Preventable Hospitalizations: Assessing Access and the Performance of Local Safety Net

1
Preventable Hospitalizations Assessing Access
and the Performance of Local Safety Net

• Presented by
• Yu Fang (Frances) Lee
• Feb. 9th, 2007

2
Outline

• Preventable hospitalizations
• Connection between local safety net capacity and
  preventable hospitalizations
• Research proposals

3
Preventable Hospitalizations

• Avoidable Hospitalizations
• Avoidable Hospital Conditions (AHCs)
• Ambulatory-Care Sensitive Conditions
• (ACSCs or ACS Conditions)

4
Preventable Hospitalizations

• Inpatient treatments of conditions for which
  timely and effective use of primary care should
  have reduced the risk of hospitalizations
• (Texas Health Care Information Collection, 2003)

5
Preventable Hospitalizations

• A list of ICD-9-CM codes based on hospital
  discharge data
• Usually measured as rates of admission to the
  hospitals
• Different technical definitions of PH
• Billings et al. (1993)
• Weissman et al. (1992)
• Agency for Healthcare Research and Quality
  (AHRQ) Prevention Quality Indicators

6
Preventable Hospitalizations
Resource http//www.qualityindicators.ahrq.gov
7
PQI 1 Diabetes Short-term Complication Admission
Rate

• Admissions for diabetic short-term complications
  per 100,000 population
• Relationship to quality Proper outpatient
  treatment and adherence to care may reduce the
  incidence of diabetic short-term complications
• Population at Risk Population in Metro Area or
  county, age 18 years and older
• Benchmark State, regional, or peer group
  average
• lower rates represent better quality care

8
PQI 1 Diabetes Short-term Complication Admission
Rate

• Numerator
• All non-maternal/non-neonatal discharges of age
  18 years and older, with ICD-9-CM principal
  diagnosis code for short-term complications
• Excluded cases
• transfer from other institution
• MDC 14 (pregnancy, childbirth, and puerperium)
• Denominator
• Population in Metro Area or county, age 18 years
  and older

9
Use of Preventable Hospitalizations

• To assess quality of health services in the
  community
• To identify unmet community health care needs
• To monitor how well complications from a number
  of common conditions are being avoided in the
  outpatient setting
• To compare the performance of local health care
  systems across communities

Resource Department of Health and Human
Services, Agency for Health Research and
Quality. (2006a). Guide to Prevention Quality
Indicators Hospital Admission for Ambulatory
Care Sensitive Conditions.
10
Preventable Hospitalizations

• Strengths
• minimal requirement for resources
• comparison across different levels
• Weaknesses
• strong relationship between PHs and SES
• limited evidence for each PH
• limited evidence on the effectiveness of
  treatments in reducing incidence of PH

Resource Department of Health and Human
Services, Agency for Health Research and
Quality. (2006a). Guide to Prevention Quality
Indicators Hospital Admission for Ambulatory
Care Sensitive Conditions.
11
Davidson et als Framework
Community Characteristics
Health care Access and Outcomes

• Safety-Net population
• Uninsured population
• Medicaid beneficiaries
• Vulnerable populations

• Potential access
• Usual source of care

• Safety-net Support
• Direct government private support of safety
  net
• Medicaid payment level

Low-income population support Medicaid
eligibility level

• Realized access
• Doctor visits
• Other health care

• Health care market
• Physician supply
• Managed care penetration
• Managed care competition

• Safety-net services
• Public hospitals
• Teaching hospitals
• Community clinics

• Access outcomes
• Preventable hospitalizations
• Other outcome indicators

Individual Characteristics

• Predisposing Need
  Enabling __
• Demographics Perceived
  Income
• Social factor Evaluated
  Health insurance
• Beliefs
  Usual source of care

Resource Davidson, P. L., Andersen, R. M., Wyn,
R., Brown, E. R. (2004). A framework for
evaluating safety-net and other community-level
factors on access for low-income populations.
Inquiry, 41(1), 21-38
12
Proposal 1 Small-Area Analysis

• Study population
• residents aged 18-64 in Harris County, Texas
  (2004 )
• Unit of Analysis
• ZIP code
• Datasets
• Project Safety Net (2004)
• Texas Health Care Information Collection (2004)
• Census 2000
• Data analysis
• linear regression

13
Proposal 1 Small-Area Analysis

• Objectives
• To investigate the rate of preventable
  hospitalizations by insurance type in Harris
  County, Texas
• To investigate the association between the
  capacity of primary care services and preventable
  hospitalizations for the low-income population
• To Investigate the association between the
  proximity to the nearest primary care services
  and preventable hospitalizations for the
  low-income population

14
Proposal 1 Local Safety Net and Preventable
Hospitalization at ZIP-code level
15
Proposal 2 Multi-Level Analysis

• Study population
• hospitalized, non-elderly (aged 18-64)
  low-income adults in Harris County, Texas (2004 )
• Unit of Analysis
• individual as level-one
• ZIP code as level-two
• Datasets
• Project Safety Net (2004)
• Texas Health Care Information Collection (2004)
• Census 2000
• Data analysis
• Multi-level logistic regression

16
Proposal 2 Multi-Level Analysis

• Objectives
• To learn about the fraction of total variability
  in preventable hospitalization at the individual
  level and at the community level for
  hospitalized, non-elderly low-income adults
• To analyze the association between the proximity
  to the nearest safety net clinic and preventable
  hospitalizations among the low-income population,
  after controlling for individual characteristics
  and community characteristics
• To analyze the association between the capacity
  of local primary care services and the
  preventable hospitalization among the
  hospitalized low-income population, after
  controlling for individual characteristics and
  area characteristics

17
Proposal 2 Multi-Level Analysis

• To compare the relative importance of health
  insurance and the proximity to the local safety
  net clinics in reducing the likelihood of
  preventable hospitalization for the hospitalized
  low-income, non-elderly adults
• To estimate the direct costs of preventable
  hospitalizations in Harris County for all
  low-income, non-elderly hospitalized population
  in 2004, and compare the average costs of
  preventable hospitalization in Harris County with
  those measured in the existing literature

18
Proposal 2 Local Safety Net and Preventable
Hospitalization at Two Different Levels
Hospitalized Safety- Net Population
Community Characteristics

• Population characteristics at zip-code level
• Uninsurance rate
• Education
• Area income at zip-code level

Access Outcome Preventable hospitalization
Safety-net Services at zip-code level
Primary-care capacity of safety net clinics

• Individual characteristics
• Predisposing Enabling
• Gender Health insurance
• Age Proximity to the nearest
  safety-net clinic
• Race

Reference Davidson, P. L., Andersen, R. M., Wyn,
R., Brown, E. R. (2004). A framework for
evaluating safety-net and other community-level
factors on access for low-income populations.
Inquiry, 41(1), 21-38
19
Summary

• Preventable hospitalization rate is based on
  hospital discharge data
• Preventable hospitalization is an access outcome,
  and is one of the main tools to monitor the local
  safety net
• Individual and community characteristics both
  contribute to preventable hospitalization