Ambulatory Care Sensitive Condition Hospitalizations Among Elderly Medicare and Medicaid Dual Enroll

1
Ambulatory Care Sensitive Condition
Hospitalizations Among Elderly Medicare and
Medicaid (Dual) Enrollees

• Marlene Niefeld, Ph.D., M.P.P.
• Johns Hopkins University
• Bloomberg School of Public Health
• June 8-9, 2004

2
Acknowledgements

• Dissertation Committee
• Judy Kasper, advisor
• Marie Diener-West
• Darrell Gaskin
• Bruce Leff
• Alternates Chad Boult, Manning Feinleib
• Jerry Anderson, Eric Slade, Lynda Burton, Chris
  Forrest, Jonathan Wiener, Wenke Huang, Andy Shore
• This work was supported by a Dissertation
  Fellowship Award from the Agency for Healthcare
  Research and Quality (AHRQ).

3
Definition of Ambulatory Care Sensitive Condition
(ACSC) Hospitalization

• ACSC hospitalizations are hospitalizations for
  conditions that, if treated in a timely and
  appropriate manner in an ambulatory care setting,
  would usually not require inpatient admission
  (Institute of Medicine, 1993)
• ACSCs defined by panels of physicians
• Conditions include angina, asthma, cellulitis,
  COPD, congestive heart failure, dehydration,
  diabetes, gastroenteritis, grand mal status,
  hypertension, hypoglycemia, kidney/urinary tract
  infections, pneumonia, severe ear, nose and
  throat infections

4
Objectives

• To evaluate the prevalence of ambulatory care
  sensitive condition (ACSC) hospitalizations among
  community-dwelling Medicare and Medicaid (dual)
  enrollees aged 65 and older
• To examine the relationship between preventable
  hospitalizations and access to care barriers
  among elderly dual enrollees

5
Background Dual Enrollees

• 2.9 million community-dwelling dual enrollees
  aged 65 and over in 1999 (Shatto 2001)
• More than three-quarters of dual enrollees had
  incomes less than 10,000 per year (Shatto 2001)
• Compared to other Medicare beneficiaries, dual
  enrollees had higher rates of self-reported
  fair/poor health status, activity of daily living
  (ADL) limitations, and certain chronic conditions
  (e.g., hypertension, stroke, diabetes) (Kasper,
  Elias, Lyons 2004 OBrien, Rowland, Keenan
  1999).

6
Why Should We Be Concerned with Access to Care
for Elderly Dual Enrollees?

• Increasing ACSC hospitalization rates over time
• (Kozak, Hall, Owings 2001).
• Evidence of access to care problems by race and
  procedure (Blustein and Weiss 1998 Miller 1992
  McMillan and Gornick 1984 Escarce, Epstein,
  Colby et al. 1993 Ayanian, Udvarhelyi, Gatsonis
  et al. 1993)
• Out-of-pocket cost burden for some dual enrollees
  (Mojtabia and Olfson 2003 Sambamoorthi, Shea,
  and Crystal 2003 Crystal et al. 2000 Gross et
  al. 1999 Lillard, Rogowski, and Kington 1999).
• Medicaid/QMB beneficiaries were less likely to
  receive needed follow-up after hospital discharge
  than other Medicare beneficiaries (Moy and Hogan
  1993).

7
Data

• 1999 Six-State Survey of Elderly Dual Enrollees
  (n2,128)
• 1999 and 2000 Medicare Inpatient Claims Data
• 2001 Area Resource File
• The six states were chosen based on regional
  variation, their level of spending on Medicaid
  home and acute care, Medicare home care
  expenditures, and the distribution of elderly
  dual enrollees in nursing home vs. community
  settings in the selected states.
• Age stratified random sample was selected in each
  state (65-79, 80)
• Overall response rate of 76.6
• Proxies answered the survey for 28 of the sample

8
Methods

• Used descriptive statistics to explore the
  characteristics of the population, prevalence of
  ACSC hospitalizations and access barriers, and
  utilization of hospital services (weighted).
• Logistic regression analyses (unweighted,
  adjusted for sample design (Korn and Graubard
  1991)).
• Missing Data

9
Hospitalizations
ACSC Hospitalizations 25
Non-ACSC Hospitalizations 75
All Hospitalizations in the 12 Months Following
Survey interview Number of Hospitalizations
Weighted (n) 100,299 Mean 4.49
Source 1999 2000 Medicare Inpatient Claims
Data matched to the 1999 Six-State Survey of
Elderly Dual Enrollees.
10
Proportion of Dual Enrollees with Any
Hospitalization and ACSC Hospitalization
ACSC Hospitalization
8
Source The 1999 Six-State Survey of Elderly
Dual Enrollees 1999 2000 Medicare Inpatient
Claims Data.
11
ACSC HospitalizationMultivariable Logistic
Regression Model Odds Ratios
Race, Educational Attainment, Living
Arrangement, Perceived Health Status, Proxy
Respondent, HPSA, Nursing Facility Beds per
person Age 65, and Predicted Organizational
Access Barrier controlled for in the
analysis. Source The 1999 Six-State Survey of
Elderly Dual Enrollees.
12
Discussion

• High numbers of comorbid conditions are
  associated with increased likelihood of
  preventable hospitalization (Wolff, Starfield,
  and Anderson 2002 Niefeld, Braunstein, Wu et al.
  2003 Bynum, Rabins, Weller et al. 2004).
• Prescription drugs (Mojtabai and Olfson 2003
  McDonnel and Jacobs 2002)
• Compliance -- poor adherence was associated with
  poorer health and higher rates of hospitalization
  
• Cost
• Drug-drug interactions
• Readmissions for diabetes, bronchial asthma, and
  congestive heart failure could indicate quality
  of care problems such as inadequate discharge
  planning and follow-up (Benbassat and Taragin
  2000 Vinson, Rich, Sperry et al. 1990).
• African-American race not significant in this
  model.

13
Limitations

• Community-based sample results cannot be
  generalized to the national population of dual
  enrollees.
• Could not identify Medicaid recipients with
  full-benefits and those receiving cost-sharing
  but no additional benefit coverage.
• Cross-sectional data (relationships should be
  interpreted as associations, not causal).

14
Conclusions

• Greater coordination of care needed for people
  with multiple chronic conditions.
• Out-of-pocket costs should be reduced or
  eliminated for dual enrollees as it is associated
  with preventable hospitalizations.