Susan D. Horn

Linear Dynamic Recursive Estimation from the Viewpoint of Regression Analysis

Abstract A large class of useful multivariate recursive time series models and estimation methods has appeared in the engineering literature. Despite the interest and utility which this recursive work has when viewed as an extension of regression analysis, little of it has reached statisticians working in regression. To overcome this we (a) present the relevant random-β regression theory as a natural extension of conventional fixed-β regression theory and (b) derive the optimal recursive estimators in terms of the extended regression theory for a typical form of the recursive model. This also opens the way for further developments in recursive estimation, which are more tractable in the regression approach and will be presented in future papers.

Practice-based evidence study design for comparative effectiveness research.

Objectives:To describe a new, rigorous, comprehensive practice-based evidence for clinical practice improvement (PBE-CPI) study methodology, and compare its features, advantages, and disadvantages to those of randomized controlled trials and sophisticated statistical methods for comparative effectiveness research. Research Design:PBE-CPI incorporates natural variation within data from routine clinical practice to determine what works, for whom, when, and at what cost. It uses the knowledge of front-line caregivers, who develop study questions and define variables as part of a transdisciplinary team. Its comprehensive measurement framework provides a basis for analyses of significant bivariate and multivariate associations between treatments and outcomes, controlling for patient differences, such as severity of illness. Results:PBE-CPI studies can uncover better practices more quickly than randomized controlled trials or sophisticated statistical methods, while achieving many of the same advantages. We present examples of actionable findings from PBE-CPI studies in postacute care settings related to comparative effectiveness of medications, nutritional support approaches, incontinence products, physical therapy activities, and other services. Conclusions:Outcomes improved when practices associated with better outcomes in PBE-CPI analyses were adopted in practice.

The Relationship Between Severity of Illness and Hospital Length of Stay and Mortality

To address the question of quantification of severity of illness on a wide scale, the Computerized Severity Index (CSI) was developed by a research team at the Johns Hopkins University. This article describes an initial assessment of some aspects of the validity and reliability of the CSI on a sample of 2,378 patients within 27 high-volume DRGs from five teaching hospitals. The 27 DRGs predicted 27% of the variation in LOS, while DRGs adjusted for Admission CSI scores predicted 38% and DRGs adjusted for Maximum CSI scores throughout the hospital stay predicted 54% of this variation. Thus, the Maximum CSI score increased the predictability of DRGs by 100%. We explored the impact of including a 7-day cutoff criterion along with the Maximum CSI score similar to a criterion used in an alternative severity of illness measure. The DRG/Maximum CSI scores predictive power increased to 63% when the 7-day cutoff was added to the CSI definition. The Admission CSI score was used to predict in-hospital mortality and correlated R = 0.603 with mortality. The reliability of Admission and Maximum CSI data collection was high, with agreement of 95% and kappa statistics of 0.88 and 0.90, respectively.

Interhospital differences in severity of illness. Problems for prospective payment based on diagnosis-related groups (DRGs).

We evaluated the ability of the diagnosis-related-group (DRG) classification system to account adequately for severity of illness and, by implication, for the costs of medical care. Hospital inpatients on medicine, surgery, obstetrics/gynecology, and pediatrics services in six hospitals were evaluated to provide a spectrum of patient and hospital characteristics. This evaluation was based on data from a generic index of severity of illness obtained by trained personnel from a review of hospital charts after patient discharge. Within each DRG, substantial differences were found in the distribution of severity of illness in different hospitals. Some hospitals treated larger proportions of severely ill patients and had a wide range of severity within each DRG, but these differences did not always agree with the teaching classification or the Health Care Financing Administrations case-mix index. These findings suggest that patient classification by means of unadjusted DRGs does not adequately reflect severity of illness, and they indicate that prospective payment programs based on DRGs alone may unfairly and adversely discriminate against certain hospitals.

RN staffing time and outcomes of long-stay nursing home residents: pressure ulcers and other adverse outcomes are less likely as RNs spend more time on direct patient care.

OBJECTIVE:A clear link has been demonstrated between lower nurse staffing levels in hospitals and adverse patient outcomes, but the results of studies of such relationships in long-term care facilities haven’t been as clear. This study explored the time nurses spent in direct care and how it affected outcomes in long-stay (two weeks or longer) nursing home residents. METHODS:In a retrospective study of data collected as part of the National Pressure Ulcer Long-Term Study (NPULS), we analyzed data on 1,376 residents of 82 long-term care facilities whose lengths of stay were 14 days or longer, who were at risk of developing pressure ulcers but had none at study entry, and who had a Braden Scale score of 17 or less. Primary data came from residents’ medical records during 12-week periods in 1996 and 1997. Dependent variables included development of pressure ulcer or urinary tract infection (UTI), weight loss, deterioration in the ability to perform activities of daily living (ADLs), and hospitalization. Independent variables included resident demographics, severity of illness, nutritional and incontinence interventions, medications, and nurse staffing time. RESULTS:More RN direct care time per resident per day (examined in 10-minute increments up to 30 to 40 minutes per resident per day) was associated with fewer pressure ulcers, hospitalizations, and UTIs; less weight loss, catheterization, and deterioration in the ability to perform ADLs; and greater use of oral standard medical nutritional supplements. More certified nursing assistant and licensed practical nurse time was associated with fewer pressure ulcers but did not improve other outcomes. CONCLUSIONS:The researchers controlled for important variables in long-stay nursing home residents at risk for pressure ulcers and found that more RN direct care time per resident per day was strongly associated with better outcomes. There’s an urgent need for further research to confirm these findings and, if confirmed, for improving RN staffing in nursing homes to decrease avoidable adverse outcomes and suffering.

Opioid Pharmacotherapy for Chronic Non-Cancer Pain in the United States: A Research Guideline for Developing an Evidence-Base

UNLABELLED This document reports the consensus of an interdisciplinary panel of research and clinical experts charged with reviewing the use of opioids for chronic noncancer pain (CNCP) and formulating guidelines for future research. Prescribing opioids for chronic noncancer pain has recently escalated in the United States. Contrasting with increasing opioid use are: 1) The lack of evidence supporting long-term effectiveness; 2) Escalating misuse of prescription opioids including abuse and diversion; and 3) Uncertainty about the incidence and clinical salience of multiple, poorly characterized adverse drug events (ADEs) including endocrine dysfunction, immunosuppression and infectious disease, opioid-induced hyperalgesia and xerostomia, overdose, falls and fractures, and psychosocial complications. Chief among the limitations of current evidence are: 1) Sparse evidence on long-term opioid effectiveness in chronic pain patients due to the short-term time frame of clinical trials; 2) Insufficiently comprehensive outcome assessment; and 3) Incomplete identification and quantification of ADEs. The panel called for a strategic interdisciplinary approach to the problem domain in which basic scientists and clinicians cooperate to resolve urgent issues and generate a comprehensive evidence base. It offered 4 recommendations in 3 areas: 1) A research strategy for studying the effectiveness of long-term opioid pharmacotherapy; 2) Improvements in evidence-generation methodology; and 3) Potential research topics for generating new evidence. PERSPECTIVE Prescribing opioids for CNCP has outpaced the growth of scientific evidence bearing on the benefits and harms of these interventions. The need for a strong evidence base is urgent. This guideline offers a strategic approach to creating a comprehensive evidence base to guide safe and effective management of CNCP.

Complications in infants hospitalized for bronchiolitis or respiratory syncytial virus pneumonia.

OBJECTIVE To characterize complications among infants hospitalized for bronchiolitis or respiratory syncytial virus (RSV). STUDY DESIGN Retrospective data from 684 infants with bronchiolitis or RSV pneumonia, < or =1 year old, admitted to 10 childrens hospitals from April 1995 to September 1996. Outcomes included complication rates and effects on hospital and pediatric intensive care unit (PICU) length of stay (LOS) and hospital costs. RESULTS Most infants (79%) had one or more complication, with serious complications in 24%. Even minor complications were associated with significantly longer PICU and hospital LOS and higher costs (P<.001). Respiratory complications were most frequent (60%), but infectious (41%), cardiovascular (9%), electrolyte imbalance (19%), and other complications (9%) were common. Complication rates were higher in former premature infants (87%), infants with congenital heart disease (93%), and infants with other congenital abnormalities (90%) relative to infants without risk factors (76%). Infants 33 to 35 weeks gestational age (GA) had the highest complication rates (93%), longer hospital LOS, and higher costs (P<.004) than other former premature infants. CONCLUSIONS Complications were common in infants hospitalized for bronchiolitis or RSV pneumonia and were associated with longer LOS and higher costs. Former premature infants and infants with congenital abnormalities are at significantly greater risk for complications. Broader use of RSV prevention should be considered for these higher-risk infants.

Estimating Heteroscedastic Variances in Linear Models

Abstract We describe an estimator of heteroscedastic variances in the Gauss-Markov linear model where E(e) = 0 and with σ i 2 and unknown. It may be thought of as an approximation to the MINQUE method which results in computational economy, positive estimates, and decreased mean square error. Properties of this almost unbiased estimator are stated. It is compared with other estimators, and extensions to more general models are discussed.

Results of a collaborative quality improvement program on outcomes and costs in a tertiary critical care unit

OBJECTIVE To demonstrate that by using the knowledge and skills of the primary care provider and by applying statistical and scientific principles of quality improvement, outcomes can be improved and costs significantly reduced. DESIGN A before and after quasi-experimentally designed trial using historical controls plus an analysis of costs in areas not influenced by intensive care unit (ICU) practice to control for possible secular changes. SETTING A tertiary ICU. PATIENTS All patients admitted to the above-mentioned ICU from January 1, 1991, through December 31, 1995. INTERVENTIONS a) A focused program that applied statistical and scientific quality improvement processes to the practice of intensive care. b) An organized effort to modify the culture, thinking, and behavior of the personnel who practice in the ICU. MEASUREMENTS Severity of illness, ICU and hospital lengths of stay, ICU and hospital mortality rates, total hospital costs as analyzed by the cost center, and measures of improvement in specific areas of care. MAIN RESULTS Significant improvement in glucose control, use of enteral feeding, antibiotic use, adult respiratory distress syndrome survival, laboratory use, blood gases use, radiograph use, and appropriate use of sedation. A severity adjusted total hospital cost reduction of

Measuring severity of illness: homogeneous case mix groups.

2,580,981 in 1991 dollars when comparing 1995 with the control year of 1991, with 87% of the reduction in those cost centers directly influenced by the intervention. CONCLUSIONS A focused quality improvement program in the ICU can have a beneficial impact on care and simultaneously reduce costs.