Towards meaningful medication-related clinical decision support: recommendations for an initial implementation.

SUMMARY: Clinical decision support (CDS) can improve safety, quality, and cost-effectiveness of patient care, especially when implemented in computerized provider order entry (CPOE) applications. Medication-related decision support logic forms a large component of the CDS logic in any CPOE system. However, organizations wishing to implement CDS must either purchase the computable clinical content or develop it themselves. Content provided by vendors does not always meet local expectations. Most organizations lack the resources to customize the clinical content and the expertise to implement it effectively. In this paper, we describe the recommendations of a national expert panel on two basic medication-related CDS areas, specifically, drug-drug interaction (DDI) checking and duplicate therapy checking. The goals of this study were to define a starter set of medication-related alerts that healthcare organizations can implement in their clinical information systems. We also draw on the experiences of diverse institutions to highlight the realities of implementing medication decision support. These findings represent the experiences of institutions with a long history in the domain of medication decision support, and the hope is that this guidance may improve the feasibility and efficiency CDS adoption across healthcare settings.

Guided medication dosing for inpatients with renal insufficiency.

CONTEXT: Usual drug-prescribing practices may not consider the effects of renal insufficiency on the disposition of certain drugs. Decision aids may help optimize prescribing behavior and reduce medical error. OBJECTIVE: To determine if a system application for adjusting drug dose and frequency in patients with renal insufficiency, when merged with a computerized order entry system, improves drug prescribing and patient outcomes. DESIGN, SETTING, AND PATIENTS: Four consecutive 2-month intervals consisting of control (usual computerized order entry) alternating with intervention (computerized order entry plus decision support system), conducted in September 1997-April 1998 with outcomes assessed among a consecutive sample of 17 828 adults admitted to an urban tertiary care teaching hospital. INTERVENTION: Real-time computerized decision support system for prescribing drugs in patients with renal insufficiency. During intervention periods, the adjusted dose list, default dose amount, and default frequency were displayed to the order-entry user and a notation was provided that adjustments had been made based on renal insufficiency. During control periods, these recommended adjustments were not revealed to the order-entry user, and the unadjusted parameters were displayed. MAIN OUTCOME MEASURES: Rates of appropriate prescription by dose and frequency, length of stay, hospital and pharmacy costs, and changes in renal function, compared among patients with renal insufficiency who were hospitalized during the intervention vs control periods. RESULTS: A total of 7490 patients were found to have some degree of renal insufficiency. In this group, 97 151 orders were written on renally cleared or nephrotoxic medications, of which 14 440 (15%) had at least 1 dosing parameter modified by the computer based on renal function. The fraction of prescriptions deemed appropriate during the intervention vs control periods by dose was 67% vs 54% (P<.001) and by frequency was 59% vs 35% (P<.001). Mean (SD) length of stay was 4.3 (4.5) days vs 4.5 (4.8) days in the intervention vs control periods, respectively (P =.009). There were no significant differences in estimated hospital and pharmacy costs or in the proportion of patients who experienced a decline in renal function during hospitalization. CONCLUSIONS: Guided medication dosing for inpatients with renal insufficiency appears to result in improved dose and frequency choices. This intervention demonstrates a way in which computer-based decision support systems can improve care.

Patient safety and computerized medication ordering at Brigham and Women's Hospital.

BACKGROUND: Medications are important therapeutic tools in health care, yet creating safe medication processes is challenging for many reasons. Computerized physician order entry (CPOE), one important way that technology can be used to improve the medication process, has been in place at Brigham and Women's Hospital (BWH; Boston) since 1993. CPOE AT BWH: The CPOE application, designed and developed internally by the BWH information systems team, allows physicians and other clinicians to enter all patient orders into the computer. Physicians enter 85% of orders, with the remainder entered electronically by other clinicians. CPOE AND SAFE MEDICATION USE: The CPOE application at BWH includes several features designed to improve medication safety--structural features (for example, required fields, use of pick lists), enhanced workflow features (order sets, standard scales for insulin and potassium), alerts and reminders (drug-drug and drug-allergy interaction checking), and adjunct features (the pharmacy system, access to online reference information). RESULTS AT BWH: Studies of the impact of CPOE on physician decision making and patient safety at BWH include assessment of CPOE's impact on the serious medication error and the preventable adverse drug event rate, the impact of computer guidelines on the use of vancomycin, the impact of guidelines on the use of heparin in patients at bed rest, and the impact of dosing suggestions on excessive dosing. CONCLUSION: CPOE and several forms of clinical decision support targeted at increasing patient safety have substantially decreased the frequency of serious medication errors and have had an even bigger impact on the overall medication error rate.

Identifying hospital admissions due to adverse drug events using a computer-based monitor.

BACKGROUND: Hospital admissions due to adverse drug events (ADEs) are expensive, and many may be preventable, yet few institutions have ongoing surveillance for these events. OBJECTIVE: To evaluate the use of a computer-based ADE monitor to identify admissions due to ADEs and to measure the associated costs. DESIGN: Prospective cohort study in one tertiary care hospital. PARTICIPANTS: All patients admitted to nine medical and surgical units in a tertiary care hospital over an 8-month period. MAIN OUTCOME MEASURE: Admissions to the hospital due to an adverse drug event. METHODS: A computer-based monitoring program generated alerts suggesting that an ADE might be present. A trained reviewer then evaluated the record. RESULTS: Among the 3238 admissions, 76 (2.3%, 1.4% after adjusting for sampling) were found to be caused by an ADE. Of these ADEs, 78% were severe and 28% were preventable. Estimated costs were $16,177 per ADE, and $10,375 per preventable ADE; annualized costs to the hospital were $6.3 million per year for all ADEs, and $1.2 million for preventable ADEs. CONCLUSIONS: Many admissions were caused by ADEs, although our point estimate undoubtedly represents a lower bound. These events were mostly severe, often preventable, and expensive. The computer-based monitoring system represents a practical approach for identifying ADEs that occur in outpatients and cause admission to the hospital.

Relationship between hematocrit and renal function in men and women.

BACKGROUND: Anemia is a known complication of renal insufficiency, but the relationship between level of renal function and magnitude of reduction in hematocrit is not well defined. Men have higher hematocrit and absolute glomerular filtration rate (GFR) than women; however, it is unknown whether the level of clearance associated with decreased hematocrit is the same in men and women. METHODS: We conducted a cross-sectional study of 12,055 adult ambulatory patients. General linear models were used to analyze the relationship between hematocrit and Cockcroft-Gault equation estimated creatinine clearance (C(Cr); mL/min) and Modification of Diet in Renal Disease (MDRD) formula estimated the GFR indexed to body surface area (mL/min/1.73 m(2)). RESULTS: The hematocrit decreased progressively below estimated C(Cr) 60 mL/min in men and 40 mL/min in women. Compared with subjects with C(Cr)> 80 mL/min, men with C(Cr) 60 to 50 mL/min, 50 to 40 mL/min, 40 to 30 mL/min, 30 to 20 mL/min, and < or =20 mL/min had mean hematocrits that were lower by 1.0, 2.4, 3.7, 3.5, and 10.0%, respectively; the corresponding reductions in women with C(Cr) 40 to 30 mL/min, 30 to 20 mL/min, and < or =20 mL/min were 1.7, 2.9, and 6.3% (all P < 0.05). This between-sex difference diminished when renal function measurement was indexed to body size. Compared with subjects with GFR> 80 mL/min/1.73 m(2), men with GFR 50 to 40 mL/min/1.73 m(2), 40 to 30 mL/min/1.73 m(2), 30 to 20 mL/min/1.73 m(2), and < or =20 mL/min/1.73 m(2) had mean hematocrits that were lower by 2.0, 4.4, 5.3, and 9.4%; the corresponding reductions in women with GFR 50 to 40 mL/min/1.73 m(2), 40 to 30 mL/min/1.73 m(2), 30 to 20 mL/min/1.73 m(2) and < or =20 mL/min/1.73 m(2) were 0.6, 1.6, 3.8, and 5.3% (all P < 0.05). CONCLUSIONS: A decrease in hematocrit is apparent even among patients with mild to moderate renal insufficiency. At any given level of renal function below estimated C(Cr) 60 mL/min, men have a larger decrease in hematocrit than women.

Blood pressure and angiotensin converting enzyme inhibitor use in hypertensive patients with chronic renal insufficiency.

BACKGROUND: Hypertension treatment is important in managing chronic renal insufficiency (CRI). Little is known, however, about the blood pressure (BP) control achieved or the pattern of antihypertensive drug prescription among CRI patients. METHODS: Using computerized medical records, we studied 3,089 adult hypertensive subjects treated at Brigham and Women's Hospital (Boston, MA) from 1990 through 1998. All subjects had at least two serum creatinine measurements 2 years apart, at least two BP readings, and online weight (to estimate Cockcroft-Gault creatinine clearance [CrCl]). RESULTS: The average mean arterial pressure over time (mean MAP) was 103 +/- 9 mm Hg among those with CrCI >60 mL/min, 102 t 9 mm Hg among those with CrCl 41 to 60 mL/min. and 101 +/- 9 mm Hg among those with CrCl 21 to 40 mL/min. There were no significant differences in mean MAP among the different categories of renal function in the multivariate analysis (P = .26 for trend). The proportion of patients with final systolic BP < 160 mm Hg and diastolic BP <90 mm Hg was 68% and did not vary with renal function (P = .68 for trend). The proportion of subjects who were prescribed ACE inhibitors was 38% among those with CrCl >60 mL/min, 36% among those with CrCI 41 to 60 mL/min, and only 27% among those with CrCl 21 to 40 mL/min (P = .003 for trend). CONCLUSIONS: The BP control achieved among hypertensive CRI subjects, although no worse than that among those without CRI, was found to be suboptimal. Patients with CrCl 21 to 40 mL/min were less likely to be prescribed ACE inhibitors than were those with CrCl >60 mL/min. Improvement is needed in the clinical management of these factors that can influence the progression of CRI.

Requirements development for a patient computing system.

Critical parts of the software development life cycle are concerned with eliciting, understanding, and managing requirements. Though the literature on this subject dates back for several decades, practicing effective requirements development remains a current and challenging area. Some projects flourish with a requirements development process (RDP) that is implicit and informal, but this approach may be overly risky, particularly for large projects that involve multiple individuals, groups, and systems over time. At Partners HealthCare System in Boston, Massachusetts, we have applied a more formal approach for requirements development to the Patient Computing Project. The goal of the project is to create web-based software that connects patients electronically with their physician's offices and has the potential to improve care efficiency and quality. It is a large project, with over 500 function points. Like most technological innovation, the successful introduction of this system requires as much attention to understanding the business needs and workflow details as it does to technical design and implementation. This paper describes our RDP approach, and key business requirements discovered through this process. We believe that a formal RDP is essential, and that informatics as a field must include proficiencies in this area.

Improving allergy alerting in a computerized physician order entry system.

Computerized physician order entry has been shown to reduce the frequency of serious medication errors. Decision support tools such as alerting functions for patient medication allergy are a key part of these applications. However, optimal performance requires iterative refinement. As systems become increasingly complex, mechanisms to monitor their performance become increasingly critical. We analyzed trend data obtained over a five-year period that showed decreasing compliance to allergy alert functions within computerized order entry. Many medication-allergy pairs were being consistently overridden. Renewal policies affecting reordering narcotics also contributed heavily to this trend. Each factor revealed a system-wide trend that could result in suggestions for policy or software change. Monitoring trends such as these is very important to maintain software correctness and ensure user trust in alerting systems, so users remain responsive to computerized alerts.

Comparison of two knowledge bases on the detection of drug-drug interactions.

This paper describes a drug ordering decision support system that helps with the prevention of adverse drug events by detecting drug-drug interactions in drug orders. The architecture of the system was devised in order to facilitate its use attached to physician order entry systems. The described model focuses in issues related to knowledge base maintenance and integration with external systems. Finally, a retrospective study was performed. Two knowledge bases, developed by different academic centers, were used to detect drug-drug interactions in a dataset with 37,237 drug prescriptions. The study concludes that the proposed knowledge base architecture enables content from other knowledge sources to be easily transferred and adapted to its structure. The study also suggests a method that can be used on the evaluation and refinement of the content of drug knowledge bases.

A clinical information systems strategy for a large integrated delivery network.

Integrated delivery networks (IDNs) are an emerging class of health care institutions. IDNs are formed from the affiliation of individual health care institutions and are intended to be more efficient in the current fiscal health care environment. To realize efficiencies and support their strategic visions, IDNs rely critically on excellent information technology (IT). Because of its importance to the mission of the IDN, strategic decisions about IT are made by the top leadership of the IDN. At Partners HealthCare System, a large IDN in Boston, MA, a clinical information systems strategy has been created to support the Partners clinical vision. In this paper, we discuss the Partners' structure, clinical vision, and current IT initiatives in place to address the clinical vision. The initiatives are: a clinical data repository, inpatient process support, electronic medical records, a portal strategy, referral applications, knowledge resources, support for product lines, patient computing, confidentiality, and clinical decision support. We address several of the issues encountered in trying to bring excellent information technology to a large IDN.

Using electronic data to predict the probability of true bacteremia from positive blood cultures.

As part of a project to help physicians make more appropriate treatment decisions, we implemented a clinical prediction rule that computes the probability of true bacteremia for positive blood cultures and displays this information when culture results are viewed online. Prior to implementing the rule, we performed a revalidation study to verify the accuracy of the previously published logistic regression model. We randomly selected 114 cases of positive blood cultures from a recent one-year period and performed a paper chart review with the help of infectious disease experts to determine whether the cultures were true positives or contaminants. Based on the results of this revalidation study, we updated the probabilities reported by the model and made additional enhancements to improve the accuracy of the rule. Next, we implemented the rule into our hospital's laboratory computer system so that the probability information was displayed with all positive blood culture results. We displayed the prediction rule information on approximately half of the 2184 positive blood cultures at our hospital that were randomly selected during a 6-month period. During the study, we surveyed 54 housestaff to obtain their opinions about the usefulness of this intervention. Fifty percent (27/54) indicated that the information had influenced their belief of the probability of bacteremia in their patients, and in 28% (15/54) of cases it changed their treatment decision. Almost all (98% (53/54)) indicated that they wanted to continue receiving this information. We conclude that the probability information provided by this clinical prediction rule is considered useful to physicians when making treatment decisions.

Effects of computerized physician order entry on prescribing practices.

BACKGROUND: Computerized order entry systems have the potential to prevent errors, to improve quality of care, and to reduce costs by providing feedback and suggestions to the physician as each order is entered. This study assesses the impact of an inpatient computerized physician order entry system on prescribing practices. METHODS: A time series analysis was performed at an urban academic medical center at which all adult inpatient orders are entered through a computerized system. When physicians enter drug orders, the computer displays drug use guidelines, offers relevant alternatives, and suggests appropriate doses and frequencies. RESULT: For medication selection, use of a computerized guideline resulted in a change in use of the recommended drug (nizatidine) from 15.6% of all histamine(2)-blocker orders to 81.3% (P<.001). Implementation of dose selection menus resulted in a decrease in the SD of drug doses by 11% (P<.001). The proportion of doses that exceeded the recommended maximum decreased from 2.1% before order entry to 0.6% afterward (P<.001). Display of a recommended frequency for ondansetron hydrochloride administration resulted in an increase in the use of the approved frequency from 6% of all ondansetron orders to 75% (P<.001). The use of subcutaneous heparin sodium to prevent thrombosis in patients at bed rest increased from 24% to 47% when the computer suggested this option (P<.001). All these changes persisted at 1- and 2-year follow-up analyses. CONCLUSION: Computerized physician order entry is a powerful and effective tool for improving physician prescribing practices.

Diabetes, hemoglobin A(1c), cholesterol, and the risk of moderate chronic renal insufficiency in an ambulatory population.

Moderate chronic renal insufficiency is common, with 12.5 million individuals in the United States estimated to have a creatinine clearance less than 50 mL/min/1.73 m(2). Little is known about the risk factors for moderate chronic renal insufficiency. We studied 1, 428 subjects with Cockcroft-Gault-estimated creatinine clearances greater than 70 mL/min in a hospital-based ambulatory population. Over a mean of 5.7 +/- 1.3 years, 86 subjects developed moderate chronic renal insufficiency, defined as a decrease in creatinine clearance to less than 60 mL/min (1.1 case/100 person-years). Risk factors for moderate chronic renal insufficiency were identified using a proportional hazards model controlling for age, sex, race, systolic blood pressure, and angiotensin-converting enzyme (ACE) inhibitor use. The risk for developing moderate chronic renal insufficiency was associated with diabetes mellitus (relative risk, 2.1; 95% confidence interval [CI], 1.3 to 3.3) and elevated hemoglobin A(1c) levels. Compared with subjects with normoglycemia (hemoglobin A(1c) 9.0%) was 2.7 (95% CI, 1.4 to 5.1). The development of moderate chronic renal insufficiency was also independently predicted by elevated maximum serum cholesterol level. Compared with subjects with maximum cholesterol levels of 250 mg/dL or less, the relative risk for those with maximum cholesterol levels greater than 350 mg/dL was 2.4 (95% CI, 1.1 to 5.2). Similar relative risks were obtained when moderate chronic renal insufficiency was defined by the development of an increase in serum creatinine level. Hypercholesterolemia was also associated with moderate chronic renal insufficiency among persons without diabetes. In conclusion, the risk for developing moderate chronic renal insufficiency is increased by diabetes and elevated hemoglobin A(1c) and serum cholesterol levels. Modification of these risk factors may decrease the incidence of moderate chronic renal insufficiency.

Improving response to critical laboratory results with automation: results of a randomized controlled trial.

OBJECTIVE: To evaluate the effect of an automatic alerting system on the time until treatment is ordered for patients with critical laboratory results. DESIGN: Prospective randomized controlled trial. INTERVENTION: A computer system to detect critical conditions and automatically notify the responsible physician via the hospital's paging system. PATIENTS: Medical and surgical inpatients at a large academic medical center. One two-month study period for each service. MAIN OUTCOMES: Interval from when a critical result was available for review until an appropriate treatment was ordered. Secondary outcomes were the time until the critical condition resolved and the frequency of adverse events. METHODS: The alerting system looked for 12 conditions involving laboratory results and medications. For intervention patients, the covering physician was automatically notified about the presence of the results. For control patients, no automatic notification was made. Chart review was performed to determine the outcomes. RESULTS: After exclusions, 192 alerting situations (94 interventions, 98 controls) were analyzed. The intervention group had a 38 percent shorter median time interval (1.0 hours vs. 1.6 hours, P = 0.003; mean, 4.1 vs. 4.6 hours, P = 0.003) until an appropriate treatment was ordered. The time until the alerting condition resolved was less in the intervention group (median, 8.4 hours vs. 8.9 hours, P = 0.11; mean, 14.4 hours vs. 20.2 hours, P = 0.11), although these results did not achieve statistical significance. The impact of the intervention was more pronounced for alerts that did not meet the laboratory's critical reporting criteria. There was no significant difference between the two groups in the number of adverse events. CONCLUSION: An automatic alerting system reduced the time until an appropriate treatment was ordered for patients who had critical laboratory results. Information technologies that facilitate the transmission of important patient data can potentially improve the quality of care.

A process to maintain the quality of a computerized knowledge base.

As part of a project to develop knowledge-based reminders for the outpatient setting, we developed a process to help maintain the quality of the knowledge base. The knowledge engineering process involved many parties, including several domain experts, a knowledge engineer, and a programmer and a process was necessary to assure that information transfer among individuals did not become confused. An MS Access database was created to store, among other data, textual versions of the rules as they evolved over time. In a 9-month period 36 rules were entered into the database. Of those, 17 are still active in their original form. The remaining 19 underwent various types of modifications; these changes were tracked in the database. Processes and tools to maintain knowledge bases are necessary if the benefits of clinical decision support systems are to be realized and investments in knowledge engineering are to be protected.

Evaluating physician satisfaction regarding user interactions with an electronic medical record system.

A limiting factor in realizing the full potential of electronic medical records (EMR) is physician reluctance to use these applications. There have been very few formal usability studies of experienced physician users of EMRs in routine clinical use. We distributed the Questionnaire for User Interaction Satisfaction (QUIS) to 75 primary care physicians who routinely use the Brigham and Women's Integrated Computing System (BICS). BICS scored highest in the area of screen design and lowest in the area of system capability. Overall user satisfaction was most highly correlated with screen design and layout, and surprisingly not with system response time. Human-computer interaction studies can help focus our design efforts as we strive to increase clinician usage of information technology.

The Brigham integrated computing system (BICS): advanced clinical systems in an academic hospital environment.

The Brigham integrated computing system (BICS) provides nearly all clinical, administrative, and financial computing services to Brigham and Women's Hospital, an academic tertiary-care hospital in Boston. The BICS clinical information system includes a very wide range of data and applications, including results review, longitudinal medical records, provider order entry, critical pathway management, operating-room dynamic scheduling, critical-event detection and altering, dynamic coverage lists, automated inpatient summaries, and an online reference library. BICS design emphasizes direct physician interaction and extensive clinical decision support. Impact studies have demonstrated significant value of the system in preventing adverse events and in saving costs, particularly for medications.

The impact of computerized physician order entry on medication error prevention.

BACKGROUND: Medication errors are common, and while most such errors have little potential for harm they cause substantial extra work in hospitals. A small proportion do have the potential to cause injury, and some cause preventable adverse drug events. OBJECTIVE: To evaluate the impact of computerized physician order entry (POE) with decision support in reducing the number of medication errors. DESIGN: Prospective time series analysis, with four periods. SETTING AND PARTICIPANTS: All patients admitted to three medical units were studied for seven to ten-week periods in four different years. The baseline period was before implementation of POE, and the remaining three were after. Sophistication of POE increased with each successive period. INTERVENTION: Physician order entry with decision support features such as drug allergy and drug-drug interaction warnings. MAIN OUTCOME MEASURE: Medication errors, excluding missed dose errors. RESULTS: During the study, the non-missed-dose medication error rate fell 81 percent, from 142 per 1,000 patient-days in the baseline period to 26.6 per 1,000 patient-days in the final period (P < 0.0001). Non-intercepted serious medication errors (those with the potential to cause injury) fell 86 percent from baseline to period 3, the final period (P = 0.0003). Large differences were seen for all main types of medication errors: dose errors, frequency errors, route errors, substitution errors, and allergies. For example, in the baseline period there were ten allergy errors, but only two in the following three periods combined (P < 0.0001). CONCLUSIONS: Computerized POE substantially decreased the rate of non-missed-dose medication errors. A major reduction in errors was achieved with the initial version of the system, and further reductions were found with addition of decision support features.

A randomized trial of a computer-based intervention to reduce utilization of redundant laboratory tests.

PURPOSE: To determine the impact of giving physicians computerized reminders about apparently redundant clinical laboratory tests. SUBJECTS AND METHODS: We performed a prospective randomized controlled trial that included all inpatients at a large teaching hospital during a 15-week period. The intervention consisted of computerized reminders at the time a test was ordered that appeared to be redundant. Main outcome measures were the proportions of clinical laboratory orders that were canceled and the proportion of the tests that were actually performed. RESULTS: During the study period, there were 939 apparently redundant laboratory tests among the 77,609 study tests that were ordered among the intervention (n = 5,700 patients) and control (n = 5,886 patients) groups. In the intervention group, 69% (300 of 437) of tests were canceled in response to reminders. Of 137 overrides, 41% appeared to be justified based on chart review. In the control group, 51% of ordered redundant tests were performed, whereas in the intervention group only 27% of ordered redundant tests were performed (P <0.001). However, the estimated annual savings in laboratory charges was only $35,000. This occurred because only 44% of redundant tests performed had computer orders, because only half the computer orders were screened for redundancy, and because almost one-third of the reminders were overridden. CONCLUSIONS: Reminders about orders for apparently redundant laboratory tests were effective when delivered. However, the overall effect was limited because many tests were performed without corresponding computer orders, and many orders were not screened for redundancy.

Using information systems to measure and improve quality.

Information systems (IS) are increasingly important for measuring and improving quality. In this paper, we describe our integrated delivery system's plan for and experiences with measuring and improving quality using IS. Our belief is that for quality measurement to be practical, it must be integrated with the routine provision of care and whenever possible should be done using IS. Thus, at one hospital, we now perform almost all quality measurement using IS. We are also building a clinical data warehouse, which will serve as a repository for quality information across the network. However, IS are not only useful for measuring care, but also represent powerful tools for improving care using decision support. Specific areas in which we have already seen significant benefit include reducing the unnecessary use of laboratory testing, reporting important abnormalities to key providers rapidly, prevention and detection of adverse drug events, initiatives to change prescribing patterns to reduce drug costs and making critical pathways available to providers. Our next major effort will be introduce computerized guidelines on a more widespread basis, which will be challenging. However, the advent of managed care in the US has produced strong incentives to provide high quality care at low cost and our perspective is that only with better IS than exist today will this be possible without compromising quality. Such systems make feasible implementation of quality measurement, care improvement and cost reduction initiatives on a scale which could not previously be considered.