Pursuing integration of performance measures into electronic medical records: beta-adrenergic receptor antagonist medications

OBJECTIVE

Electronic medical records seldom integrate performance indicators into daily operations. Assessing quality indicators traditionally requires resource intensive chart reviews of small samples. We sought to use an electronic medical record to assess use of beta-adrenergic antagonist medications (beta-blockers) following myocardial infarction, to compare a standardized manual assessment with assessment using electronic medical records, and to discuss potential for future integration of performance indicators into electronic records.

DESIGN

Cross-sectional data analysis.

SETTING

An urban academic medical center.

PARTICIPANTS

US Medicare beneficiaries 65 years of age or older, admitted to hospital with myocardial infarction between 1995 and 1999.

MEASUREMENTS AND MAIN RESULTS

Manual chart review was compared with a computer driven assessment of electronic records. Administration of beta-blockers and cases excluded from use of beta-blockers were measured, based on Medicare criteria. Among 4490 older adults, 391 (4%) of 9018 hospital admissions contained codes for myocardial infarction. In 323 (83%) of the 391 hospital admissions, criteria for excluding beta-blockers were met; 235 (60%) were excluded due to heart failure. Of 68 hospital admissions for myocardial infarction that did not meet exclusion criteria, physicians prescribed beta-blockers in 49 (72%) on admission and 42 (62%) at discharge. Compared with manual chart review, electronic review had a sensitivity of 83-100% and led to fewer false negative findings.

CONCLUSIONS

An electronic medical records system can be used instead of chart review to measure use of beta-blockers after myocardial infarction. This should lead to integration of real time automated performance measurement into electronic medical records.

Reasons for failure of the ketogenic diet

This study examined reasons why patients discontinue the ketogenic diet. A total of 46 children placed on the ketogenic diet between November 1994 and August 1996 were followed prospectively. Reasons for discontinuing the diet prior to 6 months were analyzed. Nineteen (41%) children discontinued the diet for either medical or nonmedical reasons. Nonmedical reasons were caregiver issues and patients' unwillingness to follow the diet. Noncompliance was more common in older children. The ketogenic diet, while effective, is a very stringent diet. Nonmedical reasons for discontinuation are as common as the traditional medical reasons of lack of efficacy or complications.

Healthcare costs associated with percutaneous endoscopic gastrostomy among older adults in a defined community

OBJECTIVES

The effectiveness of percutaneous endoscopic gastrostomy (PEG) in older adults remains controversial. Although prior studies have examined the safety of PEG and its impact on nutrition, there are limited data on the economic costs. The purpose of this study is to describe the healthcare costs associated with PEG tube feeding over 1 year.

DESIGN

Prospective cohort study.

SETTING

Small community of approximately 60,000 residents served by two hospital systems.

PARTICIPANTS

One hundred five (70%) of 150 patients age 60 and older receiving PEG over a 24-month period in the targeted community who permitted access to their medical records.

MEASUREMENTS

Patients were interviewed at baseline and every 2 months for 1 year to obtain information on use of enteral formula, complication rates, and health services use. We obtained inpatient charge data for all hospitalizations and PEG procedures for 1 year.

RESULTS

Censoring patients at death or 1 year post-PEG, the mean number of days of PEG tube feeding was 180 (range 5-365). The average cost for PEG tube feeding for this cohort of patients was $7,488 (median $3,691) in 1997 and 1998. The average daily cost of PEG tube feeding was $87.21 (median $33.50). The estimated cost of providing 1 year of feeding via PEG is $31,832 (median $12,227). The main components of this cost include the initial PEG procedure (29.4%), enteral formula (24.9%), and hospital charges for major complications (33.4%).

CONCLUSIONS

Direct charges associated with PEG tube feeding over 1 year are conservatively estimated at $31,832; there was considerable variation in charges because of the cost of rare but costly major complications. Also, feeding patients via PEG resulted in cost shifts in terms of the primary payor. The economic cost of PEG tube feeding is another consideration in decision making for long-term enteral feeding among older adults.

Quality improvement research on late life depression in primary care

BACKGROUND

Two million older Americans suffer from depression annually. Depression causes more functional impairment than many other common medical conditions and older adults have the highest rate of suicide in the United States. Although many of these patients fail to seek or fail to receive care for depression, the majority will be seen in primary care for the treatment of other conditions.

OBJECTIVE

To review the health services research on quality improvement for late life depression.

METHODS

Qualitative literature review.

RESULTS

During the past 30 years, multiple educational and quality improvement interventions have been designed and tested to improve the recognition and treatment of depression in primary care settings. The findings from this large body of health services research suggest that: (1) the outcome of major depression in the usual care of primary care is typically poor; this is particularly true of late life depression; (2) informational support provided to primary care physicians is necessary but insufficient to improve the outcomes of late life depression in primary care; achieving guideline-level therapy requires the substantial participation of an informed and motivated patient working in concert with a health care team and health care system designed to care for chronic conditions; (3) up to 30% of older primary care patients will fail to respond to excellent guideline-level therapy provided in primary care; and (4) the latest quality improvement efforts focus not only on the clinical skills of primary care physicians, but also on patient's self-care and on innovative strategies to improve the system of care.

CONCLUSIONS

Late life depression is often a chronic disease and outcomes research demonstrates that quality improvement efforts that focus resources on improving systems of care and the active participation of patients offer the best evidence of improved patient outcomes.

Improving primary care for depression in late life: the design of a multicenter randomized trial

BACKGROUND

Late life depression can be successfully treated with antidepressant medications or psychotherapy, but few depressed older adults receive effective treatment.

RESEARCH DESIGN

A randomized controlled trial of a disease management program for late life depression.

SUBJECTS

Approximately 1,750 older adults with major depression or dysthymia are recruited from seven national study sites.

INTERVENTION

Half of the subjects are randomly assigned to a collaborative care program where a depression clinical specialist supervised by a psychiatrist and a primary care expert supports the patient's regular primary care provider to treat depression. Intervention services are provided for 12 months using antidepressant medications and Problem Solving Treatment in Primary Care according to a stepped care protocol that varies intervention intensity according to clinical needs. The other half of the subjects are assigned to care as usual.

EVALUATION

Subjects are independently assessed at baseline, 3 months, 6 months, 12 months, 18 months, and 24 months. The evaluation assesses the incremental cost-effectiveness of the intervention compared with care as usual. Specific outcomes examined include care for depression, depressive symptoms, health-related quality of life, satisfaction with depression care, health care costs, patient time costs, market and nonmarket productivity, and household income.

CONCLUSIONS

The study blends methods from health services and clinical research in an effort to protect internal validity while maximizing the generalizability of results to diverse health care systems. We hope that this study will show the cost-effectiveness of a new model of care for late life depression that can be applied in a range of primary care settings.

Cognitive impairment and mortality in older primary care patients

OBJECTIVE

To assess the impact of cognitive impairment on mortality in older primary care patients after controlling for confounding effects of demographic and comorbid chronic conditions.

DESIGN

Prospective cohort study.

SETTING

Academic primary care group practice.

PARTICIPANTS

Three thousand nine hundred and fifty-seven patients age 60 and older who completed the Short Portable Mental Status Questionnaire (SPMSQ) during routine office visits.

MEASUREMENTS

Cognitive impairment measured at baseline using the SPMSQ, demographics, problem drinking, history of smoking, clinical data (including weight, cholesterol level, and serum albumin), and comorbid chronic conditions collected at baseline; survival time measured during the 5 to 7 years after baseline.

RESULTS

Eight hundred and eighty-six patients (22.4%) died during the 5 to 7 years of follow-up. Cognitive impairment was categorized as having no impairment (84.3%), mild impairment (10.5%), and moderate-to-severe impairment (5.2%) based on SPMSQ score. Chi-square tests revealed that patients with moderate-to-severe impairment were significantly more likely to die compared with patients with mild impairment (40.8% vs 21.5%) and those with no impairment (40.8% vs 21.4%). No significant difference in crude mortality was found between patients with no impairment and those with mild impairment. After analyzing time to death using the Kaplan-Meier method, patients with moderate-to-severe cognitive impairment were at increased risk of death compared with those with no or mild impairment (Log-rank chi(2) = 55.5; P <.0001). Even in multivariable analyses using Cox proportional hazards to control for confounding factors, compared with those with no impairment, moderately-to-severely impaired patients had an increased risk of death, with a hazard ratio (HR) of 1.70. Increased risk of death was also associated with older age (HR = 1.03 for each year), a history of smoking (HR = 1.48), having a serum albumin level <3.5 g/L (HR = 1.29), and weighing less than 90% of the ideal body weight (HR = 1.98). Outpatient diagnoses associated with increased mortality risk were diabetes mellitus, coronary artery disease, congestive heart failure, cerebrovascular disease, cancer, anemia, and chronic obstructive pulmonary disease (HR range 1.36-1.67). Factors protective of mortality risk included female gender (HR = 0.67) and black race (HR = 0.73).

CONCLUSIONS

Moderate-to-severe cognitive impairment is associated with an increased risk of mortality, even after controlling for confounding effects of demographic and clinical characteristics. Mild cognitive impairment is not associated with mortality risk, but a longer follow-up period may be necessary to identify this risk if it exists.

Outcomes of percutaneous endoscopic gastrostomy among older adults in a community setting

OBJECTIVE

Percutaneous endoscopic gastrostomy (PEG) has become the preferred method to provide enteral tube feeding to older adults who have difficulty eating, but the impact of PEG on patient outcomes is poorly understood. The objective of this study was to describe changes in nutrition, functional status, and health-related quality of life among older adults receiving PEG.

DESIGN

A prospective cohort study.

SETTING

A small community of approximately 60,000 residents served by two hospital systems.

PARTICIPANTS

One hundred fifty patients aged 60 and older receiving PEG from one of the four gastroenterologists practicing in the targeted community.

MEASUREMENTS

Patients were assessed at baseline and every 2 months for 1 year to obtain clinical characteristics, process of care data, physical and cognitive function, subjective health status, nutritional status, complications, and mortality.

RESULTS

Over a 14-month period, 150 patients received PEG tubes in the targeted community; the mean age was 78.9. The most frequent indications for the PEG were stroke (40.7%), neurodegenerative disorders (34.7%), and cancer (13.3%). All measures of functional status, cognitive status, severity of illness, comorbidity, and quality of life demonstrated profound and life-threatening impairment; 30-day mortality was 22% and 1-year mortality was 50%. Among patients surviving 60 days or more, at least 70% had no significant improvement in functional, nutritional, or subjective health status. Serious complications were rare, but most patients experienced symptomatic problems that they attributed to the enteral tube feeding.

CONCLUSIONS

PEG tube feeding in severely and chronically ill older adults can be accomplished safely. However, there are important patient burdens associated with the PEG and there was limited evidence that the procedure improves functional, nutritional, or subjective health status in this cohort of older adults. The issues raised in this descriptive study provide impetus for a randomized trial of PEG tube feeding compared with alternative methods of patient care for older adults with difficulty eating.

Association between medication supplies and healthcare costs in older adults from an urban healthcare system

OBJECTIVES

The amount of medication dispensed to older adults for the treatment of chronic disease must be balanced carefully. Insufficient medication supplies lead to inadequate treatment of chronic disease, whereas excessive supplies represent wasted resources and the potential for toxicity. We used an electronic medical record system to determine the distribution of medications supplied to older urban adults and to examine the correlations of these distributions with healthcare costs and use.

DESIGN

A cross-sectional study using data acquired over 3 years (1994-1996).

SETTING

A tax-supported urban public healthcare system consisting of a 300-bed hospital, an emergency department, and a network of community-based ambulatory care centers.

PATIENTS

Patients were >60 years of age and had at least one prescription refill and at least two ambulatory visits or one hospitalization during the 3-year period.

MEASUREMENTS

Focusing on 12 major categories of drugs used to treat chronic diseases, we determined the amounts and direct costs of these medications dispensed to older adult patients. Amounts of medications that were needed by patients to medicate themselves adequately were compared with the medication supply actually dispensed considering all sources of care (primary, emergency, and inpatient). We calculated the excess drug costs attributable to oversupply of medication (>120% of the amount needed) and the drug cost reduction caused by undersupply of medication (<80% of the amount needed). We also compared total healthcare use and costs for patients who had an oversupply, an undersupply, or an appropriate supply of their medications.

RESULTS

The cohort comprised 4164 patients with a mean age of 71 +/- 7 (SD) who received a mean of 3 +/- 2 (SD) drugs for chronic conditions. There were 668 patients (16%) who received <80% of the supply needed, 1556 patients (37%) who received between 80 and 120% of the supply needed, and 1940 patients (47%) who received >120% of the supply needed. The total direct cost of targeted medications for 3 years was $1.96 million or, on average, $654,000 annually. During the 3-year period, patients receiving >120% of their needed medications had excess direct medication costs of $279,084 or $144 per patient, whereas patients receiving <80% of drugs needed had reduced medication costs of $423,438 or $634 per patient. Multivariable analyses revealed that both under- and over-supplies of medication were associated with a greater likelihood of emergency department visits and hospital admissions.

CONCLUSIONS

More than one-half of the older adults in our study have under- or over-supplies of medications for the treatment of their chronic diseases. Such inappropriate supplies of medications are associated with healthcare utilization and costs.

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.

Symptoms in hospitalized patients: outcome and satisfaction with care

PURPOSE

Physical symptoms are responsible for substantial morbidity in outpatients. We assessed symptoms in patients admitted to a hospital to determine their frequency, persistence at discharge, and the relation between symptom outcome and satisfaction with care.

METHODS

During a 12-month period, 2,126 hospitalized medical patients completed a study interview within 2 hours of admission. More than half (n = 1,168) of the patients were re-interviewed within 24 hours of discharge. We ascertained the presence and severity of 11 physical symptoms, as well as activities of daily living, mobility, mood, self-rated health, physiologic severity of illness, satisfaction with care, and length of stay.

RESULTS

Symptoms were common at the time of hospital admission, particularly fatigue (80% of patients), dyspnea (60%), cough (51%), dizziness (51%), headache (47%), chest pain (46%), and nausea or vomiting (43%). Individual symptoms failed to resolve by hospital discharge approximately 25% to 50% of the time. The three most prominent predictors of persistence of symptoms were shorter length of stay, severity of the symptom on admission, and total symptom count. Patient satisfaction with care was associated with total symptom severity score at discharge and the degree of symptomatic improvement that had occurred during hospitalization.

CONCLUSION

Because symptoms are common at discharge and associated with decreased satisfaction with care, asking about them would be a reasonable way to enhance patient-oriented care.

Decision-making for percutaneous endoscopic gastrostomy among older adults in a community setting

OBJECTIVE

To describe clinical decision-making for percutaneous endoscopic gastrostomy from the perspective of patients, caregivers, and physicians.

DESIGN

A prospective cohort study.

SETTING AND PATIENTS

All patients aged 60 and older receiving percutaneous endoscopic gastrostomies in a defined community over a 16-month period.

MAIN OUTCOMES MEASURES

Either patients or their surrogate decision-makers completed a semistructured face-to-face interview to map out the information gathering process, expectations, and discussants involved in the decision to proceed with gastrostomy feeding. Physicians completed a written questionnaire to determine their likelihood of recommending percutaneous endoscopic gastrostomy, their involvement in the decision-making and recommendation process, and sources of perceived pressure in the decision-making.

RESULTS

We identified 100 patients who received percutaneous endoscopic gastrostomy during the study window and 82 primary care physicians who provided care in the defined community. The most common reasons for the procedure were stroke, neurologic disease, and cancer. Patients or their surrogate decision-makers reported multiple discussants, incomplete information, and considerable distress in arriving at the decision to proceed with artificial feeding. This distress was usually in the context of an acute and debilitating illness that often overshadowed the decision about artificial feeding. The decision for gastrostomy often appeared to be a "non-decision" in the sense that decision-makers perceived few alternatives. Physicians also reported considerable distress in arriving at recommendations to proceed with percutaneous endoscopic gastrostomy, including perceived pressures from families or other healthcare professionals. Physicians have clear patterns of triage for percutaneous endoscopic gastrostomy, but the assumptions underlying these patterns are not well supported by the medical literature.

CONCLUSIONS

Patients, caregivers, and physicians are often compelled to make decisions about long-term enteral feeding under tragic circumstances and with incomplete information. Decision-makers typically do not perceive any acceptable alternatives. Because data on these patients' long-term functional outcomes are lacking, decision-makers appear to focus primarily on the short-term safety of the procedure and the potential for improved nutrition.

Association of antioxidants with memory in a multiethnic elderly sample using the Third National Health and Nutrition Examination Survey

Oxidative stress has been implicated both in the aging process and in the pathological changes associated with Alzheimer's disease. Antioxidants, which have been shown to reduce oxidative stress in vitro, may represent a set of potentially modifiable protective factors for poor memory, which is a major component of the dementing disorders. The authors investigated the association between serum antioxidant (vitamins E, C, A, carotenoids, selenium) levels and poor memory performance in an elderly, multiethnic sample of the United States. The sample consisted of 4,809 non-Hispanic White, non-Hispanic Black, and Mexican-American elderly who visited the Mobile Examination Center during the Third National Health and Nutrition Examination Survey, a national cross-sectional survey conducted from 1988 to 1994. Memory is assessed using delayed recall (six points from a story and three words) with poor memory being defined as a combined score less than 4. Decreasing serum levels of vitamin E per unit of cholesterol were consistently associated with increasing levels of poor memory after adjustment for age, education, income, vascular risk factors, and other trace elements and minerals. Serum levels of vitamins A and C, beta-carotene, and selenium were not associated with poor memory performance in this study.

Accuracy of self-reported health services use and patterns of care among urban older adults

BACKGROUND

Understanding older adults' volume and patterns of health service use is fundamental to efforts to improve the quality and efficiency of services.

OBJECTIVE

To analyze the accuracy of older adults' self report of health services use and to determine the proportion of care obtained outside a defined urban academic health care system.

RESEARCH DESIGN

Telephone survey of self-reports validated against data routinely archived in an electronic medical record system.

SUBJECTS

Stratified random sample of 422 patients (> or = 60 years) who had contact with the health care system at least once in the previous 3 months.

MEASURES

Self reports of hospitalizations, emergency room visits, physicians visits, extended care visits, and home care visits over the past 12 months, health status, physical activity, and sociodemographics factors.

RESULTS

The sample population was more likely to report health services use and functional disability than was a community-based sample of older adults; 67% of the sample were women, 53.9% were African American, 71% were age 65 and over, 38.7% lived alone, and 24.6% reported poor financial resources. Based on data from the electronic medical record, 27.9% of the sample were hospitalized at least once in the prior 12 months, 54.6% had at least one emergency room visit, and the mean number of ambulatory visits was 8.1. Comparing self-report data to the electronic record data, 24.1% of older adults with a hospitalization in the prior 12 months failed to report the episode; 28.1% of those with an emergency room visit failed to report the episode as did 5.2% of those with an ambulatory care visit. The accuracy of the self reports of volume of these services were also substantially under reported. We were unable to identify any patient characteristics that were highly correlated with inaccuracy. We estimate that approximately 9.5% of health care costs are accrued outside this urban health care system.

CONCLUSIONS

These older adults substantially under-report health services use, including hospital episodes over a 12-month period. Reliance on self-reported use data over the prior year to model patterns of health care use among older adults is not supported by these data.

Validity and utility of patient-reported health measures on hospital admission

Patient perceptions of their health are often inadequately captured and explored on hospital admission where physiologic and other objective measures are the focus of attention. Therefore, we conducted a prospective study to develop and validate measures of several domains of patient-reported health status at the time of admission to a general medicine inpatient service, and to determine the value of these new measures in predicting length of stay (LOS). Within 2 hours of the time that a decision to admit a patient was made, research assistants delivered a structured interview that captured patients' current symptoms, functional status, mood, and perceived health. Interviews were conducted between 8 a.m. and 11 p.m., 5 days per week from July 1996 through June 1997. During this time, there were 3621 unique patients admitted to the medicine service; 2672 (74%) of these patients were eligible for an interview. Eighty-eight percent of the 2672 eligible patients were interviewed. In addition to the patient-reported measures captured through the structured interview, the acute physiology score (APS) of the APACHE II was calculated for all subjects. The internal consistency (i.e., Cronbach's alpha) of the scales was 0.76 or greater and concurrent validity (i.e., correlation) of the patient-reported measures with the APS was 0.01 to 0.13. Overall perceived health was correlated 0.20 to 0.45 with symptoms and functional status, and was correlated 0.07 with the APS. The patient-reported measures performed comparably to the APS in predicting LOS (R-square = 0.08). When the patient-reported measures and the APS were included in the same model, the R-square was 0.14. These analyses suggest that patient-reported measures of health and function on admission hold validity, and that responses to as few as 15 questions can provide data that may help to explain differences in length of a hospital stay.

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.

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.

Estimating risk associated with care in alternative settings: deterioration among children hospitalized

BACKGROUND

Although managed care favors use of alternative settings in an attempt to avoid hospitalization, uncertainty about possible deterioration creates concern about their safety.

OBJECTIVE

To derive preliminary estimates for the risk of adverse outcome in children hospitalized with acute illness who met criteria for admission to potentially less-expensive, alternative settings (eg, short-stay unit, home nursing).

DESIGN

Description of hospitalization outcomes for a community-wide childhood population.

SETTING AND POPULATION

All 11591 hospitalizations for residents of Monroe County (Rochester), New York, aged 1 month to 18 years in 1991 and 1992.

MEASUREMENTS

To identify potential adverse outcomes in alternative settings (numerator estimate), hospital medical records for admissions to regular inpatient units were examined. To ascertain deterioration among these admissions, detailed record reviews were conducted if the child died or was transferred to another hospital or to a critical care unit. To estimate the total number of admissions eligible for care in alternative settings (denominator estimate), hospital discharge files were analyzed.

RESULTS

Deterioration was found in 83 medical admissions. Of these 83, major chronic problems (n=53) or severe illness at presentation (n=27) precluded alternative setting eligibility, leaving only 3 in whom alternative setting care might have been considered. The total number of admissions eligible for alternative setting care was estimated between 1661 (restrictive criteria) and 3322 (inclusive criteria) for the 2-year observation period. Based on these observations, best- and worst-case estimates for the risk of deterioration in candidates for care in alternative settings were 0.6 and 1.8 per 1000, respectively. For the 3 children for whom alternative setting care might have been considered, the shortest period from first indication of deterioration to arrival in the critical care unit was 3.0 hours.

CONCLUSIONS

These preliminary estimates suggest that alternative settings may be safe for the care of many children currently hospitalized. A randomized clinical trial to evaluate directly the potential benefits and harms of alternative setting care should be considered.

Association of symptoms of depression with diagnostic test charges among older adults

BACKGROUND

Previous studies have documented greater use of health services by depressed persons and have postulated that health care costs could be reduced overall through better recognition and treatment of depression.

OBJECTIVE

To determine whether a greater burden of medical illness contributes to excess charges for diagnostic tests among older adults with symptoms of depression.

DESIGN

Prospective cohort study.

SETTING

A primary care group practice at an academic institution.

PATIENTS

3767 patients 60 years of age and older who completed testing on the Centers for Epidemiologic Studies Depression Scale (CES-D) during routine office visits.

MEASUREMENTS

Charges for all inpatient and ambulatory diagnostic testing for 2 years, including clinical pathology, diagnostic imaging, and special procedures; number of visits to the ambulatory care center or emergency department; and number of hospitalizations. The Ambulatory Care Group case-mix approach, which is based on ambulatory diagnoses, was used as a measure of health status and expected resource consumption.

RESULTS

Patients with symptoms of depression (CES-D scores > or = 16) were significantly younger (66.6 compared with 68.1 years; P < 0.001), more likely to be white (50.5% compared with 33.9%; P = 0.001), and more likely to be female (75.8% compared with 67.6%; P = 0.001) than were those without these symptoms (CES-D scores < 16). They also had more nonpsychiatric comorbid conditions, had more visits to the ambulatory care center (9.2 compared with 7.8; P < 0.001), were more likely to use the emergency department (52.3% compared with 40%; P = 0.001), were more likely to be hospitalized (22.4% compared with 17%; P = 0.002), and had greater median total diagnostic test charges for a period of 1 year ($583 compared with $387; P < 0.001). The difference in charges, most of which were clinical pathology charges (54.2%), persisted into the second year. Ambulatory Care Group assignment was independently associated with diagnostic test charges. The CES-D summary score was not independently associated with diagnostic test charges when controlling for Ambulatory Care Group assignment.

CONCLUSIONS

Patients with symptoms of depression accrue greater average diagnostic test charges. However, these data suggest that such patients also have a greater burden of comorbid nonpsychiatric illness. Efforts to improve outcome and decrease cost for patients who have late-life depression must target interventions to improve the care of psychiatric and medical illness concurrently.

Hospitalization for pneumonia among older adults

BACKGROUND

Hospitalizations and mortality due to pneumonia increase steadily with age. The purpose of this study is to describe the frequency, costs, and risk factors for hospitalization for pneumonia among older adults with particular attention to the effect of functional disability.

METHODS

The Longitudinal Study of Aging (LSOA) is the follow-up to the Supplement on Aging, which was appended to the 1984 National Health Interview Survey. Participants included a nationally representative sample of 7,527 community-dwelling adults aged 70 and older in 1984 who were followed prospectively for 8 years. The LSOA data are linked to the National Death Index and to yearly abstracts from the Medicare Automated Data Retrieval System (1984-1991).

RESULTS

From 1984 to 1991, 617 subjects (8.2%) had at least one hospitalization for pneumonia, 4,333 (57.5%) had at least one hospitalization for any reason, and 2,867 (38.1%) of the LSOA subjects died. The yearly frequency of hospitalization for pneumonia increased over time from 0.6% in 1984 to 2.4% in 1991. The median length of stay was 8 days, and the median hospital charge was $5,100. Over 8 years, median discounted charges for a pneumonia hospitalization increased 75% while length of stay remained relatively constant. Patients hospitalized for pneumonia had greater comorbidity, total hospital resource use, and mortality, but over 80% survived their first hospitalization for pneumonia. Hospitalization for pneumonia was associated with age, male gender, malnutrition, history of hip fracture, prior hospitalizations, and lower body limitations.

CONCLUSIONS

Hospitalization for pneumonia was frequent and accounted for 6% of the Medicare expenditures over 8 years among this cohort of older adults. Hospitalization for pneumonia occurred most often among subjects with prior evidence of failing health, but most subjects survived the first hospitalization for pneumonia. Even controlling for comorbidity, prior hospitalizations, and functional impairment, hospitalization for pneumonia was independently associated with age.

Suicidal ideation among older primary care patients

OBJECTIVE

To describe the prevalence and sociodemographic and clinical correlates of suicidal ideation among older primary care patients.

DESIGN

Baseline screening for depressive symptoms, functional status, and suicidal ideation, with prospective assessment of mortality.

SETTING

An academic primary care group practice at an ambulatory care clinic.

PARTICIPANTS

301 patients who screened positive for depression and a random sample of 101 patients who screened negative from among 3767 patients aged 60 and older who completed screening for depression during routine office visits.

MEASUREMENTS

Centers for Epidemiologic Studies Depression (CES-D) scale, Hamilton Depression Rating (HAM-D) scale, Sickness Impact Profile (SIP), structured psychiatric interview, sociodemographic and clinical variables, and mortality.

RESULTS

Among the 301 patients with CES-D scores > or = 16, 14 (4.6%) reported suicidal ideation and received urgent evaluation by mental health professionals. The estimated prevalence of suicidal ideation in this cohort of older primary care patients was 0.7 to 1.2%. All patients with suicidal ideation had evidence of a current affective disorder, and nearly all had moderate to severe functional impairment. However, even though all 14 patients endorsed suicidal ideation, corresponding HAM-D scores ranged from 3 to 40, and only four of 14 met diagnostic criteria for major depression. The most common suicide plan involved use of a hand gun. Depressed patients with suicidal ideation did not differ significantly from depressed non-suicidal patients on any of the following variables: age, gender, race, education, alcohol abuse, cognitive impairment, or mean CES-D, HAM-D, or SIP scores. None of the suicidal patients had died within 12 months of the screening date.

CONCLUSION

The prevalence of suicidal ideation was about 1% among this cohort of older primary care patients, and the prevalence approaches 5% among those older adults who report significant symptoms of depression. However, asking patients directly about the presence of active suicidal ideation appears to be the only effective means of identifying those at risk.

Baseball injuries: a Little League survey

OBJECTIVES

To determine the patterns of injury in youth baseball and apply the data to estimate the value of proposed safety equipment.

DESIGN

Prospective population-based injury survey.

PARTICIPANTS

2861 Little League baseball players (ages 7 to 18) for 140 932 player-hours.

MEASUREMENTS

An injury was included in the data only if it was serious enough to require medical/dental care, caused missing a game, or disallowed playing a certain position. The injuries were subdivided into acute or overuse. The acute injuries were classified as either catastrophic, severe, or minor. Injuries were categorized according to mechanism, area injured, and whether the player was on offense or defense.

RESULTS

There were 81 total injuries, of which 66 (81%) were acute and 15 (19%) were overuse. Of the acute injuries, 11 were severe and 55 were minor. The overall injury rate was .057 injuries per 100 player-hours. The severe injury rate was .008 injuries per 100 player-hours, of which 46% were ball-related injuries and 27% were collisions. The most frequent mechanism of injury was being hit by the ball, which represented 62% of the acute injuries. Of the 41 ball-related injuries, 28 (68%) occurred to players on defense. Of the 18 ball-related facial injuries, 16 occurred to players on defense.

CONCLUSIONS

1) Little League baseball is a safe activity with a low injury rate and a particularly low rate of severe injury; 2) impact by the ball causes more than half the acute injuries, thus safety interventions should be directed towards decreasing these injuries, especially on defense; and 3) facemasks on batters can safely eliminate facial injuries to offensive players, but would only moderately reduce the incidence of ball-related facial injuries as most of these injuries are sustained by defensive players.

MR imaging of the rotator cuff mechanism: comparison of spin-echo and turbo spin-echo sequences

OBJECTIVE

Although well documented in other MR imaging applications, the value of turbo spin-echo sequences in evaluating the shoulder has not been addressed. This study was designed to directly compare matched spin-echo and turbo spin-echo sequences in the MR imaging evaluation of the rotator cuff.

MATERIALS AND METHODS

Using otherwise matched double-echo sequences of equal time duration, we performed 123 paired spin-echo (TR/TE, 200/25,75; one excitation) and turbo spin-echo (3500-5000/22,90; two excitations) sequences in the paracoronal and/or parasagittal plane of the shoulders of 76 patients referred to our institution for possible rotator cuff tear. The sequences were retrospectively analyzed for cuff signal and morphology, fluid conspicuity, coracoacromial arch morphology, and bone marrow signal abnormalities. Surgical and nonsurgical clinical results were correlated when available.

RESULTS

We found 100% diagnostic correlation between spin-echo and turbo spin-echo sequences for rotator cuff integrity. Surgical data were available for 26 patients, and clinical follow-up for another 37. For complete rotator cuff tear in the surgical subpopulation, sensitivity was 89%, specificity was 94%, and diagnostic accuracy was 24 of 26 (92%). We found no diagnostically significant difference between the two imaging sequences for fluid conspicuity, coracoacromial morphology, or marrow signal. Signal-to-noise ratios were superior in the turbo spin-echo sequences.

CONCLUSION

Turbo spin-echo sequences are an accurate and efficient tool in the MR imaging evaluation of the rotator cuff. Potential benefits include time saving, increased spatial resolution, and improved signal-to-noise ratio.

Physical function among retirement-aged African American men and women

Prior attempts to identify factors associated with physical function (here, major lower body movements) among African Americans have been constrained by a narrow range of measures, small sample sizes, or both. The 1992 Health and Retirement Study (HRS) contains a substantial over-sample of African Americans (649 men and 957 women self-respondents aged 51 to 61 years), and detailed measures of high-risk behaviors, disease prevalence and severity, impairment, and physical function. We extend the natural history of disease to the natural history of functional status and model sociodemographic characteristics, high-risk behaviors, disease prevalence and severity, and impairments as direct and indirect influences on physical function in this African American sample. This natural history of functional status model fits the data well for both men (ROC = .88) and women (ROC = .83), although there are gender differences. Slightly over one-half of the women report some difficulty in physical function, compared with one-third of the men. Women also have a higher mean body-mass and report a greater prevalence and severity in 6 of 9 chronic diseases and more pain, but are less likely to smoke or abuse alcohol than men. Importantly, many of the factors with the largest direct and indirect associations with difficulty in physical function among these African American men (alcohol abuse, smoking, body mass, diabetes, heart disease, cerebrovascular disease, arthritis, and pain) and women (alcohol abuse, body mass, arthritis, and respiratory illness) are all potentially preventable or manageable.

The risk for and severity of bleeding complications in elderly patients treated with warfarin. The National Consortium of Anticoagulation Clinics

OBJECTIVE

To determine whether increasing age is associated with an increased risk for bleeding during warfarin treatment.

DESIGN

Combined retrospective and prospective cohort studies.

SETTING

6 anticoagulation clinics.

PATIENTS

2376 patients receiving warfarin for various indications.

MEASUREMENTS

Bleeding events categorized as minor (resulting in no costs or consequences), serious (requiring testing or treatment), life-threatening, or fatal.

RESULTS

812 first bleeding events (4 fatal, 33 life-threatening, 222 serious, and 553 minor) occurred during 3702 patient-years. Age was inversely related to the mean warfarin dose and dose-adjusted prothrombin time ratio. The unadjusted incidence of minor bleeding complications did not vary according to age group: 18.0 per 100 patient-years for patients younger than 50 years of age, 21.5 for patients 50 to 59 years of age, 24.0 for patients 60 to 69 years of age; 23.5 for patients 70 to 79 years of age, and 16.3 for patient 80 years of age and older. The unadjusted incidence of serious bleeding complications also did not vary according to age group: 9.3 per 100 patient-years for patients younger than 50 years of age, 7.1 for patients 50 to 59 years of age, 6.6 for patients 60 to 69 years of age, 5.1 for patients 70 to 79 years of age, and 4.4 for patients 80 years of age and older. The unadjusted incidence of life-threatening or fatal complications combined was significantly higher among the oldest patients: 0.75 per 100 patient-years for patients younger than 50 years of age, 0.97 for patients 50 to 59 years of age, 1.10 for patients 60 to 69 years of age, 0.68 for patients 70 to 79 years of age, and 3.38 for patients 80 years of age and older. Patients 80 years of age and older had a relative risk of 4.5 (95% CI, 1.3 to 15.6) compared with patients younger than 50 years of age. After adjustment for the intensity of anticoagulation therapy and the deviation in the prothrombin time ratio using Cox and Poisson regression, age was not generally associated with the occurrence of bleeding; relative risk estimates ranged from 0.99 to 1.03 per year of age (lower-bound 95% CI, 0.97 to 1.01; upper-bound 95% CI, 1.00 to 1.09). The single exception was life-threatening and fatal complications in patients 80 years of age or older (relative risk, 4.6 [CI, 1.2 to 18.1]).

CONCLUSIONS

Age did not appear to be an important determinant of risk for bleeding in patients receiving warfarin, with the possible exception of age 80 years or older. The intensity of anticoagulation therapy and the deviation in the prothrombin time ratio were much stronger predictors of risk for bleeding.

Consistency and change in functional status among older adults over time

Consistency and change between 1984 and the last reinterview (either two, four, or six years later) on 22 individual functional status markers and the five summary scales that they form are examined among the 5,986 members of the Longitudinal Study on Aging who were reinterviewed at least once. At baseline, at least three-fifths of the respondents are without limitations on any individual marker. At the last reinterview, at least 43.4% of the respondents are without such limitations. Among those who had limitations at baseline, at least one-fifth get better. For those without limitations at baseline, one-seventeenth to two-fifths get worse. Difficulties in walking and doing heavy housework were the most common, most likely to develop, and least likely to resolve of any of the ADL or IADL items, and lower body limitations were more common, more likely to develop, and less likely to resolve than upper body limitations. Linear panel analysis of the five summary scales indicates that the top predictors of increased functional limitation are baseline levels of functional limitation, older age, decedent status, and poorer perceived health (in that order). Other less consistent and less robust predictors include the length of the exposure window, being female, having a history of arthritis, lower educational attainment, having fewer nonkin social supports, higher prior physician visit levels, and not living alone.

Primary care physicians' medical decision making for late-life depression

OBJECTIVE

To describe primary care physicians' clinical decision making regarding late-life depression.

DESIGN

Longitudinal collection of data regarding physicians' clinical assessments and the volume and content of patients' ambulatory visits as part of a randomized clinical trial of a physician-targeted intervention to improve the treatment of late-life depression.

SETTING

Academic primary care group practice.

PATIENTS/PARTICIPANTS

One-hundred and eleven primary care physicians who completed a structured questionnaire to describe their clinical assessments immediately following their evaluations of 222 elderly patients who had reported symptoms of depression on screening questionnaires.

INTERVENTIONS

Intervention physicians were provided with their patient's score on the Hamilton Depression rating scale (HAM-D) and patient-specific treatment recommendations prior to completing the questionnaire regarding their clinical assessment.

MAIN RESULTS

Those physicians not provided HAM-D scores were just as likely to rate their patients as depressed, as determined by specific query of these physicians regarding their clinical assessments. A physician's clinical rating of likely depression did not consistently result in the formulation of treatment intentions or actions. Treatment intentions and actions were facilitated by provision of treatment algorithms, but treatment was received by fewer than half of the patients whom physicians intended to treat. Barriers to treatment appear to include both physician and patient doubts about treatment benefits.

CONCLUSIONS

Lack of recognition of depressive symptoms did not appear to be the primary barrier to treatment. Recognition of symptoms and access to treatment algorithms did not consistently result in progression to subsequent stages in treatment decision making. More research is needed to determine how patients and physicians weigh the potential risks and benefits of treatment and how accurately they make these judgments.

Relationship of age, education, and occupation with dementia among a community-based sample of African Americans

OBJECTIVE

To explore the relationship between age, education, and occupation with dementia among African Americans.

DESIGN

Community-based survey to identify subjects with and without evidence of cognitive impairment and subsequent diagnostic evaluation of a stratified sample of these subjects using formal diagnostic criteria for dementia.

SETTING

Urban neighborhoods in Indianapolis, Ind.

SUBJECTS

A random sample of 2212 African Americans aged 65 years and older residing in 29 contiguous census tracts.

MEASUREMENTS

Subjects's scores on the Community Screening Instrument for Dementia (CSI-D), formal diagnostic clinical assessments for dementia, years of education, rural residence, primary occupation, self-reported disease, and alcohol and smoking history. Caseness was defined by four separate criteria: (1) cognitive impairment as defined by the subject's performance on the CSI-D cognitive scale; (2) cognitive impairment as defined by the total CSI-D score that included a relative's assessment of the subject's functional abilities; (3) dementia as defined by explicit diagnostic criteria; and (4) possible or probable Alzheimer's disease as defined by explicit diagnostic criteria.

RESULTS

The mean age was 74 years (age range, 65 to 100 years), 65% of subjects were women, the mean education was 9.6 years (age range, 0 to 16 years), 98% of the subjects were literate, and 32% reported living in a rural area until age 19 years. Service, domestic, and production occupations accounted for 55.2% of the subjects' primary occupations with a mean of 25.8 years (range, 1 to 75 years) in the primary occupation. Years of education, rural residence to age 60 years, and primary occupation were highly correlated. Caseness defined by any of the four criteria was associated with functional impairment, but the frequency of impairment increased with increasing diagnostic specificity. Age, education, and rural residence to age 60 years were significantly independently associated with caseness for cognitive impairment, dementia, and Alzheimer's type dementia. White-collar occupation was independently associated only with caseness for cognitive impairment. History of stroke was associated with caseness for cognitive impairment and dementia but not Alzheimer's disease, while history of smoking was negatively correlated with Alzheimer's disease.

CONCLUSIONS

Education was independently associated with cognitive impairment and dementia among a representative community-based sample of African Americans and the association remains significant across a variety of sensitivity analyses designed to control for measurement and confounding biases. The potential protective role of education against the development of dementia among African Americans deserves further evaluation.

The recognition and treatment of late-life depression: a view from primary care

OBJECTIVE

Efforts to improve the recognition and treatment of late-life depression in primary care are often based on the assumption that primary care physicians underutilize currently available and effective treatments. This article reviews the validity of this assumption and offers recommendations for future research.

METHODS

Clinical trials designed to improve the recognition and treatment of late-life depression in primary care are reviewed. Because studies limited to older adults are rare, we also include studies enrolling younger patients. These data are reviewed in the context of recent reviews on the prevalence of depression in primary care settings and the effectiveness of available treatments.

RESULTS

Although depressive symptoms are common among older adults, there is insufficient literature documenting the proportion of these patients who respond to currently available treatments. Patients with uncomplicated major depressive disorder constitute the minority of primary care patients with depressive symptoms. Nearly all available studies of treatment effectiveness of pharmacotherapy or psychotherapy focus on older adults with uncomplicated major depression. Currently available treatment options may apply to less than 15 percent of depressed primary care patients.

CONCLUSIONS

More research is needed to help primary care providers manage their depressed patients with comorbid medical conditions, functional disability, or minor or chronic depressions. In addition, more research is needed to identify those patients who would benefit from specialized or interdisciplinary care.

Health services use and mortality among older primary care patients with alcoholism

OBJECTIVE

To describe the prevalence of alcoholism in an older primary care population and to compare rates of health services use and mortality among those with and those without evidence of alcoholism.

DESIGN

Baseline screening for alcoholism using the CAGE questionnaire and longitudinal assessment of health services use and mortality using an electronic medical record system.

SETTING

An academic primary care group practice at an urban ambulatory care clinic.

PATIENTS

A total of 3954 patients aged 60 and older who completed the CAGE alcoholism screening questionnaire during routine office visits.

MAIN OUTCOME MEASURES

Comorbidity, preventive health services use, hospital episodes and length of stay, emergency room visits, ambulatory care visits, total outpatient charges, and mortality.

RESULTS

The prevalence of current evidence of alcoholism, as defined by a CAGE score > or = 2 and alcohol use in the previous 12 months, was 10.6%. Patients with evidence of alcoholism were younger (66.2 vs 68.3 years), had fewer years of education (8.4 vs 9.1), were more likely to be male (65.8 vs 27.2%), black (71.2 vs 62.6%), smokers (40.4 vs 26.3%), and malnourished (32.5 vs 26.3%). Patients with alcoholism were more likely to have a diagnosis of obstructive lung disease (22.9 vs 18.3%), injuries (14.2 vs 8.3%), and gout (6.7 vs 2.9%) and less likely to have a diagnosis of hypertension (56.9 vs 61.8%), arthritis (23.3 vs 29.3%), and diabetes (15.9 vs 23.3%). Among those with evidence of alcoholism, 41.6% had a diagnosis of alcoholism in their outpatient medical record. Rates of completion of preventive health services did not differ between the two groups, and there was no difference in the number of ambulatory care visits, emergency room visits, or total outpatient charges. Patients with evidence of alcoholism were more likely to be hospitalized (21.5 vs 16.9%) and more likely to die within 2 years (10.6% vs 6.3%).

CONCLUSIONS

One of 10 older patients in this primary care practice had current evidence of alcoholism, fewer than half of whom had documentation of alcohol abuse in their medical records. These patients were more likely to be hospitalized and more likely to die but did not consume a greater amount of outpatient resources. Further research is needed to determine if interventions to reduce alcohol use would also reduce excess hospitalizations and mortality among these older patients.

Predicting hospital costs among older decedents over time

To explain the variation in total real hospital costs among elderly patients who died between 1984 and 1991, a cohort analytic study of the nationally representative sample of elderly subjects included in the Longitudinal Study on Aging (N = 7,527) was carried out. The cohort comprised the subset of 1,778 community-dwelling Americans who were age 70 years and older in 1984, had one or more subsequent hospital episodes, and died by 1991. Hospital charges for 1984 through 1991 were taken from the Medicare Automated Data Retrieval System. Annual hospital charges were adjusted for inflation (restated in 1984 dollars) using the hospital market basket component of the consumer price index. The natural logarithm of aggregated real charges was used in the analysis. Mean total real hospital charges were $24,956 (SD = $27,847). A standard multivariable regression model explained 9.7% of the variance in real total hospital charges. After incorporating additional measures reflecting a respondent's distribution (mean and standard deviation) of comorbidities (as measured by the number of ICD-9-CM codes [truncated at five]) during all hospitalizations in the observation window, the cause of death, and the concentration of charges in the last year of life, the explained variance increased to 29.3%. The most important explanatory factors were the two variables controlling for the distribution of comorbidity, the variable controlling for population density, and the dichotomous variable indicating that the patient's death was related to an acute myocardial infarction. Total real hospital resources consumed by elderly decedents vary substantially. The concentration of resources consumed in the last year of a respondent's life was only marginally significant in predicting total real hospital charges over an 8-year observation window.

Hospitalization for major depression among older Americans

BACKGROUND

The objective of this study was to report the pattern of hospitalization for major depression among older Americans and to examine correlates of those hospitalizations. We sought to investigate the hypothesis that hospitalization for major depression would be more common among those respondents with declining functional status whose ability to adapt to this decline was impaired by inadequate social support systems or economic stressors.

METHODS

The data were taken from Version 5 of the Longitudinal Study on Aging (LSOA), which includes 7,527 subjects who were aged 70 and older in 1984. We identified all subjects with any hospitalizations for which major depression was a discharge diagnosis (ICD9-CM codes 296.2, 296.3, 300.4, and 311), and all subjects for whom depression was the primary discharge diagnosis. Only patients with a first-listed discharge diagnosis of depression were considered to have been hospitalized for major depression.

RESULTS

The yearly incidence of hospitalization for which depression was the primary discharge diagnosis was 0.1%. The mean length of stay was 14.6 days and the mean hospital charge was $6,742. Length of stay and charges did not vary by hospital type (general vs psychiatric), but both charges and length of stay were significantly longer when major depression was the primary discharge diagnosis rather than a secondary diagnosis. Patients with a hospitalization for major depression had more hospitalizations, longer total lengths of stay, and greater total hospital charges over the seven-year period as compared to patients with at least one hospitalization for any other reason. These differences in hospital resource use dissipated when hospitalizations for depression were excluded. Hospitalization for major depression was not associated with gender, race, education, or social support. Hospitalization for major depression was independently associated with a forced residential move, a history of nursing home stays, decline in household activities of daily living, younger age, and perceived health rated as less than excellent.

CONCLUSIONS

One older American per thousand is hospitalized each year with a primary discharge diagnosis of major depression. These individuals did not have evidence of greater total hospital resource use if episodes of hospitalization for depression are eliminated. Hospitalization for major depression was more common among those with a loss of independent living.

Patient outcomes following unicompartmental or bicompartmental knee arthroplasty. A meta-analysis

The purpose of this study was to summarize the literature describing patient outcomes following unicompartmental and bicompartmental knee arthroplasty. Original studies were included in this meta-analysis if they enrolled 10 or more patients at the time of an initial knee arthroplasty and measured patient outcomes using a global knee rating scale. Forty-six studies on unicompartmental prostheses and 18 studies on bicompartmental prostheses met these criteria. For unicompartmental studies, the total number of enrolled patients was 2,391, with a mean enrollment of 47 patients and a mean follow-up period of 4.6 years. The mean patient age was 66 years; 67% were women, 75% had osteoarthritis, and 16% underwent bilateral knee arthroplasty. The mean postoperative global rating scale score was 80.9. The overall complication rate was 18.5% and the revision rate was 9.2%. Studies published after 1987 reported better outcomes, but also tended to enroll older patients and patients with osteoarthritis and higher preoperative knee rating scores. For bicompartmental studies, the total number of enrolled patients was 884, with a mean enrollment of 44 patients and a mean follow-up period of 3.6 years. The mean patient age was 61 years; 79% were women, 31% had osteoarthritis, and 29% underwent a bilateral arthroplasty. The mean postoperative global rating scale score was 78.3. The overall complication rate was 30% and the revision rate was 7.2%. Although bicompartmental studies reported lower mean postoperative global rating scale scores, these studies tended to enroll patients with worse preoperative knee rating scores. Recent improvements in patient outcomes following unicompartmental knee arthroplasty appear to be due, at least partially, to changes in patient selection criteria. Patient outcomes appear to be worse for bicompartmental arthroplasties than for other prosthetic designs; however, patients enrolled in these studies had more poorly functioning knees before surgery and actually had greater absolute improvements in global knee rating scores.

Documentation and evaluation of cognitive impairment in elderly primary care patients

OBJECTIVE

To describe the prevalence of cognitive impairment among elderly primary care patients and to compare diagnostic evaluations and use of health services among patients with and those without cognitive impairment.

DESIGN

Prospective cohort study.

SETTING

Academic primary care group practice.

PATIENTS

3954 patients aged 60 years and older who completed the Short Portable Mental Status Questionnaire during routine office visits.

MEASUREMENTS

Demographics and comorbid illness at baseline, diagnostic evaluations for cognitive impairment, use of standard and preventive health services, use of psychoactive drugs, and death in the year after the screening date.

RESULTS

The prevalence of cognitive impairment among all patients aged 60 years and older at baseline was 15.7%; 10.5% had mild impairment and 5.2% had moderate to severe impairment. Patients with moderate to severe impairment were significantly older than patients with no impairment (76.2 years and 67.4 years, respectively), were more likely to be black (85.8% and 61.3%), had fewer years of education (7.3 years and 9.2 years), and were more likely to have cerebrovascular disease (20.4% and 6.3%) and evidence of undernutrition (30.6% and 16.9%). Dementia was recorded as a diagnosis for less than 25% of patients with moderate to severe cognitive impairment, but patients with documented impairment were more likely to have been evaluated for reversible causes. In the year after screening, patients with moderate to severe impairment were more likely than those with no impairment both to be hospitalized (29.1% and 16.5%) and to visit the emergency department (55.8% and 38.5%) but had fewer outpatient visits (6.0 and 7.6) and greater mortality (8.2% and 2.8%).

CONCLUSIONS

Cognitive impairment is associated with increased use of health services and increased mortality. Patients with undocumented cognitive impairment were significantly less likely to be evaluated for reversible causes. Research is needed to determine if better documentation of cognitive impairment would improve not only diagnostic evaluations but also patient management, counseling, and outcomes.

How residents spend their time in clinic and the effects of clerical support

The authors conducted a time study of residents in clinic to determine the effects of providing clerical assistance. The residents recorded their activities at 5-minute intervals at baseline and six months after hiring three clerical assistants. Before and after introduction of the clerical assistants, approximately 40% of the time was devoted to direct interaction with patients. Statistically significant improvements were observed in the availability of medical records (89% vs 100%) and the time spent looking up test results (5% vs 3% of the clinic time). The residents felt the clerical assistants greatly improved their clinic experience and the quality of patient care.

Hospital resource consumption among older adults: a prospective analysis of episodes, length of stay, and charges over a seven-year period

After linking their administrative records and interview data, the consumption of Medicare-reimbursed hospital resources during 1984 through 1990 by the 7,527 LSOA respondents was prospectively assessed using a two-part design. First, logistic regression was used to model whether a hospital episode occurred. Second, among those having had hospital episodes, OLS regression was used to model the number of episodes, as well as the natural logarithms of the total length of stay and the total charges. The risk of hospitalization was mostly associated with being male, prior hospital and physician utilization, and lower body limitations. Among those hospitalized: (a) greater numbers of episodes were mostly associated with prior hospital and physician utilization, and poorer perceived health; (b) longer lengths of stay were mostly associated with prior hospital and physician utilization, and poorer perceived health; and, (c) higher charges were mostly associated with population density, poorer perceived health, and prior physician and hospital utilization. Decedents consistently consumed substantially more hospital resources than survivors.

Anticoagulation therapy and primary care internal medicine: a nurse practitioner model for combined clinical science

The anticoagulation clinics at the University of Virginia Health Sciences Center and the University of California at Davis Medical Center are nurse-practitioner-operated, are affiliated with the general medicine clinic, and rely on portable prothrombin time (PT) monitors that use whole blood and provide timely as well as accurate results reported in PT seconds or as the international normalized ratio (INR). On-site PT/INR testing at these clinics simplifies anticoagulation, mandates direct patient contact, and facilitates primary as well as comprehensive care for patients requiring multispecialty services in large tertiary care centers. Encounters are relatively brief, averaging 19 minutes; 72% of the encounter time involves anticoagulation care and 28% involves primary care. Anticoagulation results using portable PT/INR monitors are safe and accurate based on comparisons with results from clinics relying on standard instruments.

Improving treatment of late life depression in primary care: a randomized clinical trial

OBJECTIVE

Facilitate primary care physicians' compliance with recommended standards of care for late life depression by reducing barriers to recognition and treatment.

DESIGN

Randomized controlled clinical trial of physician-targeted interventions.

SETTING

Academic primary care group practice caring for an urban, medically indigent patient population.

PATIENTS/PARTICIPANTS

Patients aged 60 and older who exceeded the threshold on the Centers for Epidemiologic Studies Depression Scale (CES-D) and the Hamilton Depression Rating Scale (HAM-D) and their primary care physicians.

INTERVENTION

Physicians of intervention patients were provided with patient-specific treatment recommendations during 3 special visits scheduled specifically to address the patient's symptoms of depression. In general, physicians were encouraged to establish a diagnosis of depression and educate their patient about the diagnosis, discontinue medications that can cause or exacerbate depressive symptoms, initiate antidepressants when appropriate, and consider referral to psychiatry. Guidelines for prescribing antidepressants were provided. Control physicians received no intervention, and control patients received usual care.

MAIN OUTCOME MEASURES

Frequency of recording a depression diagnosis, stopping medications associated with depression, initiating antidepressant medication, and psychiatry referral; mean changes in HAM-D and Sickness Impact Profile (SIP) scores.

RESULTS

One hundred three physicians and 175 patients were involved in the clinical trial. Physicians of intervention patients were more likely to diagnose depression and prescribe antidepressants (P < 0.01). There were no differences between the groups in the frequency of stopping medications associated with depression or referrals to psychiatry. Medications with the strongest cause and effect relationship to depression were infrequently used in this cohort of patients. Although both groups showed improvement in HAM-D and SIP scores, we were unable to demonstrate significant differences in HAM-D or SIP scores between the 2 groups.

CONCLUSIONS

Intensive screening and feedback of patient-specific treatment recommendations increased the recognition and treatment of late life depression by primary care physicians. However, we were unable to demonstrate significant improvement in depression or disability severity among intervention patients despite the informational support provided to their physicians. Efforts to improve the functional status of these patients may require more integrated interventions and more aggressive attempts to target psychosocial stressors traditionally outside the purview of primary care.

Longitudinal study of depression and health services use among elderly primary care patients

OBJECTIVE

To describe the prevalence and 9-month incidence of depressive symptoms among a cohort of elderly primary care patients and to determine whether different patterns of depression are associated with different patterns of health services use.

DESIGN

Prospective study of depressive symptoms as measured by the Center for Epidemiologic Studies Depression (CES-D) scale and identification of patients' outpatient health services use through an electronic medical record system.

SETTING

An academic primary care group practice at an urban ambulatory care clinic.

PATIENTS/PARTICIPANTS

1711 patients aged 60 and older who completed the CES-D at baseline and 9 months later; 935 of these patients also completed the CES-D at 6 months.

MEASUREMENT AND MAIN RESULTS

The prevalence of significant symptoms of depression (CES-D > or = 16) was 17.1% at baseline and 18.8% at 9 months; 26.8% of patients exceeded the threshold on the CES-D either at baseline or 9 months, and the 9-month incidence was 11.7%. Among the patients re-interviewed at both 6 and 9 months, the 6-month incidence was 12%, and the incidence between the 6- and 9-month assessments was 10%. Of the 292 patients with depression at baseline, 140 (47.6%) remained depressed at the 9-month follow-up. Baseline and 6-month CES-D score, in addition to perceived health at 6 months, explained 45% of the variance in the 9-month CES-D score. Patients above the threshold on the CES-D at any time were more likely to rate their health as fair or poor (69.8% vs 43.7%) and more likely to have an emergency room visit (40.4% vs 29.4%). These patients also had 38% more outpatient visits (7.7 vs 5.6) and 61% higher total outpatient charges ($1209 vs $751) than patients who never exceeded the CES-D threshold over the 9-month window (all P values < 0.01).

CONCLUSIONS

Depressive symptoms were frequent and often persistent in this patient population. We identified patterns of oscillating severity of symptoms within individuals but relatively stable incidence and prevalence rates over a 9-month period. Patients who exceeded the threshold on the CES-D at any time during the study had significantly greater health services use and poorer perceived health.

Global rating systems used in assessing knee arthroplasty outcomes

Global, aggregated knee rating systems are commonly used to assess patient outcomes following knee arthroplasty. In this study, the authors performed a systematic literature search and found that 17% of the English-language studies addressing primary knee arthroplasty reported on patient outcomes following the procedure using a standardized global rating system. The authors describe, in detail, the rating systems' development and format. This study found 34 different rating systems represented in the literature from 1972 to 1992. Great variability was found in the rating systems' design and utilization. Additionally, these condition-specific, physician-based rating systems did not have documented studies demonstrating their reliability or validity. Future research will need to address the issues of selecting and aggregating outcome measures and of deriving any necessary weighting schemes. The ability of researchers to compare patient outcomes across studies will be enhanced when there is consistency in reported outcome measures.

Patient outcomes following tricompartmental total knee replacement. A meta-analysis

OBJECTIVE

To provide estimates of patient outcomes following tricompartmental knee replacement and to examine variation in outcomes due to patient and prosthesis characteristics.

DATA SOURCES

English-language articles identified through a computerized literature search and bibliography review.

STUDY SELECTION

Studies were included if they enrolled 10 or more patients at the time of initial knee replacement and measured patient outcomes using a global knee-rating scale.

DATA EXTRACTION

Each study was subjected to a blinded qualitative assessment and unblinded abstraction of patient characteristics, surgical techniques, and outcomes.

DATA SYNTHESIS

A total of 130 studies reporting patient outcomes on 154 cohorts satisfied inclusion criteria. The total number of enrolled patients was 9879 with a mean enrollment of 64.1 patients. The mean follow-up was 4.1 years. The mean patient age was 65.0 years, 71.7% of patients were women, 62.6% had osteoarthritis, and 26.6% underwent bilateral knee replacement. Global rating scale scores improved by 100% for the typical enrolled patient, and 89.3% of patients reported good or excellent outcomes. Anatomic classification of the prosthesis, percentage of enrolled patients with osteoarthritis, publication year, and number of enrolled patients explained 27% of the variation in reported mean postoperative global rating scale scores. The weighted mean complication rate was 18.1%, and the mean mortality rate per year of follow-up was 1.5%. The overall rate of revision during 4.1 years was 3.8%.

CONCLUSIONS

Tricompartmental knee replacement was a safe and effective procedure for the patients reported in these studies. The knee pathology and the type of prosthesis were significant predictors of outcomes. Limitations in the reporting style of these articles severely constrain the ability to explore variation in outcomes due to study, patient, or prosthesis characteristics and restrict their generalizability.

The effect of gender and race on the measurement properties of the CES-D in older adults

Having observed a three-fold difference in the prevalence of significant symptoms of depression among four race-gender groups of elderly adults attending an urban primary care practice, we investigated the extent to which these differences might be explained by variability in the measurement properties of the Centers for Epidemiologic Studies depression scale (CES-D). Although the internal consistency of the CES-D was acceptable for all groups, 5% of our patients were excluded for inability to complete the minimum required number of CES-D items, and nearly 40% of patients required response imputation for the allowable one to four items that they could not answer. Imputation was most frequently required for items tapping positive affect. Principal components factor analysis was performed separately for respondents answering all items and for respondents with imputed values. In both analyses we found important race-gender differences in factor structure. Moreover, the factor structure for those with imputed values was markedly different from that of respondents answering all items, including a dissolution of the positive affect dimension. Neither the race-gender differences in factor structure nor the differences among those with and without imputed data were resolved by eliminating respondents with poor education, cognitive impairment, or alcoholism, or by varying the assumptions for data imputation. However, the disparities in factor structure were essentially resolved by eliminating five CES-D items, suggesting the need to modify the CES-D in populations like ours. Although eliminating these five items results in a more pure factor structure, it does not resolve the differences in prevalence of depressive symptoms. These differences may, however, be partially due to differential response tendencies among the race-gender groups.