Minimizing Prematurity in a High-Risk Obstetrical Population
Summary
The health status of a high-risk newborn population
is determined by the days per thousand and its average
length of stay (ALOS) in the NICU. In 2001, the
contracted HMO’s neonatal intensive care unit (NICU)
days per thousand averaged 87 days per thousand admits.
Its average length of stay for newborns in the
NICU was 14 days.
Objective
To reduce the HMO’s days per thousand and the ALOS
of newborns in the NICU by ten percent (10%) annually
through its contracted relationship with Alpha
Maxx Healthcare Inc. (AMAXX) and its OB population
health management services.
Study Design
The organization performs annually an on-going longitudinal
populationbased
review of its birth
outcomes using the latest
data available (2002
through 2004). The year
2001 is the base for comparison.
The study focuses
on days per thousand
and the average length
of stay in the NICU. The
source used for the study
is the HMO’s Board Reports
as validation of its
success.
AMAXX’s success is also measured using the denominator
of 5,545 pregnancies, which represent the total
number of the HMO’s pregnancies throughout the region
(2,178 of those were assigned to AMAXX). The
numerator represents the total number of admits to the
NICU. However, the data is not adjusted for ‘rule-outs’
(i.e. admits to rule out sepsis, beta strep, etc.). It is
diffi cult to get an accurate measure for lowering the incidence
of prematurity using the admission data as the
only criteria. So, if the days per thousand and the ALOS
are decreasing in spite of the ‘rule-outs,’ this becomes a
true indicator of a healthier newborn population.
AMAXX is held accountable for all premature deliveries
in Shelby County, whether the member was referred
or not. Most of the HMO’s ‘critically’ high risk pregnant
patients are referred to Shelby County late in their
pregnancies to be followed by a high-risk OB provider
(perinatalogist) and deliver in a hospital with a level
three NICU. The tertiary hospital and high-risk physicians
are all located in Shelby County.
Overview of the Patient Population
Medicaid recipients have many risk factors that signifi -
cantly affect the birth outcomes of low-income women,
including medical comorbidities,
preterm birth,
substance abuse issues,
smoking, domestic violence,
low literacy, and
homelessness. According
to the latest March
of Dimes data for Shelby
County, Tennessee, the
percentage of low birth
weight African American
infants per 1,000 live
births is 14.3 compared
to 7.0 for whites. The
percentage of preterm births for African Americans is
18.9 compared to 11.4 for whites.
The Medicaid population is defi ned by the presence of
chronic conditions, disability, and poor health. There
are specifi c risk factors that are common among this
population and act as a barrier to a healthy outcome.
Most of these risk factors are exacerbated during a
woman’s pregnancy. Interventions to reduce risk factors
include early identifi cation and stratifi cation of potentially
high-risk members using a variety of disciplines.
Interventions include nursing and social work coordination
of care for all members—low-risk and high-risk.
Members are given support during their pregnancy
through our telephonic care coordination contacts and
outreach, incentive programs, patient education, home
visits, hospital rounds, transportation, and connection
with other community-based organizations.
Members receive individualized care plans based upon
risk factors. The initial and subsequent routine assessments
identify medical, mental health, economic, and
psychosocial issues that might impede the member
from experiencing the optimal outcome of a healthy
newborn. AMAXX’s software system supports its care
management program. The system is a repository for
all information necessary to manage the obstetrical system
of care for our members.
Background
Prior to the implementation of AMAXX’s population
health management program, the HMO had an outreach
program that involved indirect contact only through
mail and member services. Early in the implementation
process, AMAXX began working with the HMO’s
provider network. The goal of the organization’s outreach
efforts was to encourage each provider to follow
evidence-based clinical guidelines to ensure optimal
outcomes. AMAXX also assisted providers to better
serve their HMO patients by instituting the following
services:
- Assist with transfer of patients between levels
of care.
- Assist with non-compliant patients.
- Issue loaner pagers to transient members to
enhance communication between the physician
and his patient.
- Provide information on member’s emergent
care to her provider as a result of AMAXX
hospital rounds.
- Quarterly meetings with university-affi liated
providers (major high-risk providers in the
community) to review trends, complaints, opportunities
for improvement, and on-going
concerns.
In stark contrast, before 2002, the HMO’s contracted
providers were not required to give a notifi cation of pregnancy
or get an authorization for payment of services.
It must be noted that the Plan is co-owned (50/50) by
a university-affi liated medical group and the region’s
safety-net and the university’s teaching hospital.
Identifi cation of High-Risk Members
As stated in the preceding section, providers were not
required to have an authorization to be paid for providing
OB services to the HMO’s members. In order to
have a successful OB population health management
program, the providers had to be provided incentives to
change their procedures. The HMO instituted a timely
payment program. A global payment was paid to any
participating provider who agreed to the following:
- Follow the HMO’s evidence-based clinical
guidelines in caring for its members.
- Submit a Pregnancy Risk Notifi cation (PRN)
and report patient work-ups and lab results
within 14 days of the initial encounter.
- Submit “zero” balance claims for encounter
data to elicit treatment information from the
provider for each OB encounter.
- Complete proper paperwork for high-risk referrals
to perinatalogists or other specialists.
- Will not “drop” patients after two missed appointments,
as is the habit with many private
physicians. They must contact AMAXX to
initiate the transfer to another provider.
AMAXX also assisted the providers by pre-processing
claims to minimize the number of rejections on claims
(stop the “ping-pong” factor). If there were issues associated
with the claims, AMAXX’s staff ‘tracked down’
the problem and checked the remittance advice to make
sure the provider was paid properly. This interaction
with the physician and his staff fostered a good working
relationship with AMAXX.
AMAXX has initiated an incentive program for provider’s
offi ce staff. Appreciation luncheons are offered
semi-annually for participating providers’ offi ce staff.
Quarterly luncheons are also offered to providers’ offi
ce staff that have consistently provided updated and
correct information on members (demographics and
high-risk clinical data).
Results
The study was conducted using the HMO’s Board Reports
and independent review of hospital data. The
reduction of the HMO’s NICU days per thousand was
from 87 days per thousand admits in 2002 down to 50
days per thousand admits in 2004, a 45% reduction in
days per thousand. Correspondingly, its NICU length
of stay was 14 days in 2002 and in 2004 the average
length of stay was 9.8 days, a 31% reduction in the average
length of stay for neonates in the NICU. These
fi gures were a result of our study of the HMO’s NICU
admits, and the length of stay.
Our fi ndings were borne out by the HMO’s Board Reports
that illustrated the fact that in late 2001, when
AMAXX’s contract was implemented, over 20% of its
population was premature, and in late 2004 the prematurity
rate had dropped to six percent (6%) in Shelby
County. This 6% fi gure is almost two-thirds lower
than the county rate of 18% (according to the March
of Dimes latest—2003—data for African Americans).
In 2001 the HMO spent $20 million for OB care for
76,000 members. By 2004 they spent $6 million for
OB care for a population of 176,000. Concomitantly, a
by-product of lowering prematurity is lowering admits
to the Pediatric Intensive Care Unit. Each year that a
healthy versus a premature baby is delivered, the cost
of care is driven down.
Conclusion
AMAXX far exceeded its objective of reducing the
HMO’s days per thousand and ALOS by 10% annually.
This was accomplished primarily through AMAXX’s
relationship with the physicians, including the initiation
of the global payment. The interaction with the
physician and his offi ce staff fostered a good working
relationship and ensured more coordinated care for
AMAXX’s members. Another factor was the collaborative
relationship and quarterly meetings with the university-
affi liated OB providers, who handle the majority
of the HMO’s high risk pregnant members in Shelby
County.
A program that interfaces with the entire OB delivery
system and addresses issues of providers, hospitals,
clinics, provides face-to-face and telephonic care for all
its members, will signifi cantly affect birth outcomes for
this population. It is fair to say that you must not only
modify the behaviors of the members, but also initiate
change in the health care delivery system. AMAXX
has focused on system-wide changes to ensure care delivery
improvements for its members.
Attachments
Results of the longitudinal study:
- Neonatal Intensive Care Unit Trended Results,
2001-2004 (Days per Thousand and Average
Length of Stay)
- Trended Membership Results, 2000-2004 (Illustrates
lowering of days per thousand as
membership rises)
- Trended Cost Statistics—NICU and PICU
Costs, 2002-2004