Evaluation of the Mobile Acute Care of the Elderly (MACE) Unit

Inpatient hospitalizations for the elderly are fraught with hazards. Hospital complications such as delirium, falls, pressure ulcers are common in the elderly. These complications may be avoidable. Patient care units specializing in geriatric care, acute care of the elderly (ACE) units, were developed and tested and had demonstrated fewer patients discharged to nursing home, and improved ability of acutely ill older patients to perform basic activities of daily living at the time of discharge (1). However, ACE unit model have not been widely disseminated because of issues of financial viability and other limitations.

Mobile Acute Care of the Elderly (MACE) unit, a new model of geriatric inpatient care, was newly developed at the Mount Sinai Hospital and has been in use since July 2007.

MACE consists of a team care providers dedicated to the care of inpatient hospitalized elderly patients. The team consists of an attending geriatrician, a geriatrics fellow, a nurse practitioner, a social worker. The potential benefits of the MACE team over a general medicine medical service include specific attention to the needs of the elderly in the inpatient setting, potential complications, the avoidance of potentially harmful medications and improved communication in transition of care. Other quality outcome measures such as length of stay, re-hospitalization rates may be affected as a result of MACE care.

The purpose of the study is to evaluate patient care using quality markers including length of stay, re-hospitalization rates, patient satisfaction of patients admitted to the Mobile Acute Care of the Elderly Unit compared to a prospectively matched and propensity score matched control group from patients admitted to general medicine service.

Description of cohort group selection methods

We aim to enroll all Mobile ACE patients who satisfy our inclusion and exclusion criteria. To match patients on the General Medical Unit with similar characteristics to the Mobile ACE patients, we will match patients according to age (within 5 years), admitting diagnosis, and their ability to ambulate independently (functional status). We aim to match each Mobile ACE patient (case group) to one general medical unit patient (control group) with these matching characteristics. In order to do so, we will establish a pool of Mobile ACE patients and a pool of general medicine patients and enroll patients into each pool if they satisfy the inclusion and exclusion criteria according to a prospective matching protocol established and validated by Charpentier et al. (2)

This matching algorithm was selected because it has been demonstrated to achieve balance across prognostic factors in trials for which randomized allocation to treatment group is not possible. The method involves prospective individual matching of patients that have already been assigned to treatment groups (in this case, Mobile ACE unit and general medical unit). The method provide a methodologically rigorous alternative for achieving balance across treatment groups, with respect to important prognostic factors, in non-randomized trials where randomization is not possible.

For patients in the Mobile ACE unit who satisfy the inclusion and exclusion criteria, we will determine if there is a matching patient in the control group pool. If so, we will enroll the patient and matched the patient to the matching patient in the control group. If there is no match in the control group pool, we will enroll the patient into the case group pool. Because we aim to match patients concurrently, we will only match patients who have admission dates within 180 days of each other.

For patients in the general medical unit who satisfy the inclusion and exclusion criteria, the matching characteristics will be assessed and if there is a match in the pool of case group patients, the patient will be enrolled. If there is no successful match in the case group patients, we will enroll the patient to the pool of control group patients if the pool of control group patients is fewer than a predetermined size of 10 patients; or if the pool of control group patients is more than 10 patients, we will enroll the patient randomly by a predetermined probability to be in the pool of control group patients.