Individuals with intellectual disability (ID) represent a small but important group of hospitalized patients who often have complex health care needs. Individuals with ID experience high rates of hospitalization for ambulatory-sensitive conditions and high rates of hospitalizations in general, even when in formal community care systems; however, no research was found on the common reasons for which this population is hospitalized. Academic medical centers often treat the most complex patients, and data from these centers can provide insight into the needs of patient populations with complex needs. The purpose of this study was to analyze descriptive data from the UHC (formerly known as the University Healthsystem Consortium; an alliance of 115 U.S. academic medical centers and 300 of their affiliated hospitals) regarding common reasons for hospitalization, need for intensive care, and common hospitalization outcome measures of length of stay and complications for adult (age ≥ 18) patients with ID. Findings indicate the need for specific attention to the needs of hospitalized patients with ID.
In 2001, the surgeon general issued Closing the Gap: A National Blueprint to Improve the Health of Persons With Mental Retardation (now called intellectual disability [ID]; U.S. Public Health Service, 2001), which set a goal of improving the quality of health care for this population. One issue in improving health care is to address the needs of hospitalized patients with ID. A study conducted in Manitoba, Canada found that individuals with ID are more likely than the general population to be hospitalized for ambulatory-sensitive conditions specifically related to their histories, with an overall adjusted rate ratio of 6.1; for seizure disorders the adjusted rate ratio was 54 and for schizophrenic disorders 15 (Balogh, Brownell, Oullette-Kuntz, & Colantonio, 2010). Even when in formalized community care systems, individuals with ID experience high rates of hospitalization. A study done in two regions of New York state among adults with ID ages 40–79 living in community residential facilities of 4–15 people indicated that, in 1 year, 16% were hospitalized (Janicki et al., 2002).
Despite high rates of hospitalization, various studies have indicated that individuals with ID and their caretakers perceive hospital staff to often be uncomfortable with and indifferent to the needs of people with ID (Cumella & Martin, 2004; Webber, Bowers, & Bigby, 2010). Studies have indicated that nurses and therapists think they lack the knowledge to adequately care for individuals with ID and have worries that, despite best intentions, they may not be providing optimal care (McConkey & Truesdale, 2000; Sowney & Barr, 2006). A majority of deans of medical and dental schools indicated that graduates are not prepared to be competent in the care of individuals with ID (Holder, Hood, & Corbin, 2006). Hospital staff are further hampered as typical information gathered about individuals with ID during hospitalization is not sufficient for developing comprehensive plans for their care (Bollands & Jones, 2002).
Using MEDLINE, CINAHL, and PsychINFO databases, no research in the last 10 years was found on common reasons for which individuals with ID are hospitalized and, with the exception of a study on postoperative complications (Lin, Liao, Chang, Chang, & Chen, 2011), none on outcomes for this population when hospitalized for medical conditions. Information on hospitalizations of individuals with ID would be useful in guiding efforts to improve hospital care for individuals with ID.
The purpose of this study was to determine the most common reasons for hospitalization, need for intensive care while hospitalized, and outcomes of length of stay and complications for adult individuals (age ≥ 18) with ID at academic medical centers, using a retrospective review of UHC (formerly known as University Healthsystem Consortium) data.
The hypothesis of this study was that there would be differences in the reasons for hospitalizations and differences in percentage with intensive care unit (ICU) days, average length of stay, and percentage with complications between adults with and without ID.
Data were obtained from the UHC clinical database; the UHC is an alliance of 115 U.S. academic medical centers and 300 of their affiliated hospitals that provides benchmarking data for clinical and operational improvement purposes. Data are available on patient populations and outcomes, including intensive care use, length of stay, and complications (UHC, 2013).
The study is a descriptive study of admissions of adult individuals with ID (N = 39,397) and without ID (N = 7,847,560) reflected in UHC data for the time period of July 1, 2011, through June 30, 2013.
International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM; National Center for Health Statistics, 2011) diagnosis codes representing intellectual disability (as secondary diagnoses) were used to identify patients with ID. Diagnosis codes used in our definition of ID included: 317, mild mental retardation; 318.x, other specified mental retardation; and 319, unspecified mental retardation.
Descriptive information about hospitalized adult individuals with ID included (a) the 10 most common discharge base Medicare Severity Diagnosis Related Groups (MS-DRGs). The discharge MS-DRG is the principle diagnosis valid on the date of discharge. Base MS-DRGs collapse groups of DRGs together (the same condition without complications, with complications, and with major complications; American Hospital Directory, 2012); (b) number of patients with this discharge base MS-DRG for adult individuals with ID and without ID; (c) percentage of adult individuals with ID in total patients with this discharge base MS-DRG; (d) percentage of adult individuals with ID with this discharge base MS-DRG among all adult individuals with ID and percentage of adult individuals without ID with this discharge base MS-DRG among all adult individuals without ID; (e) mean length of stay (LOS) observed (calculated from the day of admission to the day of discharge for an inpatient episode of care); (f) mean LOS expected (expected LOS computed with a multiple regression model using severity of illness and risk of mortality levels for each patient and variables such as age, gender, ethnicity, admission source, transfer status, admission through emergency room, number and type of comorbidities, and socioeconomic status to predict LOS based on a standard population; Meurer, 2009); (g) LOS index (ratio of observed to expected LOS) with statistical significance; (h) percentage of patients who had ICU stays (Intensive care days are days in an intensive care unit during an inpatient episode of care.); and (i) percentage of patients with one or more complications. Complications are potentially avoidable conditions not present on admission such as acquired decubiti, health care–acquired infections, lung injury due to medical care, equipment left in the body during surgery, postoperative pneumonia, and others (Murray, Griswold, Sunesara, & Smith, 2012). Complication computations are risk adjusted for patient diagnoses and procedures (Meurer, 2009).
Descriptive statistics were calculated on the top 10 MS-DRGS among patients with ID and the same 10 MS-DRGs for patients without ID including the percentage of patients with the 10 base MS-DRGs among patients with and without ID, the mean observed LOS for patients with each of the 10 MS-DRGs, expected LOS, the LOS index, whether the difference between expected and observed LOS is statistically significant, percentage of patients with ICU stays, and percentage of patients with one or more complications. Chi-square statistics were used to determine difference between patients with and without ID for percentage of patients with ICU stays and percentage of patients with one or more complications.
The most frequent MS-DRG for adult individuals with ID was psychoses (16.9% of total discharge MS-DRGs of all adult individuals with ID), followed by seizure disorders (7.9% of total), septicemia (5.4% of total), respiratory infections (3.1% of total), pneumonia (2.8% of total), kidney and urinary tract infections (2.4% of total), gastrointestinal (GI) disorders (2.3% of total), nutritional and metabolic disorders (1.9% of total), renal failure (1.6% of total), and GI obstructions (1.6% of total).
The five most common discharge base MS-DRGs for adult individuals with ID compared to adults without ID for the same discharge base MS-DRGs are presented in Table 1. Together the five most common discharge base MS-DRGs accounted for 36% of hospitalizations of adult individuals with ID. For adult individuals without ID, psychosis was the second most common reason for hospitalization (following vaginal deliveries) and septicemia was the fifth most common reason for hospitalization (following joint replacements [third] and GI disorders [fourth]). For four of the five MS-DRGs, the difference between observed and expected mean LOS was statistically significant for both adult individuals without ID and adult individuals without ID. A statistically higher percentage adult individuals with ID had ICU stays compared with adults without ID for three of the five base discharge MS-DRGs (p < .01), and a statistically higher percentage had one or more complications for three of the five base MS-DRGS (p < .01).
There were some limitations to the present study. The study used UHC data to identify information about hospital admissions of adult patients with ID. It is possible that not all patients with ID had a secondary diagnosis code for ID entered that would identify their information. Thus the data may not capture all the information about adult individuals with ID.
The five most common discharge base MS-DRGs for adult individuals with ID admitted to academic medical centers in the United States were psychoses, seizure disorders, septicemia, respiratory infections, and pneumonia. Similarities to adult individuals without ID were found; two of the top five discharge base MS-DRGs were the same, psychoses and septicemia. In addition, for four of five MS-DRGs, both patients with and without ID had statistically higher observed LOS than expected LOS. The statistically higher percentages of adult individuals with ID with ICU stays and with complications for some of the base discharge MS-DRGs indicate some differences in outcomes between adult individuals with and without ID.
Health care in the United States is fragmented, with poor relationships between care recipients, providers and systems, poor information flow, competing entities, and incentives not geared to quality care (Cebul, Rebitzer, Taylor, & Votruba, 2008). The differences in outcomes of LOS and complications indicate that the challenges in the health care system are magnified for individuals with ID. Backer, Chapman, and Mitchell (2009) conducted an integrative literature review of research on issues that influence access to acute and hospital care and on evaluation of interventions to improve such access for people with ID. Access issues included factors related to individuals with ID, such as fear of going to the hospital and communication needs; factors related to the role of caregivers, such as nonrecognition of their knowledge and experience by hospital staff; the attitudes, knowledge, and communication of hospital staff; and physical environment factors, such as poor layout. Recommendations based on the integrative review included liaison systems for better coordination of care, improvement of current systems including specific protocols for this population, improvement of staff attitudes and knowledge, improvement of communication and information sharing, and improvements to the physical environment (Backer et al., 2009). The 2001 surgeon general report, Closing the Gap: A National Blueprint to Improve the Health of Persons With Mental Retardation, stressed the importance of evidence-based standards of health care for the population of individuals with ID.
Information about common reasons for hospitalization for individuals with ID may assist in identifying issues to be addressed by improved liaison systems, as well as health issues in need of standards of care and protocols. Attention to improving hospital care may include the need to improve communication and information sharing systems. Backer et al. (2009) also indicated the need for improved staff knowledge and attitudes. Health care providers receive little education about the health care needs of individuals with ID. Education about specific acute care needs of individuals with ID may provide examples for health care professionals about the overall acute care needs of individuals with ID.
The findings of this study have implications for primary care. The most common discharge base MS-DRG for adult individuals with ID was psychoses, and it was the second most common for adult individuals without ID. Treatment of co-occurring psychoses is important among individuals with ID. For psychoses, adults with ID also had a statistically higher percentage with ICU stays and a statistically higher percentage with complications than adult individuals without ID. In addition, a population-based study found the prevalence of psychoses among individuals with ID to be 2.6%–4.4%, with a 2-year first episode incidence of 0.5%; the standardized first episode incidence ratio was 10.0 compared with the general population. The authors discussed the need for further study of psychoses among individuals with ID and for educating mental health professionals about their needs (Cooper et al., 2007). The findings of this study and previous studies indicate a need for specific protocols in managing the care of individuals with ID and co-occurring psychoses.
Seizure disorders were the second most common discharge base MS-DRG for adult individuals with ID. This may be related to the findings of the Balogh et al. (2010) study, which indicated a risk ratio for hospitalization for seizures of 54 for individuals with ID versus individuals without ID. In addition, population-based studies have indicated the prevalence of seizures among individuals with ID to be 18%–26% (Matthews, Weston, Baxter, Felce, & Kerr, 2008; McGrother et al., 2006). The findings of this study also reinforce the importance of comprehensive seizure care for individuals with ID and co-occurring seizure disorders.
Septicemia was a common reason for hospitalization for adult individuals with ID. Similar to individuals who are elderly and/or have immune system complications (Baldwin, 2006), individuals with ID may be at increased risk of septicemia and may benefit from efforts at prevention and early identification. Respiratory conditions and pneumonia were also common reasons for hospitalization in this study. Of note, a population-based study in Finland that examined mortality data over a 35-year period found respiratory conditions to be the second most common cause of death among individuals with ID (Patja, Moisa, & Iivanainen, 2001). The prevention and treatment of acute respiratory conditions among individuals with ID is indicated.
As the five most common discharge base MS-DRGs account for 36% of hospitalizations of adult individuals with ID, efforts to improve primary care to prevent hospitalizations for these conditions may assist in reducing hospitalization rates for individuals with ID. Findings on the differences in hospitalization outcomes between individuals with and without ID further indicate the need for evidence-based standards of care for this population.
The authors wish to acknowledge the funding support of the Center for Clinical Research and Scholarship at Rush University Medical Center and the Gamma Phi Chapter of Sigma Theta Tau.
Sarah H. Ailey, Tricia Johnson, Louis Fogg, and Tanya R. Friese, Rush University Medical Center.