Urologic Diseases in America

1
Urologic Diseases in America

• Available Datasets

2
Urologic Diseases in America

• Mission
• 1. Define the burden of illness posed on the
  nation by the major urologic conditions
• 2. Use existing data to inform public policy,
  identify promising areas for new research,
  identify existing health care quality problems

3
Defining Burden of Illness

• Prevalence and incidence
• Inpatient stays
• Hospital outpatient visits
• Physician office visits
• Ambulatory surgery visits
• Emergency room visits
• Nursing home admissions
• Direct costs (national and Medicare)
• Indirect costs

4
Types of UDA datasets

• Nationally-representative
• Claims-based
• Special populations

5
UDA Datasets

• Nationally representative datasets
• Healthcare Cost and Utilization Project-
  Nationwide Inpatient Sample
• National Ambulatory Medical Care Survey
• National Hospital Ambulatory Medical Care Survey
• National Survey of Ambulatory Surgery
• Surveillance, Epidemiology, and End Results
• National Health and Nutrition Examination Survey
• Medical Expenditure Panel Survey
• National Nursing Home Survey

6
National Health and Nutrition Examination Survey
(NHANES)

• Maintained by the National Center for Health
  Statistics
• First released as NHANES I, II, III
• Now released every two years
• Population-based survey of households
• Mobile Examination Center allows physical and
  laboratory data collection after household
  interview

7
National Health and Nutrition Examination Survey
(NHANES)

• In-person interview provides comprehensive
  sociodemographic, dietary and medical history
• Each survey has a few urology questions
• (ED?Urinary Incontinence and BPH)
• Comprehensive labs done
• DEXA scanning, audiology, etc

8
Strengths and Limitations

• Strengths
• Clinically detailed, nationally-representative
  data
• Ability to describe minority health issues
• Environmental exposures
• Possible link to other datasets
• Limitations
• No longitudinal data
• Limited scope of urologic conditions

9
Healthcare Cost and Utilization Project
(HCUP)Nationwide Inpatient Sample (NIS)

• Maintained by the Agency for Healthcare Quality
  and Research
• Nationally representative data on hospital
  inpatient stays (20 stratified sample of
  hospitals in the US)
• Unit of analysis is the hospital discharge
• http//hcupnet.ahrq.gov/
• Can be linked to AHA and Area Resource File
  databases

10
HCUP-NIS

• Largest collection of longitudinal hospital care
  data in the United States
• Can be used to identify, track, and analyze
  national trends in access, charges, quality
• The only national hospital database with charge
  information on all patient stays, regardless of
  payer

11
HCUP-NIS

• 6-7 million stay records (37 states represented)
• Over 100 variables, including

Primary and secondary diagnoses Primary and
secondary procedures Admission and discharge
status Patient demographics Expected payment
source Total charges Length of stay Hospital
characteristics (e.g., ownership, size, teaching
status)
12
Some topics that can be illuminated by HCUP

• Access to care
• Complications of care
• Surgical volume/outcome relationships
• Diffusion of technologies
• Practice pattern variation

13
Strengths and Limitations

• Strengths
• Large sample, ability to describe inpatient
  procedure experience for many GU conditions
• Population-based
• Charge data
• Limitations
• No longitudinal data
• ICD-9 procedure coding only
• Charge data

14
Kids Inpatient Database (KID)

• HCUP-NIS for pediatric discharges
• Nationally representative sample of peds
  discharges (2-3 million discharges)
• Conducted 1997, 2000, 2003
• Strengths and Limitations similar to NIS

15
National Ambulatory Medical Care Survey (NAMCS)

• Maintained by the National Center for Health
  Statistics
• Nationally representative sample of physician
  office visits
• Unit of analysis is the visit
• Sample of patient visits is characterized
  during a
• 1-week survey period

16
National Hospital Ambulatory Medical Care Survey
(NHAMCS)

• Maintained by the national center for health
  statistics
• Nationally-representative sample of ambulatory
• care services in hospital emergency and
  outpatient departments
• Unit of analysis is the visit
• Each patient visit is characterized during a
  4-week survey period

17
NHAMCS and NAMCS

• Variables recorded include
• age, sex, race, ethnicity
• patients symptoms, complaints or other reasons
  for the visit
• physicians diagnoses
• diagnostic and therapeutic services
• medications
• expected sources of payment
• visit disposition

18
Some topics that can be illuminated by
NAMCS/NHAMCS

• Use of physician services for GU conditions by
  race and gender
• Medication practice patterns
• Treatment of GU conditions by non-urologists
• Practice pattern variations

19
Strengths and Limitations

• Strengths
• Captures physician subspecialties that may
  encounter urologic conditions
• Large, nationally representative portrait of
  outpatient care, for all types of insurance
• Limitations
• Limited data on procedures (ICD-9 coding) and
  testing
• No longitiudinal data
• Often required combining cells across demographic
  strata or years to achieve adequate counts

20
Surveillance, Epidemiology, End Results Database
21
Surveillance, Epidemiology, End Results Database
(SEER)

• Maintained by National Cancer Institute and
  Centers for Disease Control
• Covers about 26 of the population
• SEER population is somewhat more urban and
  foreign-born than the general population
• Collects patient demographics, tumor site,
  histology, stage, initial treatment, vital status

22
Strengths and Limitations

• Strengths
• Only comprehensive source of population-based
  data on cancer stage at diagnosis as well as
  cancer mortality
• Limitations
• Limited follow up data
• VA participation?

23
National Nursing Home Survey (NNHS)

• Maintained by National Center for Health
  Statistics
• National sample surveys of nursing homes, the
  providers of care, and their residents
• Sample size
• 1,500 facilities
• 8,100 residents
• Information is provided on the recipients of
  care, including demographics, health status, and
  services received
• 1995. 1997, 1999, 2004

24
Strengths and Limitations

• Strengths
• Representative data on a vulnerable population
• Many GU conditions in the elderly
• Limitations
• No longitudinal data
• Little clinical detail

25
Medical Expenditure Panel Survey (MEPS)

• Source Agency for Healthcare Research and
  Quality
• Nationally representative survey of health care
  use, expenditures, sources of payment, and
  insurance coverage for the US civilian
  non-institutionalized population
• Provides information on the financing and
  utilization of medical care in the United States
• Sample size 10,000 families (or 24,000
  individuals)
• Survey is continuous, population-based

26
MEPS

• MEPS household interview components
• health conditions, health status, use of medical
  services, charges and source of payments, access
  to care, satisfaction with care, health insurance
  coverage, income, and employment
• Followed up by confirmation/supplementation from
  providers, employers, insurers

27
Strengths and Limitations

• Strengths
• Outpatient prescription drug expenditures
• Detailed and reliable expenditure data
• Limitations
• Conditions identified at the 3-digit ICD-9 level
• Small sample to detect many GU conditions

28
National Survey of Ambulatory Surgery

• Nationally-representative data regarding
  freestanding and hospital-based ambulatory
  surgery centers
• ICD-9 diagnosis and procedure codes
• Data only from 1994-96
• HCUP has a State Ambulatory Surgery Database with
  only hospital-based surgeries

29
UDA datasets Special populations

• Special populations
• National Association of Childrens Hospitals and
  Related Institutions
• Society of Assisted Reproductive Technology
  database

30
National Association of Childrens Hospitals and
Related Institutions (NACHRI) database

• NACHRI dataset contains information on all
  inpatient stays at 58 member hospitals, including
  approximately 2 million pediatric inpatient
  discharges
• Variables of interest diagnosis, demographics,
  length of stay, total charges, and cost-to-charge
  ratio
• Limited detail for substantive analyses
• 1999- onward

31
Society for Assisted Reproductive Technologies
(SART) database

• SART is a professional society which collects
  data from fertility clinics across the nation, in
  concert with CDC
• Demographics, outcomes, indications for ART use
• 1999 data
• Access is by request

32
UDA Datasets Claims-based

• Centers for Medicare and Medicaid Services
• Marketscan
• Ingenix
• Innovus/I3 database

33
Centers for Medicare and Medicaid Services (CMS)

• Inpatient Stays/ Medicare Provider Analysis and
  Review (MedPAR) (5 sample)
• Contains claims for Medicare beneficiaries using
  hospital inpatient services
• Outpatient Hospital Claims (5 sample)
• Contains claims for Medicare beneficiaries using
  hospital outpatient services
• Physician/Supplier Part B (5 sample)
• Contains claims for Medicare beneficiaries using
  physician services

34
Strengths and Limitations

• Strengths
• Enormous database describing healthcare
  utilization for vast majority of Americans 65 and
  over
• Common Procedural Terminology (CPT) codes
• Detailed expenditure data
• Ability to follow individuals over time
• Limitations
• Lack of clinical detail
• Only captures those who receive care
• Lack of outpatient medication information
• Excludes those in HMOs

35
SEER-Medicare linkage

• Linkage available for 1991-2002 incident cases to
  2005 claims (2006 update coming)
• Links clinical data from SEER (stage, grade) with
  utilization data from CMS
• Data in house on renal, bladder, and prostate
  cancers
• Specific permission must be obtained from NCI for
  each analysis.

36
Strengths vs Limitations

• Strengths
• Ability to combine clinical detail from SEER with
  longitudinal utilization data from Medicare
• Look at costs, disparities in care, variations in
  care, technology diffusion
• Limitations
• Limited to the cancer experience of the elderly
• No quality of life data

37
MarketScan

• Dataset of claims from 100 health plans serving
  Fortune 500 employers
• Enables evaluation of productivity and pharmacy
  data and associated medical claims information
• Unique source of indirect cost data
• Patients experience may not be
  nationally-representative
• Many GU conditions not well represented

38
Ingenix

• Includes 1.8 million enrolled employees and their
  dependents
• Provides detailed financial information, such as
  procedure and diagnosis codes and plan costs
• Copays, deductibles included
• Not nationally-representative
• Used in first UDA project to model incremental
  costs associated with a diagnosis (controls for
  age, sex, zip code median income, plan type,
  comorbidities)

39
Innovus i3 database
40
Strengths and Limitations

• Strengths
• Ability to follow individuals through 5 years
• 30 million covered lives
• Unique lab data
• Limitations
• Non-representative
• Lab data are inconsistently reported