Title Page
 Table of Contents

Title: Provider profiling : implementation issues within Florida MediPass
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00091296/00001
 Material Information
Title: Provider profiling : implementation issues within Florida MediPass
Physical Description: Book
Language: English
Creator: Schaffer, Virginia A.
Publisher: Florida Center for Medicaid & the Uninsured, College of Public Health and Health Professions, University of Florida
Place of Publication: Gainesville, Fla.
 Record Information
Bibliographic ID: UF00091296
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Table of Contents
    Title Page
        Title Page
    Table of Contents
        Table of Contents
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
Full Text

Provider Profiling: Implementation Issues within Florida

Virginia A. Schaffer, MA
Allyson Hall, PhD

Florida Center for Medicaid & the Uninsured
College of Public Health and Health Professions
University of Florida

Sponsored by
The Agency for Health Care Administration

Florida Center for Medicaid and the Uninsured
Shaping a H eal care Policy

June, 2004

Table of Contents

E executive Sum m ary .......... ...... ...... ........... .. .. ..................... 1
I. Introduction and B background .......................... ...... ..................... .............. 2
II. Study M ethodology................... .. ....... .. ........ .......... .. ........ 3
III. Key Informant Views on Provider Profiling ................... .................................... 5
IV. Findings from the State Interviews.................................... .... .......... 5
A A rk an sa s......... ............................................ .......................................... 5
B Indiana ......... ... ............... ................................................................ 8
C. Louisiana .............. ... ..... ... ... ...... ...... ....... .. ................. 9
D. North Carolina ................................................... 10
E M assachu setts ..... ............................................. ....... ............... 12
F. A additional State Inform ation ........................................ .......................... 13
V. Findings from Physician Interviews .............. .... ......................... ............ 14
VI. Conclusion .......... .... ................................. .............. 15
VII. Implications for AHCA .. ..... ........................................... ............... 17
V III.R eferen ces .................................................... ... ......... ...... 2 0

Executive Summary

Provider profiling use epidemiological methods to compare physician practice
patterns across various dimensions of care. These dimensions of care may include length
of stay, procedures ordered or cost per diagnosis related group (DRG). The premise
behind provider profiling is that physicians with appropriate information will change the
way they practice in order to conform to acceptable guidelines thereby improving the
quality of care. Increasingly, managed care organizations and other delivery entities are
using provider profiling as a means to understand and control costs.

This report provides the Agency for Health Care Administration (AHCA) with
important conclusions made from reviewing literature regarding provider profiling
programs. Findings from interviews with key informants, state Medicaid agency
officials, and physicians are also provided to highlight the important conclusions and
implications of implementing provider profiles.

Specifically, areas of concern for the current AdvancedMed profiling system used
by AHCA would be the time lags associated with data used to compile provider profiles,
the current configuration of reports and its usefulness for providers, and the level of input
Medicaid physicians have on the design and implementation of the provider profiling
program. As AHCA considers the implementation of provider profiles within its
Medicaid program, the following guidelines are critical to assuring some level of success:

The initial emphasis should be on quality improvement
Performance measures should be clinically important
Accountability for performance should be limited to patients and services for
which physician is directly responsible
Adequate risk and case mix adjustment must be incorporated in the program
design with accompanying explanation of how data was adjusted
Physicians should be involved early in the design and implementation of provider
profiling strategies in order to generate 'buy-in'
Performance reports should be clear and easy to understand and coupled with
provider feedback and educational opportunities
Provider profiling programs should make every attempt to use accurate data

Page 1

I. Introduction and Background

Provider profiling is a term used to describe the process of compiling and
disseminating performance feedback information to physicians and other clinicians.
Essentially, provider profiling use epidemiological methods to compare physician
practice patterns across various dimensions of care including length of stay, drug
utilization, procedures ordered, or cost per diagnosis related group (DRG). Provider
profiling reports are used in a variety of organizations including managed care
companies, hospitals, and in large group practices. Increasingly, managed care
organizations and other delivery entities are using provider profiling as a means to
understand and control costs and improve the quality of care and as a tool to ensure that
they are 'getting value for money' (Goldfield et al, 2003).

Several years ago, the Agency for Health Care Administration (AHCA) via its
fiscal intermediary Affiliated Computer Services, Inc. (ACS) contracted with Advanced
Med to provide data management activities for Medicaid claims data. At that time,
AHCA's plan was to use AdvancedMed to generate and distribute risk-adjusted
performance reports to their primary care physicians who participate in the MediPass
program. Without understanding the complexity of patients in different groups,
comparisons made between providers (or any other variable for that matter) are futile
(Goldfield, 1996). AdvancedMed would produce formatted reports showing aggregated
actual measures of use activity, expected measures of use activity, a morbidity index, and
actual and expected costs. For any one time, claims data for a one-year period (four
quarters) would be available for analysis. New data uploads were to be performed each
quarter, at which time the oldest quarter's data would drop out of the data set and a new
quarter's data would be added. Accordingly, data should be accurate (i.e. low rate of
random errors such as the reversing of digits), unbiased (e.g., providers have not
systematically adjusted their CPT coding for payment purposes), and have a low rate of
missing values (Brand et al, 1995).

Page 2

The Agency for Health Care Administration (AHCA) contracted with the Florida
Center for Medicaid and the Uninsured (FCMU) to examine the issues around the
introduction of a provider profiling program within the Florida Medicaid program.
During the first phase of this study, a literature review revealed that as currently
configured, the AdvancedMed program may not be very effective in changing provider
behavior, improving quality of care and reducing health care disparities. One reason
has to do with the 8 to 9 month time lag between dates of service and the date when the
data become available from AdvancedMed. Physicians who modify their behavior will
not begin to see these changes reflected in data until several months later. In addition,
because the data is aggregated over 1 year, it will take even longer before doctors can
start to see the full effect of changes they may have made. One quarter's data can be
isolated in the database. This example highlights why physician profiling has been
heavily criticized because of 1) perceived adverse effects of pressures to restrict services
on quality of care, 2) reporting is thought to be based on poor quality data and 3)
inadequate risk adjustment (American College of Physicians, 2001).

It is our understanding that ACS and AHCA no longer contracts with
AdvancedMed. However, AHCA remains interested in learning whether programs such
as AdvancedMed's could be of value to the state. The study conducted by FCMU
reviewed the existing literature base, researched current provider profiling programs by
interviewing state Medicaid agency officials, and gathered pertinent information from
other sources such as key informants, organizational websites, and journal articles. To
the extent possible, clinical outcomes measures should be related to the processes of care
that can be easily modified by the physician or practice (Massachusetts Medical Society,
1999). However, it is important to recognize that provider performance measures may
never be able to accurately portray the quality of medical care delivered to patients
(Brand et al, 1995).

II. Study Methodology

1 See Dimoulas E and Hall A Literature Review on Provider Protiliing submitted to the Agency in June,

Page 3

The overall purpose of the provider profiling study was to collect and document
pertinent data from other Medicaid agencies and health plans regarding systems of
compiling and reporting provider performance. In addition, this project aimed to collect
MediPass physician views on profiling.

Initial activities focused on identifying and contacting key informants particularly
within state agencies. Preliminary inquiries revealed that only a handful of states had
implemented provider profiling programs within their primary care case management
programs. Further research revealed that states may encourage provider profiling
programs as part of Medicaid Managed Care or Primary Care Case Management
contracts, however interviews were not conducted with individual Managed Care
Organizations or other private sector groups. In addition, information searches had
shown that organizations such as the Center for Health Care Strategies/Mathematica,
Center for Studying Health System Change, Positive Health care, the Special Committee
on Physician Profiling, and American Public Human Service Association had conducted
research and were disseminating information such as issue briefs and presentations on the
area of provider profiling.

Interviews were completed with Medicaid agency officials in Arkansas, Indiana,
Louisiana, North Carolina, and Massachusetts. Information or brief interviews were
conducted with Georgia, Maryland and Texas. Please refer to Appendix 1 to review the
Provider Profiling Matrix that highlights details about each state's program. Medicaid
state contacts were asked to comment on their state's involvement with provider profiling
and to discuss problems, limitations, and successes associated with their programs. In
addition questions related to implementation and evaluation of their provider profiling
programs were asked.

Providers and their office staff were also interviewed about provider profiling. In
some instances, questions were included as part of a series of questions on their overall
experiences with MediPass. In other instances, providers were explicitly asked to
comment on the usefulness of the reports generated by AdvancedMed. Questions for

Page 4

both sets of physicians were designed to get their opinions about potential effects of
profiling and what measures they would find useful in improving the quality of their
practice. We completed 4 in depth interviews with providers about profiling. Fourteen
other physicians have been interviewed about MediPass in general.

III. Key Informant Views on Provider Profiling

Many of the key informants stressed that any kind of provider profiling activity
must be grounded in quality improvement activities, rather than as a way to play
'punitive games' with physicians that can threaten their practices and reputations. For
example, one agency said that their profiling activity stems from a goal to improve
quality of care, which can lead to better satisfaction and improved compliance among
their HIV positive patients. One informant noted that 'provider profiling is not intended
to be used to address issues of physician competency, including medical knowledge and
skills. The ultimate goal is to improve clinical outcomes'. Therefore, as one key
informant points out, provider profiling activities are challenged by ensuring data are
available and meet appropriate quality and risk adjustment standards.

Gaining provider acceptance and believability is an important aspect of any
profiling activity. Understandably many physicians are resistant to any kind of
monitoring. Physicians want to have their autonomy and there is a general sense that
profiling can interfere with this. Key informants suggested that one way to overcome this
kind of resistance is to ground performance measures in accepted practice guidelines.

IV. Findings from the State Interviews

The following section provides pertinent details from interviews conducted with
officials at five state Medicaid agencies. Interviews were conducted with officials in
Arkansas, Indiana, Louisiana, North Carolina, and Massachusetts while information was
collected on Maryland and Texas. Please refer to Appendix 1 to review the Provider
Profiling Matrix that highlights details about each state's program.
A. Arkansas

Page 5

A number of stakeholders identified Arkansas as a leader in using provider
profiling within Medicaid primary care case management programs. Arkansas Medicaid
is involved in two sets of profiling activities PCP Profiles and The Angels Program:
High Risk OB/GYN.

PCP Profiles: The Arkansas Foundation for Medical Care (AFMC) conducts
provider profiling activities for the Arkansas Medicaid program. The profiling initiative
began in 1995 and for the last six years has provided profile reports to PCPs in the
Medicaid program. Arkansas Medicaid Primary care physicians receive quarterly reports
that include graphs and tables designed to help providers compare their costs and
utilization to the statewide average. Appendix 2 contains a copy of the PCP profile that
is sent to providers. The first page of the report contains a letter from the medical
director. Center pages include the performance report and provides cost and utilization
information for the quarter and year to date. Physicians can compare their performance
to state totals using the 'index' (3rd and 5th columns). The index is the ratio for the
individual physician compared to the corresponding state value. For example, for the
quarter reported in the profile, this physician's total cost PMPM is about 12 percent
higher than the state average (see last line of the table). The report also includes
quarterly pharmacy cost and utilization patters for the physician's panel and the state
average. The last page provides some general information on pharmacy claims for the

Until recently the PCP profile was not risk-adjusted. Over time however,
physicians began to advocate for some kind of risk adjustment methodology. Beginning
in the fall of 2003, the Johns Hopkins Adjusted Clinical Group Case-Mix System (ACG)
is used to calculate a risk-adjusted efficiency ratio. This ratio is the mean cost per patient
per quarter for that PCP divided by the PCPs expected cost per quarter. The expected
cost per quarter is calculated based on the severity of the patient's caseload using the
Hopkins methodology. When asked why AFMC decided to use the Hopkins system, the
Arkansas key informant mentioned that 'it was the methodology that most people in the

Page 6

office were familiar with'.2 They had hoped to introduce the risk-adjustment sooner, but
that there was a significant learning curve associated with its introduction.

AFMC also send a yearly profile to PCPs that tracks EPSDT/preventive screening
(see Appendix 3). This profile shows the number eligible beneficiaries who are screened
and not screened during the previous year. The tables also show the potential income lost
to the provider for not performing all of the required screens. AFMC plans to do profile
for diabetes in the near future.

The format of the two reports was vetted with the board of directors of the
Arkansas Medical Care Foundation, all of whom are physicians. The board members
provided comments on the structure and design of several iterations the reports. In
addition, Medicaid provider representatives solicited feedback on report format from the
physicians. To date, the state has not conducted a formal evaluation of the program.

The Angels Program: High Risk OB/GYN: Arkansas has a targeted quality
initiative aimed at improving outcomes for mothers at high risk for having an adverse
pregnancy. The program began 2 years ago with the development of clinical guidelines
for high-risk pregnancies. The original intent was to use existing guidelines developed
by the American College of Obstetrics and Gynecology. However, early in the process
they realized that the guidelines would need to be customized to meet the needs of
Arkansas Medicaid population. For example, the rural nature of the state may limit
access to emergency rooms and that special focus needed on high-risk mothers with
asthma. As a result, Arkansas Medicaid worked to adapt the clinical guidelines to meet
the State's particular needs.

An integral component of the process was getting clinicians to agree to the
guidelines. This was an iterative process. Guideline development was the topic of
weekly conference calls (on average about 35 providers attended the calls). Guidelines

2 Another possible reason for the use of the ACG system is that it offered free of charge (for capitation and
rate-setting purposes only) to state Medicaid programs. In addition to Arkansas, the ACG system is used
in Kansas, and Oklahoma.

Page 7

were also posted on the Medicaid website. In addition nurses and other medical workers
did outreach to providers throughout the state. Throughout the process providers were
encouraged to comment and suggest revisions.

The agreed upon clinical practice guidelines were used to create claims groupings
(using ICD and CPT codes) which were incorporated into the physician profiler (done by
EDS). The profiles compare physicians to their peer groups with respect to cost and
utilization across a number of dimensions including drug utilization, inpatient
hospitalizations, outpatient referrals, and emergency room utilization.

Establishing the peer groups for the clinicians engaged in high-risk pregnancies
was somewhat challenging. In Arkansas potentially family practice, general practice,
OB/GYN, internal medicine (including sub-specialties) and pediatric specialties are all
considered PCPs. Arkansas Medicaid sent letters to all of the providers asking them
define what specialty they actually practice in and the extent to which they are involved
in pregnancy care. Based on their responses, physicians were grouped into various
categories or peer groups.

To date, providers have received the OB/GYN profile favorably. Contacts at
Arkansas Medicaid attribute this to the fact that the physicians were involved in the
guideline development and that the profiles are based on these guidelines. Furthermore,
Arkansas Medicaid worked towards securing 'buy-in' from the state medical society.
Next steps involve visits to outlier physicians needing corrective training and action.
These visits have been shown to be very time consuming and labor intensive and can take
up to 4 hours per visit.

B. Indiana

Although the Indiana Medicaid program, known as the Indiana Health Coverage
Programs (IHCP), does not create or distribute reports through an established provider
profiling program, IHCP does monitor providers. In conjunction with their contractor,
Health Care Excel, IHCP examines provider health care delivery and utilization patterns

Page 8

for purposes of detecting issues such as the over-utilization of services. The three
business functions within Health Care Excel are Medical Policy (MP), Prior
Authorization (PA), and Surveillance and Utilization Review (SUR). The objectives
listed below are examples of SUR responsibilities in regards to the IHCP contract.
Monitor utilization to identify potential misutilization of IHCP.
Identify utilization trends and patterns and develop audits.
Develop statistical profiles of health care delivery and utilization
patterns by providers and members in various categories of services.
Identify concerns in the level of care or quality of covered services that
are funded by the IHCP and make appropriate referrals.

Since approximately 1998, Health Care Excel has worked with IHCP to develop
statistical profiles of health delivery and utilization patterns by providers to detect over-
utilization. Through current approaches, SUR attempts to take a proactive approach in
monitoring and detecting areas of concern. Health Care Excel initiates appropriate action
with IHCP when findings such as over-utilization are identified. For providers who are
identified with issues such as over-utilization, it was not reported whether an incentive
component was part of the action plan in addressing these issues with providers. Health
Care Excel and IHCP work together to review the statistical profiles to determine
appropriate evaluation parameters and guidelines for the SUR function.

C. Louisiana

Since 2002, as a response to a legislative mandate, the Louisiana Department of
Health and Hospitals' Medicaid Pharmacy Program has conducted peer-based profiling
for providers. There were many components to the legislative mandate, but overall,
legislators were establishing programs such as prior authorization and provider profiling
with the hopes of monitoring costs and quality. Within the Louisiana pharmacy program,
providers or "prescribers" are sent information that compares them to a peer-based
comparison group. Selecting providers who have similar specialties and number of
patients is generally how these peer-based comparison groups are created. In addition,
attempts have been made to create comparison groups geographically. In 1996, a
separate disease management initiative including provider profiling was established and
perhaps in the future, these profiles will be linked. However, this interview concentrated

Page 9

only on the pharmacy provider profiles. An interview was not conducted with any
official working with the disease management provider profiles.

Louisiana Medicaid, an office of the Bureau of Health Services Financing in the
Department of Health and Hospitals (DHH), contracts with Unisys to compile provider
data and create profiles. Unisys works with data specialists at Louisiana Medicaid to
risk-adjust the data used for profiling. At the time of enrolling as a provider in the
Louisiana Medicaid network, providers are sent a baseline profile to see where their
prescribing patterns currently compare to others. Within a year, providers are sent an
annual review with updated information so these providers can see how they fit with
peers in their provider comparison group. The profiles are in color and contain graphs
and charts that present providers with information about their prescription/drug utilization
and how they compared to their peers with accompanying explanations. Since the
Louisiana Medicaid Pharmacy Program has only been administering the provider
profiling program for two years, it is not known at this time whether using a contractor
will become too costly. In addition, it is too early to estimate whether the
implementation of provider profiles will indeed reduce costs or prescription rates and
whether implementing a structured incentive program would also be beneficial.

A great deal of time was taken with physicians and the local health community to
implement this program. Any program like this cannot be perceived as "big brother" or
as an attacker showing which providers are doing something wrong. Bureau of Health
Services Financing and DHH develop processes to evaluate the program. This is a young
program that is still evaluating and making changes as necessary. A team of individuals
that includes Medicaid's medical director, pharmacists, and DHH staff members meet to
evaluate and discuss needed improvements to the peer-based profiling program. Officials
believe that given the youth of the pharmacy provider profiling program, it is difficult to
report whether the program is effective. However, the state is consistently analyzing
feedback to determine the success of the program as well as future directions.

D. North Carolina

Page 10

Within the Carolina Access II and III program (North Carolina's primary care
case management program) enhanced case management services are provided to
physicians in the area of diabetes and asthma. Provider profiles are currently being
developed as part of this enhanced case management activity (see Appendix 4 for
example of a mock profile). The profile will contain two graphs. The first graph will
show a physician's total per member per month cost for the last 4 quarters as well as
emergency room utilization for non-emergent conditions. The graph compares the PCP
to their peers and to the network as a whole. Peers are defined as physicians of the same
specialty within the same geographic region. The profile also contains utilization rates
per 1000 member months for the last and current quarters. The PCP is also compared to
the network and their peer. Specific utilization rates for asthma and diabetes are also
included. The profile is to be mailed quarterly.

Access II and III consist of 13 networks are in western North Carolina who
provide services to approximately 550,000 Medicaid enrollees. The board of each
network is comprised of local community providers (hospitals, physicians, FQHCs, and
substance abuse and mental health centers). Quality improvement activities originated
from the physician leaders on these boards. These physician leaders were able to
generate considerable 'buy-in' for profiling activities between the Access II and III
physicians. Much of this was achieved through frequent meetings with the physicians to
discuss quality improvement activities. Currently, numbers used to compute the profiles
were risk-adjusted. In previous years North Carolina used a proprietary company to for
risk-adjustment. However, the cost proved to be prohibitive and the current reports will
not be risk-adjusted. It is anticipated that in the future North Carolina will acquire the
technology to perform the risk adjustment themselves.

North Carolina has two additional ongoing profiling activities. To control
pharmaceutical expenditures, Access II and III created the Prescription Advantage List
(PAL). The PAL list ranks medications in each class in order from the least to most
expensive. Physicians are encouraged, but not required to prescribe the least expensive
drugs (drugs included in tier 1 of the list. Tiers 2 and 3 are more expensive drugs).

Page 11

Periodically each network receives a scorecard comparing the networks' use of drugs in
tiers 1, 2, and 3 to the rest of the networks. North Carolina also distributes to all
physicians a report card that compares a physician's utilization and cost to their peers
(Appendix 5).

E. Massachusetts

The provider profiling program for MassHealth (MassMedicaid) provides profiles
for approximately 400 Primary Care Clinician (PCC) plan providers who have 200 or
more plan members. MassMedicaid utilizes a contract with the Massachusetts
Behavioral Health Partnership, often referred to as the "The Partnership," to compile
these provider reports. HEDIS measures are used and current indicators in the profiles
are asthma, cancer, diabetes, and well-care visits. Recently, a pediatric behavioral health-
screening indicator was added. In addition, top brand prescription and emergency room
rates are measured. Risk adjustments are not used for provider profiles and there are no
financial incentives or sanctions related to profiles. MassMedicaid is constantly trying to
improve the provider profile by determining best practices to streamline the process.

Profiles have been distributed since 1995 and since then, consistent efforts have
been made to ensure that the profile is a useful tool. Profiles are viewed as educational,
interventional tools that provide useful information to providers. The reports are
approximately 25 pages long with 1 graph/chart per page detailing the pertinent indicator
and provide the necessary information and explanation of that indicator. Providers are
sent the profiles twice a year in the spring and fall. The providers are surveyed to get
their feedback about the provider profiling program. In addition, practice site visits are
conducted to both educate about the profile, but also obtain evaluation information. In
addition, RNs play a role in providing important feedback regarding the provider profiles.
Currently, longitudinal data is being reviewed to determine effects of program. It seems
that there has been positive effects, however, the provider profiling program is most
likely not the only reason.

Page 12

In the Center for Health Services Research and Policy's Negotiating the New

Health System, readers are provided excerpts of state contracts regarding Medicaid

Managed Care, Primary Care Case Management, and the State Children's Health

Insurance Program. Specifically, within the quality assurance section, there are details

about the contracts' requirements on areas such as provider profiling. Table 1 provides

the Massachusetts Medicaid Managed Care contract specifications for provider profiling

(Rosenbaum et al, 2000).

Table 1. Massachusetts Medicaid Managed Care Contract Provider Profiling
"Section 2.10 Quality Management...
E. Provider Profiling

The Contractor shall conduct PCP and other Provider profiling activities at least
annually. Such profiling activities shall include, but not be limited to:
1. Developing PCP and Provider-specific reports that include a multi-dimensional
assessment of a PCP or Provider's performance using clinical, administrative, and
member and Enrollee satisfaction indicators of care that are accurate, measurable, and
relevant to the MassHealth population.
2. Establishing PCP, Provider, group, plan or regional benchmarks for areas profiled,
where applicable, including MassHealth-specific benchmarks were appropriate.
3. Providing feedback to individual PCPs and Providers regarding the results of their
performance and the overall performance of the Provider Network, including
performance related to Enrollees.
4. Identifying areas of improvement for individual PCPs and Providers, and/or groups
of Providers and establishing quality improvement goals for priority areas in which a
Provider or Providers do not meet established Contractor benchmarks relevant to
Enrollees..." Massachusetts Contract, pages 57-61.

F. Additional State Information

Information was collected about Georgia, Maryland, and Texas through state

Medicaid agency websites, journals, and brief interviews with Medicaid staff. Utilizing

the website for the Georgia Health Partnership, it appears that the state has implemented

a provider profiling program with their contractor ACS. It appears that Georgia's main

goal in developing the profiles is somewhat like Indiana's in that the main area of

analysis is utilization. However, Georgia disseminates Primary Care Case Management

providers with quarterly reports that contain utilization rates specified for nine target

areas and the rates are developed for each participating provider, for each participating

specialty group, and for all PCPs combined (Rosenbaum et al, 2000).

Page 13

Neither the state of Maryland nor its data contractor engages in creating or
compiling provider profiles. Currently, approximately 470,000 enrollees are served 7
Managed Care Organizations (MCO) that are contracted to provide Medicaid Managed
Care. Within the contract specifications between the state of Maryland and any MCO
applicant, the following is stated, "L. The applicant's proposed written utilization
management program that specifies, at a minimum, policies and procedures for: (7)
Integration of activities with quality improvement for provider-profiling" (Rosenbaum et
al, 2000). It does not appear that provider-profiling activities are required, but strongly
encouraged. Therefore, MCOs can determine the level at which they want to create and
implement a provider profiling program.

Although there have been programmatic and contractor changes since 1998-1999,
the mock provider profile (see Appendix 6) displays what Texas has developed and
disseminated to providers. Currently, it is not known whether this report format is still
being disseminated or if provider profiling is being conducted. In reviewing the Texas
provider profile, the main focus of analysis is cost and utilization. In addition, provider
and comparison panel data is used, however, explanatory data is not supplied on the
report. Table 2 presents the language provided for applicants responding to the Texas
Request for Proposals (RFP) regarding the Medicaid Managed Care program (STAR) and
Medicaid Expansion Project (Primary Care Case Management).

Table 2. Texas RFP Provider Profiling
"Quality Improvement and Utilization Management Activities
The STAR Network Administrator will perform several activities designed to monitor the quality of
medical care being performed by network providers, and other activities designed to monitor the utilization
of services. There will be a separate Quality Monitoring contractor who will provide an external review of
quality of care. This effort must be supported by the STAR Network Administrator. The separate and
distinct effort undertaken by the STAR Network Administrator will include ... provider profiling with
analysis and corrective action plans associated with aberrant patterns of care..." Texas RFP, pages 3-4.

V. Findings from Physician Interviews

Medicaid's fiscal agent, ACS sends provider utilization summary reports to
MediPass PCPs. These reports compare physicians and/or groups within a specific
geographic location to other physicians/groups in the same PCP specialty within that

Page 14

geographic region. The summary includes several areas of comparison including office
visits, ER visits, outpatient visits, physician referrals, lab and x-ray procedures etc.
Measures that are below average relative to the peer group are identified with a 'u', while
those above average are identified with an 'o.'

Although some offices are familiar with profiles they receive from other health
plans, very few office managers or physicians recalled receiving reports from ACS or
MediPass. Among those who did, none of them found the reports particularly useful or
meaningful. In particular they were concerned about whether the comparisons with the
peer group were appropriate. The 'o' and 'u' indications provided no sense of how much
they differ from their peers and why. When asked about the kinds of information would
they want to have on a profile, most mentioned that pharmaceutical/drug utilization
information would be very important.

When asked to comment on the overall usefulness of a profile, most commented
that a well-designed and meaningful profile could be very important to their practice.
Several offices already get profiles from other programs and they use the results to
discuss problems and implement changes to make improvements. In response to
whether the AdvancedMed format would be useful, many replied that it would, although
they would like extensive training on the report. In particular, physicians want to be sure
that there are 'apples to apples' comparisons. This relates to what is found in the
literature base physicians appear more willing to accept evaluations based on peer
comparisons rather than those of expert panels (Brand et al, 1995).

VI. Conclusion

Utilizing the information collected from the interviews in addition to literature
review, particular areas of concern for the current AdvancedMed profiling system used
by AHCA would be the time lags associated with data used to compile provider profiles,
the current configuration of reports and its usefulness for providers, and the level of input
Medicaid physicians have on the design and implementation of the provider profiling

Page 15

program. Most state Medicaid agency officials that were interviewed did not specifically
discuss time lag issues with the data. However, perhaps programs that have had more
time to implement provider profiling (e.g. Arkansas and Massachusetts) have utilized
evaluative processes to address their specific data concerns. Most officials interviewed
did discuss the importance of physicians being involved in the implementation of their
program and that the providers helped to improve their state's program by making
suggestions on how the reports could be most useful to them.

All of the state officials interviewed reported that no incentive structure was tied
into the profile program. According to 1999 data from the Center for Studying Health
System Change (HSC), physicians in practices of two or more said they are less often
subject to financial incentives based on profiling (14 percent), which are more likely to
restrain use of services, than incentives based on patient satisfaction (24 percent) and
quality (19 percent), which are more likely to encourage use. Stoddard, Grossman, &
Rudell (2002), also point out that profiling or other cost-control incentives could conflict
with quality incentives, and productivity incentives with patient satisfaction incentives.
Therefore, given that the ultimate objective of any performance assessment is to translate
information into action, there must be careful consideration to how performance reports
or profiles will be created, implemented, and presented to providers.

Noren, Thibodeau, Insigna, & Landon (1999), in a summary of studies on the
effectiveness of provider profiling, argue that there are several factors that might boost
the success of provider profiling. First, feedback must be timely. Physicians will have
difficulty in applying results if there is a considerable lapse between when services were
rendered and when they receive the profile. Second, physicians must be able to control
the variables being measured. Third, physicians must be convinced of the need for
improvement within their practice. Accordingly, how and whom administers feedback is
critical to its success. Physicians may be more likely to respond favorably to feedback if
a trusted colleague presents it and if provided in an easily interpreted format. Lastly,
physicians should be provided a forum in which to discuss and ask questions about the
profiles (Koska, 1990).

Page 16

As mentioned earlier, some states may have provider profiling occurring within
Medicaid provider networks, however, these programs are solely administered by the
MCOs and are not necessarily required by the state. Currently, Florida NetPASS and
PhyTrust, the contracted organizations who work with Florida's Minority Physician
Network (MPN) Program, send periodic performance reports to their primary care
physicians (PCP). Specifically, Florida NetPASS disseminate reports to providers
showing the utilization of their MediPass beneficiaries and also compare the PCPs to
their peers on a variety of measures (Lemak et al, 2004). PhyTrust also sends PCPs
similar reports as well as quarterly newsletters and provides monthly in-services and
training for their PCPs and quarterly meetings (Lemak et al, 2004). Generally, this is
helping to make MediPass function better by offering providers timely beneficiary
utilization information (Lemak et al, 2004).

Many believe that the provider profiling efforts to measure and improve quality of
care are secondary to the goal of cost control. However, the effect of provider profiling
on health care costs is unclear. A meta-analysis of 12 studies concluded provider
profiling has a modest effect on physician behavior (Balas et al, 1996). The authors
indicated the potential cost savings are unlikely to exceed the costs profiling (Balas et al,
1996). Others have noted that rigorously designed studies are needed to assess the
effectiveness of provider profiles on physician behavior and cost of care.

VII. Implications for AHCA

The Agency for Health Care Administration contracted with AdvancedMed, Inc.
to provide risk-adjusted provider information for all Medicaid primary care providers.
The risk-adjusted data are provided via a secure Web interface. Formatted reports can be
run at anytime showing aggregated actual measures of use activity, expected measures of
activity, a morbidity index, and actual and expected costs. For any one time, claims data
for a one-year period (four quarters) is available for analysis. New data uploads were to
be performed each quarter, at which time the oldest quarter's data would drop out of the
data set and a new quarter's data would be added.

Page 17

One important result of this database's architecture is that, on average, there is an
8 to 9 month lag period between the date of service and the date when the data become
available from AdvancedMed. An important consequence of this lag time is that
physicians who modify their behavior will not begin to see those changes reflected in the
data until many months later. Furthermore, because the data is aggregated across 1 year,
it will take even longer before physicians begin to see the full effect of their behavior
change. A single quarter's data cannot be isolated in the database. For example, a
physician who changes his or her behavior on September 1, 2003 will begin to see the
effect of this change in April 2004, when the first data concerning that date is made
available. The full effect of this behavior change will not be seen until April 2005, when
a full year of changed practice is aggregated in the data.

This study conducted a literature review and interviews with key informants, state
Medicaid agency officials, and providers in order to assess the contracted provider
profiling system. These research activities provided some indication that as currently
configured, this provider profiling program will not be very effective in improving the
quality of care and/or reducing health care expenditures. First, the lag time associated
with the AdvancedMed reporting system does not encourage physician response to the
data presented to them. Second, the feedback reports as they are currently configured are
not particularly useful to providers. For example, reports provide a total count of all
services provided and compares them to an expected total count. Similarly actual and
expected per member per month figures are provided. However, the counts and the per
member per month figures are only disaggregated into broad service or diagnostic
categories. Finally, it is unclear what level of input Medicaid physicians may have had in
the design and implementation of the profiling program or in the feedback process. Of
special concern is the fact that Medicaid physicians may not understand how the data
were risk-adjusted.

Page 18

As AHCA considers the implementation of provider profiles within its Medicaid
program, the following guidelines are critical to assuring some level of success:

The initial emphasis should be on quality improvement
Performance measures should be clinically important
Accountability for performance should be limited to patients and services for
which physician is directly responsible
Adequate risk and case mix adjustment must be incorporated in the program
design with accompanying explanation of how data was adjusted
Physicians should be involved early in the design and implementation of provider
profiling strategies in order to generate 'buy-in'
Performance reports should be clear and easy to understand and coupled with
provider feedback and educational opportunities
Provider profiling programs should make every attempt to use accurate data

Page 19

VIII. References

American College of Physicians (2001). Assessing individual physician performance by
managed care organizations. A white paper prepared by the American College of
Physicians, American Society of Internal Medicine.

Balas, E.A., Boren, S.A., Brown, G.D., Ewigman, B.G., Mitchell, J.A., & Perkoff, G.T.
(1996). Effect of physician profiling on utilization: Meta-analysis of randomized
clinical profiles. Journal of General InternalMedicine, 11, 584-590.

Brand, D.A., Quam, L., & Leatherman, S. (1995). Medical practice profiling: Concepts
and caveats. Medical Care Research and Review, 52, 223-251.

Georgia Health Partnership. http://www.ghp.georgia.gov/wps/portal

Goldfield, N. (1996). Understanding your managed care practice: The critical role of case
mix systems. In N. Goldfield, & P. Borland (Eds)., Physician profiling and risk
adjustment. Aspen Publishers, Inc.: Maryland.

Goldfield, N., Gnani, S., & Majeed, A. (2003). Primary care in the United States:
profiling performance in primary care in the United States. British Medical
Journal, 326, 744-747.

Koska, M.T. (1990). Physician practices go under the microscope. Hospitals, 64, 32-37.

Lemak, C.H., Hall, A.G., Johnson, C.E., Saxena, P., Aftuck, C., & Johnson, C. (2004).
Evaluation ofFlorida's minority physician network (mpn) program. Report
prepared for the Florida Agency for Health Care Administration.

Marder R and Sheff RA (2003) 10 steps to successful physician profiling, HCPro Inc,
Marblehead, MA

Massachusetts Medical Society (1999). Stason, W.B. (Ed). Principles for profiling
physician performance. Massachusetts Medical Society: The Health Report, 1-40.

Noren, J., Thibodeau, R.M., Insinga, R., & Landon, B.M. (1999). Turning analysis into
action: Profiling as a change agent. In N.F. Piland and K.B. Lynam (Eds).,
Physician Profiling: A Source Book for Health Care Administrators. Jossey-Bass
Publishers: San Francisco.

Rosenbaum, S., Stewart, A., & Sonosky, C. (2002). Negotiating the new health system
(4th ed.). Washington, DC: Center for Health Care Strategies.
httD://www.awu.edu/-chsrD/Fourth Edition

Page 20

Page 21

Appendix 1. Provider Profiling Matrix

State or Profiling Measures or Contractor Length Distribution Incentive Profile Risk Data used for Evaluation of
Organization Mechanism Indicators Used of time of profiles Structure Categories Adjustment Profile* Profile
Used using (examples: Effectiveness
profiles ACG, CDS,
Arkansas Letter from Inpatient, Arkansas 9 years Quarterly to Not EPSDT ACG Provider, No formal
Medical Outpatient, Foundation (1995) Primary Care Reported OB/GYN State evaluation
Director, ER, Cost for Medical Provider (Separate Comparison,
Booklet PMPM, Care OB 2 Profiles) Explanatory
with Profile Psych, Rx years
and Graphs

Florida Planned to Actual and ACS Profiles Annual Not TBA Risk Provider TBA
send reports Expected contracted never distribution planned adjustments
to providers Utilization with sent out planned, time planned
and Costs, Advanced lag of data (ACG
and Med concern and/or CRG)
Georgia Utilization Utilization ACS At least Quarterly ** ** ** ** Provider
Reports since Advisory
2000 ______Committee
Indiana Statistical Delivery and Health Care Not Reports are not None None Not reported Provider Health Care
Profiles to Utilization Excel Profiles sent directly to specified Excel and
detect Patterns, providers State work on
provider Categories of Monitor determining
over- Services 6 years effectiveness
utilization 1998
Louisiana Packet with Rx Patterns Unisys 2 years Initial Receipt Young Rx Data is risk- Provider, DHH review
provider (2002) of Packet then program, adjusted Comparison to process by
information Disease Annual too soon Disease Peers, Medical
and graphs Mgmt not Review within to develop Mgmt Explanatory Director,
linked Peer Group incentives Initiatives Pharmacists,
maybe and staff

*Provider, Comparative, and/or Explanatory
Provider unique information about each physician
Comparative compares physicians to other physicians
Explanatory provides an explanation on how data were derived (Marder and Sheff, 2004)
** Information not collected

Appendix 1, Page 1

Appendix 1. Provider Profiling Matrix Continued

State or Profiling Measures or Contractor Length Distribution Incentive Profile Risk Data used for Evaluation of
Organization Mechanism Indicators Used of time of profiles Structure Categories Adjustment Profile* Profile
Used using (examples: Effectiveness
profiles ACG, CDS,
Maryland ** ** Each of the ** ** ** ** ** ** **
7 MCOs are
for profiling
Massachusetts Provider HEDIS Mass 9 years Bi-Annually Nothing Asthma No risk Provider and Visits to
Profiles Behavioral (1995) financial Cancer adjustments Explanatory Provider,
Rx Top Health Diabetes used for Provider
Brands and Partnership Well-Care provider Surveys and
ER sometimes (Recently profiles Hotline, and
called "The Pediatric Feedback from
Partnership" Behavioral RNs
North Utilization ACS rates, NC DHHS At least Quarterly to Not Asthma DxCG was Provider, Peer Meet with
Carolina and Disease Rx costs, Division of 2 years Providers Reported Diabetes previous Group and clinical
Mgmt. Asthma, Medical within Access profiling Network directors 6
Report HAlc Assistance II and III contractor. Comparisons times a year
(Managed Network Too costly,
Care) but plan to
use DXCG
Texas Refer to Refer to Contractor ** Using ** Refer to ** Refer to **
Appendix 6 Appendix 6 may have Appendix 6, Appendix 6 Appendix 6
changed Quarterly
(Medicaid (Utilization since 1998- (Provider and
Managed Care
is STAR and and Cost) 1999 Comparison)

*Provider, Comparative, and/or Explanatory
Provider unique information about each physician
Comparative compares physicians to other physicians
Explanatory provides an explanation on how data were derived (Marder and Sheff, 2004)
** Information not collected

Appendix 1, Page 2

Appendix 2. Arkansas Provider Profile

Arkansas Medicaid Pharmacy Claims
FY .tairrwFcirafrw

E .*.. ril-c rP oril .1*'1 ill' Ie I A. ror phlarniat y
clahnni lnoeaw-% bl, 1 22'r Lhai:. yoai

K It is ixtrant to reiterate that brand name
m~odecations are &Ignifkantly nw-e eorpeoslue thars
generic mediation5, oenerK mcicitiorks have a
1 '~.1 I 11 .1 III~ In ~ -ar.* In- 11M hN iV
4111 ol jlpqr-%Crl n~C]I-.Pfjn% TO 1L
Medicaid recipients In Arkansas
were generic costing over $54
million.The other 5-6% of pre-
scriptbnei diseneia were brand

Cost par preacription

P"Meripon Prie pren-1)" Wol
Wr 7?A1.



-,. Iy I
NO ,% ll:C1UIV' rX IY? TFB ~Y

Brand name vs. ge

50.1% 49.9%

neric utilization Total paid
ivermWeri amount, by year

44 5.5% ty A !'

1"4A PIN .4b

t H..U 6 s..lh i
UtY'To tlal Y7 MY^I YT4 'JX 'O Tuim Brn ar nriln

Total number Brand and generic
of prescriptions per yea paid amounts by year

C, I- i
JH^^Q~~- B b-wk "Hson
1 if, i wl 4 i r r


llt'Ill.' UT 1 I-VXFl L-.TlCrNV

P h beiwoimu Ctp~idut Kited hsbow In awarfol
hPlvu@IWdmW g(lem pn ulliltledim de
I I I I I swIM r-

1 ..... P1))( 11- Tr AK11 I I rIS 1,iA- 1
,I qJd Y T Ihaciot ,d [?,l, c h- Ail 1- tiuh

tO IrA) p111, it rMIK AO r 111rWW) pOP 01ktunj
im' -,, -h,,,,W r-4 ll ilo, cU-B 0

fI&4I f iod Mju iIm M I It vad4m1 l mIi .....

tbatfr-ie -Pff vrit :Drnuwphiirrn. V in wAte,#k
=i~* o I~iI ItlcIIIW II i n W~~ Ir~~T r
,drlu rni :,-col e jd -d vmdhjut cT-P-l a- the Iee Cd O
-w p-,k-d to -en,, -i ;- t pritm

Thh Ja', it rnt c&V .r .Nj edd .1, t dirM I h 0111h t
b-.meCt From PL fi e repbst tne past Skm

October-December 2003

M~ddka mE Mq.4n PrCPI trr bi in*MII hLab
um-le wuifded ka -dphie Pdwsq. 71. lWUE claim,
ern w,4 Irurkdddd q1Ped A illhilp fres Kr ooolm

~ It
mi-~k hkM 13. 1,~ 1,-tt, 1P ... IN, drrnhdIi pfhrJ
cnactwU I1D Ubtan a det1H er p. R
Airl,, rfulke ad othartlh irlarktdPII

ptibdl 4Wv ULr-JrI 1- h4Wb*$ I I h --I I, uibwillkritktI,

W Ify, 0/0w KP II
l017 O I' I- F I1 14 I P t M 11 Ii I -
VkJ(I IAIII I m, ~&Pi a.1. f

3r yf4litna MMIAU P-idefo 14MKKW M
41 SIr 0- W I eSJIU.i
11 ,,,1 lw& fJlle t il rnuy h~ IM

carperv a WWUK Diiecr:4 ARVI


BpA mf Ur


1"Its k. ANlidul OM, MuO
064 0%m CIkI
AI,O m. AK t AM6

Appendix 2, Page 1


Appendix 2. Arkansas Provider Profile Continued

I Anua Ris AdutdEfcec ai

Efficiency Ratio

Most Efficient

1 Least E ie
Least Efficient

Arkansas Medicaid Primary Care Physician Performance Report

Tawa FQilR1t Ef 'vi 3 a5j 41B Baa
FaD EI VolE I I dsi
Fum or n .r u irn fl an B LA

Frny ,EPCa Unr EP mI s t II I -

Fa~gi Emnfgearr.Kr Bi 1 a R
'-MsA I

--I^ I --- | I I
YOUR Adjusted Efficiency Ratio. 0.875
.*WI IPr "q I a E

Explanation of Risk Adiustment Process

)ati: Cans data for te most recent four quarters were used for
he analyses. Inl.a i-a cnlanis 'pa'i arn.a-un'rrs. eceecg J 1 DOo
Mere elirmnated from t*e data set.

'aopulion On- y PCPs (110B) with at east 10 enrollees in each o
he most recent four quarters were inc uded m the analysis.

Jnit of Analysis: The mean cost per patient per quarter per PCP
rJas r.e unlaamer.al un, usec for :.hE r sl ad,.sr'ineni criDess

4diusted Efficiency Ratio IAERm.

rhe AER is the mean :osr per par'Enr per quarter for me grer PCF
tbided by that PCP,s expected cost per patient per quarter. The
expectedd cost per patent per quarter s adjusted for te cost
ieverity of the PCP's case load using tte Johns Hopklns Adjusted
:I;n;.al 3')ursi icware For ths data set the mean AER was
I COO ard ire upper -5". confidence snit for the mean was 1.438.

An AER greater than 1.438 (noted by an asterisk) indicates
tnat PCPs resource utzation with respect t costs was
significantly greater than the average for the population.

JA A value of NA n tOe A sted Efficency Ratio area abave
irnd,:ales n rt your pai eril panel ras roo small to establish
siars.ca c.ornf ce th A risk 3r, us:rnrn


I D 5 40 I
Limv LI
PIN Es Cash 1754 nam UIJUK P.6SI

^2t Sft IcA 2A Srr1g
Sw*E If Wa ft! 12
Tal4 Pn 017cur

3n,4 opc EPA 15 10 A14 n
Thai Pnwli- IW k
A 2d u Msia

IIa~i .1 un1 *l~, USn UtIH Y~I

Ubsr~ Ptwa Codu nOII a tm. us? imTIu

. Gm. P"IC .i Coa
1XEPW meloo orta RIIH Wa n

I:eLi*..al I'm a,
U. n DaJ 1.1U".11112 .Ic...

M uml e t PeriO L me so
Taw.oeIa, Co uP I fm on m
Ibw4Wl ost PailLm M I lit tYm 1. fP!
ThManavms *pl4 4.UE Ui

Ftannae Coat PWU flu hA Mi mR R

aiman Vms.







Plarmacv Croi PEPM 71 $5i
PVsentage or Erraniia ufl lrI
a Bcipt 70 3% t1t.
SaiptB Pei EnroI I,- 57 &a

Adrage ,ilPer Script 53 31 70

" Gernic PRfcrlipln Inc l0 d3. Am

Si ,t.L ,.i', 1| l^M IiJK IMl li, lf,*Ijr

Arccoidh. to rtectl Xledicaid dar. pre:hrrtpnon xipendirure
COE nErF to rise. II k5 (mportat fr prescrnbrs to keep a miad
tir brand ame meditaionas are significaudt mre expensive
thin eraenm medliuman Bv making g tie effort to preribe
gment medLianoms when i1 would be tieiapesrtacltr aeItril (lo
the panlent phlrLuciU. rLa help Ieep zancessary epases down
irthotl tacnfKuwE qucjht of carr

Arkansas ConnectCare Program
Average cost per prescription



tsnom iiii
Q1- 02- W- 14- 01- 02- 03- 04-

2002 20 2002 B20 2003 201K 2103

I****I rtitll*** UiIIta lI lUm*lI 1 uI

,~m mha n rj ".d 1 n a MabILI m L h
i.e r d* .. in l . 'i n- A i.l n..j i i.s a ,.. .n *..-
nI.bIn it. rCll r i ft13im l .ll .s," Il- MI". |

Appendix 2, Page 2

Appendix 3. Arkansas Provider Profile (EPSDT/Preventative Health Screening)

Tools Available for You!

Well Child Tools to Encourage Care of a New Baby
AFMI( i N r pmulnte au with riy a l r rnl-mn that ac d l lp rlp~ learw yIU palirnl paoplainmi
You wrmw wder thmoolsFree of dhora by "Ping B-T77-P5 570Dor viit vww.waFRmsy

Carti FIdurUMth Prupirjad
renmisallan Remnd


WW I ....i ...

Slitiker ;

WellI Oitd PSotir

aII i ..a I ,

Fn 1A

rI Chhldelsn


Well Child Manual
(I A rIstht w ptruf ro

SFY 2004 Preventive Health Screening Procedure Codes


rfj I o )IW1 'U, .IMBi a I iP L

S;al I.t' 1


New Pattent 99"l

Pa.rsi1i 99391
awrMnt.,we flTMa wr~ rroJ

*-' IjL 'rIs .9 '.-I 9A

99aI12 9Aad 99AM4

99392 99393 99394

iF II,

- Profie

_Fr Itwm and Nrd a WreWnn Dhqutn dtlrr
rp )II .1 n sn.rwa ra Hl 111 Wr.% r Ar -11

i 1.1.. i .r) a ..* a i m- I I .I h .. T. rr I..l

.st 1._l 1 Iii .3i I..s j LqJr. u M r ..r.a i

nad9rtmi t ar k'lhmravsut idesrm-~a a ar
A.sm 4dd nruid d ar m .SYY
I., jruir-n -qr m. EPdYbT.Ln- JBi.tE-.III--E-&-

ty@h 1 tgrejtInin ff~ri2-Wfly
The mcggik II' umu. bor Iwstis wnnering win si hdiqelek

fl I.." r r r' .t 'i... s i : Fmw .u . c p emtj r Fs,

ThA e W r% m .QJ oil, .1i.n0A.ar I n la i t&dr op

i.I.E Lc~L c .-Va ..J..... 3wid ruH .0...nr. ~rlI .1
"I .... r mc, ei..g .l %e I att' t rrnt e e,

P'ren. t i hhfl alreer.w' reWrm cptrr ?mini.Y Me
to ie *fte eupi&RM da "mqtnd djd rwaM oil&

Ifl I a -s-VA Ii duI
hr d.-11- inl 1' r i. i.. I;t'nr !- A rd r V" -l' -lIr wr I
F~u ~l ~r~-IJ 4 -l D4 rI
F.CAE.k ala I. LIsEf ,,F ..tll .0--cav I," ,II1 I

joar-ifll T 9kw J P A 30 I ru ,.q X-d rp , WI eAY C..
Mt'ID- -Ef W0 MW Mr E rKI lCfell &ATIG'M&E

Ilr ffilW [mnrI id rf rrrrtl vrll.1kan~nim i n r lr i f rhr n-nw

SFY 2003
k. raMv A.MRap

wa wcrswrdduwirSFl 2Wl V u wu fare Do t ntact
rIs-, I dr.jrr-.- rn,- rdo-n,- r k.. 'r o, ,
rrv mp en~rrv~~n r aw-w a ragraq.lll s quera
sring yr wrldrrcuyrnsa
the h f de'e ec *eo:n porar rtch fe. ersob.M ng.r

knit' Vt. rltE-h.,ef.T hrty..n~liqq Ob dttl .' n s~I~
0)' EF jiL1 s5.r...Qradrelu Ir.quljceirbS IM ..1-I Ui.y.ar
)nny~ Fb7sW mnniau CrrnctLMc P421.
Anr. -, .r .j'rNs pnIir. a t .d' i t.e. d. i'r,'.M.

*Iwjh l~.jmr i-.;1ri ~ic tr.-psN rn. P nliw
p ir. :rrs. -aI..m ir. r*-i eF r eA nw P P r ih
pk.W*r,7 t'. %&QSLAXMn3 0-7
91 Z* 0 Cr A*:, l Im 'J r-e :1a 1r A I lF
Pret~arra'ul JlriA*CHIUtd
*hora mu dtafied rrepot eto rhaduear n MWS

I .nIl 'tru b e i. r I 1 r .P ... Ai W
F.Intpnto A8nt
&'mrrllir- rr



Appendix 3, Page 1

Appendix 3. Arkansas Provider Profile (EPSDT/Preventative Health Screening) Continued

ARWDS A Elrolfl Elgible ior EPSDT Screening

:. ;
% 70
,.' P 3

1, S ,



01 *' 2 6 i22 A
Ilu'lS" r.-rlh I .&pI I*% .f

Adual IOsera compaed ta R owmmMonded Scheaul

10 .22 -3ter-

12-23 F" m'

*Actua *Recorrnendea

Recipient Screening During SFY2003

i 333


0-l 1U-23 i i i1-20f
fmontis months yCan wyews wa
*Rec*piets with NO screen
*Recipients with at least 1 screen

1-1 I Moimr

-22 ir, I"- -3 I~ C.O C)

' r ears

6' "= r.? .; :" CTM
i6 i -: Iti-O

D G CE 30

. w 31

A PCP Ws pa"MW19 m carem f. r sevM or nae maV fs M c apga -1 ma-rt i an11 ar l-r m psity aw owIer
Mtgi recipi-r"is during SFY23. .
-" TnUnvTaa i MEPSDTis performed on M2abIle leaimsintatigSFY20w. Scwens prfx n rrsgiB
roapnesa, or reaaw nr asssgnegmd IMS PCP, afwe mD wn. I


H-' MoFuitre

IL-f ienrr

I -,msr

6" ^

s 1






: "...,.

0 u

Fri A

Recplints nay have had mor tan one PCP dOtNi SFY 23IS.
Tihe rclplenit whoe count are Indicated In re4 were not ecreene by ANY PCP during SFY M3
COMPLETE EPSDT acresn Ofuring SFY 203, were billd wtllh on of the Tlng procedure cwdet 2C1tziCS2t22SS or S432.|


AIGWID B Enrlles BIgMle T1r
Prermntfve tnaa Screenling


4 A

-11 12-3 2 -.5 6 10-0
irmts man" yurns yasm lIwrl

Actual Breilen compared to R coumnneid SchedulE

1Q-2j rean
6-9 ran-
2-5 W' 1
12-23 Mcrrns
$-11 VrI I
t0 aiAB 1,00o 1ie000 2000
iActual ERecomnmended

Recipient Screening During SFY2003

0-11 12-23 2-5 E-9 10-20
mongwi mon ft es ye yearw
Rec rlmrns wit iNO screen
*Retipiens with at least 1 screen

Appendix 3, Page 2



i 11 i11 L Li3 15E

Appendix 4. North Carolina Mock Provider Profile

Managed care provider type:
Administrative Entity
Address 1
Address 2

My Mutispecialty Practice
Access II
Access II Care of Western NC
12 Drs Park Drive
Ashville, NC 27110

Time Period
Peer Group
Average monthly enrollment
Eligibility 0-21
Eligibility >21

1/1/2003 through 3/31/2003
Internal Medicine

Non-Emergent ED Rate



-- "PCP Total
-'-Peer Group Total
*" Network Total

022002 03 2002 042002 01 2003


_ -*-PCP Non Emergent ED Rate

-.- -Peer Non Emergent ED Rate

-- -Network Non Emergent ED Rate

Q2 2002 Q3 2002 Q42002 012003

Utilization (rate per 1000 MM)
Hospital Inpatient
ED Total
ED Non emergent
Inpatient Mental Health
Out-patient Mental Health

Disease Management
Utilization (rate per 1000 MM)
Case Rate
Case Count
ED asthma visits
IP asthma discharges

Case Rate
Case Count
ED diabetes visits
IP diabetes discharges
Eye Exam %
Lipid Tet %

PCP Last Quarter

PCP Last Quarter


PCP Current Quarter

PCP Current Quarter



Comparison Network

Appendix 4, Page 1

Total PMPM









Appendix 5. North Carolina Provider Report Card

Peot~V: HiSR4051 WORTB 1CAROL-NA MIKS DATE: 09/26/2C03 .1
04/01/2003 06/30/2003


(1) PCP OFFICE SERVICES 576 $29.S2 339 $18.11 616 $32.69
(2, TOTAL EM/URGENT CAPR SERVICES 47 $10.74 42 $10.20 49 $11.57
A. IDENTIFIn F R,(GEWCNC 27 $7.16 25 $7.03 31 $8.13
B. UON-EMERGENT 20 $3.5B 17 $3.17 18 $2,44
(3] PBRa CY 70 $4 9.6. 512 $29.41 728 $52.45
(4) HOSPITAL INPATIENT 3 $8.07 3 $14.04 5 $15.3
(5) INPATIEKT MENTAL hEALT, 1 $2.62 1 $2.11 $2.61 t
(6i srEcrALISTS/HE-REPALS 63 $9.89 96 $10.14 4 $10.34 w
.7) LABS 53 $1.95 18 $0-.l 52 $1.8 -
(3) X-rAYS 1 $1.2 2 $0.71 9 $1.29
(B) X-RAYS 10 1.ZU 2 9a
(9) MENTAL rXSLTH OUTPATIENT 92 $9.70 189 $28.65 79 $8.11
(10) OUTPATIENT/AMBULAORY 5D $24.64 61 $22.15 55 $23.74

-------------------------------------------------- ------------------------- ---- ------- - ----
(11) PRMlARY CARE ROVIDEBR $30.92 $21.80 $34.43 $23.61
(12) ALL OTKiR S]EVICES $147.34 $157.22 $157.40 $161.48
(13) TOTA SERVTCFS S178.26 179.03 $i1.83 S185.09

I---------- --------------------------------- - ---- ---- ------- -- ----


S-------------------------------------- ---



ruc d

Appendix 5, Page 1

Appendix 6. Texas Mock Provider Profile




Membership Age and Sex Distribution

Cost by Type of Service
I Inan inned

Time Penod: 3/1/98 to 2/28/99
Specialty/Peer Group Pediatncs
Member Months: 7448
Illness Burden 0.74
Actual/Expected Ratio: 0.87

PCP Member Panel Comparison Group Panel
$13.95 $24.20
$15.54 $35.42
$33 90 $36 33
$3.28 $4.53
R1 49 5 33



PCP Svcs. Per Comp. Svcs. Per
Utilization PCP 1000 Members Weighted Comparison 1000 Members
Summary $PMPM Annualized Measure Panel $PMPM Annualized
Inpatient Maternity $061 153.1 $082 $064 2428
Inpatient All $8.98 2241.1 $12.10 $23.25 7685.7
All Professional Services $3754 13907.6 $50.57 $74.16 391583
EPSDT Dental $0.60 568 7 $0.80 $0 71 6437
EPSDT Visits $194 581.6 $2.61 $3.45 1034.4
ER Visits $4.28 3319 $5.76 $3.97 6470
Family Planning $0.11 1289 $015 $0.06 422
Immunizations $0.23 5462 $0.31 $0.55 1333.8

Specialist Utilization # of Services
Specialty Services 8817

Inpatient Utilization PCP Comparison Panel
Inpatient Bed Days/1000 members 106 374
Discharges/1000 members 39 94
Average Length of Stay 2 8 4.0

Location of Primary Visits
% of visits by following locations

Outpatient Hospital

Comparison Panel


Cost by Type of Service
PCP Member Panel

5% 2%
N Ancillary
E] Facility
0 Management
E Surgery
El Unassigned
50%4 23%

Birch & Davis Health Management Corporation


Birch & Davis Health Management Corporaton

Cost by Type of Service Comparison
Group Panel


Appendix 6, Page 1

....... d $149.......

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs