July 2001

Grant Results

SUMMARY

From 1991 to 1994, these grants provided support to the United Hospital Fund of New York to develop, validate and implement a research technique called small-area analysis, used to identify communities with high rates of unnecessary hospitalization and limited access to primary care.

During the first phase of the project investigators studied differences in hospitalization rates for "ambulatory-care-sensitive" (ACS) conditions, such as bacterial pneumonia and otitis media (a severe ear infection), which can be reduced with timely and effective outpatient care.

During the second phase, investigators further analyzed data on preventable hospitalization in New York City and applied the small-area analysis technique to urban areas throughout the country.

Key Findings
The findings validated the use of small-area analysis as a technique to identify and measure differences in hospitalization rates for ACS conditions.

  • Hospitalization for these ACS conditions was two to 10 times greater in low-income than in high-income neighborhoods in New York City.
  • Poor African-American areas in New York City had consistently higher hospitalization rates than other low-income areas.
  • More than half of all low-income patients in New York City reported access problems that delayed or prevented them from obtaining care.
  • More than half the low-income areas in urban communities of Massachusetts, New York, New Jersey, Florida, Oregon, California and Washington state had substantially higher admission rates for ACS conditions than did high-income areas.

Funding
The Robert Wood Johnson Foundation (RWJF) supported this project from August 1990 to December 1994 with two grants totaling $1,146,238.

 See Grant Detail & Contact Information
 Back to the Table of Contents


THE PROBLEM

Many low-income patients are unable to obtain needed primary care for preventable or avoidable illnesses, and they often require hospital admission for conditions that might otherwise have been controlled in an outpatient setting.

For many chronic conditions, including asthma, heart disease, and diabetes, effective management of the disease on an outpatient basis can help avoid acute flare-ups requiring hospitalization. For some acute conditions, such as pneumonia, cellulitis (a potentially severe skin infection), and otitis media (a severe inner-ear infection), timely provision of ambulatory care can often prevent further deterioration and subsequent hospitalization.

Gaps in the delivery of primary care are likely to exacerbate health problems, necessitating hospitalization. In addition, avoidable hospitalizations consume resources that could be used for more appropriate ambulatory care.

Analytical tools and methods are needed to assess the seriousness of access problems, to target programs serving the areas in the greatest need, and to evaluate the performance of the health care system in meeting those needs.

In preliminary studies in New York City and Washington, D.C., investigators used a technique called "small-area analysis" to measure the extent of avoidable hospitalizations for certain ambulatory care sensitive (ACS) conditions, such as otitis media and bacterial pneumonia, which rise in the absence of timely and effective outpatient care.

Although the investigators were able to identify utilization and potential access problems at the zip code and census tract levels, the technique had a number of limitations that could restrict its ability to identify differences in hospitalization for related diagnoses; provide meaningful comparisons between different communities; and distinguish access barriers, the effects of physician practice styles and differences in health status, lifestyle, or culture.

 Back to the Table of Contents


THE PROJECT

These grants from RWJF provided funding for a two-phase project to refine the small-area analysis technique and to test its applicability in urban and rural areas throughout the country. In addition to RWJF funding, the project also received $400,000 from the United Hospital Fund.

During the first phase of the project (ID# 016739), investigators examined the potential usefulness of small-area analysis for a study of hospital admission rates in New York City. A medical advisory panel provided guidance in identifying diagnostic groupings for ACS conditions, which are believed to be sensitive to timely and effective outpatient care.

The project also identified "marker conditions," such as appendectomies and heart attacks, in which hospitalization is less sensitive to the provision of outpatient care, and "referral-sensitive surgeries," including joint replacement and pacemaker insertion, which may not be readily available to lower income groups.

The researchers conducted a small-area analysis at the zip code level for New York City, focusing on a sample of 1988 hospital discharges in specific neighborhoods. In addition, they conducted a pilot survey of physicians using factors that may have exacerbated the patient's condition prior to admission. They also conducted structured interviews with selected patients about their health habits, usual source of health care, and their current hospital admission.

The objectives of the second phase of the project (ID# 017488) were the following:

  1. Conduct further analysis of the New York City database, including an assessment of the impact of readmissions on hospital utilization rates.
  2. Fully implement the physician survey and structured patient interviews that were piloted during the first grant in nine New York City hospitals.
  3. Apply the small-area analysis approach to urban and rural areas in Vermont, New Hampshire, Massachusetts, New Jersey, Florida, Illinois, Nevada, Oregon, Washington state, California, and Ontario.
  4. Conduct additional research on variations in practice among emergency room physicians.

All but the fourth objective was accomplished during the grant.

 Back to the Table of Contents


FINDINGS

Findings (ID# 016739)

  • Low-income areas in New York City had substantially higher admission rates for ACS conditions than high-income areas. Asthma, diabetic ketoacidosis (a severe complication of diabetes) and coma, bacterial pneumonia, and congestive heart failure were among the conditions with the widest disparity in hospitalization rates between low- and high-income areas. Hospitalization rates for asthma in low-income areas were 6.4 times higher than in high-income areas. They were 6.3 times higher for diabetic ketoacidosis and coma, 5.3 times higher for bacterial pneumonia, and 4.6 times higher for congestive heart failure. On average, hospitalization rates for all 30 ACS conditions were 3.8 times higher in low-income areas than in high-income areas. More than two-thirds of the variation in hospitalization rates can be explained by differences in income between areas.
  • Race influenced hospital utilization rates for ACS conditions for low-income areas. Predominantly African-American middle class zip codes in New York City generally had hospitalization rates for ACS conditions that were comparable to other middle class areas, but low-income African-American areas had consistently higher hospital admission rates than other low-income zip codes.
  • For marker conditions, such as heart attacks and appendectomies, no significant differences in hospitalization rates were observed between low- and high-income areas. For example, for heart attacks, the hospital admission rate for all low-income areas in New York City matched the rate for all high-income areas.
  • Hospital admission rates for ACS conditions varied by age. The difference in hospital utilization rates between low- and high-income areas in New York City peaked among the 25-to-44 age group. There were weaker correlations between income and hospitalization among the youngest patients (perhaps suggesting the impact of Medicaid) and among the oldest patients (perhaps reflecting the effect of Medicare and greater experience coping with barriers to access).
  • For several common surgical procedures, including cholecystectomies (gall bladder removal) and hysterectomies, little difference was found between low- and high-income areas. However, for higher-cost, more technology-intensive surgeries, including coronary bypass and organ transplant, referral rates in lower-income areas of New York City were about half of those in higher-income areas.
  • There was substantial variation among emergency rooms in their management of ACS conditions. The pilot survey revealed that some New York City hospitals aggressively managed patients with the goal of releasing them for management in an outpatient setting, while other hospitals were more likely to admit patients.

Findings (ID# 017488)

  • As in New York City, low-income areas in urban communities of Massachusetts, New Jersey, Florida, Oregon, California, and Washington State had substantially higher admission rates for ACS conditions than did high-income areas.
  • Disparities in health outcomes, as measured by preventable hospital admissions, remain considerable in the broad range of urban centers studied in the project. These disparities remain despite the moderate expansion of Medicaid and recent efforts at the state and local levels to improve primary care services. Much smaller differences are found in urban areas of Ontario, Canada, where universal health coverage may help to reduce barriers to care.
  • Hospital admission rates for ACS conditions in New York City rose 35 percent among low-income areas from 1982 to 1992, whereas high-income areas experienced about a 7 percent increase.
  • Although patients from low-income areas in New York City are more likely to be readmitted to the hospital following an initial hospitalization, these multiple admissions explain only a small portion of the differences in hospitalization rates between low- and high-income areas.
  • Low-income patients in New York City were more likely to have no usual source of care than were high-income patients (26 percent vs. 7 percent), and they were less likely to have any physician contact prior to their current hospital admission (22 percent vs. 63 percent).
  • More than half of all low-income patients in New York City reported access problems that delayed or prevented them from obtaining care. Factors other than direct costs were cited frequently, including difficulties getting off work, obtaining child care, and dissatisfaction with current source of health care.
  • Low-income patients under age 18 in New York City are more likely to have some usual source of health care than are low-income adults (93 percent vs. 63 percent) and are less likely to experience barriers to care, prior to admission (30 percent vs. 66 percent).

Communications

Articles based on the results of this study have appeared in the Journal of the American Medical Association, Health Affairs, HRP Reports, and in the edited volume of Health Care Delivery in the U.S. A book entitled Anatomy of a Safety Net is forthcoming. (See the Bibliography for a complete list.) Due to continued interest in issues related to access at both the local and national level, the grantee received many inquiries about the project and assisted other researchers and local planning officials in employing the small-area analysis methodology.

 Back to the Table of Contents


LESSONS LEARNED

  1. Zip codes can change, affecting research. Some zip code boundaries changed and new zip codes were added to the city over the course of the project, compelling the project team to make adjustments in order to ensure the reliability of their comparisons over time.
  2. An analysis of hospitalization rates according to census tract rather than zip code may yield more detailed information on the effects of race and ethnicity. For example, utilization rates differ among African Americans and Latinos, who comprise substantial proportions of the population in several Harlem and South Bronx zip codes. Inconsistencies in available census tract data made its use impossible at the time of the grant.

 Back to the Table of Contents


AFTER THE GRANT

The investigator plans to continue to monitor ACS rates in New York City and to explore questions related to an increase in hospitalization for asthma. Further analyses using data based on census tract rather than zip code will be used to more accurately pinpoint areas with very high utilization. The United Hospital Fund also plans to use ACS admission rates as an important monitoring tool for the evaluation of Medicaid managed care in New York City.

 Back to the Table of Contents


GRANT DETAILS & CONTACT INFORMATION

Project

Study of Barriers to Primary Care Leading to Unnecessary Hospitalization

Grantee

United Hospital Fund of New York (New York,  NY)

  • Assessment of Barriers to Care Leading to Unnecessary Hospitalization
    Amount: $ 281,831
    Dates: August 1990 to July 1991
    ID#:  016739

  • Study of Barriers to Primary Care Leading to Unnecessary Hospitalization
    Amount: $ 864,407
    Dates: August 1991 to December 1994
    ID#:  017488

Contact

John Billings, J.D.
(212) 998-7455
john.billings@nyu.edu

 Back to the Table of Contents


BIBLIOGRAPHY

(Current as of date of this report; as provided by grantee organization; not verified by RWJF; items not available from RWJF.)

Books and Reports

Billings J. Consideration of Small Area Analysis As A Tool to Evaluate Barriers To Access. Consensus Conference on Small Area Analysis: Proceedings. US Public Health Service, Health Resources and Services Administration. DHHS Publication No. HRS-A-PE (A), 1991.

Billings J and Mijanovich T. Primary Care Access and Consideration of Cost Effectiveness." Proceedings of the National Primary Care Conference. US Public Health Service and Health Care Financing Administration, 1993.

Billings J, Newman L, et al. Use of ACS Conditions for Planning and Analysis. United Hospital Fund, 1995.

Billings J. Findings for Ambulatory Care Sensitive Conditions in Michigan, 1983–1994. HRP Reports. October 1996.

Billings J, Majanovich T, Blank A, et al. Barriers to Care for Patients with Preventable Hospital Admissions. New York: United Hospital Fund, 1997.

Billings J. Access to Health Care Services in Health Care Delivery in the US. New York: Springer, 1998.

Billings J, Cantor J and Haslanger K. Anatomy of a Safety Net.

Millman M (ed.). Access To Health Care In America. Washington, D.C.: National Academy Press, 1993.

The Robert Wood Johnson Foundation. Access To Health Care: Key Indicators for Policy. Princeton, N.J.: The Robert Wood Johnson Foundation, 1993.

United Hospital Fund of New York. New York City Community Health Atlas. United Hospital Fund, New York, 1994.

Articles

Billings J, Zeitel L, Lukomnik J, Carey T, Blank AE and Newman L. "Impact of Socioeconomic Status on Hospital Use in New York City." Health Affairs, 12(1): 162–173, 1993. Abstract available online.

Billings J, Anderson GM and Newman LS. "Recent Findings on Preventable Hospitalizations." Health Affairs, 15(3): 239–249, 1996. Abstract available online.

Billings J, Krenz SE, Rose R, Rosenbaum S, Sullivan M, Fowles J and Weiss KB. "National Asthma Education and Prevention Program Working Group Report on the Financing of Asthma Care." American Journal of Respiratory Critical Care Medicine, 154(3 Pt. 2): S119–S130, 1996. Abstract available online.

Billings J, Mikanovich T, Newman L, et al. "Understanding Barriers to Access for Ambulatory Care Sensitive Conditions." Unpublished.

Bindman AB, Grumbach K, Osmond D, Komaromy M, Vranizan K, Lurie N, Billings J and Stewart A. "Preventable Hospitalizations and Access to Health Care." Journal of the American Medical Association, 274(4): 304–311, 1995. Abstract available online.

Newman LS, Blank AE and Billings J. "Health Profiles of New York City Communities." Journal of Ambulatory Care Management, 15(4): 63–76, 1992.

Survey Instruments

Billings J. "The Admitting Physician Survey." United Hospital Fund of New York, fielded 1991 and 1992.

Billings J. "The Hospital Patient Interview Survey." United Hospital Fund of New York, fielded 1991 and 1992.

 Back to the Table of Contents


Report prepared by: Eric Love
Reviewed by: Timothy F. Murray
Reviewed by: Richard Camer
Reviewed by: Robert Crum
Program Officer: James Knickman

Most Requested