The Redesigned National Hospital Discharge Survey

1
The Redesigned National Hospital Discharge Survey

• National Center for Health Statistics
• Division of Health Care Statistics
• Hospital Care Team
• Last Updated November 19, 2008

2
NHDS Background

• Nationally representative sample of inpatient
  discharges
• Longest continuously fielded survey of inpatient
  care - annually since 1965
• Nonfederal, general, and short-stay hospitals

3
Inpatient Hospitalization rates with Clostridium
difficile first and any listed 1996-2006

• Citation McDonald LC, et al. Emerg Infect Dis.
  200612(3) 410-5.

4
Coronary Artery Bypass Graft (CABG) and
Percutaneous Coronary Intervention (PCI)1980-2004

Rate per 10,000 population
PCI
CABG
Source National Hospital Discharge Survey
1980-2004.
5
Average Length of Stay by Age U.S., 1970-2006
Source CDC/NCHS, National Hospital Discharge
Survey (NHDS)
6
Context for Redesign

• Current data limited to Uniform Bill (claims
  data)
• More flexible survey needed
• To address policy and research issues
• To reflect different patient and health-care
  environment
• HCUP and others built on administrative data

7
Redesign Priorities

• Resource use/cost/billings/payments
• Clinical depth
• medications, laboratory results
• Improved patient demographics
• race, ethnicity, address for geocoding
• Health-related outcomes
• death index, readmissions, Medicare data

8
Current vs. Redesigned NHDS
9
Scope

• Current
• Non Federal
• Non institutional
• 6 beds or more
• General
• Short-stay (LOS lt30 days)

• Redesigned
• Non Federal
• Non institutional
• 6 beds or more

10
Content, Inpatient Discharges

• Current (UB 04 only)
• Patient age, gender, race/ethnicity, marital
  status
• Source and type of admission
• Diagnoses and procedures
• Discharge status
• Length of stay
• Expected payment source

• Redesigned
• All current data
• More clinical depth
• Laboratory results
• Medications
• Vital signs
• ICU use
• Payments, billings
• Patient identifiers
• Special/sponsored topics

11
Content, Hospital Level

• Current (UB 04 only)
• Not collected

• Redesigned
• Specific technologies/services
• EMR specifics
• Uncompensated care
• Staffing

12
Sampling

• Current
• 112 geographic primary sampling units (PSUs)
• 239 hospitals
• Hospitals stratified by bed size
• Annual discharges
• 250 per manual hospital
• Oversample automated
• 150,000 discharges annually

• Redesigned
• List sample stratified by service type
• 240 hospitals
• Stratification of discharges
• Observation cases
• Normal newborns (undersample)
• Acute myocardial infarction
• Deaths
• Other patients
• 24,000 annually

13
Redesigned Survey Potential

• Link to NDI, birth records, CMS data
• Risk adjustment possible
• Quality of care analyses
• Payment and cost analyses
• Facility-level analyses

14
Potential AnalysesQuality of Care/Patient Safety

• Discharge medications by patient and provider
  characteristics
• Effects of antibiotic resistance patterns within
  the hospital on patient outcomes
• Effects of surgeon specialty on health-related
  outcomes
• Readmission and post-discharge mortality rates
  for selected diagnoses or procedures by patient
  and provider characteristics

15
Potential Economic Analyses

• Variation in payment by payer
• Variation in payment for similar diagnoses or
  procedures by patient and provider
  characteristics
• Relationship between payment rates and quality of
  care for specific diagnoses or procedures
• Relationship among billings, costs, and payments
• Effects of case mix on profitability (specialty
  hospitals)

16
Potential AnalysesSponsored Topics

• True rate of hospitalized acute myocardial
  infarction
• Disparities in do-not-resuscitate use among
  in-hospital deaths
• Variation in nosocomial bloodstream infections by
  hospital characteristics

17
For more information

• Contact
• Carol DeFrances, Ph.D.
• Hospital Care Team Leader
• National Center for Health Statistics
• 3311 Toledo Road
• Hyattsville, MD 20782
• 301-458-4440
• E-mail cdefrances_at_cdc.gov