Effect of a closed drug-delivery system on the incidence of nosocomial and catheter-related bloodstream infections in infants.

We conducted a prospective, cohort study at two affiliated level III neonatal intensive care units to evaluate the effect of a closed drug-delivery system on the incidence of nosocomial and catheter-related bloodstream infections (CRBSI) in infants. A total of 300 infants (n=150 at each site) were enrolled over a 4-year study period. There was no difference in the rate of CRBSI per 1000 catheter days between the two sites (16.2+/-39 vs. 8.9+/-24, P=0.054, 95% CI-14.8 to 0.13). Infants at site A (closed drug-delivery system) had a higher rate of infectious nosocomial respiratory complications per 100 hospital days than infants at site B (open delivery system) (1.1+/-2.2 vs. 0.5+/-1.5, P=0.009), however, there was no difference in the overall number of confirmed or suspected nosocomial infection events per patient between study sites. Logistic regression revealed that the number of additional peripheral catheters, gestational age and duration of parenteral nutrition all significantly contributed to the risk of developing one or more CRSBI. The closed drug-delivery system failed to reduce the incidence of CRBSI or overall rate of nosocomial infections in premature infants.

An institution-specific heparin titration nomogram: development, validation, and assessment of compliance.

STUDY OBJECTIVE: To develop, validate, and assess compliance with a heparin titration nomogram. DESIGN: Prospective, open-label trial. SETTING: University teaching hospital. SUBJECTS: Patients admitted with heart failure who required therapy with intravenous unfractionated heparin. Intervention. An in vitro concentration-response was determined by measuring activated partial thromboplastin times (aPTTs) on normal pooled plasma containing known concentrations of heparin. The therapeutic aPTT range was determined from the concentration-response by using the therapeutic heparin concentration range of 0.2-0.4 U/ml (protamine neutralization). Patients were consecutively enrolled, and therapy was managed by using the heparin titration nomogram. Paired aPTT-heparin concentrations were obtained, and nomogram validation was performed by comparing the in vitro and the ex vivo concentration-responses with use of linear regression. Nomogram compliance also was assessed. MEASUREMENTS AND MAIN RESULTS: The therapeutic aPTT ranges based on in vitro and ex vivo data were determined to be 45-72 seconds and 47-61 seconds, respectively. The ranges were significantly different (p<0.001). Overall compliance with the nomogram was 88%. CONCLUSION: These results confirm that, even in a relatively homogeneous disease-state patient population, in vitro data do not accurately predict ex vivo data. If in vitro data are used to develop an institution-specific nomogram, a validation procedure should be used to ensure accuracy. Although 100% compliance to a nomogram may not be attainable, it should be expected. Therefore, a compliance rate of 88% is concerning and suggests a need for increased nursing and physician education.

Interpreting the findings of the IMPROVE study.

Various findings of the Impact of Managed Pharmaceutical Care on Resource Utilization and Outcomes in Veterans Affairs Medical Centers (IMPROVE) study are reviewed. Suggestions for future methodologies that will enhance this study are discussed. The IMPROVE study is one of the largest pharmaceutical care studies conducted. Although it was an intervention study that examined global outcomes following management by pharmacists, it was designed as an effectiveness study. Several new practice and research methods were developed, including a method to identify patients at high risk for drug-related problems utilizing pharmacy databases, a method to identify chronic diseases using pharmacy databases, a method to evaluate the structure and process for delivering pharmaceutical care in Veterans Affairs medical centers (VAMCs), and guidelines for providing care to patients in the IMPROVE study. Nine VAMCs participated in the study, and 1054 patients were randomized to either an intervention group (n = 523) or a control group (n = 531). Pharmacists documented a total of 1855 contacts with the intervention group patients and made 3048 therapy-specific interventions over the 12-month study period. There was no meaningful difference in patient satisfaction or quality of life in the two groups. Selected disease-specific indicators found an improved rate of measurement of hemoglobin A1c tests and better control of total and low-density-lipoprotein (LDL) cholesterol levels in the intervention group compared with the control group. Total health care costs increased in both groups over the 12-month period. The mean increase in costs in the intervention group was $1020, which was lower than the control group's value of $1313. The lessons learned from the IMPROVE study suggest to future investigators how to study and measure the effects of clinical pharmacy services on patient outcome.

Can clinical pharmacists affect SF-36 scores in veterans at high risk for medication-related problems?

BACKGROUND: An objective of pharmaceutical care is for pharmacists to improve patients' health-related quality of life (HRQOL) by optimizing medication therapy. OBJECTIVES: The objective of this study was to determine whether ambulatory care clinical pharmacists could affect HRQOL in veterans who were likely to experience a drug-related problem. RESEARCH DESIGN: This was a 9-site, randomized, controlled trial involving Veterans Affairs Medical Centers (VAMCs). Patients were eligible if they met > or = 3 criteria for being at high risk for drug-related problems. Enrolled patients were randomized to either usual medical care or usual medical care plus clinical pharmacist interventions. HRQOL was measured with the SF-36 questionnaire administered at baseline and at 6 and 12 months. RESULTS: In total, 1,054 patients were enrolled; 523 were randomized to intervention, and 531 to control. After patient age, site, and chronic disease score were controlled for, the only domain that was significantly different between groups over time was the bodily pain scale, which converged to similar values at the end of the study. Patients' rating of the change in health status in the past 12 months was statistically different between groups, intervention patients declining less (-2.4 units) than control subjects (-6.3 units) (P < 0.004). This difference was not considered clinically meaningful. However, a dose-response relationship was observed for general health perceptions (P = 0.004), vitality (P = 0.006), and change in health over the past year (P = 0.007). CONCLUSIONS: These results suggest that clinical pharmacists had no significant impact on HRQOL as measured by the SF-36 for veterans at high risk for medication-related problems.

Co-prescribing of medications used to treat obstructive lung disease, congestive heart failure and depression among users of topical beta blockers: estimates from three US Veterans Affairs Medical Centers.

PURPOSE: To estimate the frequency of co-prescribing of medications used to treat obstructive lung disease, congestive heart failure, and depression among US armed service veterans for whom topical beta blockers (TBB) have been prescribed. METHODS: Prescription data from the computerized medical record systems of three US Veterans Affairs Medical Centers were obtained during Fall 1999 to identify patients for whom active prescriptions existed for TBB and/or drugs commonly used to treat obstructive lung disease, congestive heart failure, and depression. RESULTS: Among TBB users, the estimated co-prescribing rate of medications used to treat obstructive lung disease is in the range of 7.8-9.7%, of medications used to treat congestive heart failure is in the range of 4.3-25.1%, and of medications used to treat depression is in the range of 8.8-11.7%. The total number of TBB users with a concurrent prescription for one or more of the conditions studied was in the range of 30.2-45.7%. CONCLUSIONS: Since TBB may cause systemic side-effects and thus potentially cause exacerbations of selected chronic conditions, it is important for researchers and prescribers to be aware of the frequency with which their patients with glaucoma may have other concomitant conditions.

Systemic adverse effects of ophthalmic beta-blockers.

OBJECTIVE: To review published clinical information on the systemic adverse effects of ophthalmic beta-blockers for the purpose of developing a pilot contraindication/warning system for active prescriptions in the Veterans Affairs dispensing database. DATA SOURCES: Articles were identified by searching MEDLINE (1966-October 2000) and International Pharmaceutical Abstracts (1970-October 2000). STUDY SELECTION AND DATA EXTRACTION: Article relevance was determined by review of titles, abstracts, and key words. DATA SYNTHESIS: The preponderance of the evidence suggests that ophthalmic beta-blockers may be associated with bronchospasm and adverse cardiovascular effects including bradycardia. Depression and other central nervous system effects are reported less commonly. Data are inadequate to suggest that ophthalmic beta-blocker use is routinely associated with adverse metabolic effects. CONCLUSIONS: The strongest level of evidence (grade A1) supports a contraindication for use of ophthalmic beta-blockers for respiratory disease, with a moderate level of evidence (grade B1) for cardiovascular disease. Data are insufficient to support or refute contraindications for other disease states. The search technique and classification scheme described in this article provide a methodology for evaluating, grading, and applying evidence on potential adverse effects of drug therapy.

Clinical and economic impact of ambulatory care clinical pharmacists in management of dyslipidemia in older adults: the IMPROVE study. Impact of Managed Pharmaceutical Care on Resource Utilization and Outcomes in Veterans Affairs Medical Centers.

We examined the impact of ambulatory care clinical pharmacist interventions on clinical and economic outcomes of 208 patients with dyslipidemia and 229 controls treated at nine Veterans Affairs medical centers. This was a randomized, controlled trial involving patients at high risk of drug-related problems. Only those with dyslipidemia are reported here. In addition to usual medical care, clinical pharmacists were responsible for providing pharmaceutical care for patients in the intervention group. The control group did not receive pharmaceutical care. Seventy-two percent of the intervention group and 70% of controls required secondary prevention according to the National Cholesterol Education Program guidelines. Significantly more patients in the intervention group had a fasting lipid profile compared with controls (p=0.021). The absolute change in total cholesterol (17.7 vs 7.4 mg/dl, p=0.028) and low-density lipoprotein (23.4 vs 12.8 mg/dl, p=0.042) was greater in the intervention than in the control group. There were no differences in patients achieving goal lipid values or in overall costs despite increased visits to pharmacists. Ambulatory care clinical pharmacists can significantly improve dyslipidemia in a practice setting designed to manage many medical and drug-related problems.

An economic analysis of a randomized, controlled, multicenter study of clinical pharmacist interventions for high-risk veterans: the IMPROVE study. Impact of Managed Pharmaceutical Care Resource Utilization and Outcomes in Veterans Affairs Medical Centers.

STUDY OBJECTIVE: To determine if clinical pharmacists could affect economic resource use and humanistic outcomes in an ambulatory, high-risk population. DESIGN: Prospective, randomized, controlled study. SETTING: Nine Veterans Affairs medical centers. PATIENTS: Patients who were at high risk for medication-related problems. INTERVENTION: Patients were randomized to usual medical care with input from a clinical pharmacist (intervention group) or just usual medical care (control group). MEASUREMENTS AND MAIN RESULTS: Of 1,054 patients enrolled, 523 were randomized to the intervention group and 531 to the control group. The number of clinic visits increased in the intervention group (p=0.003), but there was no difference in clinic costs. Mean increases in total health care costs were $1,020 for the intervention group and $1,313 for the control group (p=0.06). CONCLUSION: Including the cost of pharmacist interventions, overall health care expenditures were similar for patients randomized to see a clinical pharmacist versus usual medical care.

Factors associated with successful epoetin alfa therapy in premature infants.

OBJECTIVE: To determine the impact of two different recombinant human erythropoietin (epoetin alfa) dosing strategies on the number of red blood cell (RBC) transfusions, and explore relationships between specific patient and drug regimen variables with epoetin alfa therapy outcomes. DESIGN: Retrospective cohort study. SETTING: Level III university neonatal intensive care unit. METHODS: Infants who received epoetin alfa therapy three times weekly for more than one week were categorized into two epoetin alfa dosing strategy groups: group A (300-749 units/kg/wk) and group B (750-1200 units/kg/wk). The following patient variables were collected and their relationship to therapy outcomes (corrected reticulocyte count [%], hematocrit [%], and number of RBC transfusions after therapy was started) were evaluated using independent Student's t-test, correlation analysis, and stepwise linear regression: birth weight (kg), gestational age (weeks), postnatal age at therapy onset (days), duration of mechanical ventilation (days), number of RBC transfusions before epoetin alfa therapy, phlebotomy loss (mL/kg), epoetin alfa dosage (units/kg/dose), iron dosage (mg/kg/d), duration of therapy (days), and postconceptional age at therapy discontinuation (weeks). RESULTS: The charts of 44 patients were reviewed. No significant impact on outcome was attributed to overall dosing strategy (group A vs. group B). Linear regression identified postnatal age at therapy onset as a significant contributor to mean hematocrit (R2 = 2 0.116; p = 0.023) and postconceptional age at therapy discontinuation as a significant contributor to number of transfusions during and after epoetin alfa use (R2 = 0.118; p = 0.022). A significant positive correlation was found between weekly mean epoetin alfa dosage and mean reticulocyte count (r = 0.326; p = 0.046), mean iron dosage and mean reticulocyte count (r = 0.439; p = 0.006), and ventilator days and total number of transfusions (r = 0.606; p < 0.001). A significant negative correlation was found between number of transfusions and reticulocyte count (r = -0.367; p = 0.023). CONCLUSIONS: Epoetin alfa dosing strategy, as defined in our study, did not significantly affect the number of transfusions. However, postnatal age at therapy initiation, postconceptional age at therapy discontinuation, mean epoetin alfa dosage, and iron dosage correlate with specific outcomes of epoetin alfa therapy in premature infants.

Development of a chronic disease indicator score using a Veterans Affairs Medical Center medication database. IMPROVE Investigators.

OBJECTIVE: Develop a chronic disease index that approximates the number of chronic diseases a patient has using a medication database. METHODS: An expert panel determined whether specific medication classes could be indicative of a chronic disease. Those classes identified were incorporated into a computer program and then used to screen the medication records of 246 randomly selected patients to estimate the number of chronic diseases present in each patient. This number was designated as the chronic disease index (CDI). The CDI was then validated against chart review. The CDI and a measure of disease severity, the chronic disease score (CDS) also were compared. The sensitivity and specificity of the computer program was analyzed for seven common chronic diseases. RESULTS: The expert panel designated 54 drug classes containing medications used to treat chronic diseases. The CDI correlated moderately with the number of chronic diseases found via chart review (r = 0.65; P = 0.001) and highly with the CDS (r = 0.81; P = 0.001). The index predicted the presence of three common diseases with a sensitivity of > or = 75%, and of six common diseases with a specificity of > or = 75%. CONCLUSIONS: The CDI correlates moderately well with the actual number of chronic disease states present. This tool may be useful for researchers when trying to identify patients with specific diseases and also for risk adjustment.

The IMPROVE study: background and study design. Impact of Managed Pharmaceutical Care on Resource Utilization and Outcomes in Veterans Affairs Medical Centers.

An ongoing study of the impact of ambulatory care clinical pharmacists on patient outcomes at selected Veterans Affairs medical centers (VAMCs) is described. The IMPROVE (Impact of Managed Pharmaceutical Care on Resource Utilization and Outcomes in Veterans Affairs Medical Centers) study will examine the effects of referring patients at high risk for drug-related problems to a pharmacist-managed monitoring program. Nine study sites from diverse geographic locations and small and large urban areas have been selected. Investigators visited each site to evaluate the structure of care, observe pharmacist-patient interactions, and assess the level and documentation of pharmacists' activities. A coordinating center will collect and process patient-specific data from the study sites to identify high-risk patients. It is expected that 500 intervention patients and 500 control patients from the nine VAMCs will complete all portions of the study. Intervention patients will be scheduled for medication assessments by ambulatory care pharmacists and will be monitored by pharmacists for at least 12 months. The coordinating center will track refill histories for intervention patients. Investigators will assess the activities performed by ambulatory care pharmacists to determine predictors of successful patient outcomes. The two groups will be compared with respect to change from baseline in quality of life and satisfaction with health care providers. A cost-benefit analysis will be undertaken to determine the impact of pharmaceutical care relative to total patient care costs. The main outcome results of the IMPROVE study are expected to be available in 1999. The IMPROVE project will be the first study of the impact of ambulatory care clinical pharmacists on patient outcomes.

Factors associated with antibiotic prescribing in a managed care setting: an exploratory investigation.

This multi-site, cross-sectional, observational study sought to identify attitudinal and social normative factors associated with the prescribing of oral antibiotics to ambulatory patients in a managed care setting. Participants were 25 physicians specializing in internal medicine, family practice or pediatrics from five ambulatory care clinics within a large, fully integrated health care system in a major midwestern U.S. city. The main outcome measure was number of prescriptions per physician written in the fourth quarter of 1994 for each of seven selected antibiotics. Correlational and multiple regression analyses revealed that behavioral intentions were significantly associated (P < 0.05) with both attitudes and subjective norms. However, physicians' attitudes, subjective norms and intentions were not predictive of actual antibiotic prescribing behavior. Prescribing behavior may have been a function of patient-specific rather than general beliefs about antibiotics. Methodological limitations related to the sample size and the sparseness of the utilization data may also have prevented a significant effect of intentions on behavior from being detected. Alternatively, in managed care settings, it is hypothesized that prescribing behavior may have been influenced more by non-psychological factors, such as management systems, formularies and therapeutic substitution programs, than they were by internal, psychological factors such as attitudes, subjective norms and intentions. Managed care is altering the role of the physician as an autonomous decision-maker. In response, models of prescribing must either incorporate variables such as perceived behavioral control to aid in the prediction of non-volitional behavior, model the decision-making of non-physician managers, or forego psychological models in favor of structural or system-level models of drug utilization.