Mary B. McDonell

The Risk for and Severity of Bleeding Complications in Elderly Patients Treated with Warfarin

Warfarin is prescribed for more than 1 million persons in the United States, many of whom are elderly and take the drug indefinitely. The pooled results of five randomized trials document that warfarin reduced the incidence of stroke by 68% in persons with atrial fibrillation, a benefit that substantially outweighs the risk for bleeding that accompanies warfarin therapy [1-6]. Atrial fibrillation is present in 4% of persons older than 60 years of age and 10% of persons older than 80 years of age [7]. Paradoxically, many studies have found that elderly persons are at higher risk for hemorrhagic complications than are younger patients [8-17]. It has been suggested that older patients are more prone to bleeding because they metabolize warfarin more slowly [18-20]; they have an elevated risk for drug interactions because of polypharmacy [21]; and they often have chronic illnesses, such as renal insufficiency [22], heart failure [9], cancer [10, 22], and cerebrovascular disease [23], that increase the risk for bleeding during warfarin therapy. On the other hand, numerous studies have not found elderly persons to be at greater risk for bleeding [24-31]. These discrepant findings have not been adequately explained. Because of the clinical importance of this problem, we analyzed data from six sites to determine whether elderly patients who were receiving warfarin had a higher incidence of complications than similar younger patients who were also receiving warfarin. Methods We collected our data in two phases. The first phase involved a retrospective review of medical records from 1980 to 1990; other results from this review have been reported previously [30]. In the second phase, we analyzed data collected prospectively between 1990 and 1993, including the results from a multicenter, randomized trial of a computerized scheduling system for patients receiving warfarin [32]. Study Setting and Patients Our study was done in the six anticoagulation clinics that make up the National Consortium of Anticoagulation Clinics. These clinics represent a mix of geographic locations, practice settings, and patient populations. Clinic personnel were responsible for managing anticoagulation therapy for all patients throughout the study, although in two clinics, personnel also provided primary medical care [33]. The sites were the University of California at Davis, Sacramento, California; Jefferson Medical College, Philadelphia, Pennsylvania; the University of Virginia, Charlottesville, Virginia; the Veterans Affairs Medical Center, Buffalo, New York; the Veterans Affairs Medical Center, Palo Alto, California; and the Veterans Affairs Medical Center, Seattle, Washington. Because Jefferson Medical College did not join the Consortium until 1992, data from that site were used only in the prospective phase of the study, which followed completion of the randomized trial of computerized scheduling. Retrospective Data Collection During the retrospective phase of our study, we abstracted the medical records of all patients who were currently receiving anticoagulation therapy and all patients whose anticoagulation therapy had been discontinued within the previous 18 to 24 months. If patients had received multiple courses of warfarin (that is, if there were extended periods during which a patient did not receive warfarin), we abstracted data from all courses. All patients were eligible for inclusion unless they had received warfarin for 6 weeks or less. We have described our data collection methods in detail elsewhere [30]. Trained abstractors reviewed inpatient and outpatient records using standard forms that have been extensively tested for reliability. All records from inpatient admissions and visits to the anticoagulation clinic; other medical, surgical, and urgent care clinics (excluding visits made for psychiatric reasons); and emergency departments were abstracted. We also reviewed any records that were maintained separately from the formal medical record by a practitioner responsible for anticoagulation therapy. At each of the three Veterans Affairs medical centers, data were also retrieved from the hospitals information system. We excluded seven patients whose charts were missing. Indications for anticoagulation therapy were organized into seven main categories and 29 subcategories. For patients who had more than one indication, the most serious problem (the one that required the greatest intensity or longest duration of therapy) was deemed the primary indication. During the retrospective phase of our study, only one of the five medical centers reported results using the international normalized ratio. Because the international sensitivity index values of the thromboplastins used at some of the participating centers were unavailable before 1988, we analyzed all results using the prothrombin time ratio. However, we contacted each laboratory to determine whether the international sensitivity index values of the reagents used were known; this information was available from several of the study sites after 1985 and from all sites after 1990. All laboratories used standard North American thromboplastins that had international sensitivity index values ranging from 2.0 to 2.4. Practitioners typically adhered to therapeutic recommendations published by the American College of Chest Physicians (ACCP) in 1986 [34, 35], which set the target range as a prothrombin time ratio of 1.5 to 1.8 (high intensity) for patients with mechanical valves and 1.3 to 1.5 (low intensity) for most other patients. Records from outpatient visits were reviewed to ascertain the reason for the appointment, the occurrence of any intercurrent illnesses, all medications prescribed in addition to warfarin, and the dates on which therapy with each medication was started and stopped [36]. Prospective Study The prospective data collection was done over a 3-year period (1990 to 1993). During the first year of this period, we did a randomized trial of a computerized scheduling intervention [32]. Because the scheduling system produced no statistically significant differences in control of anticoagulation or frequency of complications, we have included data collected during that 1-year period in our present analysis. During the 2 years after completion of the trial, we continued to collect identical information on patients who had participated in the trial and on all eligible patients subsequently enrolled into participating clinics for the management of warfarin therapy. Jefferson Medical College did not participate in the trial of computerized scheduling. Patients who were actively enrolled in one of the six participating clinics or who were newly referred to one of the clinics were eligible for this portion of the study if the total planned duration of their anticoagulation therapy was 6 weeks or longer. Before the start of prospective data collection, we abstracted the medical records of all active eligible patients in each clinic. These data were collected according to the same protocol used in the retrospective study. At the three clinics located in university medical centers, our trial was exempted from requirements for verbal or written informed consent by the local institutional review boards, and all eligible patients were enrolled. At the three Veterans Affairs clinics, the local review boards required informed consent, and we invited the participation of all eligible patients in person or by mail; 5.2% of patients in these clinics declined to participate. At each visit to each anticoagulation clinic during the prospective phase of the study, all data, including those on the prothrombin time ratio (or international normalized ratio) results, warfarin dosage, complications, and follow-up plans, were entered directly into a notebook computer. Classification of Outcomes In both the retrospective and prospective phases of data collection, we used the same detailed scheme to classify bleeding complications as minor (no associated costs or medical consequences), serious (requiring treatment or medical evaluation), life-threatening, or fatal. Minor complications required no additional testing, referrals, or outpatient visits but were remarkable enough to report to the provider. Examples of minor bleeding included mild nosebleeds, bruising, mild hemorrhoidal bleeding, and microscopic hematuria. Examples of serious bleeding included overt gastrointestinal bleeding, occult gastrointestinal bleeding if endoscopic or radiographic studies were done, gross hematuria that prompted cystoscopy or intravenous urography or lasted more than 2 days, and hemoptysis. If blood was transfused, 2 units or fewer were given. Life-threatening bleeding was defined as that leading to cardiopulmonary arrest, surgical or angiographic intervention, or irreversible sequelae, such as myocardial infarction, neurologic deficit consequent to intracerebral hemorrhage, or massive hemothorax. Bleeding was also considered to be life-threatening if it led to at least two of the following consequences: loss of 3 or more units of blood; systolic hypotension (systolic blood pressure less than 90 mm Hg); or critical anemia (hematocrit less than equals 0.20). Fatal bleeding was defined as that leading directly to the death of the patient. All serious, life-threatening, and fatal complications were independently reviewed by a physician investigator at the local site and by three investigators at the coordinating center. Using standardized criteria, we determined whether deaths were related to bleeding caused by warfarin therapy or to thromboembolic complication. Disagreements were resolved by discussion. Deviation in the Prothrombin Time Ratio We previously described [37] a method that can be used to characterize the degree to which a patients prothrombin time ratios deviate from his or her target prothrombin time ratio over time, and we have shown that the level of variability in the prothrombin time ratio is stati

Health status predicts long-term outcome in outpatients with coronary disease

Background—Although patient-reported health status measures have been used as end points in clinical trials, they are rarely used in other settings. Demonstrating that they independently predict mortality and hospitalizations among outpatients with coronary disease could emphasize their clinical value. Methods and Results—This study evaluated the prognostic utility of the Seattle Angina Questionnaire (SAQ), a disease-specific health status measure for patients with coronary artery disease. Patients were enrolled in a prospective cohort study from 6 Veterans Affairs General Internal Medicine Clinics. All patients reporting coronary artery disease who completed a SAQ and had 1 year of follow-up were analyzed (n=5558). SAQ predictor variables were the physical limitation, angina stability, angina frequency, and quality-of-life scores. The primary outcome was 1-year all-cause mortality, and a secondary outcome was hospitalization for acute coronary syndrome (ACS). Lower SAQ scores were associated with increased risks of mortality and ACS admissions. Prognostic models controlling for demographic and clinical characteristics demonstrated significant independent mortality risk with lower SAQ physical limitation scores; odds ratios for mild, moderate, and severe limitation were 1.5, 2.0, and 4.0 versus minimal limitation (P <0.001). Odds ratios for mild, moderate, and severe angina frequency were 0.8, 1.2, and 1.6 (P =0.078). The odds ratios for ACS admission among those with mild, moderate, and severe angina frequency were 1.4, 2.0, and 2.2, respectively (P =0.016). Conclusions—SAQ scores are independently associated with 1-year mortality and ACS among outpatients with coronary disease and may serve a valuable role in the risk stratification of such patients.

Continuity of care and other determinants of patient satisfaction with primary care

AbstractOBJECTIVE: The patient-clinician relationship is a central feature of primary care, and recent developments in the delivery of health care have tended to limit continuity of care. The objective of this study was to evaluate the extent to which continuity of care and other factors are related to patient satisfaction. DESIGN: Cross-sectional, mailed questionnaire study. SETTING: Primary care clinics at 7 Veterans Affairs medical centers. PATIENTS/PARTICIPANTS: Patients (N=21,689) participating in the Ambulatory Care Quality Improvement Project who returned the baseline Seattle Outpatient Satisfaction Questionnaire (SOSQ). MEASUREMENTS AND MAIN RESULTS: We evaluated the association between self-reported continuity and satisfaction, after adjusting for characteristics of patients, clinics, and providers. The humanistic scale of the SOSQ measures patient satisfaction with communication skills and humanistic qualities of providers, whereas the organizational scale measures satisfaction with delivery of health care services. The mean adjusted humanistic score for patients who reported always seeing the same provider was 17.3 (95% confidence interval [CI], 15.5 to 19.1) points higher than for those who rarely saw the same provider. Similarly, the mean adjusted organizational score was 16.3 (95% CI, 14.5 to 18.1) points higher for patients who always saw the same provider compared to rarely. Demographic factors, socioeconomic status, health status, clinic site, and patient utilization of services were all associated with both the adjusted humanistic and organizational scores of the SOSQ. CONCLUSIONS: Self-reported continuity of care is strongly associated with higher patient satisfaction. This suggests that improving continuity of care may improve patient satisfaction with providers as well as with their health care organization.

Screening for Problem Drinking: Comparison of CAGE and AUDIT

AbstractOBJECTIVE: To compare self-administered versions of three questionnaires for detecting heavy and problem drinking: the CAGE, the Alcohol Use Disorders Identification Test (AUDIT), and an augmented version of the CAGE. DESIGN: Cross-sectional surveys. SETTING: Three Department of Veterans Affairs general medical clinics. PATIENTS: Random sample of consenting male outpatients who consumed at least 5 drinks over the past year (“drinkers”). Heavy drinkers were oversampled. MEASUREMENTS: An augmented version of the CAGE was included in a questionnaire mailed to all patients. The AUDIT was subsequently mailed to “drinkers.” Comparison standards, based on the tri-level World Health Organization alcohol consumption interview and the Diagnostic Interview Schedule, included heavy drinking (>14 drinks per week typically or ≥5 drinks per day at least monthly) and active DSM-IIIR alcohol abuse or dependence (positive diagnosis and at least one alcohol-related symptom in the past year). Areas under receiver operating characteristic curves (AUROCs) were used to compare screening questionnaires. MAIN RESULTS: Of 393 eligible patients, 261 (66%) returned the AUDIT and completed interviews. For detection of active alcohol abuse or dependence, the CAGE augmented with three more questions (AUROC 0.871) performed better than either the CAGE alone or AUDIT (AUROCs 0.820 and 0.777, respectively). For identification of heavy-drinking patients, however, the AUDIT performed best (AUROC 0.870). To identify both heavy drinking and active alcohol abuse or dependence, the augmented CAGE and AUDIT both performed well, but the AUDIT was superior (AUROC 0.861). CONCLUSIONS: For identification of patients with heavy drinking or active alcohol abuse or dependence, the self-administered AUDIT was superior to the CAGE in this population.

Validation of case-mix measures derived from self-reports of diagnoses and health

Self-reported chronic diseases and health status are associated with resource use. However, few data exist regarding their ability to predict mortality or hospitalizations. We sought to determine whether self-reported chronic medical conditions and the SF-36 could be used individually or in combination to assess co-morbidity in the outpatient setting. The study was designed as a prospective cohort study. Patients were enrolled in the primary care clinics at seven Veterans Affairs (VA) medical centers participating in the Ambulatory Care Quality Improvement Project (ACQUIP). 10,947 patients, > or = 50 years of age, enrolled in general internal medicine clinics who returned both a baseline health inventory checklist and the baseline SF-36 who were followed for a mean of 722.5 (+/-84.3) days. The primary outcome was all-cause mortality, with a secondary outcome of hospitalization within the VA system. Using a Cox proportional hazards model in a development set of 5,469 patients, a co-morbidity index [Seattle Index of Co-morbidity (SIC)] was constructed using information about age, smoking status and seven of 25 self-reported medical conditions that were associated with increased mortality. In the validation set of 5,478 patients, the SIC was predictive of both mortality and hospitalizations within the VA system. A separate model was constructed in which only age and the PCS and MCS scores of the SF-36 were entered to predict mortality. The SF-36 component scores and the SIC had comparable discriminatory ability (AUC for discrimination of death within 2 y 0.71 for both models). When combined, the SIC and SF-36 together had improved discrimination for mortality (AUC = 0.74, p-value for difference in AUC < 0.005). A new outpatient co-morbidity score developed using self-identified chronic medical conditions on a baseline health inventory checklist was predictive of 2-y mortality and hospitalization within the VA system in general internal medicine patients.

A refill adherence algorithm for multiple short intervals to estimate refill compliance (ReComp)

Background: There are many measures of refill adherence available, but few have been designed or validated for use with repeated measures designs and short observation periods. Objective: To design a refill-based adherence algorithm suitable for short observation periods, and compare it to 2 reference measures. Methods: A single composite algorithm incorporating information on both medication gaps and oversupply was created. Electronic Veterans Affairs pharmacy data, clinical data, and laboratory data from routine clinical care were used to compare the new measure, ReComp, with standard reference measures of medication gaps (MEDOUT) and adherence or oversupply (MEDSUM) in 3 different repeated measures medication adherence-response analyses. These analyses examined the change in low density lipoprotein (LDL) with simvastatin use, blood pressure with antihypertensive use, and heart rate with β-blocker use for 30- and 90-day intervals. Measures were compared by regression based correlations (R2 values) and graphical comparisons of average medication adherence-response curves. Results: In each analysis, ReComp yielded a significantly higher R2 value and more expected adherence-response curve regardless of the length of the observation interval. For the 30-day intervals, the highest correlations were observed in the LDL-simvastatin analysis (ReComp R2 = 0.231; [95% CI, 0.222–0.239]; MEDSUM R2 = 0.054; [95% CI, 0.049–0.059]; MEDOUT R2 = 0.053; [95% CI, 0.048–0.058]). Conclusions: ReComp is better suited to shorter observation intervals with repeated measures than previously used measures.

Transforming self-rated health and the SF-36 scales to include death and improve interpretability

Background.Most measures of health-related quality of life are undefined for people who die. Longitudinal analyses are often limited to a healthier cohort (survivors) that cannot be identified prospectively, and that may have had little change in health. Objective.To develop and evaluate methods to transform a single self-rated health item (excellent to poor; EVGGFP) and the physical component score of the SF-36 (PCS) to new variables that include a defensible value for death. Methods.Using longitudinal data from two large studies of older adults, health variables were transformed to the probability of being healthy in the future, conditional on the current observed value; death then has the value of 0. For EVGGFP, the new transformations were compared with some that were published earlier, based on different data. For the PCS, how well three different transformations, based on different definitions of being healthy, discriminated among groups of patients, and detected change in time were assessed. Results.The new transformation for EVGGFP was similar to that published previously. Coding the 5 categories as 95, 90, 80, 30, and 15, and coding dead as 0 is recommended. The three transformations of the PCS detected group differences and change at least as well as the standard PCS. Conclusion.These easily interpretable transformed variables permit keeping persons who die in the analyses. Using the transformed variables for longitudinal analyses of health when deaths occur, either for secondary or primary analysis, is recommended. This approach can be applied to other measures of health.

Anginal symptoms consistently predict total mortality among outpatients with coronary artery disease

BACKGROUND Age, race, education, and diabetes have been associated with differences in anginal symptoms, treatments, and outcomes among outpatients with coronary artery disease (CAD), but there is little data on whether such characteristics affect relationships between anginal symptoms and mortality. METHODS Using a prospective cohort design, we examined associations of anginal symptoms, as assessed by the Seattle Angina Questionnaire, with total mortality among 8908 outpatients with CAD to investigate whether this relationship is influenced by patient demographic or clinical characteristics. Potential effect modification was primarily assessed for age, race, education, and diabetes, and secondarily assessed for smoking, prevalent congestive heart failure (CHF), myocardial infarction, and coronary revascularization. RESULTS Over 2 years mean follow-up, there were 896 deaths. After adjustment for potential confounders, persons reporting greater physical limitation due to angina had higher mortality: 27% higher with mild limitation (hazard ratio [HR] 1.27, 95% CI 0.98-1.64), 61% higher with moderate limitation (HR 1.61, 95% CI 1.27-2.05), and 2.5-fold higher with the greatest limitation (HR 2.55, 95% CI 1.97-3.30), compared with little or no limitation (P for trend <.001). Anginal instability was also independently predictive of mortality. There was little evidence that these relationships varied by age, race, education, diabetes, smoking, or presence of CHF, prior myocardial infarction, or prior coronary revascularization (P for each interaction >.28). Anginal symptoms predicted higher mortality risk comparable to a decade of age difference, presence of diabetes, or presence of CHF. CONCLUSIONS Among outpatients with CAD, self-reported anginal symptoms consistently predict mortality irrespective of differences in age, race, education, or clinical comorbidities.

Predicting risk of hospitalization or death among patients receiving primary care in the Veterans Health Administration.

Background:Statistical models that identify patients at elevated risk of death or hospitalization have focused on population subsets, such as those with a specific clinical condition or hospitalized patients. Most models have limitations for clinical use. Our objective was to develop models that identified high-risk primary care patients. Methods:Using the Primary Care Management Module in the Veterans Health Administration (VHA)’s Corporate Data Warehouse, we identified all patients who were enrolled and assigned to a VHA primary care provider on October 1, 2010. The outcome variable was the occurrence of hospitalization or death during the subsequent 90 days and 1 year. We extracted predictors from 6 categories: sociodemographics, medical conditions, vital signs, prior year use of health services, medications, and laboratory tests and then constructed multinomial logistic regression models to predict outcomes for over 4.6 million patients. Results:In the predicted 95th risk percentiles, observed 90-day event rates were 19.6%, 6.2%, and 22.6%, respectively, for hospitalization, death, and either hospitalization or death, compared with population averages of 2.7%, 0.7%, and 3.4%, respectively; 1-year event rates were 42.3%, 19.4%, and 51.3%, respectively, compared with population averages of 8.2%, 2.6%, and 10.8%, respectively. The C-statistics for 90-day outcomes were 0.83, 0.86, and 0.81, respectively, for hospitalization, death, and either hospitalization or death and were 0.81, 0.85, and 0.79, respectively, for 1-year outcomes. Conclusions:Prediction models using electronic clinical data accurately identified patients with elevated risk for hospitalization or death. This information can enhance the coordination of care for patients with complex clinical conditions.

Alcohol Screening Results in Elderly Male Veterans: Association with Health Status and Mortality

Objectives: To evaluate the association between alcohol screening results and health status or mortality in elderly patients.