As required by state law, the Texas Comptroller studied the two state-paid health care programs: the Medicaid program and the state employees workers' compensation program. The Medicaid study includes a study of the acute care Fee-For-Service (FFS) program, the Vendor Drug Program (VDP) and the Managed Care Capitated (premium) payment program. The comptroller conducted this study with the State Auditor's Office and with the assistance of the affected state agencies, particularly the Health and Human Services Commission (HHSC) and the State Office of Risk Management (SORM).
The study found the potential overpayment rates for the Medicaid programs to be 13.7 percent for the FFS program, 21.2 percent for the Vendor Drug program and 2.2 percent for the managed care premium payment program. The overpayment rate for the state employees workers' compensation program administered by the State Office of Risk Management was 13.7 percent. The margin of error was plus or minus 0.2 percent for the Medicaid FFS study, plus or minus 0.2 percent for the Medicaid Vendor Drug study, plus or minus 0.07 percent for the Medicaid managed care premium study and plus or minus 0.26 percent for the state employees workers' compensation study.
The potential overpayments are caused by factors that range from billing for services that are not provided or documented to potential fraud or abuse. To determine if these potential overpayments are actual overpayments, the HHSC's Office of Inspector General will perform further research and investigate the sample claims. Texas workers' compensation laws do not provide a similar process for determining the potential overpayments for the state employees workers' compensation program.
The Comptroller's office published three previous study reports. The first report, published in December 1998, reviewed the Medicaid FFS program, the state employees workers' compensation program and the Employees Retirement System (ERS). ERS services were removed from previous studies through a change in state law. The second report was published in January 2001. The 2001 study enhanced the 1998 study's sampling criteria, review methodology and overpayment calculation and served as the model for the 2003 study. The March 2003 study also added the Vendor Drug Program.
In conjunction with performing the Texas Health Care Claims Study, the Comptroller's office and HHSC are participating in pilot demonstration projects sponsored by the Centers for Medicare and Medicaid Services (CMS) to develop a national measurement study methodology that all states can use to identify and measure overpayments in the Medicaid program.
The March 2003 study was also the first CMS Payment Accuracy Measurement (PAM) study. CMS compiled the methods used by eight states towards a standardized method to use with both small and large state Medicaid programs. The project team performed this study parallel to the second year of PAM studies to compare the overpayment rates from the two separate study methods since there are significant differences. These differences include sampling by the paid date and percentage of paid dollars instead of the date-of-service and a set number of clients per category. The categories of health care are also different-PAM included long-term care, community-based home health and medical transportation services. Consequently, the March 2005 study is drawn from 2002 claims, a year after the time frame used for the March 2003 report.
In March 2005, CMS allowed Texas to perform the state studies in tandem with the federal studies using the pilot grant because the same dates-of-service and resources were used for both.
The Comptroller's office was prepared to make a recommendation to the legislators on the merits of keeping the Texas study because it provides different information on overpayment errors or eliminating the Texas study in lieu of the unfunded mandated federal study. However, when CMS published the proposed regulations for the mandated measurement study, the methodology was entirely different than the methods used by Texas and other participating states for the past three years. CMS is testing this new method in a final pilot study this fourth year of the demonstration project. Both the proposed mandated study and the pilot study have been renamed from PAM to PERM for Payment Error Rate Measurement.
The Comptroller's office and HHSC are participating in the PERM pilot. The Comptroller's office will provide the 2007 Texas Legislature with the results in the next biennial report and a recommendation for the most appropriate measurement study for meeting Texas' needs.
Medicaid FFS Study
The FFS study sampled 800 Medicaid clients' randomly selected health care services, including office visits, emergency room visits and medical supplies. The Comptroller selected these health care services from paid Medicaid claims from September 1 to November 30, 2002. The study included claims from the state's managed care program, Texas Health Network, a Primary Care Case Management (PCCM) program. The claims were grouped into eight categories based on the medical provider type and specialty of service. These categories are listed in the Medicaid FFS study section. Each client's sample service was selected by a date of service within the specified time frame. The study sample included the selected client's sample service and all other services performed by the provider on the sample date.
The Medicaid FFS study had 2,202 sample services with a total paid amount of $187,708. This sample was selected from 1,075,958 claims with a total paid amount of $135,214,047, from September 1 to November 30, 2002. The total number of FFS claims paid in 2002 was 22.4 million for a total paid amount of $3.9 billion.
Both this study and the March 2003 study's overpayment rates were 13 percent. Although HHSC implemented some of the recommendations from the Comptroller's March 2003 study, the recent reorganization and consolidation of health and human service agencies slowed the progress of these cost containment projects. Additionally, new and different types of provider error were identified in this study.
If this study's error rate is applied to annual expenditures, the total amount that may have been overpaid in the Medicaid FFS program for fiscal 2003 was about $544
Medicaid Vendor Drug Program (VDP) Study
This study reviewed randomly selected prescription services paid to 800 sample clients from October 1 to December 31, 2002. The Comptroller's project team reviewed the original copy of the sample prescriptions for compliance with state and federal requirements.
The Medicaid VDP study included 4,036 paid prescription services for a total paid amount of $191,888. This sample was selected from 1.3 million prescription claims for $62 million. The state paid 31.8 million VDP claims in 2002 for $2.3 billion. This was the second overpayment measurement study for this program.
Two types of errors accounted for the majority of overpayments. Out of the 916 identified errors, 488 pharmacies did not submit an original hard copy of a prescription with the documentation sent for the study review, 125 prescriptions did not meet the federal requirements for controlled substances and 375 prescriptions did not meet the state's requirements for a legitimate prescription. If the VDP study's error rate is applied to the annual VDP claim expenditures, the total at risk for overpayments in the Medicaid VDP for fiscal 2003 was about $491 million.
Medicaid Managed Care Capitation Payment Study
The agencies performed the managed care study as part of the second PAM study. Texas was one of five states helping CMS develop a methodology for measuring the appropriateness of the managed care capitated payments. The agencies selected random premium payments made by HHSC to their contracted Health Maintenance Organizations (HMO) from October 1 to December 31, 2002. The study reviewed the sample client's eligibility application for correct Medicaid program assignment and subsequent enrollment into a managed care HMO risk group. Two components of the HMO's premium payments are based on a client's residence locality and risk group assignment related to their Medicaid program type.
The managed care study had 1,067 sample clients with a total of $148,254 paid. The total premium amount paid to HMO's in 2002 was $867.5 million for a total of 577,913 clients.
The most common error identified was delayed client eligibility recertification due to either a client's reporting noncompliance or a caseworker's untimely review. This caused the client to be disenrolled, then re-enrolled into the managed care program, potentially disrupting the client's health care continuity.
If the study's error rate is applied to the annual premium expenditures, the total at risk for overpayments for fiscal 2003 was $18.9 million.
State Employees Worker's Compensation Program Study
The State Employees Worker's Compensation program study included a review of 200 randomly sampled medical bills paid on worker's compensation injury claims. These medical bills were paid from September 1 to November 30, 2003. The study reviewed the employee's claim, medical bills and medical records.
The State Employees Worker's Compensation program study had 732 medical bills for a total paid amount of $66,996. This sample was selected from 28,823 paid medical bills for $7,769,105. The total number of paid medical bills for 2003 was 117,343 for a total paid amount of $36.8 million.
If the study's error rate is applied to the annual State Employees Worker's Compensation medical bill expenditures, the total at risk for overpayments for fiscal 2003 is about $5 million.
In both this study and the March 2003 study, the project team had the actual medical records to confirm the delivery of the paid services. This error rate of 13.7 percent is a decrease of 6 percent from the March 2003 study. The study results indicate this is probably due to the cost containment efforts by SORM and the reimbursement changes implemented since the March 2003 study.
Summary of Medicaid FFS Recommendations
To improve the integrity of the Medicaid program, HHSC should:
- include additional procedure queries to their Medicaid Fraud and Abuse Detective System and office investigation to increase potential recoupments for the Medicaid program;
- pursue efforts to prevent providers from billing for services they did not provide as noted in 52 percent of the total overpayment errors identified in this study. HHSC is testing smart card technology to deter this type of abuse; and
- follow up with Texas Medicaid Healthcare Partnership to correct system processing areas that resulted in potential overpayments.
To conserve state funds, the Comptroller's office should:
- evaluate the feasibility of performing parallel state and federal measurement studies on the Medicaid fee-for-service program.
Summary of Medicaid VDP Recommendations
To improve the integrity of the Medicaid Vendor Drug program:
- the Office of Inspector General should continue to conduct regular audits on the Vendor Drug Program to identify potential fraudulent or abusive practices by either pharmacies or prescribing providers; and
- the Vendor Drug Program department should consider targeted pharmacy provider education on the state and federal prescription requirements for the corporate and independent pharmacies identified in this study and in any studies conducted by HHSC.
To conserve state funds, the Comptroller's office should:
- evaluate the feasibility of performing parallel state and federal measurement studies on the Medicaid Vendor Drug program.
Summary of Medicaid Managed Care Program Recommendations
To improve the integrity of the Medicaid Managed Care program, the Office of Eligibility Services should:
- develop a staff of eligibility reviewers to perform regular reviews of clients' applications and enrollments into the traditional Medicaid and Medicaid Managed Care programs; and
- use the review findings from these reviews to improve the Medicaid eligibility certification process through education of the caseworkers and through evaluation of the online eligibility verification system.
To conserve state funds, the Comptroller's office should:
- evaluate the feasibility of performing parallel state and federal measurement studies on the Medicaid managed care program.
Summary of State Employees Workers' Compensation Recommendation
To improve the integrity of the State Employees Workers' Compensation program, SORM should:
- continue the Medical Management Review Team's responsibilities for targeted reviews, evaluation of prior authorizations and timely identification of cases that should be managed medically to ensure the claimant receives the appropriate treatment and medical care; and
- consider requesting the new cost containment vendor, CorVel, to add claim processing edits and audits that target the most common coding errors.
With the exception of the managed care program, all of these recommendations can be implemented within current appropriation levels. HHSC officials will need to evaluate their resources to determine if these recommendations can be implemented within current appropriations.