Statistical Brief #78: Hospitalizations in which Patients Leave the Hospital against Medical Advice (AMA), 2007

Hospitalizations in which Patients Leave the Hospital against Medical Advice (AMA), 2007
Elizabeth Stranges, MS, Lauren Wier, MPH, Chaya T. Merrill, DrPH, and Claudia Steiner, MD, MPH

Introduction

Patients who leave the hospital against medical advice (AMA) may be at increased risk for adverse health outcomes.1 Also, patients who leave AMA have significantly higher readmission rates compared to other patients.1 Patients may leave the hospital for various reasons, including financial considerations and stresses, family emergencies, self-assessment of their health status, or dissatisfaction with their treatment.1 Understanding the characteristics of hospital stays that result in patients leaving AMA is critical to designing strategies to prevent premature hospital departures that could result in adverse health outcomes.

This Statistical Brief presents data from the Healthcare Cost and Utilization Project (HCUP) on hospital stays that ended with the patient leaving AMA in 2007. Utilization, resource, and patient characteristics of these hospitalizations are presented and compared to characteristics of all other non-maternal, non-newborn hospital stays.2 The most common reasons for AMA stays, as well as variations by patient location and geographic region, are provided. Differences between estimates noted in the text are statistically significant at the 0.05 level or better.

Findings

General Findings
Hospitalizations in which the patient left AMA accounted for 368,000 hospital stays in 2007 (1.2 percent of all hospitalizations) (table 1). In 1997, they accounted for only 264,000 discharges. The 39 percent increase in AMAs between 1997 and 2007 far exceeded the growth in all other hospital stays during the same period (13 percent) (figure 1). On average, AMA stays were about 2.5 days shorter (2.7 versus 5.1 days) and about half as expensive ($5,300 versus $10,400) than all other hospital stays.

AMA hospital stays, by patient characteristics, 2007
Characteristics of patients who left the hospital AMA differed from those of all other patients. The average age of the two groups differed by more than ten years (46 years for the AMA group compared to 58 years for all other hospital stays), and the pattern of stays by age group was different. Hospital stays ending AMA occurred most frequently among persons 45–64 years old (1.8 hospitalizations per 1,000 population); all other hospital stays occurred most frequently among individuals 65 years and older. In terms of gender, AMA stays were 1.6 times more common among males than females (1.5 versus 0.9 AMA stays per 1,000 population, respectively). The inverse was true of all other hospitals stays: women were 1.1 times more likely to be hospitalized than men (102.8 versus 91.5 stays per 1,000 population, respectively). Highlights
The number of hospital stays in which patients left AMA grew by 39 percent between 1997 and 2007 to 368,000 stays. The growth in these stays exceeded that of all other stays combined during this period (13 percent).


Uninsured and Medicaid stays accounted for nearly half of all AMA stays, but less than 20 percent of all other stays.


Hospital stays in which the patient left AMA were 1.6 times more common among men than among women (1.5 versus 0.9 AMA stays per 1,000 population, respectively).


AMA stays were 2.7 times greater among patients living in the poorest communities than in the wealthiest communities (compared to 1.5 times greater in the poorest communities among all other hospital stays).


Three of the top five reasons for hospitalization among AMA stays were for mental health and substance abuse conditions. Patients hospitalized for alcohol- and substance-related disorders were 11.6 and 10.8 times more likely, respectively, to leave the hospital AMA than other patients.


Nonspecific chest pain and diabetes with complications were 3.6 and 2.7 times more common in AMA stays than other hospital stays.


Hospitalization rates were highest in the poorest communities across all hospital stays; however, the magnitude of difference was greater among stays that ended AMA. AMA stays were 2.7 times greater among patients living in the poorest communities than in the wealthiest communities (compared to 1.5 times greater in the poorest communities among all other hospital stays).

Patients living in large urban areas were nearly twice as likely to leave the hospital AMA compared to patients living in all other areas (1.8 AMA stays in urban areas versus about 1.0 stays in all other areas per 1,000 population). However, all other hospital stays—those in which the patient did not leave against medical advice—were not more likely to occur in urban areas. In fact, such stays were 20 to 30 percent more likely to occur among people living in rural areas than in other (111.8 stays in rural areas compared to about 90.0 stays in other areas per 1,000 population).

AMA hospital stays, by region, 2007
The rate of AMA stays in the Northeast was double that of any other region (2.0 versus about 1.0 AMA stays per 1,000 population, respectively) (table 2 and figure 2). This pattern was not consistent with the rate of all other hospital stays, where all regions were similar (ranging from 103.4 to 108.0 stays per 1,000 population) except the West. The West had the lowest rate of non-AMA hospital stays at 73.0 stays per 1,000 population.

AMA hospital stays, by payer, 2007
Uninsured stays and those billed to Medicaid accounted for a disproportionate share of hospitalizations in which patients left AMA, collectively accounting for about half of all AMA stays (figure 3). Uninsured and Medicaid stays accounted for 21.7 and 27.3 percent of AMA stays; whereas, they represented only 5.9 and 11.7 percent of all other hospital stays, respectively. Stays billed to Medicare and private insurance showed the opposite pattern: a smaller portion of these stays resulted in patients leaving the hospital AMA compared to their share of all other hospital stays.

Most frequent reasons for AMA hospital stays, 2007
Table 3 highlights the five most frequent health conditions for which patients who left the hospital AMA were hospitalized. Three of the top five reasons for AMA hospital stays were mental health and substance abuse-related (alcohol-, substance-, and mood-related disorders). These three conditions collectively accounted for 16.4 percent of AMA stays (60,200 stays), while representing 3.0 percent of stays for all other hospitalizations. Patients hospitalized for alcohol- and substance-related disorders were 11.6 and 10.8 times more likely, respectively, to leave the hospital AMA; while, those hospitalized for mood disorders were twice as likely to leave AMA.

The remaining two common conditions for which patients left the hospital AMA were for potentially preventable conditions3,4 or conditions for which a hospital stay may be avoided with effective and timely ambulatory care: nonspecific chest pain and diabetes with complications. These two conditions collectively accounted for 10.4 percent of AMA stays (38,100 stays) and were 3.6 and 2.7 times more common in AMA stays than other hospital stays, respectively.

Data Source

The estimates in this Statistical Brief are based upon data from the HCUP 2007 Nationwide Inpatient Sample (NIS). Many statistics were generated from HCUPnet, a free, online query system that provides users with immediate access to the largest set of publicly available, all-payer national, regional, and State-level hospital care databases from HCUP. A supplemental source included population data from Claritas (2007) to create population-based rates.

Definitions

Case Definition
For this report, discharge against medical leave (AMA) was determined by the disposition of the patient at discharge as indicated on the medical record. Any discharge not defined as against medical advice (routine, in-hospital deaths, or discharge to short-term hospital, skilled nursing facility, intermediate care facility, another type of facility, home health care) was categorized as a non-AMA discharge.

Diagnoses, Procedures, ICD-9-CM, and Clinical Classifications Software (CCS)
The principal diagnosis is that condition established after study to be chiefly responsible for the patient’s admission to the hospital. All-listed procedures include all procedures performed during the hospital stay.

ICD-9-CM is the International Classification of Diseases, Ninth Revision, Clinical Modification, which assigns numeric codes to diagnoses and procedures. There are about 13,600 ICD-9-CM diagnosis codes and 3,500 ICD-9-CM procedure codes.

CCS categorizes ICD-9-CM diagnoses into a manageable number of clinically meaningful categories.5 This "clinical grouper" makes it easier to quickly understand patterns of diagnoses and procedures.


Types of hospitals included in HCUP
HCUP is based on data from community hospitals, defined as short-term, non-Federal, general and other hospitals, excluding hospital units of other institutions (e.g., prisons). HCUP data include OB-GYN, ENT, orthopedic, cancer, pediatric, public, and academic medical hospitals. They exclude long-term care, rehabilitation, psychiatric, and alcoholism and chemical dependency hospitals, but these types of discharges are included if they are from community hospitals.

Unit of analysis
The unit of analysis is the hospital discharge (i.e., the hospital stay), not a person or patient. This means that a person who is admitted to the hospital multiple times in one year will be counted each time as a separate "discharge" from the hospital.

Costs and charges
Total hospital charges were converted to costs using HCUP Cost-to-Charge Ratios based on hospital accounting reports from the Centers for Medicare and Medicaid Services (CMS). Costs will tend to reflect the actual costs of production, while charges represent what the hospital billed for the case. For each hospital, a hospital-wide cost-to-charge ratio is used because detailed charges are not available across all HCUP States. Hospital charges reflect the amount the hospital charged for the entire hospital stay and does not include professional (physician) fees. For the purposes of this Statistical Brief, costs are reported to the nearest hundred.

Median income of the patient’s ZIP Code
Median community-level income is the median household income of the patient's ZIP Code of residence. The cut-offs for the quartile designation are determined using ZIP Code demographic data obtained from Claritas. The income quartile value is missing for homeless and foreign patients. In 2006, the lowest income quartile ranged from $1–$37,999, while the highest income quartile was defined as $62,000 or above.

Place of residence
Urban-rural measurement for patient residence was based on the U.S. Office of Management and Budget (OMB) definitions of Core-Based Statistical Areas. OMB classifies counties into metropolitan and micropolitan areas. For this Statistical Brief, the metropolitan areas were further divided into large and small metropolitan areas using the Urban Influence Codes (UIC). Thus, for this report, counties were classified into one of four categories:
Large central metropolitan includes metropolitan areas with 1 million or more residents.
Large fringe metropolitan includes counties of metropolitan areas with 1 million or more residents.
Median and small metropolitan includes areas with 50,000 to 999,999 population.
Micropolitan and noncore includes micropolitan and nonmetropolitan counties, i.e., counties with no town greater than 50,000 residents.

abrir aquí para acceder al documento HCUP-US AHRQ completo, del cual se reproduce el 20%:
Statistical Brief #78