The purpose of this research study is to evaluate the effect of a health information exchange (HIE)-supported care coordination package on 30-day readmission rates in a frail elderly population.

BACKGROUND

Reducing hospital readmission rates is a top national priority. Unplanned hospital readmission is estimated to have accounted for more than $17 billion of the roughly $103 billion hospital payments made by Medicare in 2004.1 For patients in Medicare fee-for-service programs, the 30-day hospital readmission rates was recently found to be 19.6% nationally, and 20.7% in New York State (Jencks et al., 2009). Hospitals have urgent incentives to address readmission rates: readmission rates have been added to the National Quality Forum performance metrics (National Quality Forum, 2007); readmission rate comparisons are posted on www.hospitalcompare.hss.gov as public indicators of hospital quality; and provisions in health care reform legislation will soon mean that hospitals will not receive payment for many readmissions within 30 days of discharge.

Targeted transitional programs and better coordination of care between inpatient and outpatient settings have the potential to reduce hospital readmission rates (Naylor et al, 2004; Coleman et al, 2006; Peikes et al, 2009). Successful care coordination measures depend upon the effective transmission of health information between the inpatient and outpatient settings.

The Brooklyn Health Information Exchange (BHIX) is a regional health information organization (RHIO) that provides secure health information exchange (HIE) services among participating health-care organizations in Brooklyn, Queens, and other parts of New York City. HIE allows the meaningful sharing of health information of locations where a patients may receive care or healthcare services and can be used to help improve the effective transmission of health information between inpatient and outpatient settings. Maimonides Medical Center is working with BHIX to offer a health information technology- and HIE-based care coordination program (CCP) to help improve the care of frail elderly patients upon discharge. The CCP includes: (1) access to a secure online personal health record (PHR) that people can logon and manage their health information, as well as receive alerts and reminders about action items for them to take on their healthcare; and (2) depending on the patient's health care needs, nursing support (either in-person or by phone).

The main objective of this study to determine the impact of the CCP in a frail elderly population.

SPECIFIC AIMS

Weill Cornell Investigators will be analyzing a HIPAA-defined de-identified dataset from BHIX to evaluate the impact of the CCP. The two main outcomes we will be addressing in our data analysis are:

1. Readmission to any BHIX hospital within 30 days of hospital discharge from Maimonides;
2. Number of inpatient days within 30 days after being discharged from Maimonides Hospital.

See CITATIONS, for references.

Inclusion Criteria:

• Weill Cornell Investigators will be receiving a HIPAA-compliant de-identified dataset from the Brooklyn Health Information Exchange (BHIX) that includes:

  • Demographic data information
  • Diagnoses (admission, discharge, readmission)
  • Whether the patient was readmitted readmission, # of inpatients days if the patients was readmitted
  • Care coordination program statistics (e.g. usage of the personal health record, and frequency of contact with nursing support staff).

• The data set will include data of the following individuals:

  1. Intervention Dataset (Group 1): Those age 65 or older who are discharged from Maimonides to home during the study period and enrolled in the Care Coordination Program.
  2. Control Dataset (Group 2): Those age 65 or older who are discharged from Maimonides to home during the study period.

Exclusion Criteria:

• The exclusion criteria for this study for both the intervention & control dataset is anybody who does not fall into the above inclusion category and anybody who was:

  1. Transferred on the day of discharge to another acute care hospital, admitted to a hospital specialty unit, admitted to an inpatient rehabilitation facility, or admitted to a long-term care hospital;
  2. Approached and declined to participate in the Care Coordination Program.

A fully deidentified data set listing outcomes for the included patients is available by contacting the study PI (Jessica Ancker, jsa7002@med.cornell.edu)