Hemophilia Disease Management in Florida · Hemophilia Disease Management in Florida Final report to Florida Agency for Health Care Administration Prepared and Submitted by Abraham

Hemophilia Disease Management in Florida

Final report to

Florida Agency for Health Care Administration

Prepared and Submitted

by

Abraham G. Hartzema PharmD, MSPH, PhD, FISPE

Richard Segal, PhD

Jinghua He, MSc, PhD candidate

Dawn Alayon, MPH, CPH

Department of Pharmaceutical Outcomes and Policy

College of Pharmacy, University of Florida

And

Heather Steingraber

Florida Center for Medicaid and the Uninsured

College of Public Health and Health Professions

Table of Contents Executive Summary ...................................................................................................................................... 4

Background ................................................................................................................................................... 5

Agency for Health Care Administration Disease Management Program................................................. 5

Epidemiology of Hemophilia .................................................................................................................... 5

Hemophilia Overview ........................................................................................................................... 5

Disease Progression and Outcomes ......................................................................................................... 6

Severity Levels of Hemophilia .................................................................................................................. 6

Treatment Options .................................................................................................................................... 6

Prophylactic Treatment......................................................................................................................... 6

On-Demand Treatment ......................................................................................................................... 8

Inhibitors ............................................................................................................................................... 8

Economic Factors in the Treatment of Hemophilia .................................................................................. 9

Costs and Economic Data ..................................................................................................................... 9

Patient Adherence to Treatment Plans .................................................................................................. 10

Interdisciplinary Provider Team............................................................................................................ 11

Comprehensive Hemophilia Disease Management Program in Florida ................................................ 11

Methods ...................................................................................................................................................... 11

Data Sources ........................................................................................................................................... 11

Analysis ................................................................................................................................................... 12

Descriptive Analysis ............................................................................................................................ 12

Evaluation of the Overall DMP .......................................................................................................... 13

Evaluation of the DMP Vendor ........................................................................................................... 13

Statistical Analysis Plan...................................................................................................................... 13

Findings ...................................................................................................................................................... 13

Part I. Descriptive Analysis .................................................................................................................... 13

Demographic Characteristics ............................................................................................................. 13

Health Care Resource Utilization ....................................................................................................... 17

Part II. Overall DMP Evaluation ........................................................................................................... 24

Demographics Characteristics ........................................................................................................... 24


Health Care Expenditures ................................................................................................................... 28

Part III. Between-Vendor Comparison ................................................................................................... 30

Demographic Characteristics ............................................................................................................. 30


Health Care Expenditures ................................................................................................................... 35

Discussion .................................................................................................................................................. 39

Results ..................................................................................................................................................... 39

Conclusion .................................................................................................................................................. 40

References ................................................................................................................................................... 41

Appendices ..................................................................................................................................................... i

Appendix A ................................................................................................................................................. i

ICD-9-CM diagnosis codes used for outcomes and baseline characteristics identification. ................ i

Appendix B ................................................................................................................................................ ii

NDC Codes Used For Factor Product Identification ........................................................................... ii

Appendix C ............................................................................................................................................... iv

Coding Algorithms for Health Care Resource Utilization and Cost Identification ............................. iv

Appendix D............................................................................................................................................... vi

Appendix E .............................................................................................................................................. vii

Cost per category without outlier exclusion (bucket code) ($) ........................................................... vii

Appendix F ............................................................................................................................................. viii

4

Executive Summary

A contract between the Agency for Health Care Administration and two Hemophilia

Vendors was executed on January 1, 2008. Due to contract implementation, fiscal agent activities,

and recipient choice periods, the program became fully operational on June 1, 2008. Florida

Medicaid started the two-year Comprehensive Hemophilia Disease Management Program (DMP)

which is mandatory for all Florida Medicaid non-HMO recipients with at least one pharmacy

record for factor products. Participants in the DMP chose either Caremark or CuraScript as their

service provider for comprehensive care, including disease management services, patient

education, and specialty pharmaceutical care for factor products. This study is an evaluation of

the first year of the DMP implementation. The observation period was one year prior to DMP

implementation (June 1, 2007 – May 31, 2008) and the first year of the DMP (June 1, 2008 –

May 31, 2009). The objectives of the program evaluation are to (1) describe the DMP population;

(2) evaluate the impact of the DMP on health care resource utilization and expenditures; and (3)

compare and contrast the two DMP vendors.

A total of 288 Florida Medicaid non-HMO recipients were identified as having

hemophilia or von Willebrand disease (vWD). Of the 288 members, 281 recipients were

receiving factor products through Medicaid during the pre-DMP period (2,880 member-months).

During the DMP period, there were 241 participants (2,126 member-months). The population

included mainly young men with varying racial/ethnic backgrounds, and few were identified

with HIV/AIDS or Hepatitis infections. Most were diagnosed with Hemophilia A or B, followed

by vWD and Factor X. This population was costly to Florida Medicaid: the average monthly

total cost was $18,316 PMPM, ranging from $3 PMPM to $335,030 PMPM. Hemophilia-related

costs accounted for 94.0% of the total cost, and the factor product cost accounted for 92.9% of

the total cost.

The DMP program was evaluated among 117 hemophilia participants with continuous

Florida Medicaid enrollment over the two-year observation period. The number of hemophilia-

related emergency and hematologist visits increased during the DMP period, while the number of

home health services and orthopedic surgeon visits decreased. These changes had little influence

on the total health care expenditure. After excluding outliers, the amount of dispensed factor

products increased approximately six percent during the DMP period compared to the pre-DMP

period. This result may be due to the increased body weight of this young population as they

mature and Medical and Scientific Advisory Council (MASAC) guidelines issued that promote

increased use of prophylactic dosing; however, the associated expenditures remained nearly

unchanged, presumably due to the discounted contract price. Mainly influenced by the factor

product cost, the total health care expenditure of the DMP period remained nearly the same as

compared to the pre-DMP period. The two vendors were also compared. The results showed no

difference in utilization or expenditures.

Overall, the comprehensive disease management program appears to have little impact on

health care resource utilization except for reducing home health service. The expenditure of

factor products is the major cost driver for the Florida Medicaid hemophilia recipients. The

discounted factor product price appears to offset the cost associated with the increased factor

product use.

5

Background

Agency for Health Care Administration Disease Management Program

The Florida Agency for Health Care Administration (AHCA) implements and manages

the Florida Medicaid Program. Florida Medicaid is a state and federal partnership that provides

health coverage for selected categories of people with low incomes and disabilities. Since 1970,

its purpose is to improve the health of people who might otherwise go without medical care for

themselves and their children. (AHCA, 2009)

In 1998, Florida legislature passed the Medicaid Disease Management Initiative. This

initiative offers systematic approaches for managing the health care needs of recipients who are

at risk of or diagnosed with a specific disease, using various methods – such as best practices,

prevention strategies, and outcomes research – to reduce overall costs and improve measurable

outcomes. This study focuses on the Comprehensive Hemophilia Disease Management Program

(DMP). In 1997, the Medicaid Reform Task Force found that approximately 10% of all Medicaid

recipients accounted for more than 60% of program expenditures. They found that the average

annual cost for a MediPass recipient with hemophilia exceeded $75,000. (Note: MediPass is the

care case management program for Medicaid recipients.) (AHCA, 2009) Key components of the

DMP are (AHCA, 2000):

clinical practice guidelines;

provider and recipient profiling;

specialized (disease-specific) physician and other practitioner care;

intensive care management;

provider education;

recipient education;

claims analyses;

quarterly and yearly outcome measurement and reporting.

Medicaid identifies eligible recipients for disease management which include all

Medicaid eligible recipients who have been prescribed factor replacement products and are

enrolled in MediPass , a Minority Physician Network, a Provider Service Network, or Fee-For-

Service. Eligible recipients are notified by Medicaid for participation in the program. Eligible

recipients must choose one of the two contracted hemophilia disease management providers, but

may change between the two providers at any time. Approximately 200 recipients receive

services each month from the Hemophilia DMPs. (AHCA, 2009)

Epidemiology of Hemophilia

Hemophilia Overview

Hemophilia is a rare X- linked hereditary bleeding disorder, primarily occurring in males.

It has two main types: Hemophilia A (Factor VIII Deficiency) and Hemophilia B, also known as

the Christmas Disease (Factor IX Deficiency), both of which can be classified as severe,

6

moderate and mild depending on the coagulation factor activity level. Hemophilia A is more

prevalent than Hemophilia B. As a result, there are more studies on the treatment of the former.

The incidence of hemophilia is about 1 in 5,000 live male births in the United States. All races

and economic groups are affected equally. The national population of people living with

hemophilia in the United States has been estimated to be 18,000, and 400 children are born each

year with the disease (Blankenship, 2008).

Disease Progression and Outcomes

Hemophilia has a significant negative impact on patient health. Its complications include

excessive bleeding, joint disease, severe pain due to bleeding into joints, and risk of infectious

diseases transmitted through blood products. If left untreated, health consequences include nerve

damage, loss of musculoskeletal function and death (Shapiro, 2007). Children may experience

internal bleeding into the joints, muscles and soft tissues as well as nose and mouth bleeds (The

Haemophilia Society, 2008). von Willebrand's Disease (vWD) is related to hemophilia (Miners

et al, 2009) and mimics Hemophilia A. This is an autosomal inherited disorder in which there is

deficiency or dysfunction of von Willebrand factor (vWF). (Treatment of Hemophilia, 2008).

Historically, people with hemophilia did not survive past their third decade. Currently, advanced

medical care, particularly the introduction of factor replacement therapy, has significantly

prolonged life expectancy of people with hemophilia (Soucie et al, 1998).

Severity Levels of Hemophilia

The three categories for hemophilia severity are based on clotting levels: severe,

moderate and mild. The severe form is defined as frequent spontaneous bleedings predominantly

associated with the joints and muscles (World Federation, 2005). Specifically, for people with

severe hemophilia there is a greater chance of internal bleeding into the joints, muscles and soft

tissues (The Haemophilia Society, 2008), which is associated with high expenditures for

managing these patients. The moderate form involves occasional spontaneous bleeding although

severe bleeding may occur from surgery or trauma; while the mild form can also be associated

with severe bleeding from surgery or trauma, but does not incur the other problems noted with

severe disease due to better clotting factors. As a result, patients with moderate and mild disease

may not be diagnosed until later in life (The Haemophilia Society, 2008). (Note: For the

Hemophilia DMP, clotting levels are not used to set severity; severity is associated with frequent

bleeding).

Treatment Options

Prophylactic Treatment

In 1952, Professor Nillson in Sweden was one of the first to introduce prophylactic

treatments to hemophilia patients (Fischer et al, 2002). Prophylactic treatments are preventive,

while on-demand or episodic treatments are administered when bleeding occurs or bruises are

7

observed (Berntorp, 2009). Prophylactic treatments appear helpful in preventing joint bleeding

and hemophiliac arthropathy. These treatments are mostly administered to those with severe

cases of hemophilia to maintain factor levels in the blood and to decrease the incidence of

spontaneous bleeding. Mild and moderate hemophilia patient’s factor levels are not as critical as

severe cases and usually only require on-demand treatments. Prophylactic treatments assist in

maintaining factor levels in the blood (Kern et al, 2004), and they have been shown to reduce

bleeds and prevent severe joint damage. As a result, timing of prophylactic treatment is critical;

however, no age has been specified as the gold standard of care for initiating therapy. Berntorp

(2009) followed-up with a group of boys with severe hemophilia and found that those who

started prophylactic treatment between the ages of one and two developed normal joints and led

normal lives. Other studies agree that treatment initiation should be at a young age based on

diagnosis (Fischer et al (2002a), Berntrop (2009), Kern et al (2004)). Prophylactic dosages are

based on weight. Berntorp et al (2009) recommends for Hemophilia A, 25-40 IU of factor VIII

per kg body weight and administered three times weekly, and for Hemophilia B, 25-40 IU of

factor IX two times weekly.

Prophylactic treatment is strongly encouraged for severe hemophilia patients by the U.S.

National Hemophilia Foundation Medical and Scientific Advisory Committee and the World

Health Organization (Fischer et al, 2002b). The World Federation of Hemophilia (WFH) created

a comprehensive guide based on published consensus guidelines which encourages the

prevention of bleeding and the use of prophylactic treatment. Moderate cases for hemophilia may

require prophylactic treatment based on the level of factor present in the blood (WFH, 2005).

Mild and moderate cases of hemophilia do not necessarily need prophylactic treatment according

to these organizations. For those cases, desmopression (DDVAP) is administered when a

bleeding event occurs (on-demand therapy) (NHLBI); this is a hormone that releases the stored

factor already present in the blood stream. DDAVP is provided to those who have five percent or

greater FVIII and are responsive to pre-tests (WFH, 2005). Further, general consensus among

studies indicates that prophylactic treatments are effective in ameliorating the effects of

hemophilia. Manco-Johnson conducted a multi-center, randomized study, observing the effects

of prophylactic treatment versus on-demand care. The study followed boys with Hemophilia A

from the ages of less than 30 months to six years of age, comparing prophylactic and on-demand

treatment. The study found boys treated with prophylactic treated experienced less joint and total

hemorrhages as compared to the on-demand group.

Fischer et al (2002) analyzed data from two retrospective studies involving patients with

severe disease: on-demand treatment in France and prophylactic treatment in the Netherlands.

The findings demonstrated that those with prophylaxis had better health outcomes such as fewer

incidents of joint bleeds per year, arthropathy and history of orthopedic surgery. The on-demand

group has a greater number of incidents of bodily pain and limited physical functioning and

social functioning.

Despite the advances in prophylactic treatments, the majority of severe hemophilia

patients do not take advantage of this therapy. In 1995, only 33% of all children with hemophilia

in the United States received prophylactic treatments (Manco-Johnson, 2007). In 2004, 51.5%

children under six years old with severe hemophilia received prophylaxis (Manco-Johnson,

2007).

Prophylaxis has been shown to decrease the likelihood of the development of inhibitors

in patients compared to patients treated with on-demand treatment (Petrini, 2007); regardless of

8

whether plasma-derived or recombinant derived factors are used (Goew et al, 2007). An inhibitor

is a type of antibody which attempts to destroy substances, such as following treatment to

replenish factor XIII or IX, they do not recognize. The antibody attaches to the factor VIII or IX

and inhibits its ability to stop bleeding (WHF, 2009).

Now, that so many children with hemophilia have become adults, one argument is to

whether or not to continue prophylactic treatment into adulthood in terms of effectiveness. For

adults, the prophylactic interventions are secondary rather than primary treatments; the results

are better with children from ages 9 to 12 (Tagliaferri et al, 2008). Finally, regardless of whether

a patient is treated with prophylactic treatment, occasionally on-demand treatments are necessary

for emergent bleeds.

On-Demand Treatment

Currently, there is more literature on the efficacy of prophylactic treatments than on-

demand treatments alone for severe hemophilia patients. This form of treatment is commonly

given when the clotting factor concentrate is administered before surgical procedures, after an

injury, or once a bleed has started (Haemophilia Society, 2008), or when a bruise is observed,

and for mild or moderate cases (NHLBI). On-demand is an acute form of therapy, managing,

rather than preventing Hemophilia associated health problems. Among its disadvantages is that

one waits until the patient experiences a bleeding event before administering the treatment. Thus

patients may have repeated bleeding, which could lead to orthopedic problems such as muscular

atrophy and hemophilic arthropathy. (Miners et al, 2009). In the Manco-Johnson study, the on-

demand, episodic care group developed life-threatening hemorrhages, including intracranial and

gastrointestinal; the prophylactic treated patients did not experience those adverse events. Other,

common health problems associated with on-demand treatment are central nervous system

(CNS) hemorrhages (Hoots WK, 2007).

Inhibitors

A major consequence of prophylactic treatment is the development of inhibitors, which is

common among Hemophilia A patients (20-30% of severe cases), and occurs in Hemophilia B at

five percent or less (Tjønnfjord et al, 2007; The Hemophilia Society, 2008). Antibodies are

created against the factor treatment, and may render this form of treatment ineffective. Inhibitors

manifest early, approximately within 10 to 20 days of exposure. Causes for inhibitors

development include gene mutations, ethnicity, and family history (Abshire, 2007).

Upon developing higher inhibitor levels, therapy changes may be made to either provide

higher doses of a given product or use of a more expensive activated clotting factor. When high

inhibitor levels occur, alternative therapies are available, including NovoSeven, FEIBA, and

immune tolerance induction (ITI). NovoSeven, recombinant factor VIIa (rfVIIa), is a genetically

engineered clotting factor that was introduced in the early 1990s (The Haemophilia Society,

2008; Tjønnfjord et al, 2007). FEIBA is a factor eight inhibitor bypassing agent, plasma-derived

treatment. Currently, bloodborne pathogens have not been reported with FEIBA (Tjønnfjord et

al, 2007). Both of these treatments have been shown to be effective in preventing bleeds but not

as effective as non-inhibitor treatments (Tjønnfjord et al, 2007). Immune tolerance induction

(ITI), or immune tolerance therapy (ITT) (The Haemophilia Society, 2008), is an alternative

treatment option to rid inhibitors. For those with FIX inhibitors, ITI is difficult to implement

9

(Mannucci, 2008). This therapy would last for long periods of time, ranging from 9 to 24 months

with an 80% success rate (Hemophilia Society, 2008). Generations of recombinant factor

therapies have evolved. One of the newer treatment options is recombinant plasma/albubin-free

FVIII concentrate (rAHF-PFM, Advate). Advate was developed from Chinese hamster ovaries

(CHO), thus there may be a risk of inducing antigenicity from non-human (trans-species)

sourced products develop inhibitors (Shapiro, 2007).

Economic Factors in the Treatment of Hemophilia1

Hemophilia has become a costly chronic disease because of the life-time dependence on

expensive factor products and extra demands for health care resources. It has been reported that

the total annual cost of hemophilia care in 1998 was $30,820 for those patients receiving on-

demand treatment and $87,865 for patients receiving prophylactic treatment (Globe et al, 2004).

The factor prices have not decreased over time. In 2004, the reported median cost for factor

prophylactic inhibitor treatments was $55,853/year and non-inhibitor treatments $2,760 per year

less (Bohn et al, 2004). Please note that there is no comparison price for prophylactic treatment

reported. This study consisted of twelve patients with inhibitors to FVII to FIX (cases) identified

in the hemophilia surveillance system (HSS) at two centers were matched on age, severity of

hemophilia descriptive matched analysis was conducted to examine the annual differences in

cost of product used and hospitalizations. Thus, factor product utilization is the major cost driver,

accounting for approximately 45 to 93% of the total health care costs in the treatment of people

with hemophilia. Further, the total health care cost of people with hemophilia is influenced by

disease severity, inhibitor history, HIV and Hepatitis viral infection, and route of factor product

administration (intravenous infusion vs. through a port) (Globe et al, 2003).

Costs and Economic Data

All hemophilia treatments are costly and represent a significant financial burden for the

payers. Different countries have conducted cost analyses to explore the relative total costs of

treating patients with prophylactic versus on-demand therapy. Though prophylactic treatment is

more costly in terms of drug costs, the overall health of the people with hemophilia appears to be

better compared to patients given on-demand treatment, and thus may reduce long-term health

care costs.

In the United Kingdom, Miners et al. (2009) conducted a cost effectiveness study for

prophylactic treatment compared to on-demand treatment for severe Hemophilia A. The analysis

showed a mean expected life-time cost of £644,000 ($1,037,677.19 USD) for on-demand

treatment and £858,000 ($1,382,495.39) for prophylactic therapy, respectively. Quality of

Adjusted Life Years (QALY) was reviewed as well, which showed that on-demand therapy was

13.95 QALY, whereas for prophylactic therapy was 19.58, thus a 5.63 QALY increase.

In Canada, a cost study was conducted involving Canadian boys with severe Hemophilia

A for a year prior to, and a year after target joint bleeding (Kerns et al, 2004). Target joint

bleeding may lead to chronic deformities, leading to severe pain along the joint. Specifically, the

1 Unable to find studies about the Medicaid hemophilia population.

10

study focused on the pre- and post-target joint (TJ) to review costs incurred by the patient once

TJ occurs when administered prophylactic treatment. The study defined TJ as three bleeds into

any single joint within a consecutive three-month period. The enrolled population had an average

age of 54 months (range, 15–94 months), with ankles being most often affected, followed by

elbows and knees (46% vs. 28% and 23%, respectively). Each of the boys in the study developed

TJ bleeding, as defined by the study protocol, which then increased cost of care. The time period

was one year before and after treatment. The pre-TJ mean overall cost, included the infusion

costs, clinic visits, diagnostic tests/blood work, per boy was $20,091 Canadian Dollar vs.

$42,612 (CND) vs. the post-TJ, the mean cost was $43,891. Factor VIII use accounted for 87%

of the total cost the year before development of a target joint and 93% the year after the target

joint was affected. The factor cost was $18,381 pre-treatment vs. $38,538 post-treatment.

In the United States, a cost utilization study was performed for 336 Hemophilia A

patients, recruited from five comprehensive Hemophilia treatment centers during 1995. Nearly

half of the study sample was covered by Medicaid, while the others had commercial insurance.

(Globe et al, 2004). The study took into account the different cost components of care (e.g.

physician visits, factor VIII concentrate, emergency room, hospitalization), and found the total

annual cost of care was $139,102, with a mean cost of factor VIII of $130,438 for the entire

study sample, in 1995. The same study also found annual on-demand costs to be $69,656 for

severe and $18,890 per patient for mild and moderate hemophilia.

In another U.S. study, Manco-Johnson performed a randomized, clinical, multi-center

trial to discover an effective way to prevent arthropathy in people with severe hemophilia in the

United States (Manco-Johnson et al, 2007). During the course of this trial, Manco-Johnson found

that at a price of $1 per unit of recombinant factor VIII, the cost of prophylaxis for a child

weighing 50 kg (110 pounds) could reach $300,000 per year.

Patient Adherence to Treatment Plans

The patient’s ability to follow prescribed plans is a contributing factor to the

effectiveness of the treatment. Bentrop et al conducted a study for the hemophilia population in

Sweden, based on data from randomized, controlled trials comparing prophylaxis with on-

demand treatment. For those who use prophylactic treatment, 90% of patients, ages of 12 and

under follow the treatment plans and the percentage drops considerably to 36% for ages 19 to 29

(Berntrop, 2009). Physicians’ decisions to prescribe the treatments are based on patient’s ability

to adhere. An option is to ensure patient adherence by tailoring the treatment to individual needs.

Additional suggestions include psychosocial training for people with hemophilia to become more

independent, and to engage in physical activity.

11

Interdisciplinary Provider Team

According to the World Federation of Hemophilia, successful treatment of patients with

hemophilia involves a team of health care professionals, not just physician contacts. It is

recommended by the World Federation of Hemophilia (WFH) that a comprehensive team would

include professionals well-versed in bleeding disorders such as hematologists, nurse

coordinators, physiotherapists and social workers. In the United States alone, there are 142

federally funded hemophilia treatment centers (HTC) (Linney, 2010). According to Linney, there

is a 40% higher mortality rate among those who do not seek treatment from HTCs for the study

period 1993-1995 as observed in records on 2,950 males reviewed in the study. Treatment plans

improve the life expectancy of Hemophilia patients and reduce costs.

Comprehensive Hemophilia Disease Management Program in Florida

A contract between the Agency for Health Care Administration and two Hemophilia

Vendors was executed on January 1, 2008. Due to contract implementation, fiscal agent activities,

and recipient choice periods, the program became fully operational on June 1, 2008. Florida

Medicaid started the two-year Comprehensive Hemophilia Disease Management Program (DMP)

which was mandatory for all Florida Medicaid non-HMO recipients with at least one pharmacy

record for a factor product within the 18 month period prior to identification of eligibility for the

program. Florida non-HMO Medicaid recipients with vWD were also eligible for enrollment in

the DMP. Florida Medicaid HMO recipients were excluded from the program, and dual eligible

recipients (Medicaid and Medicare) were enrolled voluntarily. Participants in the DMP chose

either Caremark or CuraScript as their service provider for comprehensive care, including

disease management services, care management, 24/7 toll-free access to a nurse/pharmacist

helpline, physician and recipient education, and dispensing/home delivery of factor products.

Florida Medicaid reimburses both providers at Average Wholesale Price minus 39% (AWP-39%)

for the factor products they dispense to DMP participants. The care management component of

the program is provided by the vendors at no cost to the state or recipients.

Methods

Data Sources

Two data sources were used in the data analysis: (1) participant records provided by the

DMP vendors; and (2) Florida Medicaid claims database. Vendor-provided participant records

were available for the DMP period which were used to identify disease type and disease severity.

The Florida Medicaid claims database provides comprehensive data including health care

resource utilization and associated costs during both the pre-DMP and DMP periods. The pre-

DMP period was defined as the one-year period immediately prior to the DMP program

implementation (June 1, 2007 to May 31, 2008). The DMP period was defined as the one-year

period that participants actually received DMP interventions following the completion of their

enrollment (June 1, 2008 to May 31, 2009). The demographic information was summarized for

12

age, gender, race, disease type and co-infections of HIV/AIDS or Hepatitis. Based on the

information available in the claims data, the following utilization and cost categories were

examined:

Total health care resource utilization and cost for the following categories:

hospital inpatient (hospital admission and length of hospitalization)

emergency room visit

outpatient visit

physician office service

home health service

pharmacy (number of prescription fillings regardless of type of medications)

other health care service (health care service that did not belong to any categories above,

such as dental, laboratory, physical therapy, etc).

Hemophilia-related health care resource utilization and cost for the following categories:

hospital inpatient (hospital admission and length of hospitalization)

emergency room visit

outpatient visit

physician office visit

home health service

factor product amount dispensed

See Appendix A for the ICD-9-CM codes for hemophilia claims and co-infections.

See Appendix B for NDC codes used for factor product identification (MOE or MOF classes).

See Appendix C for the detailed coding algorithms for each outcome of interest.

Analysis

Descriptive analyses were conducted to characterize the Florida Medicaid hemophilia

population with regard to the patterns and frequencies of health care events and expenditures.

Next, the overall DMP program was evaluated by using a pre-post study design. Finally, the, two

DMP vendors were compared against each other.

Descriptive Analysis

The primary descriptive analyses consisted of all recipients with hemophilia enrolled in

Florida Medicaid non-HMO program for at least one month between June 1, 2007 and May 31,

2009. The recipients with hemophilia were identified by AHCA according to pharmacy claims of

factor products (MOE or MOF classes). Findings from the claims data are presented for the two-

year study period, as well as the one-year pre-DMP period (June 1, 2007 to May 31, 2008) and

the one-year DMP period (June 1, 2008 to May 31, 2009). Since Florida Medicaid recipients

could drop in and out of the Medicaid program at any time during the observation period, the

findings for both utilization and costs were summarized on the basis of the annual total and per

member per month (PMPM), (i.e. 20,000 factor product units PMPM and $2,000 PMPM for

factor product costs). All cost values were adjusted to the Florida Medicaid fiscal year 2008-

2009, based on the Florida Medicaid specific inflation factors provided by AHCA.

13

Evaluation of the Overall DMP

A pre-post design was implemented to evaluate the DMP program. Each patient served as

his/her own control (self-controlled, case series design), which helped to adjust for factors that

did not change over time, such as disease severity and chronic co-morbid conditions. Another

advantage of this design is its capability to answer questions about changes over time for the

outcomes of interest.

Florida Medicaid non-HMO Hemophilia enrollees were included in this analysis only if

they had

Continuous Florida Medicaid non-HMO enrollment for at least 12 months immediately

prior to the DMP period (June 1, 2007 – May 31, 2008), and

Continuous Florida Medicaid non-HMO DMP enrollment for 12 months during the DMP

period (June 1, 2008 – May 31, 2009).

Evaluation of the DMP Vendor

A concurrent, non-equivalent comparison group design was used to compare the two

DMP vendors, Caremark and CuraScript. Florida Medicaid non-HMO Hemophilia participants

were included if they had

Continuous Florida Medicaid DMP enrollment for 12 months during the DMP period

(June 1, 2008 – May 31, 2009).

No vendor switching during the DMP period (June 1, 2008 – May 31, 2009).

Continuous Florida Medicaid non-HMO enrollment for one year immediately prior to the

DMP period (June 1, 2007 – May 31, 2008). The purpose of this criterion is to collect

baseline demographic and clinical characteristics.

Statistical Analysis Plan

See Appendix D for descriptions of the statistical testing procedures used for the DMP

evaluation.

Findings

Part I. Descriptive Analysis

This section describes all identified recipients with hemophilia who enrolled in the

Florida Medicaid non-HMO program for at least one month between June 1, 2007 and May 31,

2009 in terms of their demographic characteristics, health care resource utilization, and cost.

Demographic Characteristics

The demographic characteristics of the study population are summarized in Table 1a. A

total of 288 Florida Medicaid recipients were included in the descriptive analysis. These

individuals contributed a total of 5,006 member-months of observation during the two-year study

period. Analyses were also conducted for the pre-DMP and DMP periods, separately. Of the 288

recipients, 281 of these recipients received care during the pre-DMP period between June 1,

2007 and May 31, 2008, (2,880 member-months). These recipients would have participated in

14

the DMP had the program been available. Further, 241 recipients (2,126 member-months) of the

288 recipients were provided care during the DMP period. They were the actual DMP

participants.

Gender:

Most enrollees were male in the entire population. The proportion was similar for both

the pre-DMP (89.0%) and DMP period (90.5%).

Age:

The average age in the entire population was 13.0 years old (SD=11.4). The median age

was 10.4 years old (IQR=10.4). Approximately 80% recipients were 18 years old or

younger. The age distribution was similar for both the pre-DMP and DMP period.

Race:

In the entire population, Whites accounted for 29.9%, Blacks accounted for 14.2%,

Hispanics counted for 26.0%, and other races accounted for 29.9%. The race distributions

were similar for both the pre-DMP and DMP periods.

Disease Type:

Hemophilia A was the most prevalent disease type, accounting for 68.8% in the two year

period for the entire population. The prevalence rates of Hemophilia B and vWD were

13.9% and 13.5%, respectively. Only 3.8% recipients had other clotting disorders, such

as factor X deficiency. The disease type distributions were similar for both the pre-DMP

and DMP period.

Co-infections:

A small proportion of the recipients had infection history of HIV/AIDS (2.1%), Hepatitis

4.9%), or both (2.4%). The distributions were similar for both the pre-DMP and DMP

period.

15

Table 1a. Demographics and Co-infections of Florida Hemophilia Population (Overall)

Medicaid Hemophilia Population

2-Year Period Pre-DMP Period DMP Period

(06/01/2007-

05/31/2009)

(06/01/2007-

05/31/2008)

(06/01/2008-

05/31/2009)

Total Recipients 288 100.0% 281 100.0% 241 100.0%

Member-Month (Average per member) 5006 17.4 2880 10.2 2126 8.8

Gender (%) Male 257 89.2% 250 89.0% 218 90.5%

Female 31 10.8% 31 11.0% 23 9.5%

Age

Average (Year) SD 13.0 11.4 13.0 11.4 13.7 11.6

Median (Year) IQR 10.4 4.7-17.4 10.4 4.8-17.3 10.8 5.5-18.3

0-4 Years (%) 69 24.0% 66 23.5% 47 19.5%

5-10 Years (%) 76 26.4% 75 26.7% 70 29.0%

11-13 Years (%) 23 8.0% 23 8.2% 26 10.8%

14-18 Years (%) 60 20.8% 59 21.0% 43 17.8%

19-59 (%) 58 20.1% 56 19.9% 53 22.0%

≥ 60 (%) 2 0.7% 2 0.7% 2 0.8%

Race

White (%) 86 29.9% 84 29.9% 70 29.0%

Black (%) 41 14.2% 40 14.2% 35 14.5%

Hispanic (%) 75 26.0% 72 25.6% 59 24.5%

Other (%)1 86 29.9% 85 30.2% 77 32.0%

Disease Type

Hemophilia A 198 68.8% 191 68.0% 165 68.5%

Hemophilia B 40 13.9% 40 14.2% 36 14.9%

vWD 39 13.5% 39 13.9% 30 12.4%

Other Clotting Disorder 11 3.8% 11 3.9% 10 4.1%

Co-infection

HIV/AIDS only 6 2.1% 7 2.5% 5 2.1%

Hepatitis only 14 4.9% 11 3.9% 8 3.3%

Both 7 2.4% 6 2.1% 3 1.2% 1 “Other” includes Asians, Native Americans and any other unspecified races

Table 1b compares the demographic characteristics between two vendor groups,

Caremark and CuraScript, during the DMP period. There were 117 participants served by

Caremark for at least one month. They contributed a total of 926 member-months. There were

133 participants served by CuraScript for at least one month. They contributed a total of 1,200

member-months. Nine participants switched between these two vendors during the DMP period

and contributed to both groups.

16

Table 1b. Demographics and Co-infections of Florida Hemophilia Population by Vendor

Medicaid Hemophilia Population Caremark CuraScript

Total Participants1 117 926 member-month 133

1200 member-

month

Gender

(%)

Male 100 85.5% 127 95.5%

Female 17 14.5% 6 4.5%

Age

Average (Year) SD 13.8 (12.0) 13.3 (11.1)

Median(Year) IQR 12.0 (5.4-18.5) 10.7 (5.2-17.9)

0-4 Years (%) 21 17.9% 35 26.3%

5-10 Years (%) 35 29.9% 38 28.6%

11-13 Years (%) 10 8.5% 11 8.3%

14-18 Years (%) 22 18.8% 24 18.0%

19-59 (%) 27 23.1% 25 18.8%

≥ 60 (%) 2 1.7% 0 0.0%

Race

White (%) 26 22.2% 46 34.6%

Black (%) 17 14.5% 21 15.8%

Hispanic (%) 40 34.2% 22 16.5%

Other (%) 34 29.1% 44 33.1%

Disease

Type

Hemophilia A

Mild 11 9.4% 9 6.8%

Moderate 4 3.4% 13 9.8%

Severe 49 41.9% 75 56.4%

Unknown 6 5.1% 3 2.3%

Total 70 59.8% 100 75.3%

Hemophilia B

Mild 6 5.1% 5 3.8%

Moderate 8 6.8% 2 1.5%

Severe 9 7.7% 5 3.8%

Unknown 0 0.0% 2 1.5%

Total 23 19.6% 14 10.6%

vWD 20 17.1% 12 9.0%

Other Clotting Disorder 3 2.6% 7 5.3% 1 Nine patients switched vendors during the one-year period and are included in the findings of each group.

Gender:

Most participants were male for both vendor groups. Caremark had more female

participants than CuraScript (14.5% vs. 4.5%).

Age:

The Caremark group was slightly older than the CuraScript group. The average age was

13.8 years old vs. 13.3 years old. The median age was 12.0 years old vs. 10.7 years old.

Caremark participants were more likely to be above 18 years old (24.8% vs. 18.8%), and

less likely to be below 4 years old (17.9% vs. 26.3%) than CuraScript participants.

Race:

Caremark had a larger proportion of Hispanics (34.2% vs. 16.5%) and a smaller

proportion of Whites (22.2% vs. 34.6%) than CuraScript. The proportion of blacks and

other races were nearly the same between the two vendors.

17

Disease Type and Severity:

Hemophilia A was the most prevalent disease type for both vendor groups. The prevalent

rate of Hemophilia A was relatively lower in the Caremark group than the CuraScript

group (59.8% vs. 75.3 %). Caremark had relatively higher prevalent rates of Hemophilia

B (19.6% vs. 10.6%) and vWD (17.1% vs. 9.0%) than CuraScript. Taking both

hemophilia types together, 49.6% Caremark and 60.2% CuraScript participants had

diseases at severe level.

Health Care Resource Utilization

The total and hemophilia-related health care resource utilizations were examined for all

recipients with hemophilia who were enrolled in Florida Medicaid non-HMO program for at

least one month between June 1, 2007 and May 31, 2009. The analyses were performed for the

two year study period as a whole, as well as the pre-DMP period (June 1, 2007 – May 31, 2008)

and the DMP period (June 1, 2008 – May 31, 2009) separately.

Total health care resource utilization

The descriptive analysis findings on total health care resource utilization for the 288

Florida Medicaid non-HMO enrollees are summarized in Tables 2a.

Health Care Resources Utilization and Expenditure in Florida Non-HMO Hemophilia Population

Table 2a. Total Health Care Resource Utilization

Medicaid Hemophilia

Population

2-year Period1 Pre-DMP Period

2 DMP Period

3

(06/01/2007-

05/31/2009)

(06/01/2007-

05/31/2008)

(06/01/2008-

05/31/2009)

Total PMPM Total PMPM Total PMPM

Inpatient

Hospital

Admission 311 0.06 166 0.06 145 0.07

Hospitalization

Day 1,509 0.30 764 0.27 745 0.35

Emergency Visit 580 0.12 324 0.11 256 0.12

Outpatient Visit 1,845 0.37 986 0.34 859 0.40

Physician Office Visit 2,828 0.56 1,712 0.59 1,116 0.52

Home Health Service 3,266 0.65 2,399 0.83 867 0.41

Pharmacy Claims 10,789 2.16 5,980 2.08 4,809 2.26

Other Service 9,635 1.92 5,334 1.85 4,301 2.02 1 Based on 288 patients and 5,006 member-months

2 Based on 281 patients and 2,880 member-months


Hospital Inpatient:

o There were a total of 311 hospital admissions (0.06 PMPM) and 1,509 hospitalization

days (0.3 days PMPM) during the two-year study period.

o The average monthly rates were similar between the pre-DMP and DMP periods. The

hospital admission rate was 0.06 PMPM vs. 0.07 PMPM. The hospitalization day rate

18

was slightly lower in the pre-DMP period than the DMP period, which was 0.27

PMPM vs. 0.35 PMPM.

Emergency Room Visits:

o There were 580 emergency room visits during the two-year study period. The overall

average monthly rate was 0.12 PMPM.

o The average rate monthly was similar for the pre-DMP and DMP periods, which was

0.11 PMPM vs. 0.12 PMPM.

Outpatient Visits:

o There were 1,845 Outpatient visits during the two-year study period. The overall


o The average monthly rate was slightly lower for the pre-DMP period than the DMP

period (0.34 PMPM vs. 0.40 PMPM).

Physician Office Visit:

o There were 2,828 physician office visits during the two-year study period. The overall


o The average monthly rate was similar between the pre-DMP and DMP periods, 0.59

PMPM vs. 0.52 PMPM.

Home Health Services:

o There were a total of 3,266 home health services provided to the study population

during the two-year period. The overall average monthly rate was 0.65 PMPM.

o The total home health service utilization reduced significantly, from 2,399 events in

the pre-DMP period to 867 events in the DMP period. The average monthly rate

dropped approximately 50% from 0.83 PMPM to 0.41 PMPM.

Pharmacy:

o There were 10,789 prescription claims during the two-year study period. The overall


o The average monthly rate was similar between the pre-DMP and DMP period, which

was 2.08 PMPM vs. 2.26 PMPM.

Other Services:

o There were 9,635 other health care services provided during the two-year study

period, which did not belong to any categories above. The overall average monthly

rate was 1.92 PMPM.

o The average monthly rates were 1.85 PMPM in the pre-DMP period and 2.02 PMPM

in the DMP period.

19

Hemophilia-related health care resource utilization

The descriptive analysis findings on hemophilia-related health care resource utilization were

summarized in Table 2b. Medical claims were included only if the primary diagnosis codes were

related to hemophilia, vWD or bleeding disorders. Pharmacy claims were included for analysis only

if the prescribed medications were for factor products.


Table 2b. Hemophilia-Related Health Care Resource Utilization

Medicaid Hemophilia

Population


2 DMP Period

3

(06/01/2007-05/31/2009) (06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

Total

%

within

Category

PMPM Total % within

Category PMPM Total

%

within

Category

PMPM

Inpatient

Hospital

Admission 66 21.2% 0.01 38.00 22.9% 0.01 28.00 19.3% 0.01

Hospitalization

Day 253 16.8% 0.05 141.00 18.5% 0.05 112.00 15.0% 0.05

Emergency Visit 82 14.1% 0.02 38.00 11.7% 0.01 44.00 17.2% 0.02

Outpatient Visit 1,007 54.6% 0.20 486.00 49.3% 0.17 521.00 60.7% 0.25

Physician Office Visit 1,070 37.8% 0.21 637.00 37.2% 0.22 433.00 38.8% 0.20

Specialist Hematologist 172 -- 0.03 97.00 -- 0.03 75.00 -- 0.04

Orthopedist 69 -- 0.01 69.00 -- 0.02 0.00 -- 0.00

Home Health Service 2,481 76.0% 0.50 1,845.00 76.9% 0.64 636.00 73.4% 0.30

Factor Product Units 79,074,881 -- 15,796.02 40,757,289 -- 14,151.84 38,317,592 -- 18,023.33 1 Based on 288 patients and 5,006 member-months



Hemophilia-Related Hospital Inpatient:

o There were a total of 66 hemophilia-related hospital admissions (0.01 PMPM) during the

two-year study period, which accounted for 21.2% total hospital admissions.

o There were 253 hemophilia related hospitalization days (0.05 days PMPM) during the

two-year study period, which accounted for 16.8% total hospitalization days.

o The average monthly rates were similar between the pre-DMP and DMP periods. The

hospital admission rate was 0.01 PMPM for both periods. The hospitalization day rate

was 0.05 PMPM for both periods.

Hemophilia-Related Emergency Room Visits:

o There were 82 hemophilia-related emergency room visits during the two-year study

period with an overall average monthly rate of 0.02 PMPM, which accounted for 14.1%

total emergency room visits.

o The monthly rate in the pre-DMP period appeared slightly lower than that in the DMP

period, which was 0.01 PMPM vs. 0.02 PMPM.

Hemophilia-Related Outpatient Visits:

o There were 1,007 outpatient visits during the two-year study period with an overall

monthly average of 0.20 PMPM, which accounted for 54.6% total outpatient visits.

o The average rate was slightly lower in the pre-DMP than the DMP period, which was

0.17 PMPM vs. 0.25 PMPM.

20

Hemophilia-Related Physician Office Visits:

o There were 1,070 physician office visits during the two-year study period with an overall

average monthly rate of 0.21 PMPM, which accounted for 37.8% total physician office

visits.

o The average rate was similar between the pre-DMP and the DMP period, which was 0.22

PMPM vs. 0.20 PMPM.

Specialist Visits:

o There were 172 hematologist visits during the two-year study period with an overall

average monthly rate of 0.03 PMPM in the pre-DMP and 0.04 PMPM in the DMP

periods.

o There were 69 orthopedic surgeon visits during the two-year observation period. These

visits all occurred during the pre-DMP period. No such visits occurred during the DMP

period.

Hemophilia-Related Home Health Services:

o There were 2,481 hemophilia related home health services during the two-year study

period with an overall average monthly rate of 0.5 PMPM, which accounted for 76.0% of

total home health services.

o The average rate decreased approximately 50% from 0.64 PMPM in the pre-DMP to 0.30

PMPM in the DMP period.

Factor Product:

o More than 79 million factor product units were dispensed during the two year study

period with an overall average rate of 15,796 units PMPM.

o The average rates were lower in the pre-DMP than that in the DMP period, which was

14,151 units PMPM vs. 18,023 units PMPM.

Health Care Expenditures

The total and hemophilia-related health care expenditures were examined. All expenditures

were adjusted to fiscal year 2008-2009 values based the Florida Medicaid inflation factors that

AHCA provided.

21

Total health care expenditure

The descriptive analysis findings on total health care expenditures are summarized in Table

2c.


Table 2c. Total Health Care Expenditure ($)1

Medicaid

Hemophilia

Population


3 DMP Period

4

(06/01/2007-05/31/2009) (06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

Total % Total

Cost

PMPM

Cost Total

% Total

Cost

PMPM

Cost Total

%

Total

Cost

PMPM

Total 91,688,149 100.0% 18,316 49,985,331 100.0% 17,356 41,702,818 100.0% 19,616

Inpatient

2,880,867 3.1% 575 1,467,718 2.9% 510 1,413,149 3.3% 665

Emergency

Visit 233,029 0.3% 47 138,177 0.3% 48 94,852 0.2% 45

Outpatient

Visit 527,781 0.6% 105 289,595 0.6% 101 238,185 0.6% 112

Physician

Office Visit 173,811 0.2% 35 107,119 0.2% 37 66,692 0.2% 31

Home Health

Service 181,253 0.2% 36 129,230 0.3% 45 52,023 0.1% 24

Pharmacy

86,124,598 93.9% 17,204 46,945,394 93.9% 16,300 39,179,204 92.6% 18,429

Other Service 1,273,796 1.4% 254 728,011 1.5% 253 545,786 1.3% 257

Cost Per

Bucket

Category

See Appendix E

1. 1All expenditures were adjusted to fiscal year 2008-2009 values

2. 2 Based on 288 patients and 5,006 member-months



Total Health Care Cost:

o Approximately, $91.7 million were spent for the 288 recipients during the two-year study

period with an overall average monthly cost of $18,316 PMPM, ranging from $3 PMPM

to $335,030 PMPM.

o The average monthly cost was approximately $2,000 lower during the pre-DMP than the

DMP period, which was $17,356 PMPM vs. $19,616 PMPM

Pharmacy Cost:

o Pharmacy cost was the largest cost category in this hemophilia population. A total of

$86.6 million were spent during the two-year study period, accounting for about 94% of

the total health care cost.

o The average monthly total cost was also lower in the pre-DMP than the DMP period,

which was $16,300 PMPM vs. $18,429 PMPM.

Hospital Inpatient Cost:

22

o Hospital inpatient cost was the second largest cost category, accounting for 3.1% of the

total health care cost. Approximately, $2.9 million were spent for hospital inpatient

services in this population during the two-year study period.

o The average monthly cost was $510 PMPM for the pre-DMP and $665 PMPM for the

DMP period.

Other Cost Categories:

o The other cost categories collectively accounted for only about 4% of the total health care

cost, including emergency visits, outpatient visits, physician office visits, home health

services, and other services.

Hemophilia-related health care expenditure

The descriptive analysis findings on hemophilia-related health care expenditures are summarized in

Tables 2d.

Total Hemophilia-Related Health Care Cost:

o More than 94% of the total health care costs were hemophilia-related.

o The average monthly cost was $16,380 during the pre-DMP and $18,534 during the DMP

period.

Factor Product Cost:

o Factor product cost was the main cost driver for this hemophilia population. A total of

$86.6 million were spent during the two-year study period, accounting for about 92.9% of

the total health care cost, 98.5% of the total hemophilia-related health care costs, and 99.0%

of the total pharmacy cost.

o The average monthly cost was $16,131 PMPM in the pre-DMP and $18,233 PMPM in

the DMP period.

Other Hemophilia-Related Cost Categories:

The other hemophilia related cost categories collectively accounted for only

approximately 1.5% of the total hemophilia-related health care cost, including hemophilia

related hospital inpatient, emergency visit, outpatient visit, physician office visits, home

health service, hematologist visits, orthopedic surgeon visits, and others.

23

Health Care Resources Utilization and Expenditure in Florida Non-HMO Hemophilia Population1

Table 2d. Hemophilia-Related Health Care Expenditure ($)

Medicaid Hemophilia

Population


3 DMP Period

4

(06/01/2007-05/31/2009) (06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

Total

% Total

Hemo

Cost

% within

Category PMPM Total

% Total

Hemo

Cost

% within

Category PMPM Total

% Total

Hemo

Cost

% within

Category PMPM

Total 86,577,929 100.0% 94.4% 17,295 47,174,964 100.0% 94.4% 16,380 39,402,965 100.0% 94.5% 18,534

Inpatient 517,037 0.6% 17.9% 103 272,691 0.6% 18.6% 95 244,346 0.6% 17.3% 115

Emergency Visit 26,732 0.0% 11.5% 5 12,132 0.0% 8.8% 4 14,599 0.0% 15.4% 7

Outpatient Visit 268,259 0.3% 50.8% 54 131,100 0.3% 45.3% 46 137,159 0.3% 57.6% 65

Physician Office Visit 63,966 0.1% 36.8% 13 37,499 0.1% 35.0% 13 26,466 0.1% 39.7% 12

Home Health Service 101,776 0.1% 56.2% 20 78,102 0.2% 60.4% 27 23,675 0.1% 45.5% 11

Specialist Hematologist 1,143,265 1.3% -- 228 403,165 0.9% -- 140 740,100 1.9% -- 348

Orthopedist 8,610 0.0% -- 2 8,610 0.0% -- 3 0 0.0% -- 0

Factor Product 85,220,835 98.4% 99.0% 17,024 46,458,264 98.5% 99.0% 16,131 38,762,571 98.4% 98.9% 18,233

5. 1All expenditures were adjusted to fiscal year 2008-2009 values 6. 2 Based on 288 patients and 5,006 member-months

7. 3 Based on 281 patients and 2,880 member-months 4 Based on 241 patients and 2,126 member-months

24

Part II. Overall DMP Evaluation

This section evaluated the overall DMP program by using a pre-post design. The pre-

DMP (June 1, 2007 – May 31, 2008) was compared to the DMP period (June 1, 2008 – May 31,

2009) in terms of demographics, health care resource utilization, and costs. A total of 117 Florida

Medicaid recipients with hemophilia were included in this analysis. They all were continuously

enrolled in the Medicaid non-HMO program for 12 months immediately prior to and 12 months

after the DMP implementation.

Demographics Characteristics

Table 3a shows the demographic characteristics of the 117 participants included in the

pre-post study.

Overall DMP Evaluation

Table 3a. Baseline Demographics (06/01/2007-05/31/2008)

Total Participants

117

(2,808 member-

months)

Gender (%) Male 107 91.5%

Female 10 8.5%

Age at the

beginning of the

study period

Average (Year) SD 12.1 1.6

Median(Year) IQR 8.3 5.1-15.9

0-4 Years (%) 23 19.7%

5-10 Years (%) 42 35.9%

11-13 Years (%) 9 7.7%

14-18 Years (%) 23 19.7%

19-59 (%) 20 17.1%

≥ 60 (%) 0 0.0%

Race

White (%) 29 24.8%

Black (%) 18 15.4%

Hispanic (%) 18 15.4%

Other (%) 52 44.4%

Disease Type

Hemophilia A

Mild 5 4.3%

Moderate 11 9.4%

Severe 73 62.4%

Unknown 1 0.9%

Total 90 76.1%

Hemophilia B

Mild 2 1.7%

Moderate 1 0.9%

Severe 9 7.7%

Total 12 11.1%

vWD 9 7.7%

Other Clotting Disorder 6 5.1%

Co-infection

HIV/AIDS only 4 3.4%

Hepatitis only1 6 5.1%

Both 2 1.7% 1 “Other” includes Asians, Native Americans and any other unspecified races

25

Gender

Most included participants were male (91.5%)

Age:

The average age was 12.1 years old (SD=1.6) and a median age was 8.3 years old

(IQR=5.1-15.9). Most participants were 18 years old or younger (82.9%).

Race

White, Black, Hispanic and other races accounted for 24.8%, 15.4%, 15.4%, and 44.4%,

respectively.

Disease Type:

Hemophilia A was the most prevalent disease type, accounting for 76.1%. Hemophilia B,

vWD and other clotting disorders accounted for 11.1%, 7.7%, and 5.1%, respectively.

Co-infections

10.1% of the participants had claims containing diagnosis codes of HIV/AIDS or

Hepatitis or both. All these diagnosis codes first appeared during the pre-DMP period

except for one Hepatitis case, which first appeared in the DMP period.


Total and hemophilia-related health care service utilizations were compared between the

pre-DMP and DMP periods for the 117 participants included in the pre-post study.

Overall DMP Evaluation with Pre-Post Design (n=117)

Table 3b. Total Health Care Resources Utilization

Pre-Post Comparison

Pre-DMP Period DMP Period P

value1

(06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

Total PMPM Total PMPM

Inpatient Hospital Admission 66 0.05 84 0.06 0.69

Hospitalization Day 268 0.19 425 0.30 0.79

Emergency Visit 149 0.11 173 0.12 0.55

Outpatient Visit 494 0.35 572 0.41 0.69

Physician Office Visit 886 0.63 740 0.53 0.17

Home Health Service 1,680 1.20 622 0.44 <0.01

Pharmacy Claim 3,223 2.30 3,209 2.29 0.40

Other Service 2,532 1.80 2,962 2.11 0.03

Number of Beneficiaries with at

Least 1 Visit Per Study Period Total Percentage Total Percentage

P

value2

At Least 1 Hospital Admission 33 28.2% 33 28.2% 1.00

At Least 1 Emergency Visit 59 50.4% 64 54.7% 0.51

At Least 1 Outpatient Visit 100 85.5% 96 82.1% 0.53

At Least 1 Physician Office Visit 104 88.9% 95 81.2% 0.03

At Least 1 Home Health Service 38 32.5% 27 23.1% 0.05 1 Wilcoxon Signed Rank Test

2 McNemar’s Test

26


Table 3b summarizes the results for the pre-post comparisons for total health care

resource utilization in the 117 hemophilia participants.

No statistically significant differences were found for hospital admission, hospitalization

day, outpatient visit, or pharmacy claims between the pre-DMP and DMP periods.

The proportion of participants that visited physician offices decreased from 88.9% during

the pre-DMP to 81.2% during the DMP period (P=0.03). The number of physician office

visits also decreased from 0.63 PMPM to 0.53 PMPM, although difference was not

statistically significant (P=0.17).

There was a statistically significant decrease in the utilization of home health services.

During the pre-DMP period, 38 participants (32.5%) had 1,680 home health services. The

numbers decreased to 27 participants and 622 home health services during the DMP

period. The differences were statistically significant for both the number of home health

services (P<0.01) and the proportion of participants that ever had home health service

(P=0.05).

The number of “other service” increased from 1.80 PMPM in the pre-DMP to 2.11

PMPM in the DMP period (P=0.03).


The hemophilia-related health care resource utilization was also compared in the 117

participants by including only claims with a primary diagnosis code for hemophilia. The results

are presented in Table 3c.

27


Table 3c. Hemophilia-Related Health Care Resources Utilization

Pre-Post Comparison

Pre-DMP Period DMP Period

P value1 (06/01/2007-05/31/2008) (06/01/2008-05/31/2009)


Inpatient Hospital Admission 21 0.01 15 0.01 0.41

Hospitalization Day 79 0.06 60 0.04 0.22

Emergency Visit 16 0.01 37 0.03 0.01

Outpatient Visit 556 0.40 745 0.53 0.86

Physician Office Visit 434 0.31 407 0.29 0.68

Specialist Visit Hematologist 10 0.01 47 0.03 <0.01

Orthopedist 33 0.02 0 0.00 <0.01

Home Health Service 1,510 1.08 456 0.32 0.00

Factor Product Units 27,109,794 19,309 30,480,655 21,710 0.64

Number of Recipients with at Least

1 Visit Per Study Period Total Percentage Total Percentage P value

2

At Least 1 Hospital Admission 15 12.8% 11 9.4% 0.45

At Least 1 Emergency Visit 8 6.8% 21 17.9% 0.02

At Least 1 Outpatient Visit 85 72.6% 88 75.2% 0.72

At Least 1 Physician Office Visit 64 54.7% 69 59.0% 0.50

At Least 1 Hematologist Visit 8 6.8% 37 31.6% <0.01

At Least 1 Orthopedist Visit 19 16.2% 0 0.0% <0.01

At Least 1 Home Health Service 28 23.9% 12 10.3% <0.01 1Wilcoxon Signed Rank Test

2 McNemar’s Test

No statistically significant differences were detected for hemophilia-related hospital

admission, hospitalization day, outpatient visit, physician office visit, or amount of

dispensed factor product.

The average dispensed factor product amount increased 12.4% from 19,309 units PMPM

to 21,710 PMPM, although the difference was not statistically significant (P=0.64). Since

factor product was the major intervention and cost driver in this population, further

analyses were conducted to determine from where the 12.4% difference came. Three

participants were factor product outliers, whose pre-post differences were more than three

standard deviations from the mean. One outlier was dispensed 1.2 million units more in

the pre-DMP than the DMP period. The other two participants was dispensed 2.1 million

and 1.1 million units more in the DMP period than the pre-DMP period, respectively.

After the three outliers were excluded, the amount of dispensed factor product increased

about six percent during the DMP period compared with the pre-DMP period (16,671

units PMPM vs. 17,681 units PMPM). The difference remained statistically insignificant

(P=040). (See Appendix F)

Significant differences were seen in hemophilia-related emergency room, home health

service and specialist visits.

28

o A total of 8 (6.8%) participants had 16 emergency room visits during the pre-DMP,

which increased to 21 participants (17.9%) and 37 visits during the DMP period.

o A total of 8 participants (6.8%) had 10 hematologist visits during the pre-DMP,

which increased to 37 participants (31.6%) and 47 visits.

o A total of 19 participants (16.2%) had 33 orthopedic surgeon visits during the pre-

DMP period, which decreased to 0 during the DMP period. Please recheck the figures

in this bullet – they don’t seem to match the table.

o A total of 28 participants (23.9%) had 1,510 home health services during the pre-

DMP period, which decreased to 12 (10.3%) participants and 456 services.


The total and hemophilia-related health care expenditures were compared between the

pre-DMP and DMP periods for the 117 recipients. All costs were adjusted to Florida Medicaid

fiscal year 2008-2009 values based on Florida Medicaid specific inflation factors.

Total health care expenditure

The findings on total health care expenditures are presented in Table 3d.

There was no significant difference in the total cost, inpatient cost, emergency room cost,

hospital outpatient cost, or pharmacy cost.

Significant differences were seen in physician office visit, home health service, and other

service cost. However, the differences had little impact on the total health care cost since

the major cost driver was factor product cost.

o The average physician office visit cost decreased from $39 PMPM in the pre-DMP to

$32 PMPM in the DMP period (P <0.01).

o The home health service cost decreased by approximately 60% from $54 PMPM to

$23 PMPM (P<0.01).

o The other service cost increased from $204 PMPM to $223 PMPM (P=0.04).


Table 3d. Total Health Care Expenditure ($)1

Pre-Post Comparison


P value2 (06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

Total $/PMPM Total $/PMPM

Total 32,591,881 23,214 33,074,243 23,557 0.59

Inpatient 496,105 353 751,471 535 0.57

Emergency Visit 65,841 47 65,792 47 0.34

Outpatient Visit 136,791 97 145,116 103 0.12

Physician Office Visit 55,423 39 44,830 32 <0.01

Home Service Health 76,431 54 31,633 23 <0.01

Pharmacy 31,430,363 22,386 31,661,788 22,551 0.44

Other Service 287,099 204 312,400 223 0.04 1 All expenditures were adjusted to fiscal year 2008-2009 values

2 Wilcoxon Signed Rank Test

29


Table 3e summarized the hemophilia-related health care expenditure.


Table 3e. Hemophilia-Related Health Care Expenditure ($)1

Pre-Post Comparison


P value2 (06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

Total $/PMPM Total $/PMPM

Total 31,605,946 22,511 31,807,529 22,655 0.44

Inpatient 149,545 107 120,820 86 0.38

Emergency Visit 5,045 4 13,087 9 0.18

Outpatient Visit 61,317 44 92,382 66 0.63

Physician Office Visits 26,906 19 21,671 15 0.02

Specialist Hematologist 261,995 187 570,920 407 <0.01

Orthopedist 3,818 3 0 0 <0.01

Home Health Service 61,320 44 18,065 13 <0.01

Factor Product 31,193,776 22,218 31,398,965 22,364 0.40 1 All expenditures were adjusted to fiscal year 2008-2009 values


There was no difference in hemophilia-related total, hospital inpatient, Outpatient Visit,

emergency room visit, or factor product costs.

Significant differences were seen in hemophilia related physician office visit, specialist visit,

and home health service.

o The hemophilia related physician office visit cost decreased from $19 PMPM in the pre-

DMP period to $15 PMPM in the DMP period (P=0.02).

o The hematologist cost increased from $187 in the pre-DMP period to $407 in the DMP

period (P < 0.01).

o The orthopedic surgeon cost decreased from $3 PMPM in the pre-DMP period to $0

PMPM during the DMP period (P<0.01).

o The home health service cost decreased from $44 PMPM in the pre-DMP period to $13

in the DMP period (P<0.01).

Table 3f presented the changes in the number of beneficiaries who used particular brand

name factor products before and after the DMP. It should be noted that a single beneficiary

might have used more than one brand name factor product in the same period. Eleven brand

name factor products were used by the 117 beneficiaries. The distribution of types of factor

products dispensed did not change much between the two comparison periods. The greatest

change involved the use of Recombinate, for which the number of users decreased from 16

(13.7%) to 11 (9.4%) (P=0.063).

30

Overall DMP Evaluation with a Pre-Post Design

Table 3f. Factor Product Use

Pre-Post Comparison Pre-DMP Period DMP Period

P value1

(06/01/2007-05/31/2008) (06/01/2008-05/31/2009)

ADVATE 39 33.3% 38 32.5% 1.000

BENEFIX 15 12.8% 15 12.8% 1.000

BEBULIN VH IMMUNO 2 1.7% 2 1.7% 1.000

FEIBA VH IMMUNO 10 8.5% 9 7.7% 1.000

HUMATE-P 9 7.7% 6 5.1% 0.453

KOATE-DVI 1 0.9% 3 2.6% 0.500

KOGENATE FS 22 18.8% 22 18.8% 1.000

NOVOSEVEN 11 9.4% 11 9.4% 1.000

RECOMBINATE 16 13.7% 11 9.4% 0.063

XYNTHA 0 0.0% 4 3.4% 0.125

REFACTO 2 1.7% 2 1.7% 1.000

HELIXATE FS 6 5.1% 8 6.8% 0.500

MONOCLATE-P 1 0.9% 0 0.0% 1.000

HEMOFIL M 1 0.9% 0 0.0% 1.000 1 McNemar's Test

Part III. Between-Vendor Comparison

This section compares the two DMP vendors, Caremark and CuraScript, in terms of

demographics, health care resource utilization, and associated expenditures. A total of 115

participants were included in the comparison, including 45 Caremark and 70 CuraScript

participants. These participants had continuous DMP enrollment during the one year DMP

period without switching vendors. They also had one year of continuous Medicaid non-HMO

participation immediately prior to DMP enrollment in order to assess pre-DMP characteristics.

Demographic Characteristics

Demographic characteristics of the two comparison groups were summarized in Table 4a.

The two vendor groups are similar in terms of gender, age, and disease type. The Caremark

group enrolled more Hispanic participants while the CuraScript group enrolled more white

participants. The CuraScript participants also had slightly more co-infections.

Gender:

The majority of recipients were male. For both groups, the proportion of male recipients

was slightly lower in the Caremark than CuraScript which was 86.7% vs. 94.3%

Age:

The average age was approximately 13 years old for both of the two vendors. Caremark

had a smaller proportion of 0-4 years old which was 8.9% vs. 17.1%. Caremark had a

larger proportion of 19-59 years old which was 24.4% vs. 17.1%.

31

Race:

Significant differences were observed between the two vendors. Caremark served more

Hispanic recipients which was 24.4% vs. 8.6%. CuraScript served more White recipients

which was 31.4% vs. 13.3%.

Disease Type:

Hemophilia A was the most prevalent disease for both vendors which was 69.0% for

Caremark and 80% for CuraScript. Caremark had a higher proportion of Hemophilia B

than CuraScript which was 17.8% vs. 7.1%. Caremark also had a higher proportion of

vWD than CuraScript which was 11.1% vs. 5.7%. CuraScript had a relatively higher

proportion of other clotting disorders than Caremark which was 2.2% vs. 7.1%. The

disease severity distribution was similar between the two vendors. Recipients with severe

hemophilia accounted for approximately 70% of participants for both vendors.

Co-Infections:

HIV/AIDS and Hepatitis were slightly lower for Caremark than CuraScript which was

6.6% vs. 12.9% These diagnoses first appeared during the pre-DMP period except for one

Caremark participant, who had Hepatitis diagnosis code in the DMP but not the pre-DMP

period.

32

Table 4a. Characteristics At The Beginning Of The Study Period (06/01/2007)

Caremark CuraScript P value

1

Total Participants 45 70

Gender (%) Male 39 86.7% 66 94.3%

0.141 Female 6 13.3% 4 5.7%

Age at the

beginning of

the study

period

Average (Year) SD 13.3 8.9 13 10.7 0.501

Median (Year) IQR 11.3 5.0-18.0 8.8 6.3-16.2

0-4 Years (%) 4 8.9% 12 17.1%

0.715

5-10 Years (%) 17 37.8% 27 38.6%

11-13 Years (%) 5 11.1% 7 10.0%

14-18 Years (%) 8 17.8% 12 17.1%

19-59 (%) 11 24.4% 12 17.1%

≥ 60 (%) 0 0.0% 0 0.0%

Race

White (%) 6 13.3% 22 31.4%

0.036 Black (%) 8 17.8% 10 14.3%

Hispanic (%) 11 24.4% 6 8.6%

Other (%) 20 44.4% 32 45.7%

Disease Type

Hemophilia A

Mild 3 6.7% 2 2.9%

0.14

Moderate 3 6.7% 8 11.4%

Severe 25 55.6% 46 65.7%

Total 31 69.0% 56 80.0%

Hemophilia B

Mild 1 2.2% 1 1.4%

Moderate 1 2.2% 0 0.0%

Severe 6 13.3% 3 4.3%

Total 8 17.8% 4 7.1%

vWD 5 11.1% 4 5.7%

Other Clotting Disorder 1 2.2% 5 7.1%

Co-infection2

HIV/AIDS only 2 4.4% 2 2.9%

-- Hepatitis only 1 2.2% 5 7.1%

Both 0 0.0% 2 2.9% 1 Chi-Square Test

2 “Other” includes Asians, Native Americans and any other unspecified races



Table 4b summarizes the total health care resources consumed by the 115 participants by

vendor and the presence of DMP.

33

Between-Vendor Comparison (06/01/2007-05/31/2008)

Table 4b. Total Health Care Resources Utilization

Between-Vendor

Comparison

Pre-DMP period (6/1/2007-5/31/2008) DMP period (6/1/2008-5/31/2009)

Caremark (n=45) CuraScript (n=70) P

value1

Caremark (n=45) CuraScript

(n=70) P

value1 Total PMPM Total PMPM Total PMPM Total PMPM

Inpatient

Hospital

Admission 27 0.05 38 0.05 0.67 27 0.05 47 0.06 0.93

Hospitalization

Day 120 0.22 131 0.16 0.65 92 0.17 278 0.33 0.85

Emergency Visit 76 0.14 72 0.09 0.03 86 0.16 85 0.10 0.11

Outpatient Visit 149 0.28 341 0.41 0.33 211 0.39 336 0.40 0.62

Physician office visit 322 0.60 558 0.66 1.00 237 0.44 481 0.57 0.62

Home Health Service 429 0.79 1,250 1.49 0.44 24 0.04 596 0.71 0.10

Pharmacy Claim 1,125 2.08 2,033 2.42 0.51 1,123 2.08 1,974 2.35 0.65

Other Service 800 1.48 1,707 2.03 0.70 1,075 1.99 1,865 2.22 0.40

Number of Recipients with

at Least 1 Visit Per Study

Period

Total Percentage Total Percentage P

value2 Total Percentage Total

Percent

age

P

value2

At Least 1 Hospital

Admission 11 24.4% 21 30.0% 0.33 12 26.7% 19 27.1% 0.57

At Least 1 Emergency Visit 28 62.2% 30 42.9% 0.03 28 62.2% 35 50.0% 0.14

At Least 1 Outpatient Visit 36 80.0% 62 88.6% 0.16 35 77.8% 59 84.3% 0.26

At Least 1 Physician office

visit 39 86.7% 63 90.0% 0.40 34 75.6% 59 84.3% 0.18

At Least 1 Home Health

Service 13 28.9% 24 34.3% 0.35 7 15.6% 19 27.1% 0.11

1 Mann-Whitney Test 2 Chi-Square Test

There was no statistically significant difference between the two vendor groups except

that the Caremark had more emergency visits than the CuraScript during the pre-DMP

period. Both the average monthly rate (0.14 PMPM vs. 0.09 PMPM, P=0.03) and the

proportion of participants that ever had emergency room visits (62.2% vs. 42.9%, P=0.03)

were significantly higher for Caremark than CuraScript. The differences were reduced in

the DMP period and not statistically significant.


Table 4c summarizes the hemophilia-related health care resources consumed by the

included 115 participants by vendor and the presence of DMP.

There was no significant difference between the two vendor groups except hematologist

visits. Both the average monthly rate (0.01 PMPM vs. 0.005 PMPM, P=0.04) and the

proportion of participants that ever visited hematologists (13.3% vs. 2.9%, P=0.04) were

significantly higher for Caremark than CuraScript. There were no differences in the

DMP period.

34


Table 4c. Hemophilia-Related Health Care Resources Utilization

Between-Vendor

Comparison


Caremark (n=45) CuraScript (n=70)

P value1


value1 Total PMPM Total PMPM Total PMPM Total

PMP

M

Inpatient

Hospital

Admission 7 0.01 13 0.02 0.80 5 0.01 8 0.01 0.95

Hospitalization

Day 25 0.05 37 0.04 0.81 14 0.03 33 0.04 0.92

Emergency Visit 8 0.01 8 0.01 0.18 11 0.02 26 0.03 0.95

Outpatient Visit 81 0.15 167 0.20 0.95 107 0.20 240 0.29 0.38

Physician Service 144 0.27 282 0.34 0.87 90 0.17 229 0.27 0.87

Specialist Hematologist 6 0.01 4 0.00 0.04 18 0.03 28 0.03 0.79

Orthopedist 13 0.02 21 0.03 0.60 0 0.00 0 0.00 1.00

Home Health Service 330 0.61 1179 1.40 0.61 15 0.03 441 0.53 0.59

Factor Product (unit) 9,538,452 17,664 16,120,358 19,191 0.07 12,268,724 22,720 15,627,448 18,604 0.08

At Least 1 Visit Per Study

Period Total % Total % P value

2 Total % Total %

P

value2

At Least 1 Hospital

Admission 5 11.1% 9 12.9% 0.51 4 8.9% 6 8.6% 0.60

At Least 1 Emergency Visit 5 11.1% 3 4.3% 0.15 8 17.8% 13 18.6% 0.56

At Least 1 Outpatient Visit 33 73.3% 50 71.4% 0.50 33 73.3% 53 75.7% 0.47

At Least 1 Physician Service 30 66.7% 51 72.9% 0.31 26 57.8% 43 61.4% 0.42

At Least 1 Hematologist

Visit 6 13.3% 2 2.9% 0.04 15 33.3% 21 30.0% 0.43

At Least 1 Orthopedist Visit 10 22.2% 13 18.6% 0.40 0 0.0% 0 0.0% --

At Least 1 Home Health

Service 10 22.2% 17 24.3% 0.49 4 8.9% 8 11.4% 0.46

1 Mann-Whitney Test 2 Chi-Square Test

The average amount of the factor products dispensed to the Caremark recipients were

more than those dispensed to the CuraScript participants during the pre-DMP period

which was 17,664 units PMPM vs. 19,191 units PMPM, P=0.07. The situation became

opposite during the DMP period, in which the average factor product units dispensed to

the Caremark participants were more than those dispensed to the CuraScript participants

which was 22,720 units PMPM vs. 18,604 units PMPM, P=0.07.

Further examinations revealed that two participants were outliers. Compared to the pre-

DMP period, one Caremark recipient was dispensed about two million more units during

the DMP period, and one CuraScript recipient was dispensed one million units less

during the DMP period. After exclusion of the two outliers, 44 Caremark participants and

69 CuraScript participants were compared. The average factor product units dispensed to

Caremark were slightly higher than those dispensed to the CuraScript participants for

both the pre-DMP period (16,977 PMPM vs. 16,618 PMPM, P=0.05) and the DMP

period (18,050 PMPM vs. $17,554 PMPM, P=0.06) (Appendix F).

35


Total health care expenditures

Table 4d summarizes the total health care expenditures for the included 115 participants by

vendor and the presence of DMP.


Table 4d. Total Health Care Expenditure ($)2

Between-

Vendor

Comparison



value1


value1 Total PMPM Total PMPM Total PMPM Total PMPM

Total 11,644,865 21,565 19,004,158 22,624 0.05 12,600,898 23,335 17,542,241 20,884 0.05

Inpatient 239,303 443 238,416 284 0.65 147,249 273 538,210 641 0.94

Emergency

Visit 32,687 61 32,850 39 0.04 31,060 58 33,900 40 0.15

Outpatient

Visit 17,249 32 37,945 45 0.21 58,785 109 79,389 95 0.91

Physician

Office Visit 44,266 82 91,368 109 0.66 12,493 23 31,521 38 0.40

Home Health

Service 18,488 34 57,775 69 0.40 4,378 8 25,956 31 0.11

Pharmacy 11,206,418 20,753 18,303,144 21,789 0.07 12,189,082 22,572 16,621,254 19,787 0.06

Other Service 66,444 123 219,017 261 0.05 129,430 240 181,118 216 0.23 1 Mann-Whitney Test

2 Chi-Square Test

During the pre-DMP period, the average total total health care cost was higher for

Caremark than CuraScript, which was $21,565 PMPM vs. $22,624 PMPM, P=0.05. The

numbers reversed during the DMP period which was $23,335 PMPM vs. $20,884 PMPM,

P= 0.05. This pattern was similar to that observed for factor product utilization, the main

cost driver in this population. Analysis was redone after exclusion of the two outliers, in

which the average total health care cost was slightly higher for Caremark than

CuraScript during both the pre-DMP which was $20,604 PMPM vs. $19,362 PMPM,

P=0.04, and the DMP period $19,557 PMPM vs. $19,383 PMPM, P=0.04. (Appendix F).

The pharmacy cost had the same pattern as observed for total health care cost. The

average cost for Caremark was lower than CuraScript in the pre-DMP period which was

$20,753 PMPM vs. $21,789 PMPM, P=0.07 and was reversed in the DMP period which

was $22,572 PMPM vs. $19,787 PMPM, P=0.06. After the exclusion of the two factor

product outliers, Caremark had slightly higher average pharmacy costs than CuraScript

for both periods which was $19,840 PMPM vs. $18,539 PMPM, P=0.05 and $18,818

PMPM vs. $18,272 PMPM, P=0.04. (Appendix F).

During the pre-DMP period, Caremark had the higher average emergency room costs

than CuraScript which was $61 PMPM vs. $39 PMPM, P=0.04. The difference

diminished during the DMP period.

During the pre-DMP period, Caremark had the lower other service costs than the

CuraScript which was $123 PMPM vs. $ 261 PMPM, P=0.05. The difference diminished

during the DMP period.

36

No significant statistical differences were detected for hospital inpatient, hospital

outpatient, physician office visits, and home health service costs.


Table 4e summarizes the hemophilia-related health care expenditures for the included

115 participants by vendor and the presence of DMP. No statistically significant differences were

detected for any of the categories.

The total hemophilia-related health care cost was lower for Caremark than CuraScript in

the pre-DMP period which was $20,848 PMPM vs. 21,914 PMPM, P=0.06. In the DMP

period, the numbers reversed which was $22,729 PMPM vs. $19,828 PMPM, P=0.07.

The same pattern was observed for factor product cost which was $20,646 vs. $21,591,

P=0.08 for the pre-DMP and $22,421 vs. $19,573, P=0.07 for the DMP period. These

findings were similar to the previously reported observations on total health care and

pharmacy costs; these categories were mainly driven by factor product costs.

After exclusion of the two factor product outliers, these two cost categories were

consistently higher for Caremark than CuraScript participants in both the pre-DMP and

DMP periods. The total hemophilia-related health care cost was $19,913 PMPM vs.

$18,647 PMPM (P=0.05) in the pre-DMP period, and $18,941 PMPM vs. $18,314

PMPM (P=0.05) in the DMP period. The factor product cost was $19,710 PMPM vs.

$18,339 PMPM (P=0.06) in the pre-DMP period, and $18,664 PMPM vs. $18,055

PMPM in the DMP period (P=0.05) (Appendix F).

No statistically significant differences were detected for hemophilia-related hospital

inpatient, emergency room, hospital outpatient, physician office, and home health service

costs.

Caremark had higher hematologist costs than CuraScript during the pre-DMP period

which was $485 PMPM vs. $17 PMPM. The difference diminished during the DMP

period.

37


Table 4e. Hemophilia-Related Health Care Expenditure ($)1

Between-Vendor

Comparison



value1

Caremark (n=45) CuraScript (n=70) P value

2

Total PMPM Total PMPM Total PMPM Total PMPM

Total 11,258,162 20,848 18,407,410 21,914 0.06 12,273,783 22,729 16,655,452 19,828 0.07

Inpatient 56,283 104 74,876 89 0.64 29,111 54 67,467 80 0.56

Emergency Visit 2,777 5 2,267 3 0.25 3,388 6 9,699 12 0.86

Outpatient Visit 16,032 30 44,128 53 0.29 37,957 70 49,754 59 0.71

Physician Office Visit 6,227 12 20,617 25 1.00 4,086 8 16,938 20 0.92

Specialist Hematologist 247,469 458 14,526 17 0.03 272,778 505 298,008 355 0.72

Orthopedist 1,901 4 1,792 2 0.61 0 0 0 1.00 1.00

Home Health Service 13,674 25 47,477 57 0.58 3,622 7 14,444 17 0.65

Factor Product 11,136,913 20,624 18,136,362 21,591 0.08 12,107,540 22,421 16,441,402 19,573 0.07 1 All expenditures were adjusted to fiscal year 2008-2009 values

2 Mann-Whitney Test

Pre-post change comparison

For variables different between vendors during the pre-DMP period, the changes over the

two-year period were further compared to adjust for baseline differences. The pre-post changes

were calculated as the DMP minus the pre-DMP value. The results were presented in Table 4f.

No statistically significant differences were detected. The results do not change after exclusion of

the two factor product outliers (Appendix F).


Table 4f. Comparison of Changes in Health Care Resources Utilization And Expenditures Between The

Two Vendor

Between-Vendor Comparison


P

value1

pre-post

change

pre-post

change

PMPM

pre-post

change

pre-post

change

PMPM

Health Care

Resource

Utilization

Total Emergency Visit 10 0.02 13 0.02 0.71

Hemophilia-Related

Emergency Visit 3

0.01 18

0.02 0.29

Hematologist Visit 12 0.02 24 0.03 0.57

Factor Product (unit) 2,730,272 5,056.06 -492,910 -586.80 0.74

Health Care

Expenditure ($)2

Total Health Care Cost 956,032 1,770

-

1,461,917 -1,740 0.68

Total Emergency Cost -1,627 -3 1049 1 0.33

Pharmacy Cost 982,664 1,820

-

1,681,890 -2,002 0.96

Total Other Cost 62,986 117 -37,899 -45 0.17

Total Hemophilia-Related

Cost 1,015,620 1,881

-

1,751,958 -2,086 0.78

Hemophilia-Related

Emergency Cost 611 1 7,431 9 0.26

Hematologist Cost 25,310 47 283,482 337 0.75

Factor Product Cost 970,627 1,797

-

1,694,960 -2,018 0.87

1 Mann-Whitney Test

2 All expenditures were adjusted to fiscal year 2008-2009 values

38

Figures 1 and 2 below show the major cost categories discussed above by vendor and

DMP time periods before and after factor product outlier exclusion.

Figure 1. Major health care expenditure categories by vendor and comparison period before outlier exclusion

Figure 2. Major health care expenditure categories by vendor and comparison period after outlier exclusion

$0

$5,000

$10,000

$15,000

$20,000

$25,000

Total all-causehealth care cost

Totalhemophiliarelated cost

Total pharmacycost

Total FactorProduct Cost



Total pharmacycost


Pre-DMP DMP

PMPM

CareMark

CuraSCript

$0

$5,000

$10,000

$15,000

$20,000

$25,000



Total pharmacycost




Total pharmacycost


Pre-DMP DMP

PMPM

CareMark

CuraSCript

39

Discussion

Results

This study evaluated the impact of the Comprehensive Hemophilia Disease Management

Program (DMP) for the Florida Medicaid recipients who need factor product treatments for

hemophilia, vWD or other bleeding disorders. The study period was for a two year period: one

year immediately prior to the DMP implementation (June 1, 2007 to May 31, 2008) and one year

after the DMP implementation (June 1, 2008 to May 31, 2009). Analyses were conducted to

describe the target population eligible for this program. This population included predominantly

young males under the age of 18 and all races were represented. These characteristics are

representative of the broader population of persons who have Hemophilia A or B, which

primarily occurs in male children equally across all races. Despite the small size of 288 enrollees,

Florida Medicaid recipients with hemophilia consumed considerable resources which largely

reflected the cost of factor products. More than half of the Medicaid hemophilia recipients have

severe disease, which explains the large amount of factor product dispensed. The per member per

month expenditure (PMPM) for the factor product during the DMP period was more than

$17,000, accounting for more than 90% of the total health care expenditures.

To further examine the impact of the DMP on health care resource utilization and

expenditures, a pre-post design was used for 117 recipients who had continuous enrollment over

the two-year observation period. Notably, the total factor product units dispensed to this group

increased by about six percent after exclusion of three outliers. This increase was not surprising

because factor product doses are proportional to weight. Most recipients were below 18 years of

age, in an age group who tends to have an annual weight gain of 10%. Despite the increase in the

dispensed factor product units, the total amount of dollars that Florida Medicaid paid remained

nearly unchanged, after all numbers were adjusted to fiscal year 2008-2009 values based on the

Florida Medicaid inflation factor provided by AHCA. This cost savings may result from the new

contract DMP pricing on factor product, which was 39% of the AWP price. Dominated by the

factor product costs, the total expenditures for this patient population remained about the same

between the two comparison periods.

The utilization of some services increased during the DMP period. The number of

hemophilia-related emergency room (ER) visits was higher with seven percent of the recipients

having at least one ER encounter during the pre-DMP period compared to 18% during the DMP

period. Based on the information provided, the reasons for the increased number of ER visits

could not be identified. Importantly, the number of home health services billed to Medicaid

decreased by approximately 60% when participants were served by the vendors, and hemophilia-

related home health service decreased by about 70%. This finding was not surprising since the

vendors provided home health visits for infusion at their own cost. Another possible explanation

for the lower number of claims for home health service is that the participants may have been

more likely to administer the blood factor infusion themselves rather than relying on nurses

during the DMP period. This result was expected due to vendor-provided, self-care training to

program enrollees. Despite this substantial decrease, the impact of this reduction on total

Medicaid expenditures was relatively small, accounting for only 0.2% of the total health care

cost. More hematologist visits were observed during the DMP period and concentrated within

one month immediately following DMP enrollment. These visits appeared to be purposeful for

treatment planning and allowed the vendors to conduct an accurate assessment of the new

40

patients. There were few orthopedic surgeon visits during the DMP period; this type of visit

occurred only in the pre-DMP period.

Because two different vendors, Caremark and CuraScript, provided services to this

population, they were compared against each other in terms of resource utilization by recipients

enrolled with Hemophilia and expenditures. The final comparison included 45 Caremark and 70

CuraScript participants who were continuously enrolled one year prior and one year after the

implementation of DMP without switching vendors. After removing two factor-product outliers,

Caremark recipients were dispensed slightly more factor products than CuraScript. Since factor

product was the main cost driver, the total health care, pharmacy, and hemophilia-related total

health care costs were also higher for Caremark than CuraScript. Considering the fact that the

Caremark recipients also received more factor product in the pre-DMP period and the

incremental changes were not different between the two vendors, the higher factor product

consumption and cost of Caremark was likely due to the nature of the participants cared for

rather than the interventions provided. These two vendors did not appear to affect the health care

utilization or expenditures differently.

Conclusion

Overall, the Comprehensive Hemophilia Disease Management Program did not appear to

change the utilization of health resources for this population, with the exception of the finding of

a lower use of home health services which could particularly be due to cost shifting to the vendor

and not a true reduction in the use of home health services. There was a trend towards an

increased number of hemophilia-related emergency room visits, during the DMP period but this

difference had little impact on the total costs for caring for these patients. The expenditure of

factor products is the major cost driver for the Florida Medicaid hemophilia population. Any

future cost saving interventions will likely be driven by strategies to lower the cost of factor

products. Finally, there were no discernable differences between the two vendors in terms of the

heath care resources consumed by the participants served.

41

References

1. Abshire, T. Inhibitors in hemophilia: nature or nurture? Blood. 2007 June 1;109(11):4596.

2. An Introduction to Haemophilia. The Haemophilia Society. 2008 Retrieved on 15 Jan 2009.

http://www.haemophilia.org.uk/UserFiles/An%20Introduction%20to%20Haemophilia%20.p

df

3. Bohn RL, Aledort LM, Putnam KG, Ewenstein BM, Mogun H, Avorn J. The economic

impact of factor VIII inhibitors in patients with haemophilia. Haemophilia. 2004

Jan;10(1):63-8.

4. Berntorp, E. Joint outcomes in patients with haemophilia: the importance of adherence to

preventive regimens. Haemophilia (2009), 15, 1219–1227

5. Blankenship C. To Manage Costs of Hemophilia, Patients Need More Than Clotting Factor.

Accredo Health Services. Biotechnol Healthc. 2008 Nov–Dec; 5(4): 37–40.

6. Disease Management. Florida Agency for Healthcare Administration. 2009. Retrieved on 31

May 2010. http://ahca.myflorida.com/medicaid/disease_management/index.shtml

7. Factor eight inhibitor bypass activity (FEIBA) in the management of bleeds in hemophilia

patients with high-titer inhibitors. Vasc Health Risk Manag. 2007;3(4):527-31.What Are

Inhibitors? World Federation of Hemophilia. 2009 Retrieved on 30 Jan 2010.

http://www.wfh.org/2/docs/Publications/Inhibitors/Inhibitors_booklet_ENG.pdf

8. Fischer K, van der Bom JG, Mauser-Bunschoten EP, Roosendaal G, Prejs R, de Kleijn P,

Grobbee DE, van den Berg M. The effects of postponing prophylactic treatment on long-term

outcome in patients with severe hemophilia. Blood. 2002(a) Apr 1;99(7):2337-41.

9. Fischer K, van der Bom JG, Molho P, Negrier C, Mauser-Bunschoten EP, Roosendaal G, De

Kleijn P, Grobbee DE, van den Berg HM. Prophylactic versus on-demand treatment

strategies for severe haemophilia: a comparison of costs and long-term outcome.

Haemophilia. 2002(b) Nov;8(6):745-52.

10. Florida Medicaid. Florida Agency for Healthcare Administration. 2009. Retrieved on 31 May

2010. http://ahca.myflorida.com/Medicaid/flmedicaid.shtml

11. Globe DR, Cunningham WE, Andersen R, Dietrich SL, Curtis RG, Parish KL, Miller RT,

Sanders NL, Kominski G; Hemophilia Utilization Group Study.The Hemophilia Utilization

Group Study (HUGS): determinants of costs of care in persons with haemophilia A.

Haemophilia. 2003 May;9(3):325-31.

12. Globe DR, Curtis RG, Koerper MA; HUGS Steering Committee. Utilization of care in

haemophilia: a resource-based method for cost analysis from the Haemophilia Utilization

Group Study (HUGS).Haemophilia. 2004 Mar;10 Suppl 1:63-70.

13. Gouw SC, van der Bom JG, Auerswald G, Ettinghausen CE, Tedgård U, van den Berg HM.

Recombinant versus plasma-derived factor VIII products and the development of inhibitors

in previously untreated patients with severe Hemophilia A: the CANAL cohort study. Blood.

2007 Jun 1;109(11):4693-7. Epub 2007 Jan 11.

14. Guidelines for the Management of Hemophilia. World Federation of Hemophilia. 2005.

Retrieved on 23 Jan 2010.

http://www.ehc.eu/fileadmin/dokumente/Gudelines_Mng_Hemophilia.pdf

15. Hemophilia of Georgia (USA). Protocols for the Treatment of Hemophilia and Von

Willebrand Disease. Treatment of Hemophilia. 2008 Apr; 14. Retrieved on 23 Jan 2010.

42

http://www.wfh.org/2/docs/Publications/Diagnosis_and_Treatment/TOH-

14_English_von_Willebrand.pdf

16. Hoots, KW. Emergency Care Issues in Hemophilia. Treatment of Hemophilia. 2007 Nov; 43.


http://www.wfh.org/2/docs/Publications/Diagnosis_and_Treatment/TOH-43-Emergency-

Care.pdf

17. Invitation To Negotiate: For Participation In The Hemophilia Disease Management Initiative

Of The Florida Medicaid Program. Florida Agency for Healthcare Administration. 24 August

1998

18. Kern M, Blanchette V, Stain AM, Einarson TR, Feldman BM. Clinical And Cost

Implications Of Target Joints In Canadian Boys With Severe Hemophilia A. J Pediatr. 2004

Nov;145(5):628-34.

19. Linney, D. Medical And Pharmacy Cost Management Of The Patient With Hemophilia And

Utilization Of Federally-Funded Hemophilia Treatment Center. Great Lakes Hemophilia

Foundations. Retrieved on 26 Jan 2010.

http://www.glhf.org/Advocacy%20Articles/Cost%20management.htm

20. Mannucci, P.M. Back to the future: a recent history of haemophilia treatment. Haemophilia

(2008), 14 (Suppl. 3), 10–18

21. Medipass Overview. Florida Agency for Healthcare Administration. 2009. Retrieved 31 May

2010. http://ahca.myflorida.com/Medicaid/MediPass/index.shtml

22. Miners A. Revisiting the cost-effectiveness of primary prophylaxis with clotting factor for

the treatment of severe haemophilia A. Haemophilia (2009), 15, 881–887

23. National Heart Lung and Blood Institute: Diseases and Conditions Index. How is Hemophilia

Treated? Retrieved on 30 Jan 2010.

http://www.nhlbi.nih.gov/health/dci/Diseases/hemophilia/hemophilia_treatments.html

24. National Institute of Neurological Disorders and Stroke (NINDS). Creutzfeldt-Jakob Disease

Fact Sheet. Retrieved on 25 Jan 2010. http://www.ninds.nih.gov/disorders/cjd/detail_cjd.htm

25. Oldenburg J, Dolan G, Lemm G. Haemophilia care then, now and in the future. Haemophilia.

2009 Jan;15 Suppl 1:2-7. Review.

26. Petrini, P. Identifying and overcoming barriers to prophylaxis in the management of

haemophilia. Haemophilia (2007), 13 (Suppl. 2), 16–22

27. Pipe S. Antihemophilic factor (recombinant) plasma/albumin-free method for the

management and prevention of bleeding episodes in patients with Hemophilia A. Biologics.

2009;3:117-25. Epub 2009 Jul 13.

28. Shapiro, A. Anti-hemophilic factor (recombinant), plasma/albumin-free method (octocog-

alpha; ADVATE®

) in the management of Hemophilia A. Vasc Health Risk Manag. 2007

October; 3(5): 555–565. Published online 2007 October.

29. Soucie JM, Evatt B, Jackson D. Occurrence of hemophilia in the United States. The

Hemophilia Surveillance System Project Investigators.Am J Hematol. 1998 Dec;59(4):288-

94.

30. Specialty Pharmacy: Specialty Pharmacies Seek Ways to Deliver Financial and Clinical

Outcomes to Payers. The AIS Guide to Medicare Compliance Risks & Strategies. 2005 July.


http://www.aishealth.com/DrugCosts/specialty/SPN_SPs_Seek_Ways_to_Deliver_Outcomes

.html

43

31. Tjønnfjord GE, Holme PA. Factor eight inhibitor bypass activity (FEIBA) in the

management of bleeds in hemophilia patients with high-titer inhibitors. Vascular health and

risk management. 2007;3(4):527-31.

32. Tagliaferri A, Di Perna C, Rivolta GF. Secondary prophylaxis in adolescent and adult

haemophiliacs. Blood Transfus. 2008 Sep;6 Suppl 2:s17-20.

33. The Florida Medicaid Disease Management Initiative: A Report on the Florida Medicaid

Disease Management Program – Historical Perspective, Start–Up Activities, Current

Operations, Future Operations and Expectations. Florida Agency for Healthcare

Administration. 2000 February

i

Appendices

Appendix A

ICD-9-CM diagnosis codes used for outcomes and baseline characteristics identification.

Hemophilia A: 286.0

Hemophilia B: 286.1

vWD: 286.4

Acquired coagulation factor deficiency: 286.7

Other clotting factor deficiency: 286.2, 286.3

HIV/AIDS: 042.xx 079.53 V80

Hepatitis: 070.xx: 573.3: unspecified 571.4: Chronic 571.49: Other

ii

Appendix B

NDC Codes Used For Factor Product Identification

00026037220 - KOGENATE FS

00169705001 - NOVOSEVEN

RT 13533066530 - KOATE-DVI

00026037230 - KOGENATE FS 00169706001 - NOVOSEVEN 13533066550 - KOATE-DVI

00026037250 - KOGENATE FS 00169706101 - NOVOSEVEN 52769046001 - MONARC-M

00026037920 - KOGENATE FS 00169706201 - NOVOSEVEN 52769046402 - GENARC

00026037930 - KOGENATE FS 00944058101 - PROPLEX T 52769046405 - GENARC

00026037950 - KOGENATE FS 00944130101 - MONARC-M 52769046410 - GENARC

00026066520 - KOATE-DVI 00944130110 - MONARC-M 55688010602 - HYATE:C

00026066530 - KOATE-DVI 00944130201 - MONARC-M 58394000101 - BENEFIX

00026066550 - KOATE-DVI 00944130210 - MONARC-M 58394000105 - BENEFIX

00026378220 - KOGENATE FS 00944130301 - MONARC-M 58394000106 - BENEFIX





00944283110 -

RECOMBINATE 58394000301 - BENEFIX


00944283210 -



00944283310 -


00026379550 - KOGENATE FS 00944293001 - HEMOFIL M 58394000502 - REFACTO



00053761505 - HUMATE-P 00944293301 - HEMOFIL M 58394000604 - REFACTO



00053762005 - HUMATE-P 00944293503 - HEMOFIL M 58394000802 - BENEFIX

00053762010 - HUMATE-P 00944293504 - HEMOFIL M 58394000803 - BENEFIX

00053762020 - HUMATE-P

00944293801 -

RECOMBINATE 58394001102 - REFACTO

00053765601 - MONOCLATE-

P

00944293802 -

RECOMBINATE 58394001104 - REFACTO

00053765602 - MONOCLATE-

P

00944293803 -

RECOMBINATE 58394001201 - XYNTHA

00053765604 - MONOCLATE-

P 00944294001 - ADVATE 58394001301 - XYNTHA

00053765605 - MONOCLATE-

P 00944294002 - ADVATE 58394001401 - XYNTHA

00053766801 - MONONINE 00944294003 - ADVATE 58394001501 - XYNTHA

00053766802 - MONONINE 00944294004 - ADVATE 59730605907 - AUTOPLEX T

00053766804 - MONONINE 00944294010 - ADVATE UH 64193022203 - FEIBA VH IMMUNO

00053813001 - HELIXATE FS 00944294110 - ADVATE L 64193022204 - FEIBA VH IMMUNO

00053813002 - HELIXATE FS 00944294210 - ADVATE M 64193022205 - FEIBA VH IMMUNO

00053813004 - HELIXATE FS 00944294310 - ADVATE H

64193024402 - BEBULIN VH

IMMUNO

00053813005 - HELIXATE FS 00944294410 - ADVATE SH 68516320002 - PROFILNINE SD

00053813502 - HELIXATE FS 00944294510 - ADVATE UH 68516320003 - PROFILNINE SD

iii

00169701001 - NOVOSEVEN

RT 00944294610 - ADVATE 68516320004 - PROFILNINE SD

00169702001 - NOVOSEVEN

RT 13533066520 - KOATE-DVI 68516320005 - PROFILNINE SD

68516360002 - ALPHANINE

SD

68516360006 - ALPHANINE

SD 68516460101 - ALPHANATE

68516360004 - ALPHANINE

SD 68516460001 - ALPHANATE 68516460201 - ALPHANATE

68516360005 - ALPHANINE

SD 68516460002 - ALPHANATE 68516460302 - ALPHANATE

68516460402 - ALPHANATE

iv

Appendix C

Coding Algorithms for Health Care Resource Utilization and Cost Identification

Categories: Utilization

Cost

(adjusted to 2008-2009 fiscal year

values)

Inpatient

Number of

hospital

admission

Number of claims that have

bucket code “01” or ”02”

duplicates by “recip ID” and “date

of service” were considered a single

event

Amount in all inpatient facility claims

( bucket code “01” or ”02” )

And

Amount in physician service claims

(bucket 12) that match any inpatient

claims by “recip ID” and the range of

inpatient dates Inpatient

days

The value of the “Unit” variable in

claims with bucket code ”01” or “02”

ER visit

Number of claims that

have bucket code “03” or “04” or

“30”

have matches in the ER dataset by

“recip ID” and “date of service”



event

Amount in all outpatient facility claims

that are identified as ER visits

And


(bucket 12) that match any facility

claims above by “recip ID” and “date of

service”

Outpatient


bucket code “03” or “04” or “30”

Not an ER visit



event

Amount in all outpatient facility claims

( bucket code“03” or “04” or “30”) that

are not ER claims ;

And


(bucket 12) that match any claims

above by “recip ID” and “date of

service”

Physician office visit

Number of claims that

Have matches in the “new” dataset

by Recip ID, Pay to Prov ID, Date

of Service, and Amount.

Not ER/inpatient/outpatient

physician service claims

CPT codes: 59430, 99201-99215,

99241-99245, 99354, 99381-99429,

99432, 99450, 99455, ,99456,

99499, H1000, H1001TGP, H1001,

99395EP, or 99385EP

Place of service: 11, 49, 71, 72, or

99

duplicate claims by “recip ID”,

“date of service” were considered a

single event or multiple events

according to AHCA reviews

All claims that meet the criteria

above regardless of bucket code

Amount in all claims identified as

“Physician office visit”

v

Specialist visit

Number of hematologist visits:

Claims that have TRSPEC="15"

Number of orthopedic surgeon

claims:

Claims that have TRSPEC="58”



event

Amount in claims that have

TRSPEC="15"

Amount in claims that have

TRSPEC="58”

Home health service


Bucket code “20” and “21”

duplicates on “recip ID” and “date


event

Amount in all home health service

claims ( bucket code “20” and “21”)

Factor product Total units of claims with NDC

codes for factor products

Amount in all claims with NDC codes

for factor products

Pharmacy Number of claims that have bucket

code “14”

Amount in all pharmacy claims (bucket

code “14”)

All other type

All other claims that do not belong

to any outcome categories above

duplicates by “bucket code”, “recip

ID” and “date of service” were

considered a single event

Amount in all other type claims that do

not belong to any outcome categories

above

vi

Appendix D

Data Analyses:

Non-parametric statistical methods were used, which did not require data distribution

assumptions and the results were robust to outliers. A p-value of 0.05 or less was considered

statistically significant.

Overall DMP Evaluation

A descriptive data analysis was performed to show the trend of outcome change over

time on a monthly basis (PMPM). Outcome variables were compared before and after

implementation of the DMP. Since the pre-post samples were dependent, McNemar’s test was

used to compare categorical variables and two-tailed paired t-test was used to compare

continuous variables that were approximately normally distributed. Two-tailed Wilcoxon signed

ranks test was used for continuous variables that were not normally distributed. A p-value of

0.05 or less was considered statistically significant. All expenditures were adjusted to fiscal year

2008-2009 values based on the Florida Medicaid inflation factor provided by AHCA. Outliers

for factor product dispensed were determined as ± 3 standard deviations away from the mean.

Between Vendor Comparison

Univariate analysis was performed to compare the baseline characteristics and outcome

variables between the two groups. Chi-square statistic was used for categorical variables.

Fisher’s exact test was used when any expected frequency was less than 5. Two-tailed t-test was

used for continuous variables that were approximately normally distributed. Two-tailed

Wilcoxon Rank Sum test was used for continuous variables that are not normally distributed. For

variables that were found to be different during the pre-DMP period at a significance level of P <

0.20, the changes across the two-year period would be further compared between the two groups.

A p-value of 0.05 or less was considered statistically significant. All expenditures were adjusted

to fiscal year 2008-2009 values based on the Florida Medicaid inflation factor provided by

AHCA.

vii

Appendix E

Cost per category without outlier exclusion (bucket code) ($)

Bucket code Description Amount 1 Percentage Rank

14 PRESCRIBED MEDICINE 86124597.92 93.9% 1

1 HOSPITAL INPATIENT 2670124.24 2.9% 2

3 HOSPITAL OUTPATIENT 760739.07 0.8% 3

12 PHYSICIAN SERVICES 590210.06 0.6% 4

42 PERSONAL CARE SRVCES 362269.25 0.4% 5

63 PREPAID MH 259421.1 0.3% 6

20 HOME HEALTH SERVICS 181253.29 0.2% 7

43 PRIVATE DUTY NURSE 132105.27 0.1% 8

36 DEVELOPMENTAL SERVCS 127691.02 0.1% 9

7 INTRMD CARE LEVEL I 81204.38 0.1% 10

6 SKILLED NURSING HOME 56065.5 0.1% 11

23 CHILD DENTAL 30784.43 0.0% 12

11 ICF-MR PRIVATE FACLT 30538.02 0.0% 13

45 SPEECH THERAPY 30243.41 0.0% 14

49 FQHC 29761.47 0.0% 15

46 OCCUPATIONAL THERPY 24532.32 0.0% 16

15 LAB AND X-RAY 22296.8 0.0% 17

17 TRANSPORTATION 21050 0.0% 18

76 Unknown 20326.11 0.0% 19

40 RURAL HEALTH SERVICS 19527.94 0.0% 20

53 CLINIC SVCS 15696.14 0.0% 21

22 EPSDT SCREENING 14715.94 0.0% 22

37 AIDS WAIVER SERVICES 13148.08 0.0% 23

52 MEDIPASS 9684.01 0.0% 24

44 PHYSICAL THERAPY 9459.01 0.0% 25

47 RESPIRATORY THERAPY 7912.5 0.0% 26

29 ADULT HEARING 7538.75 0.0% 27

64 SCHOOL BASED SVCS 6303.08 0.0% 28

24 CHILD VISION 5024.88 0.0% 29

26 ADULT DENTAL 3922.71 0.0% 30

31 NURSE PRACTITIONER 3741.85 0.0% 31

19 FAMILY PLANNING 3366.79 0.0% 32

72 Unspecified 2861.76 0.0% 33

59 TFC COMM MENTL HLTH 2565.46 0.0% 34

99 Unspecified 1825.57 0.0% 35

27 ADULT VISION 1210.85 0.0% 37

68 PSN 945 0.0% 38

25 CHILD HEARING 893.1 0.0% 39

54 DS SLAMS 747.76 0.0% 40

8 INTRMD CARE LEVEL II 721.59 0.0% 41

62 PHYSICIAN ASSISTANT 547.45 0.0% 42

28 ADULT HEARING 442.44 0.0% 43

4 HOSPITAL OP CROSSVER 70.68 0.0% 44

57 DEVELOPMENT E & I 53.06 0.0% 45

74 Unspecified 9 0.0% 74 1 All expenditures were adjusted to fiscal year 2008-2009 values

viii

Appendix F

Evaluation Without Factor Product Outliers

Overall DMP Evaluation with Pre-Post Design (n=114) Without 3 Factor Product Outliers

Pre-Post Comparison


P

value1

(06/01/2007-05/31/2008) (06/01/2008-

05/31/2009)


Factor Product (Unit) 22,807,226 19,245 24,188,101 18,886 0.40

Total Health Care Cost3 27,019,526 18,471 26,516,040 17,948 0.47

Pharmacy Cost3 25,933,758 18,563 25,198,600 18,019 0.33

Hemophilia-Related Total Health Care Cost3 26,062,195 18,304 25,299,262 17,761 0.33

Factor Product Cost3 25,699,188 19,245 24,937,000 18,886 0.30

Pre-DMP Period Between-Vendor Comparison (06/01/2007-05/31/2008)

Between-Vendor Comparison Caremark (n=44) CuraScript (n=69) P

value2 Total PMPM Total PMPM



Pharmacy Cost3 10,475,652 19,840 15,350,427 18,539 0.05



DMP Period Between-Vendor Comparison (06/01/2008-05/31/2009)

Between-Vendor Comparison Caremark (n=44) CuraScript (n=69) P

value2 Total PMPM Total PMPM



Pharmacy Cost3 9,936,125 18,818 15,12,9224 18,272 0.04




2 Mann-Whitney Test


Incremental Health Care Resources Utilization And Expenditure Comparison Between The Two Vendor

Groups Period

Between-Vendor Comparison


P value1

Total

pre-post

change

pre-post

change/member

Total

pre-post

change

pre-post

change/mem

ber

Factor Product (unit) 566,895 47,241 775,233 64,603 0.99

Total Health Care Cost ($)2 -553,123 -1,024 17,622 21 0.93

Pharmacy Cost ($) -539,527 -999 -221,203 -263 0.78

Total Hemophilia-Related Cost ($) 2 -513,501 -951 -276,155 -329 0.96

Factor Product Cost ($)2 -552,121 -1,022 -235,081 -280 0.86

1 Mann-Whitney Test


Documents

Hemophilia Disease Management in Florida · Hemophilia Disease Management in Florida Final report to Florida Agency for Health Care Administration Prepared and Submitted by Abraham