DATA MINING IN PROGRAM DATA MINING IN PROGRAM INTEGRITY AND FRAUD INTEGRITY AND FRAUD

CONTROLCONTROLPHARMACEUTICAL REG AND PHARMACEUTICAL REG AND

COMPLIANCE CONGRESS COMPLIANCE CONGRESS 2008 2008

Jim SheehanJim Sheehan

New York Medicaid Inspector New York Medicaid Inspector GeneralGeneral

(518) 473-3782(518) 473-3782

jgs05@omig.state.ny.usjgs05@omig.state.ny.us

USUAL DISCLAIMERSUSUAL DISCLAIMERS

• GOOD IDEAS FROM MANY SOURCESGOOD IDEAS FROM MANY SOURCES

• ADOPTION IS COMPLIMENT, NOT ADOPTION IS COMPLIMENT, NOT PLAGIARISMPLAGIARISM

• FEEL FREE TO USE THESE SLIDES FEEL FREE TO USE THESE SLIDES WITHOUT ATTRIBUTIONWITHOUT ATTRIBUTION

• WHISTLEBLOWER CALLS CHEERFULLY WHISTLEBLOWER CALLS CHEERFULLY ACCEPTED-518 473-3782, ask for Jim ACCEPTED-518 473-3782, ask for Jim

DATA MINING-WHAT IS IT?DATA MINING-WHAT IS IT?

• ““Data miningData mining is the process of is the process of sortingsorting through through large amounts of data and picking out relevant large amounts of data and picking out relevant information. “ Wikipediainformation. “ Wikipedia

• Data mining for health care program integrity Data mining for health care program integrity combines claims data, encounter data, combines claims data, encounter data, demographic enrollment data, external database demographic enrollment data, external database data (e.g., Vital Statistics, licensing, provider-data (e.g., Vital Statistics, licensing, provider-internal data) with training, experience and internal data) with training, experience and intuition of auditors, investigators, health intuition of auditors, investigators, health professionals, compliance professionals, and professionals, compliance professionals, and (rarely) attorneys(rarely) attorneys

• Finding things you weren’t looking forFinding things you weren’t looking for• Goal-every professional a data miner Goal-every professional a data miner

MEDICAIDMEDICAID

• 2007 spending:$315 billion in US (state and federal)2007 spending:$315 billion in US (state and federal)• 2007 spending: $46 billion in New York (Kaiser 2007 spending: $46 billion in New York (Kaiser

Family Foundation)Family Foundation)• New York Medicaid spending growth has slowed New York Medicaid spending growth has slowed

recently, remains our safety net program for 4.5 recently, remains our safety net program for 4.5 millionmillion

• Majority of adult Medicaid enrollees in New York Majority of adult Medicaid enrollees in New York work; 1/3 of New York City residents (2.8 million) work; 1/3 of New York City residents (2.8 million) are enrolled in Medicaid (United Hospital Fund) are enrolled in Medicaid (United Hospital Fund)

• $4 billion direct prescription drug payer, even after $4 billion direct prescription drug payer, even after Part D Part D

• Significant prescription use in snfs and hospitals Significant prescription use in snfs and hospitals bundled into institutional ratesbundled into institutional rates

THE STATES FACE A FISCAL THE STATES FACE A FISCAL CRISIS CRISIS • NEXT YEAR-$12.5 Billion budget deficit-NEXT YEAR-$12.5 Billion budget deficit-

10% of current spending, in New York.10% of current spending, in New York.• Medicaid is 25% of state budgetsMedicaid is 25% of state budgets• Must cut at least $2 of Medicaid spending Must cut at least $2 of Medicaid spending

to save $1 to save $1 • Medicaid enrollees increase in a recessionMedicaid enrollees increase in a recession• Growing public concern-are these $ being Growing public concern-are these $ being

properly and wisely spentproperly and wisely spent• Elected officials-how can we cover the Elected officials-how can we cover the

deficit without voting to reduce benefits? deficit without voting to reduce benefits?

MEDICAID PROGRAM MEDICAID PROGRAM INTEGRITY-A QUALITY AND INTEGRITY-A QUALITY AND DATA PROBLEMDATA PROBLEM• MEDICAID IMPROPER PAYMENT RATE-18% MEDICAID IMPROPER PAYMENT RATE-18%

(PRELIMINARY PERM REPORT, 2007-17 state (PRELIMINARY PERM REPORT, 2007-17 state review) (NEW YORK WAS NOT REVIEWED)review) (NEW YORK WAS NOT REVIEWED)

• MEDICAID RATE PROBABLY OVERSTATED, BUT. . .MEDICAID RATE PROBABLY OVERSTATED, BUT. . .• CREDIT CARD LOSS RATE FROM IMPROPER CREDIT CARD LOSS RATE FROM IMPROPER

PAYMENTS-.07%PAYMENTS-.07%• USING DATA TOOLS AND SYSTEMS TO REDUCE USING DATA TOOLS AND SYSTEMS TO REDUCE

IMPROPER PAYMENTS AND UNNECESSARY OR IMPROPER PAYMENTS AND UNNECESSARY OR HARMFUL SERVICESHARMFUL SERVICES

• USING DATA TOOLS TO FOCUS ENFORCEMENTUSING DATA TOOLS TO FOCUS ENFORCEMENT

FISCAL CRISIS ADDS FISCAL CRISIS ADDS URGENCY URGENCY • INCREASED AUDITS OF PROCESSESINCREASED AUDITS OF PROCESSES• NO PAYMENT FOR NEVER EVENTSNO PAYMENT FOR NEVER EVENTS• WHAT IS THE OUTCOME WE ARE WHAT IS THE OUTCOME WE ARE

PAYING FOR?PAYING FOR?• WHAT TREATMENTS AND WHAT TREATMENTS AND

ORGANIZATIONS ARE EFFECTIVE?ORGANIZATIONS ARE EFFECTIVE?• WHAT TREATMENTS AND WHAT TREATMENTS AND

ORGANIZATIONS ARE EFFICIENT? ORGANIZATIONS ARE EFFICIENT?

SUCCESSFUL PROGRAM SUCCESSFUL PROGRAM INTEGRITY-ENCOURAGE THE INTEGRITY-ENCOURAGE THE LEADERS, REPORT DATA ON LEADERS, REPORT DATA ON THE LAGGARDS THE LAGGARDS • Between 1990 and 2005, the proportion of Between 1990 and 2005, the proportion of

residents physically restrained in nursing homes residents physically restrained in nursing homes dropped from 40% to 4%. dropped from 40% to 4%.

• Cystic fibrosis survival rates/Iraq battlefield Cystic fibrosis survival rates/Iraq battlefield survival discussed by Dr. Atul Gawande in Better.survival discussed by Dr. Atul Gawande in Better.

• Quality in hospitals (as measured by avoidable Quality in hospitals (as measured by avoidable deaths) has consistently improved in the best deaths) has consistently improved in the best institutions over past five yearsinstitutions over past five years– Focus on specific issues-central line infections, Focus on specific issues-central line infections,

pneumonia vaccine, antibiotics at surgery, medication pneumonia vaccine, antibiotics at surgery, medication errors, bedsores errors, bedsores

– Focus on never events and never paymentsFocus on never events and never payments– Focus on discharge and readmission with same diagnosis Focus on discharge and readmission with same diagnosis – Report cards and comparisons Report cards and comparisons

MEDICAID-GOVERNMENT MEDICAID-GOVERNMENT LAGGARDS IN PROGRAM LAGGARDS IN PROGRAM INTEGRITY AND DATAINTEGRITY AND DATA• 2003-GAO REPORT “ GAO added Medicaid 2003-GAO REPORT “ GAO added Medicaid

to its list of high-risk programs, owing to the to its list of high-risk programs, owing to the program's size, growth, diversity, and fiscal program's size, growth, diversity, and fiscal management weaknesses.” See,GAO, Major management weaknesses.” See,GAO, Major Management Challenges and Program Management Challenges and Program Risks: Department of Health and Human Risks: Department of Health and Human Services, GAO-03-101 (Washington, D.C.: Services, GAO-03-101 (Washington, D.C.: January 2003). “ We noted that insufficient January 2003). “ We noted that insufficient federal and state oversight put the Medicaid federal and state oversight put the Medicaid program at significant risk for improper program at significant risk for improper payments. “payments. “

2006 DEFICIT REDUCTION 2006 DEFICIT REDUCTION ACTACT• CMS-SIGNIFICANT MEDICAID FUNDS AND CMS-SIGNIFICANT MEDICAID FUNDS AND

STAFFSTAFF• STATES-SUPPORT AND OVERSIGHTSTATES-SUPPORT AND OVERSIGHT• CONTRACTORS-FOR AUDITS, EVALUATION, CONTRACTORS-FOR AUDITS, EVALUATION,

INVESTIGATIONSINVESTIGATIONS• DATA MINING GROUP AT CMS-ALGORITHM DATA MINING GROUP AT CMS-ALGORITHM

DEVELOPMENTDEVELOPMENT• MOVE FOCUS FROM LAW ENFORCEMENT MOVE FOCUS FROM LAW ENFORCEMENT

(OIG) TO PROGRAM AGENCY (OIG) TO PROGRAM AGENCY

PROGRAM INTEGRITY MEANS A PROGRAM INTEGRITY MEANS A FOCUS ON EFFECTIVE FOCUS ON EFFECTIVE COMPLIANCE PROGRAMSCOMPLIANCE PROGRAMS• NY-mandatory “effective” compliance programs NY-mandatory “effective” compliance programs • Failure to have effective compliance program is Failure to have effective compliance program is

basis for exclusion basis for exclusion • ““effective” compliance program requires disclosure effective” compliance program requires disclosure

to state of overpayments received, when identifiedto state of overpayments received, when identified• ““effective” compliance program requires risk effective” compliance program requires risk

assessment, audit and data analysis,remedial assessment, audit and data analysis,remedial measures measures

• ““effective” compliance program requires response effective” compliance program requires response to issues raised through hotlines, employee issues to issues raised through hotlines, employee issues

Program Integrity and Data Program Integrity and Data Mining SystemsMining Systems

• Data mining is a developing area – Data mining is a developing area – processing speed doubles every two years, processing speed doubles every two years, software and analytic approaches move at software and analytic approaches move at same speed. same speed.

• Existing state data systems, at best, reflect Existing state data systems, at best, reflect reliable, tested systems and the state-of-reliable, tested systems and the state-of-the-art at the time of procurement. Existing the-art at the time of procurement. Existing New York systems procured five years ago, New York systems procured five years ago, began operating three years ago.began operating three years ago.

• Significant opportunities for post-payment Significant opportunities for post-payment recoveries recoveries

PHARMA PROVIDES STRONG PHARMA PROVIDES STRONG DATA MINING OPPORTUNITIESDATA MINING OPPORTUNITIES

• STANDARDIZED PRODUCTS/CODESSTANDARDIZED PRODUCTS/CODES• STANDARDIZED INDICATIONS (FDA STANDARDIZED INDICATIONS (FDA

AND COMPENDIA)AND COMPENDIA)• HIGHLY AUTOMATED REAL-TIME HIGHLY AUTOMATED REAL-TIME

BILLING AND PAYMENTBILLING AND PAYMENT• THREE PARTICIPANTS IN EVERY THREE PARTICIPANTS IN EVERY

PRESCRIPTION TRANSACTION-PRESCRIPTION TRANSACTION-PHYSICIAN, PHARMACIST, PATIENTPHYSICIAN, PHARMACIST, PATIENT

EARLY DATA MINING PHARMA EARLY DATA MINING PHARMA OPPORTUNITIESOPPORTUNITIES

• Prescriptions for deceased enrolleesPrescriptions for deceased enrollees

• Returns to stock not reportedReturns to stock not reported

• Use of incorrect physician identifier “to get Use of incorrect physician identifier “to get claim paid”claim paid”

• False billing of usual and customary chargeFalse billing of usual and customary charge

• Prescribing to patients whose demographics Prescribing to patients whose demographics and diagnoses far from approved useand diagnoses far from approved use

• Unbelievable persistence rates Unbelievable persistence rates

DATA MINING TECHNIQUESDATA MINING TECHNIQUES

• Claims analysis-5 years, $200 billion in claims in data Claims analysis-5 years, $200 billion in claims in data warehousewarehouse

• Patient demographic feed and match-age, sex, marital status, Patient demographic feed and match-age, sex, marital status, addresses, licensing, ssnsaddresses, licensing, ssns

• Electronic diagnostic and treatment feeds-ICD-9s, DRGs, key Electronic diagnostic and treatment feeds-ICD-9s, DRGs, key words-claims, managed care encounters, authorizationswords-claims, managed care encounters, authorizations

• Meta-analyses of previous studies for given disease conditions. Meta-analyses of previous studies for given disease conditions. Dr. Charles Bennett, Northwestern compiled and examined Dr. Charles Bennett, Northwestern compiled and examined data from 51 prior studies in more than 13,000 patients –found data from 51 prior studies in more than 13,000 patients –found 10% higher death rates for cancer patients using epo 10% higher death rates for cancer patients using epo

• Geographic analysis for sales, patients, providers, relationshipsGeographic analysis for sales, patients, providers, relationships• Modality analysis-which physicians use injectibles? Which Modality analysis-which physicians use injectibles? Which

physicians are early adopters? Which physicians use lab and physicians are early adopters? Which physicians use lab and physiological diagnostic tests? physiological diagnostic tests?

• Who should suggest, fund, review data mining? Who should suggest, fund, review data mining?

DATA MINING TECHNIQUESDATA MINING TECHNIQUES

• PROVIDER ANALYSIS-IBM FAMS CLAIMS SURGES AND PROVIDER ANALYSIS-IBM FAMS CLAIMS SURGES AND OUTLIERS (this provider behaves differently from similar OUTLIERS (this provider behaves differently from similar providers)providers)

• PROVIDER ANALYSIS-”FAIR ISAAC” type RISK SCORING (use PROVIDER ANALYSIS-”FAIR ISAAC” type RISK SCORING (use proprietary models with multiple tools)proprietary models with multiple tools)

• REGRESSION ANALYSIS AND ALGORITHM BUILDING-(If it REGRESSION ANALYSIS AND ALGORITHM BUILDING-(If it happened this way the last time, we predict it will happen happened this way the last time, we predict it will happen this way again) this way again)

• FUZZY LOGIC-NOT BINARY (yes/no) BUT “SOMEWHAT” (190 FUZZY LOGIC-NOT BINARY (yes/no) BUT “SOMEWHAT” (190 lbs. is “somewhat” heavy, “somewhat” normal for adult lbs. is “somewhat” heavy, “somewhat” normal for adult male, total cholesterol of 220 is “somewhat” high)male, total cholesterol of 220 is “somewhat” high)

• NEURAL NETWORKS-SYSTEM THAT “LEARNS” THROUGH NEURAL NETWORKS-SYSTEM THAT “LEARNS” THROUGH PATTERN RECOGNITION AND NONLINEAR SYSTEM PATTERN RECOGNITION AND NONLINEAR SYSTEM IDENTIFICATION AND CONTROL. (“BLINK” by Malcolm IDENTIFICATION AND CONTROL. (“BLINK” by Malcolm Gladwell, elected officials and risk tolerance ) Gladwell, elected officials and risk tolerance )

Data Mining ApproachesData Mining ApproachesData Matches/DemographicsData Matches/Demographics

• Men having babiesMen having babies

• Fillings in crownsFillings in crowns

• Deceased enrolleesDeceased enrollees

• Drugs used for patients with primary Drugs used for patients with primary diagnoses of anxiety or depressiondiagnoses of anxiety or depression

Data Mining for PatientsData Mining for Patients

• Unanticipated deaths in hospital, snf Unanticipated deaths in hospital, snf and prescription historyand prescription history

• Off-label use and better outcomesOff-label use and better outcomes• Experimental use and consent-by Experimental use and consent-by

facility, by ordering physicianfacility, by ordering physician• Third party liability for treating Third party liability for treating

adverse eventsadverse events• Capturing adverse eventsCapturing adverse events

Data Mining Quality ToolsData Mining Quality ToolsProviders Not Meeting Minimum Providers Not Meeting Minimum

StandardsStandards• Never eventsNever events

• Unreported adverse eventsUnreported adverse events

• Unreported adverse Unreported adverse outcomes/unanticipated deathsoutcomes/unanticipated deaths

• Ranking/rating facilities-audit focus Ranking/rating facilities-audit focus

• Condition of participation failures Condition of participation failures (structure)(structure)

• Drug outcomes in populations and in Drug outcomes in populations and in facilitiesfacilities

DATA MINING FOR DATA MINING FOR RELATIONSHIPS AND RELATIONSHIPS AND DISCLOSUREDISCLOSURE• FIVE KINDS OF REPORTINGFIVE KINDS OF REPORTING

• IRS FORM 990 (2008 version) and reporting questions IRS FORM 990 (2008 version) and reporting questions Does the organization have a written conflict of interest policy? If “Yes”:, how many transactions did the organization review under this policy and related procedures during the year?” If “Yes”: Are officers, directors or trustees, and key employees required to disclose annually interests that could giver rise to conflicts? Does the organization regularly and consistently monitor and enforce compliance with the policy?

• Company websites (voluntary or required by CIA)Company websites (voluntary or required by CIA)• State law required disclosuresState law required disclosures• patient consent disclosurespatient consent disclosures• IRB disclosuresIRB disclosures

PUBLIC INFORMATION AND PUBLIC INFORMATION AND REPORTINGREPORTING

• Using data mining for CMS review, Using data mining for CMS review, MFCU work, public informationMFCU work, public information

• State legislatures, budget offices, State legislatures, budget offices, meeting budget shortfalls meeting budget shortfalls

• What happens if . . .What happens if . . .

• HIPAA compliance HIPAA compliance

THE FUTURE OF MEDICAID THE FUTURE OF MEDICAID INVESTIGATION THROUGH INVESTIGATION THROUGH DATA MININGDATA MINING• TEST WITNESS ALLEGATIONSTEST WITNESS ALLEGATIONS• PROJECT WITNESS ALLEGATIONS IN PROJECT WITNESS ALLEGATIONS IN

TARGET TARGET • USE WITNESS’ SUPPORTED ALLEGATIONS USE WITNESS’ SUPPORTED ALLEGATIONS

AGAINST SIMILAR ENTITIES WHERE DATA AGAINST SIMILAR ENTITIES WHERE DATA SUPPORTS SUPPORTS

• PERSUADE DECISION MAKERS AND PERSUADE DECISION MAKERS AND DEFENSE OF MERITSDEFENSE OF MERITS

• DEVELOP EVIDENTIARY SUPPORT FOR DEVELOP EVIDENTIARY SUPPORT FOR LITIGATION LITIGATION

THE FUTURE OF MEDICAID THE FUTURE OF MEDICAID PROGRAM INTEGRITY PROGRAM INTEGRITY THROUGH DATA MININGTHROUGH DATA MINING• IDENTIFY AND COMMUNICATE BEST IDENTIFY AND COMMUNICATE BEST

OUTCOMES FROM DATA MININGOUTCOMES FROM DATA MINING• IDENTIFY AND COMMUNICATE IDENTIFY AND COMMUNICATE

COMPLIANCE DATA ANALYSIS PROCESSES COMPLIANCE DATA ANALYSIS PROCESSES WHICH WILL IDENTIFY PROBLEM AT WHICH WILL IDENTIFY PROBLEM AT SOURCESOURCE

• IDENTIFY AND COMMUNICATE ISSUES IDENTIFY AND COMMUNICATE ISSUES IDENTIFIED THROUGH DATA MININGIDENTIFIED THROUGH DATA MINING

• TRAIN AND EQUIP EMPLOYEES AND TRAIN AND EQUIP EMPLOYEES AND ORGANIZATIONS IN DATA ANALYSIS ORGANIZATIONS IN DATA ANALYSIS TECHNIQUES TECHNIQUES