Search PubMed⌕ Search

Biomedical subjects

W M Tierney

Publications and source records attributed to W M Tierney.

At least 37 records · Page 2Linked to original sources
PubMed2000

The clinical and economic impact of alternative staging strategies for adenocarcinoma of the pancreas.

OBJECTIVE: Several innovative imaging modalities, including endoscopic ultrasound, have increased the number of available preoperative staging methods in patients with adenocarcinoma of the pancreas. Our goal was to estimate the clinical outcomes and cost-effectiveness of alternative staging strategies for pancreatic adenocarcinoma. METHODS: Decision analysis was used to simulate alternative staging strategies. Cost inputs were based on Medicare reimbursements; clinical inputs were obtained from the available literature. Model endpoints of interest were cost per curative resection and appropriateness of treatment allocation based on pathological stage. RESULTS: Endoscopic ultrasound followed by laparoscopy yielded the lowest cost per curative resection ($37,600) and minimized the number of unnecessary surgical explorations (5.4 per 100 patients staged). Requiring angiographic confirmation when endoscopic ultrasound demonstrated an unresectable tumor yielded an intermediate cost-effectiveness ratio and virtually eliminated the risk of overstaging. Laparoscopy alone maximized the resection rate, but each additional resection would cost approximately $2 million relative to a strategy employing both endoscopic ultrasound and angiography. CONCLUSIONS: Staging strategies incorporating endoscopic ultrasound may improve treatment allocation and are cost-effective relative to angiography-based strategies. A staging protocol that does not incorporate an imaging modality to detect vascular invasion dramatically increases the cost per additional curative resection compared with more comprehensive staging protocols.

Adenocarcinoma↗
PubMed1999

Use of complementary therapies for arthritis among patients of rheumatologists.

BACKGROUND: Use of complementary and alternative medicine (CAM) is common among persons with chronic conditions. OBJECTIVE: To identify correlates of and describe patients' perspective on use of CAM for rheumatologic conditions. DESIGN: Telephone survey. SETTING: Three university practices and three private rheumatology practices. PATIENTS: 232 of 428 eligible consecutive patients (54%) with scheduled appointments. MEASUREMENTS: Patients answered questions on CAM use, functional status, pain, provider satisfaction, and health services utilization. Chart reviews provided demographic information and rheumatologic diagnoses. Bivariate analyses identified correlates of four CAM outcomes (history, magnitude, and frequency of CAM use and communication about CAM use with a physician), and multiple logistic regression identified independent correlates of regular CAM use. RESULTS: Approximately two thirds of the respondents (n = 146) had used CAM. Of these 146 respondents, 82 (56%) currently used CAM and 132 (90%) regularly used CAM or had done so in the past. Fifty-five respondents (24%) had used three or more types of CAM. In multivariate analyses, persons who used CAM regularly were more likely to have osteoarthritis (odds ratio, 5.6 [95% CI, 1.9 to 16.8]), severe pain (odds ratio, 2.5 [CI, 1.4 to 4.8]), and a college degree (odds ratio, 2.6 [CI, 1.3 to 5.4]) than patients who had never used CAM. Nearly half of the respondents discussed CAM use with their physicians. The most common reasons for not disclosing CAM use were that the physician had not asked about it and that the patient forgot to tell the physician; fear of disapproval was rarely cited. Discussions about CAM use between patient and physician occurred more frequently among patients with fibromyalgia and persons who regularly used CAM or used several types of CAM. CONCLUSIONS: Patients with rheumatologic conditions frequently use CAM. Severe pain and osteoarthritis predict regular use of CAM but do not predict a greater likelihood of discussing CAM use with physicians. Routine inquiry by physicians will probably detect CAM use.

Arthralgia↗
PubMed1999

Work patterns of ambulatory care pharmacists with access to electronic guideline-based treatment suggestions.

The effects of the electronic display of guideline-based, patient-specific treatment suggestions on pharmacist work patterns were studied. A total of 28 pharmacists at a hospital-based ambulatory care pharmacy were randomly assigned to intervention and control groups. The intervention group had access to electronic treatment suggestions for heart failure, ischemic heart disease, reactive airways disease, and uncomplicated hypertension, while the control group did not. Starting 9 and 19 months after the initial display of treatment suggestions, all pharmacists recorded the time they spent on a variety of activities, the purpose of each activity, and persons contacted during the activity; these observations were recorded in response to a pager-like device that randomly buzzed four times an hour. A total of 11,102 observations were recorded. Pharmacists in the intervention group spent significantly more of their time discussing information, advising and informing, and solving problems than pharmacists in the control group but significantly less of their time checking and filling prescriptions. Pharmacists in both groups completed a majority of their work alone, but pharmacists in the intervention group worked significantly less by themselves and significantly more with other pharmacy personnel, patients, and physicians and nurses than control-group pharmacists. The delivery of patient-specific information to pharmacists at the time of dispensing had a significant positive impact on pharmacist work patterns.

Clinical Pharmacy Information Systems↗
PubMed1999

Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life.

This study used the standard error of measurement (SEM) to evaluate intra-individual change on both the Chronic Respiratory Disease Questionnaire (CRQ) and the SF-36. After analyzing the reliability and validity of both instruments at baseline among 471 COPD outpatients, the SEM was compared to established minimal clinically important difference (MCID) standards for three CRQ dimensions. A value of one SEM closely approximated the MCID standards for all CRQ dimensions. This SEM-based criterion was then validated by cross-classifying the change status (improved, stable, or declined) of 393 follow-up outpatients using the one-SEM criterion and the MCID standard. Excellent agreement was achieved for all three CRQ dimensions. Although MCID standards have not been established for the SF-36, the one-SEM criterion was explored in these change scores. Among SF-36 scales demonstrating acceptable reliability and reasonable variance, the percent of individuals within each change category was consistent with those seen in the CRQ dimensions. These results replicate previous findings where a value of one SEM also closely approximated MCIDs for all dimensions of the Chronic Heart Disease Questionnaire among cardiovascular outpatients. The one-SEM criterion should be explored in other health-related quality of life instruments with established MCIDs.

Adult↗
PubMed1999

The Regenstrief Medical Record System: a quarter century experience.

Entrusted with the records for more than 1.5 million patients, the Regenstrief Medical Record System (RMRS) has evolved into a fast and comprehensive data repository used extensively at three hospitals on the Indiana University Medical Center campus and more than 30 Indianapolis clinics. The RMRS routinely captures laboratory results, narrative reports, orders, medications, radiology reports, registration information, nursing assessments, vital signs, EKGs and other clinical data. In this paper, we describe the RMRS data model, file structures and architecture, as well as recent necessary changes to these as we coordinate a collaborative effort among all major Indianapolis hospital systems, improving patient care by capturing city-wide laboratory and encounter data. We believe that our success represents persistent efforts to build interfaces directly to multiple independent instruments and other data collection systems, using medical standards such as HL7, LOINC, and DICOM. Inpatient and outpatient order entry systems, instruments for visit notes and on-line questionnaires that replace hardcopy forms, and intelligent use of coded data entry supplement the RMRS. Physicians happily enter orders, problems, allergies, visit notes, and discharge summaries into our locally developed Gopher order entry system, as we provide them with convenient output forms, choice lists, defaults, templates, reminders, drug interaction information, charge information, and on-line articles and textbooks. To prepare for the future, we have begun wrapping our system in Web browser technology, testing voice dictation and understanding, and employing wireless technology.

Computer Terminals↗
PubMed1999

The use of an electronic patient record system to identify advanced cancer patients and antidepressant drug use.

PURPOSE: Psychiatric morbidity is common in cancer patients toward the end of life. In this study, we demonstrate the use of an electronic patient record system to identify patients with advanced cancer, and then analyze practice patterns regarding the use of antidepressants. PATIENTS AND METHODS: Using electronic patient records from January 1986 to December 1996, we identified 17,476 patients with possible cancer. Patients were identified by virtue of having any one of eight markers. We used an iterative process to modify the specificity of these markers, and an advanced cancer cohort was assembled consisting of 1185 patients. RESULTS: A random sample of 200 written medical records were reviewed, of which 157 records (78.5%) were retrieved. Extracted information was reviewed by an oncologist, and patients were classified as follows: (1) no evidence of cancer; (2) evidence of cancer with an expected survival of less than or equal to 24 months; or (3) evidence of cancer with an expected survival of more than 24 months. Overall, 86% of the advanced cancer sample assembled from electronic records was correctly classified as advanced cancer by the review of the written records. Overall, 16% of all 1185 patients with advanced cancer were exposed to at least one antidepressant, with 3% of patients exposed to a selective serotonin-reuptake inhibitor, 10% exposed to a tricyclic antidepressant at a dose of greater than 25 mg, and 4% exposed to a low dose of a tricyclic antidepressant. CONCLUSION: The electronic patient record can be used to assemble an advanced cancer cohort for the purpose of studying palliative care practice patterns. Antidepressants are seldom part of the palliative management of this population.

Journal Article↗
PubMed1999

Linking clinical relevance and statistical significance in evaluating intra-individual changes in health-related quality of life.

OBJECTIVE: To compare the standard error of measurement (SEM) with established standards for clinically relevant intra-individual change in an evaluation of health-related quality of life. DESIGN: Secondary analysis of data from a randomized controlled trial. SUBJECTS: Six hundred and five outpatients with a history of cardiac problems attending the general medicine clinics of a major academic medical center. MEASURES: Baseline and follow-up interviews included a modified version of the Chronic Heart Failure Questionnaire (CHQ) and the SF-36. The SEM values corresponding to established standards for minimal clinically important differences (MCIDs) on the CHQ were determined. Individual change on the SF-36 was explored using the same SEM criterion. RESULTS: One-SEM changes in this population corresponded well to the patient-driven MCID standards on all CHQ dimensions (weighted kappas (0.87; P < 0.001). The distributions of outpatients who improved, remained stable, or declined (defined by the one-SEM criterion) were generally consistent between CHQ dimensions and SF-36 subscales. CONCLUSIONS: The use of the SEM to evaluate individual patient change should be explored among other health-related quality of life instruments with established standards for clinically relevant differences. Only then can it be determined whether the one-SEM criterion can be consistently applied as a proxy for clinically meaningful change.

Adult↗
PubMed1999

Measuring patient satisfaction for quality improvement.

BACKGROUND: Surveys used for health plan quality reporting are generally administered annually to health plan enrollees to assess satisfaction with both the health plan and health care services. Therefore, surveys may lack sensitivity to measure the effects of patient-focused, quality improvement initiatives that could demonstrate results in a shorter time period. OBJECTIVES: We describe the development and testing of a multidimensional, visit-specific measure of satisfaction with primary care that may be used in quality improvement. METHODS: Conducted in five adult and pediatric primary care sites serving a commercial, largely managed-care population, the survey includes the Medical Outcomes Study Visit-Specific Questionnaire, the American Board of Internal Medicine Patient Satisfaction Questionnaire, and locally developed items. We assessed the instrument's reliability, validity, and utility for quality improvement. RESULTS: For both adult and pediatric samples, three factors emerged: satisfaction with the provider, satisfaction with access, and satisfaction with the office. Satisfaction with the provider and with the office were independently correlated with overall satisfaction in both samples; satisfaction with access was significantly correlated with overall satisfaction only for adults. For adults, patients who disenrolled from the health plan were less satisfied with the office compared with patients who remained with the health plan. Finally, for adults, we detected significant differences across practice sites in terms of satisfaction with office and access; for children, there were intersite differences in terms of satisfaction with provider, office, and access. CONCLUSIONS: We have support for the reliability and validity of this instrument that has identified differences in satisfaction between practice sites that may be used for quality improvement.

Adult↗
PubMed1998

Racial disparity in cardiac decision making: results from patient focus groups.

BACKGROUND: While numerous studies suggest that African Americans receive fewer invasive cardiac procedures than whites, the basis for these treatment differences is not understood. METHODS: We conducted focus group sessions with patients who had received treatment in the hospital or the emergency department within the preceding 3 months for ischemic heart disease at 2 urban, university-affiliated hospitals. RESULTS: Discussions with patients identified the following factors that influenced their decision making: clarity, simplicity, and consistency of treatment recommendations; advice from friends and family about whether to accept recommendations; availability to speak with others who accepted similar recommendations; and having honest and caring physicians. African American patients identified the following additional factors that influenced their decision making: perceptions of health care discrimination; perceptions of undesirable physician behavior; faith in God to control one's destiny; and patient-physician camaraderie. CONCLUSIONS: Participants identified common issues influencing health care decision making, regardless of race. However, additional factors were expressed only by African American participants. These factors conveyed racial differences in perceptions of the health care system that may, in part, contribute to differences in health care decision making and treatment.

Black or African American↗
PubMed1998

Effectiveness of computer-generated reminders for increasing discussions about advance directives and completion of advance directive forms. A randomized, controlled trial.

BACKGROUND: Physicians can increase the rate of completion of advance directive forms by discussing directives with their patients, but the means by which physicians can be induced to initiate these discussions are unclear. Computer-generated reminders have been shown to increase physician compliance with practice guidelines. OBJECTIVE: To determine the effects of computer-generated reminders to physicians on the frequency of advance directive discussions between patients and their primary caregivers and the frequency of consequent establishment of advance directives. DESIGN: Randomized, controlled trial with a 2 x 2 factorial design. SETTING: An outpatient general medicine practice associated with an urban public hospital. PARTICIPANTS: Participants were 1) 1009 patients who were at least 75 years of age or were at least 50 years of age with serious underlying disease and 2) 147 primary care physicians (108 housestaff and 39 faculty). INTERVENTION: Computer-generated reminders that recommended discussion of one or both of two types of advance directives compared with no reminders. MEASUREMENTS: Discussions about advance directives, determined by patient interviews after all scheduled patient-physician outpatient encounters, and completed advance directive forms. The study period was approximately 1 year. RESULTS: Physicians who did not receive reminders (controls) discussed advance directives with 4% of the study patients compared with 24% for physicians who received both types of reminders (adjusted odds ratio, 7.7 [95% CI, 3.4 to 18]; P < 0.001). Physicians who did not receive reminders completed advance directive forms with only 4% of their study patients compared with 15% for physicians who received both types of reminders (adjusted odds ratio, 7.0 [CI, 2.9 to 17]; P < 0.001). Overall, 45% of patients with whom advance directives were discussed completed at least one type of advance directive. CONCLUSIONS: Simple computer-generated reminders aimed at primary caregivers can increase the rates of discussion of advance directives and completion of advance directive forms among elderly outpatients with serious illnesses.

Advance Care Planning↗
PubMed1998

Effects of multidisciplinary case management in patients with chronic renal insufficiency.

PURPOSE: Though case management has been recommended to improve the outcomes of patients with costly or morbid conditions, it has seldom been studied in controlled trials. We performed a randomized, controlled clinical trial of an intensive, multidisciplinary case management program for patients with chronic renal insufficiency and followed patients for 5 years. PATIENTS AND METHODS: We enrolled 437 primary-care patients (73% of those eligible) with chronic renal insufficiency (estimated creatinine clearance consistently < 50 mL/min with the last serum creatinine level > 1.4 mg/dL) who were attending an urban academic general internal medicine practice. The intensive case management, administered during the first 2 years after enrollment, consisted of mandatory repeated consultations in a nephrology case management clinic staffed by two nephrologists, a renal nurse, a renal dietitian, and a social worker. Control patients received usual care. Primary outcome measurements included serum creatinine level, estimated creatinine clearance, health services use, and mortality in the 5 years after enrollment. Secondary measures included use of renal sparing and potentially nephrotoxic drugs. RESULTS: There were no differences in renal function, health services use, or mortality in the first, second, or third through fifth years after enrollment. There were significantly more outpatient visits among intervention patients, mainly because of the added visits to the nephrology case management clinic. There were also no significant differences in the use of renal sparing or selected potentially nephrotoxic drugs. The annual direct costs of the intervention were $89,355 ($484 per intervention patient). CONCLUSION: This intensive, multidisciplinary case-management intervention had no effect on the outcomes of care among primary-care patients with established chronic renal insufficiency. Such expensive and intrusive interventions, despite representing state-of-the-art care, should be tested prospectively before being widely introduced into practice.

Aged↗
PubMed1998

Can raters consistently evaluate the content of focus groups?

Focus groups are increasingly being used to provide insights to researchers and policy makers. These data complement quantitative approaches to understanding the world. Unfortunately, quantitative and qualitative methodologies have often been viewed as antithetical, rather than complementary, strategies. While focus groups can clearly generate rich information that is unobtainable through other quantitative methods, it is important to determine the degree to which different raters can consistently extract information from transcripts. Thus, our goal was to quantify agreement in the interpretation of transcripts from patient and physician focus groups, using decision-making in ischemic heart disease as a model. We used data from focus groups with both patients and physicians that sought to identify factors affecting diagnostic and treatment decisions in ischemic heart disease. Three raters independently reviewed transcribed audiotapes from focus groups of patients with ischemic heart disease, as well as focus groups of physicians who care for these patients. We found that raters could not distinguish between major and minor factors reliably. More troubling, however, is that consistency regarding the apparently straightforward judgment as to the mere presence or absence of a factor was difficult to achieve. In particular, the three raters of each transcript failed to agree on between one third and one half of the factors. This reasonably high level of disagreement occurred despite the raters: (1) having generated the individual factors themselves based upon their reading a random sample of actual transcripts and (2) being trained in the use of rating forms (including standard definitions of themes). These data suggest that if a single rater evaluates focus group transcripts, as is commonly done, judgments may not be reproducible by other raters. Moreover, a single rater may not extract all important information contained in the transcripts.

Female↗
PubMed1998

Mortality, symptoms, and functional impairment in late-life depression.

OBJECTIVE: To determine whether depressive symptoms measured at baseline are associated with mortality and to describe the course of depressive symptoms and their relation to physical decline in patients over a 6-year period. DESIGN: Prospective cohort study conducted from 1990 through 1996. SETTING: Urban academic primary care group practice. PATIENTS: A cohort of 3,767 patients aged 60 years and older screened for depressive symptoms during routine office visits using the Centers for Epidemiologic Studies Depression Scale (CES-D) participated in the mortality study. A subsample of 300 patients with CES-D scores 16 or above and a subsample of 100 patients with CES-D scores less than 16 participated in the study of the course of depressive symptoms and physical decline. MEASUREMENTS AND MAIN RESULTS: Mortality by December 1995 was measured for all screened patients; reinterviewed patients completed the CES-D and the Sickness Impact Profile (SIP). The mean follow-up period was 45 months (+/- SD 12.2 months); 561 (14.9%) of the patients died by December 1995. In proportional hazards models, age, gender, race, history of smoking, serum albumin value, and an ideal body weight in the lowest 10% were significant correlates of time to death, but the baseline CES-D was not. Patients with depressive symptoms had significantly worse physical and psychosocial functioning scores on the SIP than did patients without depressive symptoms. Using the generalized estimating equation method, the strongest predictor of the current CES-D score was the patient's prior CES-D score. However, worsening physical functioning score on the SIP was also independently correlated with worse CES-D scores p < or = .001). CONCLUSIONS: Symptoms of depression were not associated with mortality in this cohort of older adults. However, patients with depressive symptoms reported greater functional impairment than did those without depressive symptoms. Moreover, decline in physical functioning was independently correlated with a concurrent increase in depressive symptoms.

Activities of Daily Living↗
PubMed1998

Self-rated health and adverse health outcomes: an exploration and refinement of the trajectory hypothesis.

OBJECTIVES: To examine the hypothesis that the relationship between poor self-rated health and adverse health outcomes simply reflects self-assessments of declining health trajectories rather than current health status. METHODS: A 12-month follow-up of 786 disadvantaged adults aged 50-99 years old was conducted. Baseline markers of poor self-rated health and declining health trajectory were used to predict 12-month follow-up reports of the expectations of being hospitalized within one year, being placed in a nursing home within five years, and dying within ten years. Hierarchical multivariable logistic regression was used with poor self-rated health entered first, standard epidemiologic covariates entered next, and declining health trajectory entered last. RESULTS: Poor self-rated health was not independently associated with expectations for being hospitalized, but declining health trajectory independently increased the risk of such expectations by 65-88%. Neither poor self-rated health nor declining health trajectory were independently associated with expectations for being placed in a nursing home, but both were independently associated with expectations for dying, increasing such expectations by 70-105%. DISCUSSION: The relationship between poor self-rated health and adverse outcomes is not a simple reflection of unmeasured self-assessments of impending decline or doom. Rather, the effects of poor self-rated health and declining health trajectory appear to be independent and complementary.

Aged↗
PubMed1998

Dimensions and correlates of physician work satisfaction in a midwestern city.

OBJECTIVES: The purpose of this study was to examine the dimensions of physician work satisfaction across a variety of medical specialties and practice settings. METHODS: A modified version of the Scheckler et al survey instrument was mailed to all physicians in Marion County, Indiana. Forty-two percent (777) of the eligible physicians responded. Exploratory factor analysis and internal consistency measures were used to assess the instrument's validity and reliability. Multivariable linear regression was used to predict global and summary scale scores. RESULTS: Four dimensions of physician work satisfaction were identified: relationships with patients (k = 6, alpha = 0.81), autonomy in clinical decision-making (k = 8, alpha = 0.81), office resources (k = 7, alpha = 0.87), and professional relationships (k = 5, alpha = 0.82). Most (73%) of the physicians were satisfied with their overall practice, and the majority were also satisfied with their income. Significant differences were observed in the sources and magnitude of physician work satisfaction across medical specialty, practice setting, and financial arrangement. Physicians in private practice were most satisfied with their overall practice and office resources, whereas physicians in health maintenance organizations (HMOs) were most satisfied with their autonomy in clinical decision-making. Physicians not working in HMOs but having a large percentage of patients with capitated reimbursement were not enthusiastic about the effect of managed care on their medical practice. Among primary care physicians, family practitioners and general internists were generally less satisfied, and general pediatricians were generally more satisfied with most aspects of their medical practices. CONCLUSIONS: The modified version of the Scheckler et al instrument is a reliable and valid measure of physician work satisfaction. Increases in the market share of managed care have differentially affected the work satisfaction of physicians based on their medical specialty, practice setting, and financial arrangements.

Adult↗
PubMed1998

Changes in the SF-36 in 12 months in a clinical sample of disadvantaged older adults.

OBJECTIVES: The present study assessed changes in the Medical Outcomes Study 36-item short-form health survey (SF-36) during a 12-month period and examined the relation of those changes to selected baseline characteristics. METHODS: The study was a 12-month follow-up evaluation of 786 disadvantaged adults aged 50 to 99 years old who had participated in a randomized controlled clinical trial in the general medicine outpatient clinics of a major academic medical center. Descriptive and psychometric analyses of changes in the SF-36 scale scores during a 12-month period were performed, and two series of multivariable logistic regressions of increases or decreases greater than one standard error of measurement (SEM) versus stability on selected baseline characteristics were done. Measures were the eight SF-36 scales. RESULTS: Mean baseline scores on the SF-36 scales were substantially below age-specific national norms. Problematic floor and/or ceiling effects were found for the bodily pain, social function, role--physical, and role--emotional scales, consistent with age-specific national norms. Internal consistency was unacceptable for the general health perceptions scale, adequate for the social function scale, and good for all the other SF-36 scales. Improvements greater than one standard error of measurement were found for between one fifth and one third of the patients, and declines greater than one standard error of measurement were found for between one fifth and one third of the patients. Selected baseline characteristics generally were unrelated to either improvements or declines on the SF-36 scales. CONCLUSIONS: The SF-36 scales appear to be sufficiently sensitive for measuring changes in health outcomes during a 1-year period in older patients with debilitating disease. Little of the measured change, however, was predictable.

Adult↗
PubMed1998

Cost of health care for a community of older adults in an urban academic healthcare system.

BACKGROUND: Urban academic medical centers provide care for large populations of vulnerable older adults. These patients often suffer a disproportionate share of chronic illnesses, disabilities, and social stressors that may increase health care costs. OBJECTIVE: To describe the distribution and content of total healthcare costs accrued over a 4-year period by a community of older adults cared for in an urban academic healthcare system and to describe high-cost patients and utilization patterns. DESIGN: A cohort study. SETTING: A tax-supported public healthcare system consisting of a 450-bed hospital and seven community-based ambulatory care centers. PATIENTS: 12,581 patients aged 60 years and older who had at least two ambulatory visits and/or one hospitalization within the healthcare system from 1993 through 1995. MEASUREMENTS: Patient demographic and clinical characteristics, hospital and ambulatory utilization rates, and all healthcare costs accrued from 1993 through 1996 were determined. Costs were estimated from the perspective of the healthcare system using cost to charge ratios. MAIN RESULTS: The mean patient age was 70 years, 60% were women, 44% were Black, and 83% were covered by Medicare and/or Medicaid. Nearly 25% of patients were obese, 15.8% had a history of smoking, and 15.5% had evidence of malnutrition. The mean number of ambulatory visits per year was 4.3 (+/-7.2), and 38.1% of patients had been hospitalized one or more times. Within the 4-year window, 24.1% of patients had missed five or more appointments with their primary care physicians, 32.7% of patients had five or more unscheduled clinic visits, and 12.5% had five or more emergency room visits. Total health care costs for 4 years for this cohort of older adults was $125.2 million dollars, with per capita annual mean costs of $3893. Expenditures associated with hospitalizations accounted for 63.6% of healthcare costs. Total inpatient and outpatient costs for the 38% of patients hospitalized at least once accounted for 85.3% of all health care expenditures. Patients who died in the hospital did not accrue significantly greater costs than patients who died out of the hospital. Simulations of a random 5% adverse selection of high-cost patients among two capitated systems resulted in cost shifts of $11.1 million. Recorded smoking history, obesity, and low serum albumin were significantly associated with excess costs. CONCLUSIONS: Healthcare costs are concentrated in a significant minority of older adults. Costs accrued in conjunction with hospital stays dominate healthcare expenditures for this cohort of older adults. However, most older adults (83%) have one or fewer hospital episodes in a 4-year period. Although patients who died accrued greater healthcare costs, these costs were not higher when the death occurred in the hospital. Self-care behaviors are an important target for interventions to reduce costs.

Academic Medical Centers↗