Factors influencing hospital high length of stay outliers

ABSTRACT: BACKGROUND: The study of length of stay (LOS) outliers is important for the management and financing of hospitals. Our aim was to study variables associated with high LOS outliers and their evolution over time. METHODS: We used hospital administrative data from inpatient episodes in public acute care hospitals in the Portuguese National Health Service (NHS), with discharges between years 2000 and 2009, together with some hospital characteristics. The dependent variable, LOS outliers, was calculated for each diagnosis related group (DRG) using a trim point defined for each year by the geometric mean plus two standard deviations. Hospitals were classified on the basis of administrative, economic and teaching characteristics. We also studied the influence of comorbidities and readmissions. Logistic regression models, including a multivariable logistic regression, were used in the analysis. All the logistic regressions were fitted using generalized estimating equations (GEE). RESULTS: In near nine million inpatient episodes analysed we found a proportion of 3.9 % high LOS outliers, accounting for 19.2 % of total inpatient days. The number of hospital patient discharges increased between years 2000 and 2005 and slightly decreased after that. The proportion of outliers ranged between the lowest value of 3.6 % (in years 2001 and 2002) and the highest value of 4.3 % in 2009. Teaching hospitals with over 1,000 beds have significantly more outliers than other hospitals, even after adjustment to readmissions and several patient characteristics. CONCLUSIONS: In the last years both average LOS and high LOS outliers are increasing in Portuguese NHS hospitals. As high LOS outliers represent an important proportion in the total inpatient days, this should be seen as an important alert for the management of hospitals and for national health policies. As expected, age, type of admission, and hospital type were significantly associated with high LOS outliers. The proportion of high outliers does not seem to be related to their financial coverage; they should be studied in order to highlight areas for further investigation. The increasing complexity of both hospitals and patients may be the single most important determinant of high LOS outliers and must therefore be taken into account by health managers when considering hospital costs.     

Rural Implications of Medicare''s Post-Acute-Care Transfer Payment Policy

CONTEXT: Under the Medicare post-acute-care (PAC) transfer policy, acute-care hospitals are reimbursed under a per-diem formula whenever beneficiaries are discharged from selected diagnosis-related groups (DRGs) to a skilled nursing facility, home health care, or a prospective payment system (PPS)-excluded facility. Total per-diem payments are below the full DRG payment only when the patient's length of stay (LOS) is short relative to the geometric mean LOS for the DRG; otherwise, the full DRG payment is received. This policy originally applied to 10 DRGs beginning in fiscal year 1999 and was expanded to additional DRGs in FY2004. The Secretary may include other DRGs and types of PAC settings in future expansions. PURPOSE: This article examines how the initial policy change affected rural and urban hospitals and investigates the likely impact of the FY2004 expansion and other possible future expansions. METHODS: The authors used 1998-2001 Medicare Provider Analysis and Review (MEDPAR) data to investigate changes in hospital discharge patterns after the original policy was implemented, compute the change in Medicare revenue resulting from the payment change, and simulate the expected revenue reductions under expansions to additional DRGs and swing-bed discharges. FINDINGS: Neither rural nor urban hospitals appear to have made a sustained change in their discharge behavior so as to limit their exposure to the transfer policy. Financial impacts from the initial policy were similar in relative terms for both types of hospitals and would be expected to be fairly similar for an expansion to additional DRGs. On average, including swing-bed discharges in the transfer policy would have a very small financial impact on small rural hospitals; only hospitals that make extensive use of swing beds after a short inpatient stay might expect large declines in total Medicare revenue. CONCLUSION: Rural hospitals are not disproportionately harmed by the PAC transfer policy. An expanded policy may even benefit rural hospitals by recognizing their lower use of post-acute-care and readjusting DRG weights so that they are paid more appropriately when providing the full course of inpatient care.     

The variance of length of stay and the optimal DRG outlier payments

Prospective payment schemes in health care often include supply-side insurance for cost outliers. In hospital reimbursement, prospective payments for patient discharges, based on their classification into diagnosis related group (DRGs), are complemented by outlier payments for long stay patients. The outlier scheme fixes the length of stay (LOS) threshold, constraining the profit risk of the hospitals. In most DRG systems, this threshold increases with the standard deviation of the LOS distribution. The present paper addresses the adequacy of this DRG outlier threshold rule for risk-averse hospitals with preferences depending on the expected value and the variance of profits. It first shows that the optimal threshold solves the hospital’s tradeoff between higher profit risk and lower premium loading payments. It then demonstrates for normally distributed truncated LOS that the optimal outlier threshold indeed decreases with an increase in the standard deviation.      

Rural Trauma: Is Trauma Designation Associated With Better Hospital Outcomes?

ABSTRACT:  Context: While trauma designation has been associated with lower risk of death in large urban settings, relatively little attention has been given to this issue in small rural hospitals. Purpose: To examine factors related to in-hospital mortality and delayed transfer in small rural hospitals with and without trauma designation. Methods: Analysis of data from the Nationwide Inpatient Sample for discharges between 1998 and 2003 of patients hospitalized with moderate to major traumatic injury in nonfederal, short-stay rural hospitals with annual discharges of 1,500 or fewer patients (N = 9,590). Logistic regression was used to control for patient and hospital characteristics, stratifying by hospital volume. Main outcome measures were in-hospital death and transfer to another acute care facility after initial admission. Findings: A total of 333 patients (3.5%) died in-hospital. After adjusting for patient, injury and hospital characteristics, in-hospital death was more likely among patients treated at the non-designated hospitals with fewer than 500 discharges per year (OR 2.35; 95% CI 1.25-4.41) than among patients treated at similar trauma-designated hospitals. Patients admitted to non-designated hospitals were more likely to be transferred after admission, although this finding was significant only in the larger-volume hospitals with discharges of 500-1,500 per year (OR 1.41, 95% CI 1.08-1.83). Conclusions: Associations between trauma designation and outcomes in rural hospitals warrant further study to determine whether expanding designation to more rural hospitals might lead to further improvement in trauma outcomes.     

Do Hospital Length of Stay and Staffing Ratio Affect Elderly Patients' Risk of Readmission? A Nation‐wide Study of Norwegian Hospitals

OBJECTIVE: To test whether there is an association between hospital operating conditions such as average length of stays (LOS) and staffing ratio, and elderly patients' risk of readmission. DATA SOURCES: The main data source was a national patient database of admissions to all acute-care Norwegian hospitals during the year of 1996. STUDY DESIGN: It is a cross-sectional study, where Cox' regression analysis was used to test the factors acting on the probability of early unplanned readmission (within 30 days), and later occuring ones. The principal hospital variables included average hospital LOS and staffing ratio (discharges per man-years of personnel). Adjusting patient variables in the model included age, gender, and cost-weights of the Diagnosis Related Groups (DRGs). DATA EXTRACTION METHODS: The selected material included discharges from 59 hospitals, and 113,055 elderly patients (> or = 67 years). Multiple admissions to the same hospital were linked together chronologically, and additional hospital data were matched on. To maximize the association between the index stay and the defined outcome (unplanned readmission), no intervening planned admission was accepted. PRINCIPAL FINDINGS: Being admitted to a hospital with relatively short average LOS increased the patient's risk of early readmission significantly. In addition it was found that more intensive care (more staff) could have a compensatory effect. Furthermore, the predictive factors were shown to be time dependent, as hospital variables had much less impact on readmissions occurring late (within 90-180 days). CONCLUSIONS: The results give support to the assumption of a link between hospital operating conditions and patient outcome.     

A comparison of the resource intensity of inpatients in urban and rural nonteaching hospitals

PL 98-21 mandated a prospective payment system based on diagnosis related groups (DRGs) for all Medicare inpatients. The predetermined payment for each DRG is intended to reflect the resources used to treat patients within the DRG. Eventually, the system will allow for one payment level for each DRG in rural hospitals and a higher payment level for the same DRG in urban hospitals. This represents an equitable approach, provided there is not a predominance of high severity cases in rural hospitals and that higher costs in urban hospitals are reflective of higher priced exogenous factors beyond the control of the hospital. Equitability also requires that DRGs capture the resource intensity of treatment for a given classification of patients, equally for urban and rural patients. This work compares the pediatric population of urban hospitals without a pediatric residency program with that of rural hospitals in terms of major diagnostic category, DRG, disease severity, length of stay, and charges. It also compares the capacity of DRGs to explain the variation in resource consumption in urban and rural hospitals. A sample of 116,721 discharges from 130 urban hospitals and a sample of 54,073 discharges from 97 rural hospitals are used in this work. The results indicate that there is no difference in the patient populations of these two hospital groups. The results also indicate that DRGs explain only 50 percent of the variance in the resource variables, but this obtains equally for both populations.     

Association of hospital ownership with patient transfers to outpatient care under a prospective payment system in Taiwan

Case payment, a prospective payment system akin to diagnosis-related groups (DRGs) has in-built incentives for hospitals to transfer inpatients to their own ambulatory care units following early discharge. This study used nation-wide inpatient claims data on a total of 100,730 patients treated in 2000 in (Taiwan): cesarean section (59,364 cases), femoral/inguinal hernia operation (18,675 cases), and hemorrhoidectomy (22,691 cases), all reimbursed by case payment, to explore the relationship between hospital ownership and patient transfers to outpatient treatment. For all three diagnoses, for-profit (FP) hospitals not only had lower lengths of stay (LOS) compared to public hospitals, but also showed very high odds of patient transfer to their own outpatient units, after controlling for institutional variables, (hospital level, teaching status, and geographic location), hospital competitive environment (the Herfindal-Hirschman index), and patient variables (gender, age, length of stay, and number of secondary diagnoses, a proxy for severity of illness). Similar, though slightly lower odds were observed with not-for-profit (NFP) hospitals relative to public hospitals. The findings support the property rights theory, suggesting that in Taiwan, institutional profit maximization motives may be driving patient transfers under the case payment diagnoses, rather than medical care needs. In NFP hospitals, their physician compensation mechanism, driven largely by care volumes provided by each physician, appears to be driving the disproportionately greater likelihood of patient transfer to outpatient care.     

Contribution of case‐mix classification to profiling hospital characteristics and productivity

Case‐mix classification has made it possible to analyze acute care delivery case volumes and resources. Data arising from observed differences have a role in planning health policy. Aggregated length of hospital stay (LOS) and total charges (TC) as measures of resource use were calculated from 34 case‐mix groups at 469 hospitals (1 721 274 eligible patients). The difference between mean resource use of all hospitals and the mean resource use of each hospital was subdivided into three components: amount of variation attributable to hospital practice behavior (efficiency); amount attributable to hospital case‐mix (complexity); and amount attributable to the interaction. Hospital characteristics were teaching status (academic or community), ownership, disease coverage, patients, and hospital volume. Multivariate analysis was employed to determine the impact of hospital characteristics on efficiency. Mean LOS and TC were greater for academic than community hospitals. Academic hospitals were least associated with LOS and TC efficiency. Low disease coverage was a predictor of TC efficiency while low patient volume was a predictor of unnecessarily long hospital stays. There was an inverse correlation between complexity and efficiency for both LOS and TC. Policy makers should acknowledge that differentiation of hospital function needs careful consideration when measuring efficiency. Copyright © 2010 John Wiley & Sons, Ltd.     

Analysis of cost outliers within APR-DRGs in a Belgian general hospital: two complementary approaches

CONTEXT AND OBJECTIVES: The objective of this study was to find factors that could explain high and low resource use outliers, by associating an explanatory analysis with a statistical analysis. METHOD: High resource use outliers were selected according to the following rule: 75th percentile + 1.5* inter-quartile range. Low resource use outliers were selected according to: 25th percentile - 1.5* inter-quartile range. The statistical approach was based on a multivariate analysis using logistic regression. A decision tree approach using predictors from this analysis (intensive care unit (ICU) stay, high severity of illness and social factors associated with longer length of stay) was also tested as a more intuitive tool for use by hospitals in focussing review efforts on "not explained" cost outliers. RESULTS: High resource use outliers accounted for 6.31% of the hospital stays versus 1.07% for low resource use outliers. The probability of a patient being a high resource use outlier was higher with an increase in the length of stay (odds ratios (OR) = 1.08), when the patient was treated in an intensive care unit (OR = 3.02), with a major or extreme severity of illness (OR=1.46), and with the presence of social factors (OR = 1.44). The probability of being a low outlier is lower for older patients (OR = 0.98). The probability of being a low outlier is also lower without readmission within the year (OR = 0.55). The more intuitive decision tree method identified 92.26% of the cases identified through residuals of the regression model. One quarter of the high cost outliers were flagged for additional review ("not justified" on the basis of the model), with nearly three-quarters "justified" by clinical and social factors. CONCLUSION: The analysis of cost outliers can meet different aims (financing of justifiable outliers, improvement of the care process for the outliers not justifiable on medical or social grounds). The two methods are complementary, by proposing a statistical and a didactic approach to achieve the goal of high quality care using fewer resources.     

Factors influencing length of hospital stay in community-acquired pneumonia: a study in 27 community hospitals

We did a retrospective study of 1920 episodes of community-acquired pneumonia (CAP) in 27 community hospitals and analysed inter-hospital variability in length of hospital stay (LOS), mortality and readmission rates. The overall adjusted LOS (mean+/-S.D.) was 10.0+/-9.8 days. LOS increased according to the Pneumonia Severity Index (PSI) risk class: 7.3 days for class I to 11.3 days for class V (P<0.001). In a multiple regression model, LOS increased (P<0.001) according to the hospital (inter-hospital variability), PSI risk class, complications during hospitalization, admission to ICU, need of oxygen and transfer to a nursing home. Hospitals with shorter LOS did not show an increased readmission rate (adjusted OR 1.02, 95% CI 0.51-2.03, P = 0.97) and post-discharge mortality (adjusted OR 1.20, 95% CI 0.70-2.05, P=0.51). There are significant inter-hospital variations in LOS in patients with CAP which are related to differences in clinical management. The reduction of these differences will further improve efficiency and quality of care.     

Predicting hospital choice for rural Medicare beneficiaries: the role of severity of illness.

Previous research has confirmed that desirable hospital attributes as well as increased distance, or travel time, have an impact on hospital choice. These studies have become increasingly sophisticated in modeling choice. This study adds to the existing literature by estimating the effect of both hospital and individual characteristics on hospital choice, using McFadden's conditional logit model. Some patient characteristics have not previously been accounted for in this type of analysis. In particular, the effect of a patient's complexity of illness (as measured by Disease Staging) on the choice of hospital is taken into account. The data consist of over 12,000 Medicare discharges in three overlapping rural market areas during 1986. The hospital choice set was aggregated into seven groups of urban and rural hospitals. Results indicate that rural Medicare beneficiaries tend to choose hospitals with a large scope of service and with teaching activity over those with a lower scope of service and no teaching activity, holding other factors constant. Distance is a deterrent to hospital choice, especially for older Medicare beneficiaries. The more complex cases tend to choose larger urban and rural hospitals over small rural hospitals more often than less complex cases do.     

Association of hospital resource use with comorbidity status and patient age among hip fracture patients in Japan

OBJECTIVES: This study examined the association of resource use with comorbidity status and patient age among hip fracture patients who underwent surgical treatment. DESIGN: We used a database from the Voluntary Hospitals of Japan Quality Indicator Project that involved 10 privately owned leading teaching hospitals in Japan. SETTING: Four of these hospitals in Japan. PARTICIPANTS: We selected 778 operable hip fracture patients aged 65 or older who were admitted to these hospitals between January 1996 and August 2000 (mean age: 80.3 +/- 7.3 years). MEASUREMENTS: A linear mixed model was performed to identify factors associated with the resource use, such as total length of stay (LOS), LOS before surgery, LOS after surgery, total hospital charges, charges for diagnostic examinations, charges for surgery, and length of theater time, among operable hip fracture patients. RESULTS: The mean LOS was 45.9 days, and the mean total hospital charges were US dollars 14,495.0. Results from linear mixed models revealed that higher age was significantly associated with shorter length of theater time (P < 0.01), and that the presence of comorbidity among hip fracture patients was significantly associated with longer total LOS (P < 0.01), longer LOS after surgery (P < 0.001), higher charges for diagnostic examinations (P < 0.001), and shorter length of theater time (P < 0.01). CONCLUSION: These results suggest that the presence of comorbidity among operable hip fracture patients requires greater resource use during their hospital stay, but higher age is not significantly associated with greater resource use at all.     

Effects of payment incentives, hospital ownership and competition on hospitalization decisions for ambulatory surgical procedures

OBJECTIVE: To test the hypotheses that (1) for-profit (FP) and not-for-profit (NFP) hospitals are less likely than public hospitals to admit cases reimbursed by prospective payment favoring ambulatory over inpatient care; (2) admission odds of public, FP and NFP hospitals will converge under increasing hospital competition. METHODS: Retrospective, population-based, cross-sectional study covering 29,699 cases of unilateral, femoral/inguinal hernia operation (major surgical procedure) and 60,626 cases of cataract surgery (local surgical procedure), from Taiwan's National Health Insurance database was used. Diagnosis-wise logistic regression analysis were done to examine associations between admission propensities of FP versus public and NFP hospitals (large teaching hospitals with > or = 250 beds versus district hospitals with < 250 beds) under high and low competition, adjusted for clinical complications, and patient as well as physician demographics. RESULTS: Large public teaching hospitals are significantly more likely than FP district hospitals to admit hernia patients (ORs = 1.9 and 2.6, respectively, under high and low competition), and cataract surgery patients (ORs = 5.0 and 5.4, respectively, under high and low competition). The corresponding odds ratios for public district hospitals (relative to FP district hospitals) are 1.2 and 3.9 for hernia and 4.9 and 2.7 for cataract surgery. Odds ratios show convergence of admission odds across hospital ownership under high competition relative to low competition for hernia (OR range for different hospital types under high competition, 1.0-1.9; and under low competition, 1.0-3.9). Cataract cases show high divergence of admission odds between public and FP/NFP hospitals regardless of competition level (OR range for different hospital types under high competition, 0.3-5.0; and under low competition, 0.3-5.4). CONCLUSION: Overall, our data support the study hypotheses. Differences in the relevance of inpatient care for hernia and cataract surgery may account for the lack of admission convergence of public hospitals and FPs under high competition among cataract surgery group.     

Does severity explain differences in hospital length of stay for pneumonia patients?

OBJECTIVES: In the USA, the role of patient severity in determining hospital resource use has been questioned since Medicare adopted prospective hospital payment based on diagnosis-related groups (DRGs). Exactly how to measure severity, however, remains unclear. We examined whether assessments of severity-adjusted hospital lengths of stay (LOS) varied when different measures were used for severity adjustment. METHODS: The complete study sample included 18,016 patients receiving medical treatment for pneumonia at 105 acute care hospitals. We studied 11 severity measures, nine based on patient demographic and diagnosis and procedure code information and two derived from clinical findings from the medical record. For each severity measure, LOS was regressed on patient age, sex, DRG, and severity score. Analyses were performed on trimmed and untrimmed data. Trimming eliminated cases with LOS more than three standard deviations from the mean on a log scale. RESULTS: The trimmed data set contained 17,976 admissions with a mean (S.D.) LOS of 8.9 (6.1) days. Average LOS ranged from 5.0-11.8 days among the 105 hospitals. Using trimmed data, the 11 severity measures produced R-squared values ranging from 0.098-0.169 for explaining LOS for individual patients. Across all severity measures, predicted average hospital LOS varied much less than the observed LOS, with predicted mean hospital LOS ranging from about 8.4-9.8 days. DISCUSSION: No severity measure explained the two-fold differences among hospitals in average LOS. Other patient characteristics, practice patterns, or institutional factors may cause the wide differences across hospitals in LOS.     

Public versus private hospital maternity length of stay: a gamma mixture modelling approach.

Application of a gamma mixture model to obstetrical diagnosis-related groups (DRGs) revealed heterogeneity of maternity length of stay (LOS). The proportion of long-stay subgroups identified, which can account for 30% of admissions, varied between DRGs. The burden of long-stay patients borne was estimated to be much higher in private hospitals than public hospitals for normal delivery, but vice versa for Caesarean section. Such differences highlights the impact of DRG-based casemix funding on inpatient LOS and have significant implications for health insurance companies to integrate casemix funding across the public and private sectors. The analysis also benefits hospital administrators and managers to budget expenditures accordingly.     

Influence of length of stay on patient satisfaction with hospital care in Japan.

OBJECTIVE: The objective of this study was to identify specific patient satisfaction items related to overall satisfaction by different length of stay (LOS) for patients in Japanese hospital settings. METHODS: This cross-sectional study involved a participant sample, drawn from 77 voluntarily participating hospitals throughout Japan, of in-patients discharged to the community. Older patients and psychiatric, pediatric, obstetric and gynecologic patients were excluded. The 1050 respondents analyzed (response rate > or = 5l.1%) were divided into three groups based on their LOS: group 1, LOS < or = 1 week; group 2, LOS < or = 1 month; and group 3, LOS > 1 month. Using stepwise multiple regression analysis, we explored for each LOS group the relationship between overall patient satisfaction and satisfaction with 33 individual items, including three regarding perceived reputation of the hospital in question. RESULTS: Some unique satisfaction items for each group (e.g. 'skill of nursing care' in group 1, 'Recovery of physical health', 'skill of nursing care', and 'respect for patients opinions and feelings' in group 2, and 'relief from pain' and 'respect for patients' opinions and feelings' in group 3) were significantly associated with overall satisfaction. In all three groups, common items (e.g. 'recovery from distress and anxiety' and 'doctor's clinical competence') also related significantly to overall satisfaction. Two items pertaining to the hospital reputation dimension (e.g. 'family member's evaluation of the hospital' and 'hospital reputation among other patients') were also significant predictors of overall satisfaction in all three groups. CONCLUSION: The findings show that according to LOS, unique items could determine significantly the achievement of overall satisfaction, while some common predictors across all three LOS groupings also seemed to be indispensable for inpatient's assessment of hospital care. It was also confirmed in this study that a positive perception of hospital reputation might have an important role in patient satisfaction in Japan.     

Case shifting and the Medicare Prospective Payment System.

We assessed impacts of the Medicare Prospective Payment System (PPS) during its first two years of operation (1984-85) on 467 hospitals using data from the Commission on Professional and Hospital Activities and from the American Hospital Association. Medicare discharges as a per cent of total discharges remained constant between 1983 and 1985, but the per cent of uninsured patients increased, especially at large public hospitals. The number of Medicare and total discharges per hospital declined. The number of complex diagnosis related groups (DRGs) increased, both for Medicare and non-Medicare. This trend began before the implementation of PPS and affected all types of hospitals. There was also an appreciable increase in case mix types of hospitals. There was also an appreciable increase in case mix severity within specific DRGs during 1980-85. The proportion of total patients received from or transferred to other hospitals rose after 1983, but these increases were very small. The per cent of Medicare patients admitted through the emergency room increased, especially after 1983. By contrast, the share of total non-Medicare admissions through the emergency room (ER) remained stable. Although the growth of the number of uninsured and Medicare patients admitted through the ER predate PPS, they may be influenced by it and warrant further monitoring.     

Hospital Length of Stay: Social Work Services as an Important Factor

In earlier studies it was found that the severity of patients' psychosocial problems was a significant predictor of length of stay (LOS). This current study compared predictors of LOS for samples of patients referred to social services in three large urban hospitals in June–October 2002 (n?=?176) and 2006 (n?=?147), and examined changes in patient characteristics and the nature of social work practice. A significant relationship between psychosocial severity and LOS was again found, confirming the important role that social services can potentially play in controlling hospital costs. Some significant changes were also found in the pattern of social work practice; this was generally in the direction of more community consultation and collaboration, suggesting a greater emphasis on multidisciplinary teamwork.     

Comments on HCFA hospital death rate statistical outliers. Health Care Financing Administration.

In March 1986, the Health Care Financing Administration (HCFA) released ten lists of death-rate "outlier" hospitals, one for all 1984 Medicare discharges and nine for specific DRGs. Recent Medicare hospital discharge abstracts have substantially undercounted in-hospital deaths, with large variations by state. Apart from the proportion of a hospital's cases in 80 DRGs, the predictive models had no measures of case severity based on diagnosis or procedure. Having DRG 123 (all deaths from acute myocardial infarction) as an independent variable in the all-death regression model probably accounted for much of its high r2. Inclusion of an independent variable for average length of stay (ALOS) favored hospitals in higher ALOS states by higher predicted death rates. Model bias also favored lower-risk hospitals. Small numbers of predicted deaths for specific DRGs limited low-volume hospitals on these outlier lists to those with high ratios of actual to predicted deaths. On six of the nine DRG-specific outlier lists, a total 1,222 hospitals had unfavorable residuals, while only 8 were favorable. Ten recommendations are given to increase reliability of future outcome analyses.