Cost Effectiveness of Case Management Programs for the Elderly
by Michael J. Long, M.A., Ph.D.
| Geriatric Times |
 |
May/June 2001 |
|
Vol. II |
|
Issue 3 |
Case management has often been adopted as a means to reduce or constrain health
care service use and costs for a given enrolled population (Capitman, 1988;
Freund et al., 1984; Yordi, 1988). It is important, however, to recognize that
there are different models of case management, each with differing dynamics and
designed to achieve different goals.
Case Management Models
At the risk of oversimplifying the categorization of the numerous models,
the underlying dynamics suggest that they can be generically categorized into
one of three types: interrogative, patient advocacy and consolidated. The
interrogative model emphasizes intense oversight at the interface of the
clinical and fiduciary perspectives. In this model, although the
appropriateness of patient care may predominate in determining a treatment
regimen, the cost of care is recognized as a legitimate argument in the
decision-making process. Hurley and Fennell's (1990) medical case management
model, Zawadski and Eng's (1988) prior authorization model and Capitman's
(1988) gatekeeper model would all fit under this rubric.
The patient advocacy model emphasizes the coordination of services on the
continuum of care from the client perspective. This model suggests that all of
the patient's circumstances are recognized as legitimate arguments in the
decision-making process. That is to say, the treatment regimen is determined
not only by the medical needs of the patient, but also by the financial,
psychological and social circumstances of the patient.
It is important to note that the classification of case management models in
this manner is not intended as support for the growing tendency in the managed
care arena to view cost and quality as philosophically opposed concepts. On the
contrary, Donabedian (1980) suggested that costs play a vital role in the
quality of care. He defined quality as a measure of the extent to which the
care provided maximizes the benefits to health while minimizing the risk, as
follows: Quality=Benefits-(Risk+Cost). By this definition, both patient
advocacy and interrogative case management models can be further characterized
as addressing the quality of care, because if cost and risk remain constant and
benefits increase, quality is improved. Similarly, if benefits and risk remain
constant and costs decrease, quality is improved. Thus, the interrogative model
of case management addresses quality through the cost variable, and the patient
advocacy model addresses quality through the benefit variable.
A third model, identified by Zawadski and Eng (1988) and exemplified by the
On Lok Senior Health Services in San Francisco, is not seen as a separate and
distinct model but rather a combination of both the interrogative and patient
advocacy models. In the consolidated model, the decision-making process might
be thought of as consensus building among the providers that comprise the team.
It is reasonable to suggest that each member of the multidisciplinary team is
influenced by their own internal implicit controls and the explicit controls
imposed by the managed care organization. The resulting treatment regimen might
then be considered the result of the prevailing consensus of the team. That is
to say, the group's decision could reflect a position of any point along the
continuum of a strong interrogative approach to a strong patient advocacy
approach.
Measuring Cost Effectiveness
We conducted a post hoc evaluation of the benefits and costs associated with
a case management program for an elderly, functionally impaired population
(Long and Marshall, 2000). Clients of a large managed care organization who
were 75 years or older and functionally disabled (n=317) were randomly assigned
to a regular-care group or a case-managed group.
In this study, the case managers were health care professionals with prior
geriatric case management experience. They became integral members of a patient
care team, which comprised the client's personal physician and the physician
adviser. This team developed the initial care plan for each client, with the
case managers responsible for making periodic home visits, reporting back to
the care team and participating in the revision of the care plan as necessary.
While case managers made at least one home visit every six months, weekly
visits to some clients were not uncommon. In addition, case managers scheduled
medical appointments; accompanied patients on these appointments; and arranged
for non-medical services such as respite care, meals-on-wheels, nursing home
placement, Medicaid eligibility, and transport to and from the physician.
The goal of this intervention was to eliminate fragmented care,
inappropriate utilization, unnecessary costs and client confusion frequently
associated with chronic care. It was also expected that less costly outpatient
visits would be substituted for hospital and emergency department (ED) care.
These goals would suggest a more interrogative model, but the model adopted
was, in fact, heavily biased toward a patient advocacy model. It could be
argued that the model of case management adopted was inconsistent with the
goals. This could have been due to a lack of specification in the protocols
established by the team, a failure to reconcile protocols with goals, or a lack
of fit between the prevailing team consensus and the goals of the program.
Given this conflict, it was conceivable that the benefits of the program might
manifest themselves in financial or quality terms.
In this study, we identified two types of costs: program costs and service
costs. Program costs were then identified as the salaries and fringe benefits
of the two case managers and the research assistant. Service cost data were
taken from the MCO's 1992 Medicare Financial Performance Report. Hospital costs
were calculated using diagnosis-related groups for the appropriate fiscal year,
and outpatient and ED visit costs were based on an average cost per visit
calculated by service, medical specialty and provider type.
Benefits were calculated in two ways. We first considered the service use
and related costs with the understanding that, all things being equal, reduced
service use and cost represented a benefit. We also examined the number of days
of coverage (exposure) with the understanding that a greater number of days of
exposure represented a greater benefit. The potential number of days of
exposure during the two-year study for each group was 730 days. In the event of
death during the study, the number of days of exposure was reduced by the
number of days from the date of death to the end of the study.
The actual cost of services used by the two groups during the study period
was remarkably similar ($2,414,043 for the case-managed group and $2,473,221
for the regular-care group). However, when the program costs were allocated to
the case-managed group, the average cost per person totalled $18,210 for the
case-managed group and $13,973 for the regular-care group. This difference was
statistically significant (p=0.05).
The average number of days of exposure was 656 days for the case-managed
group and 550 for the regular-care group. This difference was statistically
significant (p=0.01).
The cost per additional day of life was calculated by dividing the
difference in average cost per person ($18,210-$13,973=$4,237) by the
difference in the average number of days of exposure (656-550=106). This
resulted in a cost of $39.97, which can be interpreted as the cost of each
additional day of life.
Discussion
Several caveats are in order. This study examined one particular case
management intervention, in one particular setting with a unique population
and, as such, the results cannot be generalized to any other population. It is
also important to note that, in this study, an additional day of life is
considered to be positive and, therefore, a positive outcome. It is entirely
possible, particularly with this population, that an additional day of life may
not be perceived by the individuals involved as positive. Since death can only
be determined retrospectively, any attempt to measure the quality of an
additional day of life using any of the available validated instruments when
death is eminent would not only require an unacceptable intrusion, it would
probably be impossible. Given the data available and the methodology adopted
for this study, an independent measure of the quality of life inherent in
additional days of life (on average) in this study cannot be determined. Under
the circumstance, it seems reasonable to accept the unobtrusive measure of an
additional day of life as a positive outcome of a case management
intervention.
The case management model adopted in this program was clearly a patient
advocacy model. The case managers, acting in what they perceived to be the best
interest of the clients, encouraged and facilitated the use of the system, so
it is not surprising that the average cost per person was greater for the
case-managed group. A simple evaluation of the difference in the cost per
person would suggest, erroneously, that the program was a failure, but when
costs are considered relative to the benefits received, a very different
picture emerges. The actual death rate in both groups was very small and very
similar. On the other hand, the cost for an additional day of life was only
approximately $40, which, it could be argued, is a small price to pay.
Consistent with the dynamic inherent in the Donabedian (1980) definition of
quality, although costs were increased, quality was improved. From this
perspective, the program might very well be considered a success.
Implications
From a management practitioner perspective, these results point to the need
to carefully evaluate the various models of case management relative to the
outcome desired. In other words, the underlying dynamic of the case management
model adopted must be consistent with the reasons for implementing it. They
also point to the need for management practitioners to evaluate both the
benefits and the costs of a program, regardless of the impetus behind its
implementation, before condemning it as a failure or praising it as a
success.
Regardless of the rationale for implementing a case management program, it
is the dynamic inherent in the model adopted that will dictate whether the
ultimate effect manifests itself in costs savings, benefit improvement, neither
or both. Unfortunately, organizations very often engage in the case management
process assuming a priori that cost savings will be realized by simply imposing
any model. This is obviously not the case. It is critical, therefore, that
before embarking on a case management endeavor, the individual or team
responsible for determining the treatment regimen fully understands the goal.
Guidelines, protocol or treatment parameters that reflect this goal must be
established and agreed upon by all concerned.
Dr. Long is professor in the department of public health
sciences at Wichita State University. He has been published widely in national
and international journals as a health economist and has written two
textbooks.
References
Capitman JA (1988), Case management for long-term and acute medical care.
Health Care Financ Rev Dec(Spec No):53-55.
Donabedian A (1980), The definition of quality and approaches to its
assessment. Ann Arbor, Mich.: Health Administration Press.
Freund DA, Ehrenhaft PM, Hackbarth M (1984), Medicaid Reform: Four Studies
of Case Management. Washington, D.C.: American Enterprise Institute for Public
Policy Research.
Hurley RE, Fennell ML (1990), Managed-care systems as governance structures:
a transaction-cost interpretation. In: Innovations in Health Care Delivery:
Insights for Organization Theory, Mick SS and associates, eds. San Francisco:
Jossey-Bass Publishers, pp241-268.
Long MJ, Marshall BS (2000), What price an additional day of life? A
cost-effectiveness study of case management. Am J Manag Care 6(8):881-886.
Yordi CL (1988), Case management in the social health maintenance
organization demonstrations. Health Care Financ Rev Dec(Spec No):83-88.
Zawadski RT, Eng C (1988), Case management in capitated long-term care.
Health Care Financ Rev Dec(Spec No):75-81.