The evaluation of screening methods to identify medically unnecessary hospital stay for patients with pneumonia.

OBJECTIVES: This study compared three methods to screen charts of pneumonia patients for excess days. METHODS: A derivation data set was used to statistically derive a severity measure to predict length of stay for pneumonia patients and to refine a clinical algorithm for identifying excess stay. A validation data set was used to compare three computerized methods to screen for unnecessary hospital days: (1) an observed length of stay greater than a target value; (2) an observed stay greater than predicted for the specific patient; and (3) an algorithm that tested whether there were clinical justifications for the entire hospital stay. RESULTS: The sensitivity and specificity for detecting excess stay for the three methods were (1) 0.48 and 0.85 for the observed stay greater than the target value; (2) 0.56 and 0.73 for observed stay greater than predicted; and (3) 0.83 and 0.85 for the algorithm. CONCLUSIONS: These results suggest that computerized clinical algorithms may provide a useful method to detect unnecessary hospital stay.

Severity of illness measures derived from the Uniform Clinical Data Set (UCDSS).

The Health Care Financing Administration (HCFA) plans to use the Uniform Clinical Data Set System (UCDSS) to collect data on hospitalized Medicare patients. This study examined the value of UCDSS data for creating severity of illness measures. UCDSS data were obtained from a study hospital and from a national data set for patients with pneumonia (n = 528) and stroke (n = 565). Models to predict length of stay or an adverse event were derived for each condition using HCFA claims data alone, UCDSS data alone, and UCDSS data supplemented with additional information also abstracted from charts. The models were derived from one set of patients and validated on another. The R2 for predicting length of stay in the validation data for the UCDSS model was 0.29 for pneumonia and 0.19 for stroke compared to R2 values from the claims model of 0.09 for stroke and 0.06 for pneumonia. UCDSS models also were better than claims models for predicting adverse events. The best UCDSS models included International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes and other information requiring clinical judgment, and were improved by adding more information on patient functional status. Some findings were more strongly associated with outcome for the study hospital than for the national data. These results suggest that UCDSS models will predict outcome much better than the claims based models currently used by HCFA for the analysis of hospitalization-related mortality; more functional status information should be added to UCDSS; and despite an extensive objective database, the most predictive UCDSS models require clinician-assigned diagnostic codes.

The value of the Uniform Clinical Data Set System (UCDSS) in a hospital setting.

BACKGROUND: The Health Care Financing Administration designed the Uniform Clinical Data Set System (UCDSS) to help peer review organizations (PROs) identify problems with patient care. The system currently is being piloted in five states and may be used for the PRO review of Medicare patients from all states by January 1995. This study tested whether UCDSS could be used with modifications for effective internal hospital quality review. METHODS: The UCDSS includes a computer program for collecting information from the medical record and 359 computerized algorithms that evaluate quality of care. For this study, 2,313 randomly chosen medical records from a tertiary care teaching hospital were abstracted with the UCDSS at the time of discharge. Cases flagged by the UCDSS algorithms as having potential quality-of-care problems were referred to a clinical reviewer to evaluate whether the flag was a true positive and to identify reasons for false-positive flags. The algorithms were modified based on this hospital's experience by adding the reasons for false-positive flags as exceptions to the algorithm rules. RESULTS: To abstract the data with the UCDSS required a median time of 45 minutes and a mean time of 55 minutes per medical record. The percentages of algorithm flags that were confirmed upon physician review to have a quality problem were estimated to be 21% for the UCDSS and 43% for the modified UCDSS. The confirmed problem rate varied substantially by algorithm. Confirmed problems were the source of numerous departmental and individual discussions and led to changes in five departmental procedures. CONCLUSIONS: Although the results of this study are preliminary and require further verification, they suggest that with modifications and careful attention to implementation, the UCDSS may be an expensive but potentially useful tool for in-hospital quality review. In-hospital employment of the UCDSS offers more opportunities for practical use of algorithms for continuous quality improvement rather than the sometimes punitive use of system findings by PROs.

The natural history of carotid bruits in elderly persons.

STUDY OBJECTIVE: To determine the relative risk for cerebrovascular events in elderly patients with carotid bruits. DESIGN: Population-based prospective study. SETTING: Community-based home for the aged. PATIENTS: All patients were residents of a home for the aged and were at least 75 years old in 1985. MEASUREMENTS AND MAIN RESULTS: Two hundred and forty-one residents were examined for carotid bruits and signs of previous stroke. The mean age of the residents was 86 years. Twelve percent of residents had asymptomatic carotid bruits. The prevalence of asymptomatic carotid bruits was 8% in residents who were 75 to 84 years of age, 10% in residents who were 85 to 94 years of age, and 13% in residents who were at least 95 years old. Interval cerebrovascular events and cause of death were ascertained from chart review, and recorded events in survivors were confirmed by a repeat examination. The 3-year cumulative incidence of cerebrovascular events in asymptomatic residents with carotid bruits was 10%, compared with 9% in residents without carotid bruits, yielding a relative risk of 1.1 (95% CI, 0.45 to 2.7). In 60% of surviving residents, baseline carotid bruits were no longer present at the time of follow-up examination. The disappearance of these bruits was not associated with the occurrence of interval cerebrovascular events. CONCLUSIONS: The prevalence of asymptomatic carotid bruits increases with advanced age. Carotid bruits do not greatly increase the risk for subsequent stroke in elderly patients. Carotid bruits may often disappear without clinical sequelae.

Bedside case presentations: why patients like them but learners don't.

STUDY OBJECTIVE: To determine current attitudes of patients, medical students, housestaff, and clinical faculty toward bedside case presentations. DESIGN: Survey using multiple-choice questionnaire and open comments for students, housestaff, and faculty, and a structured interview of patients. SETTING: Major teaching hospitals on the campus of a midwestern medical school, staffed by full-time faculty. PARTICIPANTS: 136 medical students, 58 housestaff, 66 faculty, and 73 patients. MEASUREMENTS AND MAIN RESULTS: 85% of patients liked the case presentation discussion at the bedside, but 95% of both students and housestaff felt more comfortable with such discussion away from the patient. Attending faculty were about evenly divided in preference, with the younger staff preferring the conference room setting. Most patients (88%) opposed rounds in the hallway. Duration of rounds of one to two hours was felt desirable by most, but 50% of students preferred a duration of less than one hour. For length of new patient case presentation, 60% of learners again favored brevity, less than 5 minutes. CONCLUSIONS: Bedside rounds are an opportunity to sharpen diagnostic skills and to demonstrate the art of medicine. They are undervalued by learners and younger faculty but appreciated by patients. The authors recommend that faculty improve bedside rounds by assessing team members' educational needs, by cultivating sensitivity and respect for the needs of all parties, and by assuring pertinence and brevity of bedside discussion.

Comparison of the compression strength of human vertebral bodies with the mass and density of apatite: a study by 31P NMR spectroscopy.

The force needed to fracture individual human thoracic and lumbar vertebral bodies is compared with the mass and density of apatite. 31P NMR spectrometry was used to quantify the apatite, because it permits the mineral content of bone to be determined noninvasively with minimal nonspecific interference from the organic matrix or from variations in composition of the marrow. Experiments were performed with bones of similar structure and function from a single individual with no history of trabecular fractures, to compensate for the effects of the other variables that affect bone strength. The coefficient of correlation between compression strength and the volume density (i.e., g/cm3) of apatite was 0.95. The correlation of strength with the mass (i.e., grams) of apatite in a vertebral body also was reasonably good, r = 0.82, but correlations with areal density (i.e., g/cm2) and linear density (i.e., g/cm) were much poorer.

In vivo 31P nuclear magnetic resonance spectroscopy of bone mineral for evaluation of osteoporosis.

The mineral content of stationary bone samples can be quantified by 31P nuclear magnetic resonance (NMR) spectroscopy. The assay can be performed in regions of the anatomy that pose problems for absorptiometric techniques, because the mineral content is measured within a selected volume without concern for the geometry of the bone. In vivo 31P NMR spectra of the bones in human fingers and wrist are reported. Soft tissue such as marrow and skeletal muscle contributes little to the 31P NMR spectra of human fingers and wrist and thus should not seriously affect the accuracy of the mineral assay. 31P NMR spectrometry should prove helpful for confirming rapid bone mineral loss in those at risk and for monitoring response to treatment.

Effect of resident turnover on patients' appointment-keeping behavior in a primary care medical clinic.

In this study of 208 patients attending a Primary Care Medical Clinic of a metropolitan teaching hospital, objective measurements were used to assess the effect of discontinuous care as a result of resident turnover on patient appointment-keeping behavior. A change of resident resulted in no significant difference in return visit failure rate, subsequent appointment-keeping rate, or number of patients expressing new problems or requiring chronic therapy plan changes.

Evaluating cost-effectiveness using episodes of care.

To test the feasibility of defining episodes of care as a cost-effectiveness measure, a pilot study was carried out in conjunction with an ongoing quality assessment program which involved abstracting prospective data from charts of patients treated for hypertension in the Primary Care Clinic of Hahnemann Medical College and Hospital. For comparison, data were abstracted retrospectively on hypertensive patients treated by faculty general internists in a fee-for-service private practice. The 12-month course of each patient was divided into controlled and uncontrolled episodes for which visit frequency rate and mean laboratory test utilization was calculated. Patient cost for each type of episode in each setting was calculated using standard charges. Results indicate that the episode definition is feasible and provides a measure for comparing the cost-effectiveness of different delivery systems treating the same health care problem. Factors omitted from the study that could affect cost-effectiveness are also discussed.