The Role of an Acute Care for the Elderly Unit in Achieving Hospital Quality Indicators While Caring for Frail Hospitalized Elders

Abstract

Acute Care for the Elderly (ACE) units have successfully decreased length of stay, hospital costs, and readmission rates. Furthermore, patients return home with increased functional capacity and improved satisfaction with their hospital stay. The ACE unit concept was geared toward patients returning to independent living, but the average hospitalized geriatric patient is increasingly more frail, vulnerable, and dependent. The purpose of this study is 2-fold: (1) to determine if the ACE unit continues to offer the same benefit to the frail, often bedbound elderly, and (2) to determine if such a unit is able to maintain standard hospital quality indicators. A total of 1096 cases discharged from the Memorial-Hermann ACE unit between July 2008 and June 2010 were compared to a sample of 383 patients with similar illness severity who were discharged between July 2007 and June 2008. Metrics measured include: average length of stay (ALOS), case mix index (CMI), case mix adjusted average length of stay (CMI adj ALOS), average direct costs per case, and readmission rate. Patient satisfaction was measured using Hospital Consumer Assessment of Healthcare Providers and Systems and Press-Ganey surveys; quality and safety data were provided by Memorial-Hermann's Quality and Safety Department. The ACE unit resulted in a statistically significant decrease in ALOS and CMI adj LOS with a simultaneous increase in Health Care Financing Administration CMI, indicating that the unit was serving a sicker, more frail population. The readmission rate was 11.95%. The decrease in length of stay, readmission rate, and direct cost translates into a decrease in cost per case. Furthermore, the ACE unit successfully met hospital quality indicators. (Population Health Management 2012;15:236–240)

Introduction

I n 2000, patients older than age 65 made up only 15% of the population, but required 4 times more hospital days and were responsible for more than 50% of hospital costs.¹ By 2027, the admission rate for patients older than age 65 will increase 78%, compared to only 16% in the younger population.² Older patients historically have a longer length of stay, which is further prolonged in patients with physical and cognitive impairment.³ A prolonged hospital stay results in functional decline,⁴ making hospitalization a pivotal event that often results in a downward spiral of decline for the older person.

The Acute Care for the Elderly (ACE) concept was developed to disrupt the trajectory of decline often experienced by geriatric patients admitted to the hospital. The first ACE unit, developed at University Hospitals of Cleveland in the 1990s, included 4 components: a specialized environment, patient-centered care, pharmaceutical review, and interdisciplinary team plans of care. Care was geared toward patients who would return to independent living.⁵ ACE units have been successful in decreasing length of stay, hospital costs, and readmission rates. Furthermore, patients return home with increased functional capacity and improved satisfaction with their hospital stay.⁶

However, as life expectancy inevitably increases, the average hospitalized geriatric patient is increasingly more frail and vulnerable. A large population of such patients exists in the Texas Medical Center and often presents to Memorial-Hermann Hospital. The purpose of this study is 2-fold: (1) to determine if the ACE unit model continues to offer the same benefit to the frail, vulnerable, and often bedbound elderly, and (2) to determine if such an ACE unit is able to maintain standard hospital quality indicators.

Methods

The ACE Unit

Memorial-Hermann Hospital, located in the Texas Medical Center, is a 900-bed, nonprofit hospital that is a Level I Trauma Center and one of the primary teaching hospitals for the University of Texas Medical School in Houston. The ACE unit at Memorial-Hermann Hospital opened in July 2008 and is a 14-bed unit.

One geriatric physician serves as the primary physician for the majority of the patients on the unit as well as the medical director. The geriatric team (ie, an internal medicine resident, geriatric fellow, and/or geriatric nurse practitioner) assist with medical care delivery. A full unit is staffed by 3 nurses, 1 patient care assistant, 1 unit clerk, and 1 clinical nurse manager. A nutritionist, physical/occupational therapist, social worker, case manager, and pharmacist also are partially dedicated to the ACE unit. Lastly, the unit is heavily supported by volunteer services; the volunteers undergo the same training as the patient care assistants and help tremendously with the custodial care needed on the unit. The entire team, along with respiratory therapy and chaplain services, attends daily interdisciplinary team meetings during which each patient's plan of care is reviewed. Furthermore, when the unit first opened, all staff were offered a 12-week interdisciplinary curriculum pertaining to the care of the older hospitalized patient (Table 1). The unit uses the same forms as the rest of the hospital; however, there is a standardized admission order set, applicable to all disease processes, that ensures patients get an appropriate diet and daily physical therapy.

Table 1.

Interdisciplinary Curriculum

Topic	Given By
Delirium/restraint reduction	Physician/nurse practitioner/nurses
Pain management
Falls
Insomnia
Incontinence/Foley catheter use
Constipation
Elder abuse/APS referrals	Physician/social worker
Wound care	Wound care nurse
Gait evaluation/gait belt use/transfers	Physical/occupational therapist
Different diets and calorie counts	Nutrition/speech therapist
Recognizing swallowing impairments
Oxygenation and nebulized treatments	Respiratory therapist

APS, adult protective services.

After interdisciplinary rounds, the geriatric team, along with the pharmacist, complete bedside rounds. The nurse manager speaks with patients and families individually, after bedside rounds are complete, to ensure that there are no further questions. The nursing staff continues to check on the patients every hour. After discharge, the unit clerk calls patients or their families to ensure that their hospital experience was a good one.

The patients on the ACE unit are direct admissions from the Geriatric and Palliative Medicine clinic/home visit program, transfers from the intensive care unit/cardiac care unit, or from the emergency room (ER). ER admission criteria are (1) age > 70 or age > 65 with multiple comorbidities, and (2) appropriate for a med/tele floor. Referral to the unit, however, is a random event, depending on bed availability and ER/referring physician preference; not all persons older than age 70 are sent to the unit. Patients are admitted to the ACE unit during the day by the geriatric team; in the evenings, the internal medicine float resident admits patients to the unit, transferring care in the morning. Furthermore, private physicians also are allowed to admit their older patients to this unit, room permitting; a geriatric consult is done on these patients.

The unit, which is physically isolated, consists of 2 perpendicular hallways; the only entrance into the unit is in front of the nurses' station, which is located at the intersection of the 2 hallways. With help from donors, the unit was remodeled to incorporate environmental changes necessary for caring for the older population. The single occupancy rooms have matte-finished, sound-absorbent, laminate floors; translucent shades; large clocks; and dry erase boards. The bathrooms have grab bars, raised black-rimmed toilet seats, widened doors, and showers. Each room is furnished with a recliner that converts into a bed, encouraging family to spend the night. There also are extra walkers and wheelchairs available for patients to use. A WanderGuard system is in place and low beds with bed alarms are available for patients at risk for falls. As part of the environmental changes, the unit does not accept unit-to-unit transfers, perform procedures, imaging studies, blood draws, vital signs, or pass medication between the hours of 11 pm and 5 am, unless there is an emergency. All new admissions are given a folder that explains the ACE unit concept, identifies the members of the interdisciplinary team (with photographs), and provides a time frame for physician rounds; they are also given a contact number in case they have questions for the physician after discharge.

Data collection

This was an observational cohort study. All cases discharged from the ACE unit were aggregated to formulate the metrics in Table 2; year 1 represents cases discharged between July 1, 2008 and June 30, 2009, while year 2 represents cases discharged between July 1, 2009 and June 30, 2010. Baseline data include cases from July 1, 2007 to June 30, 2008 for patients ≥ 70 yrs old who have the same diagnosis-related group (DRG) codes and case mix index (CMI) as the cases discharged from the ACE unit. Key metrics measured include: average length of stay (ALOS), defined as the total number of patient days divided by the number of discharges during the specified time period, resulting in the average number of days in the hospital for each case; CMI, defined as the average DRG weight for all cases using the Centers for Medicare and Medicaid Services Medicare relative weights and indicative of the severity of the patient population; case mix adjusted average length of stay (CMI adj ALOS), defined as the length of stay for each case adjusted for the CMI. Average direct costs are calculated as total direct costs divided by the number of discharges during the specified time period resulting in the average direct costs per case. Total direct costs include salaries and benefits; drugs; supplies; fees, interest, and depreciation; and all other controllable direct costs. Minitab statistical software (Minitab Inc., State College, PA) was used for all metrics to determine statistical validation in differences between populations using the Mood Median test. Additionally, readmission rate was calculated from year 2 data using the formula: # of unique patients readmitted/total # of unique patients admitted to the unit over 1 year.

Table 2.

Performance Metrics Overview Table

				Yr1 vs. Baseline		Yr2 vs. Baseline		Yr2 vs. Yr1
Metric	Baseline	Yr1	Yr2	Variance	% Variance	Variance	% Variance	Variance	% Variance
Cases	383	606	490
ALOS	7.76	7.37	5.55	−0.39	−5.0%	−2.21	−28.5%	−1.82	−24.7%
CMI adj ALOS	6.40	6.55	5.16	0.15	2.4%	−1.24	−19.4%	−1.39	−21.2%
HCFA CMI	1.236	1.294	1.244	0.06	4.7%	0.01	0.6%	−0.05	−3.9%
Avg direct costs	$6,971	$7,133	$6,538	$161.94	2.3%	−$433.41	−6.2%	−$595.35	−8.3%

ALOS, average length of stay; CMI adj ALOS, case mix adjusted average length of stay; HCFA CMI, Health Care Financing Administration case mix index.

Patient satisfaction was measured by the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey as well as Press Ganey. Quality and safety data were provided by Memorial Hermann's Quality and Safety Department. Data provided are for 2010.

Results

Demographic information as well as diagnosis by service line for all patients are provided in Tables 3 and 4, respectively. The baseline comprised 61% female patients vs. 70% in year 1 and 69% in year 2. The average CMI for ACE unit patients was 1.23. The ALOS, CMI, and CMI adj LOS for each of the 3 years are shown in Table 3.

Table 3.

Demographics Overview

Sex	Race	Baseline	Yr 1	Yr 2
Female	White	103	222	151
	African American/black	111	149	135
	All other	18	36	43
	Asian Pacific Islander	2	9	2
	Unknown		5	6
Female total		234	421	337
Male	White	70	96	77
	African American/black	67	65	55
	All other	10	18	15
	Asian Pacific Islander	1	5	2
	Unknown		1	4
	American Indian/Eskimo/Aleut	1	0	0
Male Total		149	185	153
Grand Total		383	606	458

Table 4.

Patient Mix by SubService Line Overview

Patient Type	Baseline	Yr 1	Yr 2
Cardiology-Medical Cardiology	22%	12%	17%
General Medicine-Nephrology	12%	13%	15%
General Medicine-Pulmonary	10%	12%	10%
General Medicine-Gastroenterology	9%	8%	9%
Orthopedics-Trauma (Postop)	5%	14%	6%
General Medicine-Endocrinology	7%	5%	4%
General Medicine-Infectious Disease	4%	3%	5%
Neurology-Stroke/Transient Ischemic Attack	5%	3%	3%
Orthopedics-General Medical	4%	2%	2%
General Medicine-Dermatology	3%	2%	3%
Neurology-Other Neurological	2%	2%	2%
Neurology-Degenerative Disorders	3%	2%	2%
Neurology-Epilepsy	3%	1%	2%
General Medicine-Rheumatology	2%	1%	2%
General Medicine-Trauma	2%	2%	1%
General Medicine-Hematology	2%	2%	1%
All other	6%	14%	16%

Using the Mood Median Test, ALOS, CMI adj ALOS, and average direct costs were not statistically different when comparing year 1 to baseline (P=0.476, 0.974, 0.143, respectively). Evaluating Health Care Financing Administration (HCFA) CMI demonstrates there is a statistically significant difference between year 1 and baseline (P=0.011).

In the second year, the ACE unit observed a decrease in ALOS when compared to baseline; additionally, CMI adj LOS also decreased while HCFA CMI was relatively flat. Using the Mood Median Test ALOS, CMI adj ALOS, and HCFA CMI were statistically significant when comparing year 2 to baseline (P=0.001, 0.007, P=0.0001, respectively). The decrease in average direct costs in year 2 was not statistically significant (P=0.314).

Comparing year 1 with year 2, the ACE unit observed a decrease in ALOS and CMI adj LOS. Using the Mood Median Test, the decrease in the ALOS and CMI adj ALOS was statistically significant (P<0.0001, P=0.009, respectively). There is no statistical difference in HCFA CMI and average direct costs between years 1 and 2 (P=0.165, 0.495, respectively).

In examining the demographic data, there was no statistical difference in CMI adj. ALOS or average direct costs when comparing gender across the 3 years (P=0.857 and P=0.689, respectively). In looking at ethnicity, there was no statistical difference in CMI adj ALOS or average directs costs across the 3 years (P=0.398 and P=0.363, respectively; data not shown).

Combining year 1 and year 2 performance for readmission rates for all diagnoses was 11.95%, while comparison data prior to the opening of the ACE unit was 14.04%, which reflects a 2.09 percentage point improvement (data not shown).

Press Ganey scores have remained greater than 80% and HCAHPS scores are consistently 9 or 10 over the last year. The unit's quality indicators for falls, rapid response calls/codes, blood culture contamination rates, unit-acquired pressure ulcers, and restraint use have met hospital goals. This cannot be compared to data prior to the ACE unit opening because quality data cannot be categorized by age or demographic variables.

Discussion

Our results show that from baseline to 2-year follow-up, establishing the ACE unit model at Memorial Hermann Hospital resulted in a significant decrease in ALOS and CMI adj LOS. Additionally, there was a simultaneous increase in HCFA CMI, indicating that the unit was serving a sicker and more frail population. The decrease in length of stay also was statistically significant when comparing year 2 to year 1. The readmission rate for the 2 years was 11.95%, an improvement from before the ACE unit opened and significantly lower than the national average of approximately 20%.⁷ The decreased length of stay and readmission rate, combined with a decrease in direct costs, translates into a decrease in cost expenditures per case, representing dollars saved by the hospital. Lastly, our ACE unit paradigm successfully maintained quality indicators determined by the hospital.

ACE units consistently have been shown to decrease length of stay^8,9 and readmission rates.^4,10 Shorter lengths of stay not only lower hospital costs, but allow the hospital to care for more patients. In our study, the length of stay continued to decline in the second year as compared to the first year, reflecting an increased efficiency in patient management. Hospital administrators should be aware that an ACE unit will continue to improve on these markers after the first year. Furthermore, readmission rates, which often are an indicator of quality of care and discharge planning,¹¹ were extremely low for our unit; this is consistent in other ACE unit studies. Lastly, ACE units have historically been linked to improved patient, nursing, and physician satisfaction¹²; this is shown in our consistently high HCAHPS and Press Ganey scores as well as by the fact that family members and patients often request the ACE unit when they require hospitalization. To our knowledge, this is the first study to publish hospital quality indicators for an ACE unit.

As the national trend continues to push for improved quality of care with decreased cost, developing ACE units should be a focus of hospital administrators. Most studies show that there is a decreased cost per patient for a patient cared for on an ACE unit. Furthermore, although there are initial start-up costs to establish an ACE unit, the decreased length of stay and readmission rates over time compensate for this expense.¹³ Traditionally, ACE units have been recommended for geriatric patients who are able to return to independent living; however, our ACE unit served a very unique purpose in the Memorial-Hermann Hospital System. Patients admitted to the ACE unit reflected a cohort of the hospital's most vulnerable and dependent general medicine patients who previously were treated in units throughout the hospital. With the help of trained, dedicated staff and good communication through daily interdisciplinary rounds, this grouping enabled the team to focus on the needs of the geriatric patient, improving patient care while continuing to meet hospital quality indicators.

CMI is used to measure disease severity in a patient population. Our average CMI over the course of the 2 years was 1.23; in ACE unit literature, there is only 1 other hospital that reports a CMI of 1.04, significantly lower than ours.¹⁴ Patient acuity also can be measured using Severity of Illness (SI) and Risk of Mortality (RM) scores. During the 2 years, patients in the ACE unit had an average SI score of 2.56 and an average RM score of 2.37, compared to 2.30 (11.3% lower) and 1.87 (26.7% lower), respectively, in our general medicine population. A separate study that was conducted to show the frailty of the patients on the unit revealed that the average age of ACE patients was 81—almost 30 years older than patients on the general medicine wards—and more than 50% of the ACE patients were incontinent and needed assistance with feeding.

In the first year, there was an increase in CMI, indicating that the average patient admitted to the ACE unit was increasingly sicker and more frail. Comparing the average CMI with general medicine patients, ACE patients have a 2% increase in CMI, reflecting an additional $227,000 in overall reimbursement. These monies, combined with the decrease in direct costs and dollars saved as a result of decreased length of stay, represent a considerable amount in hospital revenue. Thus, this may be the first study to show that the ACE unit is a successful way to provide care for the frail and vulnerable elderly; these patients often are the most costly for Medicare. This is the only ACE on the Texas Medical Center campus and, based on our experience, it is becoming well known in the community as a site to care for frail elders with and without dementia.

This study has some limitations similar to those seen in other cohort studies. The baseline data are from patients who were cared for throughout the hospital by multiple private and teaching physicians. Furthermore, the ACE unit data were collected over a period of 2 years during which the patient population may have changed. Although patients admitted to the unit continued to be similar in race and service line diagnoses, they may have been sicker, as indicated by the increasing HCFA CMI. Furthermore, over the course of the 2 years, there have been 2 different nurse practitioners, 2 different clinical nurse managers, and some changes in nursing, volunteers, and ancilliary staff, as well as a change in hospital leadership. However, despite these changes, the structure of interdisciplinary and bedside rounds have remained the same, as has the attending physician. The increasing number of family requests for admission to the ACE unit indicates that perhaps our patient population, while frail and vulnerable, may have better family support and thus better outcomes than others. Lastly, Memorial-Hermann Hospital is located in the Texas Medical Center; although our readmission rates are low, we are unable to account for patients who are readmitted to other hospitals.

In conclusion, our ACE unit has filled an important niche in the Memorial-Hermann Hospital System. It has successfully decreased length of stay, hospital cost, and readmission rates, while maintaining hospital indicators for an extremely frail and vulnerable population. Although additional studies must be done to confirm our findings, hospital administrators should consider establishing ACE units to reduce costs and improve quality of care. Specialized models of care, such as ACE units, will help care for our burgeoning older population, especially with upcoming Medicare funding and policy changes.

Footnotes

Author Disclosure Statement

Drs. Ahmed and Dyer and Ms. Taylor and Ms. McDaniel disclosed no conflicts of interest or financial ties.

References

Elixhauser

, Yu

, Steiner

, Bierman

. Hospitalization in the United States, 1997. HCUP Fact Book No. 1; AHRQ Publication No. 00-0031. Rockville, MD: Agency for Healthcare Research and Quality, 2000.

Lyons

, Landefeld

. Improving care for hospitalized elders. Ann Long-Term Care, 2001; 9:35–40.

Lang

, Heitz

, Hedelin

et al. Early markers of prolonged hospital stays in older people. J Am Geriatr Soc, 2006; 54:1031–1039.

Covinsky

, Palmer

, Kresevic

et al. Improving functional outcomes in older patients: Lessons from an acute care for elders unit. Jt Comm J Qual Improv., 1998; 24:63–76.

Palmer

, Landefeld

, Kresevic

, Kowal

. A medical unit for the acute care of the elderly. J Am Geriatr Soc, 1994; 42:545–552.

Ahmed

, Pearce

. Acute Care for the Elderly (ACE): A literature review. Popul Health Manag, 2010; 13:219–225.

Jencks

, Williams

, Coleman

. Rehospitalizations in the Medicare fee-for-service program. N Engl J Med, 2009; 360:1418–1428.

Covinsky

, King

Jr , Quinn

et al. Do acute care for elders units increase hospital costs? A cost analysis using the hospital perspective. J Am Geriatr Soc., 1997; 45:729–734.

Asplund

, Gustafson

, Jacobsson

et al. Geriatric-based versus general wards for older acute medical patients: A randomized comparison of outcomes and use of resources. J Am Geriatr Soc., 2000; 48:1381–1388.

10.

Counsell

, Holder

, Liebenauer

et al. Effects of a multicomponent intervention on functional outcomes and process of care in hospitalized older patients: A randomized controlled trial of Acute Care for Elders (ACE) in a community hospital. J Am Geriatr Soc., 2000; 48:1572–1581.

11.

Ashton

, Del Junco

, Souchek

, Wray

, Manysur

. The association between the quality of inpatient care and early readmission: A meta-analysis of the evidence. Med Care, 1997; 35:1044–1059.

12.

Palmer

, Counsell

, Landefeld

. Clinical intervention trials: The ACE unit. Clin Geriatr Med., 1998; 14:831–849.

13.

Jayadevvapa

, Chhatre

, Weiner

, Raziano

. Health resource utilization and medical care cost of acute care elderly unit patients. Value Health, 2006; 9:186–192.

14.

Flood

, McGrew

, Green

, Dodd

, Lyman

, Brown

. Cost savings from an acute care for elders (ACE) unit versus usual care for hospitalized patients. [abstract] http://www.americangeriatrics.org/files/documents/pdfs/2011.abstracts.pdf. 2011 September 28.