morbidity and Mortality Weekly Report
(1994) 43:57–60.
10. See the National Committee for Quality Assurance’s Health Plan Employer Data and
Information Set (HEDIS 2.5) for use in managed care plans. Available from the National
Committee for Quality Assurance Publications Center, P.O. Box 533, Annapolis Junction,
MD 20701-0533.
11. Luft, H.S. Health maintenance organizations: Dimensions of performance. New York: Wiley,
1981.
12. Iglehart, J.K. Health policy report: The American health care system: Managed care. The New
England Journal of Medicine (1992) 327:742–47.
13. Miller, R.H., and Luft, H.S. Managed care: Past evidence and potential trends. Frontiers of
Health Services Management (1993) 9:3–37.
14. Szilagyi, P.G., Rodewald, L.E., and Roghmann, K.J. Managed health care for children. Journal
of Ambulatory Care Management (1993) 16:57–70.
15. Horowitz, S.M., and Stein, R.E.K. Health maintenance organizations versus indemnity
insurance for children with chronic illnesses. American Journal of Diseases of Children (1990)
144:581–86.
16. Fox, H.B., Wicks, L.B., and Newacheck, P.W. Health maintenance organizations and children
with special health needs. A suitable match? American Journal of Diseases of Children (1993)
147:546–52.
17. Perrin, J.M., and Ireys, H.T. The organization of services for chronically ill children and their
families. Pediatric Clinics of North America (1984) 31:235–57.
18. Hurley, R.E., Kirschner, L., and Bone, T.W. Medicaid managed care. In The managed care
handbook. 3rd edition. P.R. Kongstvedt, ed. Gaithersburg, MD: Aspen Publications, 1996,
chapter 48.
19. Freund, D.A., and Lewit, E.M. Managed care for children and pregnant women: Promises
and pitfalls. The Future of Children (Summer/Fall 1993) 3,2:92–122.
20. Hellinger, F.G. Selection bias in HMOs and PPOs: A review of the evidence. Inquiry (Summer
1995) 32:135–42.
21. Hellinger, F.J. Selection bias in health maintenance organizations: Analysis of recent evidence.
Health Care Financing Review (1987) 9:55–63.
22. Brown, R.S., and Hill, J.W. Costs and use of services: The effects of Medicare risk HMOs on
Medicare costs and service utilization. In HMOs and the elderly. H.S. Luft, ed. Ann Arbor, MI:
Health Administration Press, 1994, pp. 13–54.
23. Morgan, R.O., Virnig, B.A., DeVito, C.A., and Persily, N.A. The Medicare-HMO revolving
door: The healthy go in and the sick go out. The New England Journal of Medicine (1997)
337:169–75.
24. Strumwasser, I., Paranjpe, N.W., Ronis, D.L., et al. The triple option choice: Self-selection bias
in traditional coverage, HMOs, and PPOs. Inquiry (1989) 26:432–41.
57
58
THE FUTURE OF CHILDREN – SUMMER/FALL 1998
25. Leibowitz, A., Buchanan, J.L., and Mann, J. A randomized trial to evaluate the effectiveness of
a Medicaid HMO. Journal of Health Economics (1992) 11:235–57.
26. Research Triangle Institute. Evaluation of the cost effectiveness and utilization of health services by
Medicaid recipients enrolled in health maintenance organizations. Final report to the Bureau of
Medical Assistance, Managed Health Care Section. Columbus, OH: Department of Human
Services, 1992.
27. West, D.L., Stuart, M.E., Duggan, A.K., and DeAngelis, C.D. Evidence for selective health
maintenance organization enrollment among children and adolescents covered by Medicaid.
Archives of Pediatrics and Adolescent Medicine (1996) 150:503–07.
28. Szilagyi, P.G., Roghmann, K.J., Foye, H.R., et al. The effect of independent practice
association plans on use of pediatric ambulatory care in one group practice. Journal of the
American Medical Association (1990) 263:2198–203.
29. Aday, L.A., and Anderson, R. A framework for the study of access to medical care. Health
Services Research (1974) 9:208–20.
30. Holl, J.L., Szilagyi, P.G., Rodewald, L.E., et al. Profile of uninsured children in the United
States. Archives of Pediatrics and Adolescent Medicine (1995) 149:398–406.
31. Newacheck, P.W., Stoddard, J.J., Hughes, D.C., and Pearl, M. Children’s access to health care:
The role of social and economic factors. In Health care for children: What’s right, what’s wrong,
what’s next. R.E.K. Stein, ed. New York: United Hospital Fund, 1997.
32. Commonwealth Fund. Report on the results of a 1994 survey of patients’ experiences with managed
care. New York: Commonwealth Fund, 1995.
33. Davis, K., Collins, K.S., Schoen, C., and Morris, C. Choice matters: Enrollees’ views of their
health plans. Health Affairs (1995) 4:93–106.
34. Miller, R.H., and Luft, H.S. Does managed care lead to better or worse quality of care? Health
Affairs (1997) 16:7–25.
35. Rubin, H.R., Gandek, B., Rogers, W.H., et al. Patients’ ratings of outpatient visits in different
practice settings: Results from the Medical Outcomes Study. Journal of the American Medical
Association (1993) 270:835–40.
36. Safran, D.G., Tarlov, A.R., and Rogers, W.H. Primary care performance in fee-for-service and
prepaid health care systems: Results from the Medical Outcomes Study. Journal of the American
Medical Association (1994) 271:1579–86.
37. Mark, T., and Mueller, C. Access to care in HMOs and traditional insurance plans. Health
Affairs (1996) 15:81–87.
38. Valdez, R.B., Ware, J.E., Manning, W.G., et al. Prepaid group practice effects on the utilization
of medical services and health outcomes for children: Results from a controlled trial.
Pediatrics (February 1989) 83:168–80.
39. Newacheck, P.W. Improving access to health services for adolescents from economically
disadvantaged families. Pediatrics (December 1989) 84:1056–63.
40. Freeman, H., and Kirkman-Liff, B.L. Health care under AHCCCS: An examination of
Arizona’s alternative to Medicaid. Health Services Research (1985) 20:245–66.
41. Kirkman-Liff, B.L. Refusal of care: Evidence from Arizona. Health Affairs (1985) 4:15–24.
42. Hurley, R.E., Freund, D.A., and Gage, B.J. Gatekeeper effects on patterns of physician use.
Journal of Family Practice (1991) 32:167–74.
43. Carey, T.S., Weis, K., and Homer, C. Prepaid versus traditional Medicaid plans: Lack of effect
on pregnancy outcomes and prenatal care. Health Services Research Journal (1991) 26:165–81.
44. Goldfarb, N.I., Hillman, A.I., Eisenberg, J.M., et al. Impact of a mandatory Medicaid case
management program on prenatal care and birth outcomes. Medical Care (1991) 29:64–71.
45. Krieger, J.W., Connell, F.A., and LoGerfo, J.P. Medicaid prenatal care: A comparison of use
and outcomes in fee-for-service and managed care. American Journal of Public Health (1992)
82:185–90.
46. Sisk, J.E., Gorman, S.A., Reisinger, A.L., et al. Evaluation of Medicaid managed care:
Satisfaction, access and use. Journal of the American Medical Association (1996) 276:50–55.
47. Freund, D.A., Rossiter, L.F., Fox, P.D., et al. Evaluation of the Medicaid Competition
Demonstrations. Health Care Financing Review (Winter 1989) 11:81–97.
48. Temkin-Greener, H., and Winchell, M. Medicaid beneficiaries under managed care: Provider
choice and satisfaction. Health Services Research (1991) 26:509–29.
Managed Care for Children: Effect on Access to Care and Utilization of Health Services
49. Reid, R.J., Hurtado, M.P., and Starfield, B. Managed care, primary care, and quality for
children. Current Opinion in Pediatrics (1996) 8:164–70.
50. Rouilidis, Z.C., and Schulman, K.A. Physician communication in managed care organizations:
Opinions of primary care physicians. Journal of Family Practice (1994) 39:446–51.
51. Leatherman, S., and McCarthy, D. Opportunities and challenges for promoting children’s
health in managed care organizations. In Health care for children: What’s right, what’s wrong,
what’s next. R.E.K. Stein, ed. New York: United Hospital Fund, 1997, pp. 199–236.
52. Hahn, B., and Lefkowitz, D. Annual expenses and sources of payment for health care services. U.S.
Department of Health and Human Services publication 93-0007. Rockville, MD: Agency for
Health Care Policy and Research, 1992.
53. Miller, R.H., and Luft, H.S. Managed care plan performance since 1980: A literature analysis.
Journal of the American Medical Association (1994) 271:1512–19.
54. Hodge, M.J., Dougherty, G.E., and Pless, I.B. Pediatric mortality and hospital use in Canada
and the United States, 1971 through 1987. American Journal of Public Health (1995)
85:1276–79.
55. DeClercq, E., and Simmes, D. The politics of “drive-through deliveries”: Putting early postpartum discharge on the legislative agenda. The Milbank Quarterly (1997) 75:175–202.
56. Braveman, P., Egerter, S., Pearl, M., et al. Early discharge of newborns and mothers: A critical
review of the literature. Pediatrics (October 1995) 96:716–26.
57. Gazmararian, J.A., and Koplan, J.P. Length-of-stay after delivery: Managed care versus fee-forservice. Health Affairs (1996) 15:74–80.
58. Hurley, R.E., Freund, D.A., and Paul, J.E. Managed care in Medicaid: Lessons for policy and
program design. Ann Arbor, MI: Health Administration Press, 1993.
59. Nadel, V. Emergency departments: Unevenly affected by growth and change in patient use. Washington,
DC: U.S. General Accounting Office, Human Resources Division, 1993.
60. Bonham, G.S., and Barber, G.M. Use of health care before and during Citicare. Medical Care
(1987) 25:111–19.
61. Davidson, S.M., Manheim, L.M., Werner, S.M., et al. Prepayment with office-based physicians
in publicly funded programs: Results from the Children’s Medicaid Program. Pediatrics (April
1992) 89:761–67.
62. Hurley, R.E., Freund, D.A., and Taylor, D.E. Gatekeeping the emergency department: Impact
of a Medicaid primary care case management program. Health Care Management Review (1989)
14:63–71.
63. Gadomski, A.M., Perkis, V., Horton, L., et al. Diverting managed care Medicaid patients from
pediatric emergency department use. Pediatrics (February 1995) 95:170–78.
64. Shaw, K.N., Selbst, S.M., and Gill, F.M. Indigent children who are denied care in the
emergency department. Annals of Emergency Medicine (1990) 19:59–62.
65. Szilagyi, P.G., Zwanziger, J., Rodewald, L.E., et al. Evaluation of Child Health Plus in New York
State. Report submitted to the New York State Department of Health. Rochester, NY:
Rochester Child Health Studies Group, 1996.
66. Cherkin, D.C., Grothaus, L., and Wagner, E.H. The effect of office visit copayments on
utilization in health maintenance organizations. Medical Care (1989) 27:1036–45.
67. Hohlen, M.M., Manheim, L.M., Fleming, G.V., et al. Access to office-based physicians under
capitation reimbursement and Medicaid case management. Medical Care (1990) 28:59–68.
68. Cartland, J., and Yudkowsky, B. Barriers to pediatric referral in managed care systems.
Pediatrics (February 1992) 89:183–88.
59