The reduced availability of sophisticated tests and proceduresin hospitals on weekends (the so-called "weekend effect") delayscare. Addressing this problem requires hospital managers to balancethe desire for timeliness with the need for efficient operations.We illustrate how a hospital can profile timeliness, demand, andcapacity utilization across the week for multiple testing areas. Thissimple, practical method, using data extracted from the hospital'saccounting system, makes visible the pattern and magnitude ofdelays caused by reduced availability on weekends, while alsoshowing how capacity is deployed. We combined the analyticaltool with a process of transparent feedback and local problem solvingthat engages multiple stakeholders in the hospital. The goal isto optimally configure capacity so as to balance the imperatives oftimely availability and efficient resource utilization.

For a large hospital that must deploy expensivecapacity while containing costs, the desire fortimely care and efficient operations may requirebalancing competing aims. The care in US hospitalsincreasingly revolves around sophisticated, resource-intensivetests and procedures (eg, computed tomography[CT], magnetic resonance imaging [MRI], invasivecardiovascular procedures, interventional radiology,and gastrointestinal [GI] endoscopy).

On weekends,hospitals often reduce the availability of these services,thereby producing a "weekend effect" that has beenshown to prolong hospitalization,

and may contributeto increased mortality observed in patients admitted onweekends.

In this article, we describe how the desire for across-the-week timely availability of sophisticated tests andprocedures relates to (and may compete with) the questfor efficient use of resources. We then describe how weare balancing those aims in our institution by using apractical analytical tool that gives hospital managers acomprehensive, reproducible view of timeliness,demand, and capacity utilization for multiple testingareas throughout the hospital. We then explain how weuse the tool to support continuous improvement byengaging the providers and requestors of these proceduresas stakeholders. We believe our methods areadoptable by other organizations.

Delays that prolong hospitalization increase the timethat patients are exposed to hazards known to occur inhospitals, such as medication errors and infection.

Delays also negatively affect a hospital's operating economicsby adding days of hospitalization that are notreimbursed under prospective, case-rate payment.Hospitals balance their wish for "24-by-7" availabilityagainst the high costs of that availability. Most do itinformally, relying on traditional working-day hours toestablish a base schedule. Capacity has to be plannedand processes designed in ways that meet the needs ofhospitalized patients for timely care, while avoidingsuboptimal deployment of scarce, expensive resources(labor and equipment). In other words, hospitals needto efficiently translate these inputs into the outputs ofcompleted plans of care, while avoiding waste.

In any service operation (eg, a hospital's testing orprocedural laboratory), the amount of infrastructureand the associated fixed cost are functions of peakworkload.

If time-sensitive capacity is mismatched totime-sensitive demand, artificially high peak workloadsmay require coverage with fixed-cost resources.Avoidable peaks represent waste.

Another source of waste is batch production. For aparticular testing or procedural area, the volume and timingof demand often necessitate intermittent operations,where volume is served in batches, interspersed withperiods of shutdown. If a procedural laboratory can beconceptualized like a manufacturing operation, startingeach batch from a state of shutdown incurs costs of "settingup." Personnel, machines, and materials must bepositioned and prepared for the batch, regardless ofsize.

Excessive setups because of small batch sizes representwaste.

Our main hospital is a 1157-bed, general, acute-carefacility with approximately 40 000 admissions per year.We developed an automated data-extraction tool fromour hospital's accounting analytical system (SunriseDecision Support Manager, Eclipsys Corporation, BocaRaton, Fla). At our institution each testing area is anaccounting cost center.

For an inpatient served by a particular testing area,our accounting system counts the days elapsingbetween admission and the first billed service from thatcost center. Procedures performed the same day asadmission are counted as zero days of waiting. A patientis assumed to receive services only once from that costcenter during a hospitalization. Calculations of averagewaiting times use only the first 6 days of hospitalization.

By using the day of admission (rather than the day ofrequest) to estimate waiting time, we intended to measurethe waiting involved in executing a plan of care formulatedupon admission. This is an importantsimplifying assumption that we believe is more objectivethan measuring the interval from request to procedure,which could be influenced by the requester's perceptionsand expectations of availability. By using theadmission-to-procedure interval, we include the waitinginvolved in the communication and receipt of requests,which are elements of the overall service process.

Twice a year we download data for all the inpatientsserved by a particular testing area in the immediatelypreceding 6 months, with each data record representing1 hospital discharge. Included in each record arefields for the diagnosis-related group, principal procedure,length of stay (LOS, in days), inpatient wardservice, and emergency department admission status(yes or no).

For each 6-month dataset from each testing area(typically containing 400 to more than 3000 cases), weuse the weekday of admission to divide the patientsserved into 7 cohorts. We assume cohorts are similar interms of medical needs. Within each cohort, we counthow many patients receive procedures on the day ofadmission (called day 0) and on subsequent days, therebygenerating a time-to-performance profile. We lookfor effects of differential availability, which appeargraphically as a weekend "gap" and post-weekend "cluster"(

To estimate the opportunity to improve timeliness,we calculate days of potentially avoidable waiting byequalization (DPAWE) by estimating how many daysof waiting could be avoided if patients admitted onFriday, Saturday, and Sunday (who would likelyencounter weekend delays) received their procedureswith the same timeliness as patients admitted onMonday, Tuesday, and Wednesday (Thursday admissionsare omitted from the calculation). To focus on theexecution of the initial plan of care, the calculationignores procedures occurring beyond the sixth day ofhospitalization.

We produce a graphical "dashboard" for each testingarea (

), showing the profiles for the 7 admission-day cohorts, as well as the area's admission andprocedural volume by weekday. The admission volumeby weekday portrays the pattern of demand. The proceduralvolume by weekday portrays capacity utilizationrelative to the weekday with the highest volume. Thedashboard allows users to quickly see patterns of volumeand timing in order to spot gaps, lags, peaks, andsurges that may be opportunities for reconfiguringcapacity. The dashboard includes a summary "boxscore" that shows the average days of waiting and volumesused to calculate DPAWE. Dashboards also can berun for subsets of patients who were served by a testingarea—for example, particular procedure(s), diagnosis-relatedgroup(s), admission site (emergency department),or ward service.

To enable problem solving at the local level, analystswithin each testing area (typically administrators) runthe spreadsheet tool to create the dashboards, whichare then shared freely among testing areas and the hospital'sclinicians. The local analysts also perform targetedsubset analyses for their clinical leaders andmanagers, who plan capacity, make policy, and designoperating processes. A transparent, participatoryapproach allows stakeholders to find a balance betweentimeliness of service and efficient use of resources.

An institutional committee maintains oversight, setsexpectations for improvement, and facilitates dialoguebetween the testing areas ("suppliers") and the clinicians("customers"). Managers are asked to focus onareas with high volumes and high DPAWE scores. Thecommittee also oversees the exchange of best practicesand innovation for ordering, scheduling, and staffing.Typical questions addressed in these dialogues include:

"Do the data confirm prior impressions of timelinessand availability?"

"Could more timely admission assessment andordering mitigate delays?"

"Would a Saturday procedural session (eg, a half-day)substantially mitigate delays? For the anticipatedvolume, what would be required for staffing andupstream/downstream coordination?"

The tool provided our hospital leaders and managerswith their first comprehensive view of timeliness acrossthe week for high-impact testing procedures throughoutthe hospital. In some areas, waiting time for testsordered for patients admitted late in the week (Friday,Saturday, Sunday) appeared to improve modestly afterthe tool was implemented in 2002. For example, forechocardiography, the average number of days fromadmission to procedure decreased from 1.7 in 2002to 1.5 in 2004. For body CT, the number of daysdecreased from 1.2 in 2002 to 0.9 in 2004; for GIendoscopy, the decrease was from 2 days in 2002 to1.7 days in 2004. In most areas timeliness remainedabout the same or improved slightly from 2002 to2004, despite growth in volume in some areas (datanot shown). For many areas (echocardiography,cardiac catheterization, neurological MRI, GI endoscopy)the disparity in average waiting timesbetween late-week and early-week admissionsappeared to modestly decrease.

The profiling technique allowed us to see how wellthe admission volume by weekday that used a particulartesting area ("demand") aligned with its proceduralvolume by weekday ("capacity," relative tothe peak weekday). In areas with excellent timelinessacross the week, the procedural-volume profile closelyresembled the admission-volume profile. However,in areas with weekend-related delays, the procedural-volume profile looked very different from theadmission-volume profile. We found a relative burstof activity on Mondays (from cases held over theweekend) and a burst on Fridays (presumably tryingto "clear the decks" ahead of the weekend)(

Rather than mandating specific weekend schedules,we designed a more locally driven process thatengages stakeholders who have clinical-leadership,managerial, provider, and front-line productionroles. Nearly all users have commented that thetool is easy to understand and helps them appreciatethe magnitude of opportunities. In response toour method, several areas plan to expand availabilityon weekends or evenings, guided by cost-benefitanalysis made possible by using the tool to betterunderstand differential waiting, demand, and capacity.Areas also are looking at ways to do proceduresearlier in order to alleviate high peak-volume weekdays(typically Mondays) and reduce associatedfixed-cost resources.

Using administrative data, we developed an analyticaltool that profiles timeliness, demand, and capacityutilization for sophisticated tests and procedures acrossthe week, thereby making visible the pattern and magnitudeof delays caused by reduced availability on weekends.We deployed the tool within the cooperativemanagement structure of our hospital to engage thestakeholders who would produce change. Because thedata are readily accessible from the hospital's accountingsystem, we believe other hospitals could easily performthis profiling.

Previously published efforts to improve availabilityof inpatient testing and procedures to expedite carehave addressed 1 testing area at a time.

In contrast,we have introduced a more comprehensive system ofassessing and improving timeliness and capacity managementfor multiple areas. The transparency, objectivity,and simplicity of our method help identifyproblems and facilitate a cooperative approach towardsolving them.

Our analytical method is similar to that of Bell andRedelmeier, who documented weekend-related delaysfor selected urgent procedures for emergently hospitalizedpatients in the Canadian province of Ontario.

Using administrative data, they measured the days waitingfrom admission to procedure and showed how waitingvaried according to weekday of admission.

Our administrative data captured timing detail at thelevel of the day of service. We did not use medicalrecord data or hourly service logs (which could paint amore detailed picture of testing timeliness), althoughwe encourage managers to tap into (or prospectivelygenerate) these types of data locally to facilitate problemsolving.

A sufficiently long run of administrative data shouldreveal patterns of timeliness, demand, and capacity utilizationthat are meaningful enough to justify changes incapacity and availability. Because follow-up measurementusing administrative data is fast and inexpensive,managers can implement changes, assess results, makeadjustments, and later repeat the measurement.

Although reducing the waiting for a given test or procedureshould contribute to shortening LOS, it may bedifficult to predict the aggregate LOS reduction attributableto reconfiguring 1 area. The relationship of 1testing area to LOS is sometimes straightforward (eg, apatient admitted with angina who needs only cardiaccatheterization and angioplasty). More often, the relationshipof the testing area to LOS is more indirect,because it is part of a sequence of contingent steps (eg,a patient admitted with GI bleeding who is found tohave a gastric cancer, but then needs surgery). Waitingfor testing has been shown to be an important cause ofdelayed care in a large teaching hospital, but there aremany other causes.

Removing a "bottleneck" in 1 area may reveal anotherdownstream impediment. Multiple areas of the hospitalmay need to improve simultaneously to accelerate thecare of large numbers of patients. We believe our methodhas the advantage of being able to examine multiple areassimultaneously using a uniform, systemic approach.

Because every hospital has its own combination ofoperating configuration, cost functions, and stakeholderdynamics, finding the optimal balance between timelinessof service and efficient resource utilization probablydefies a formulaic "cookbook" solution. We areusing a local approach to problem solving, in which thepersons most intimately familiar with operations developsolutions and implement change, guided by a toolthat provides easy-to-understand feedback and reinforcement.We believe that the insights revealed by thistransparent, intuitive analysis help clinical leaders,managers, and front-line personnel work together todevelop optimal solutions.

We wish to thank David A. Asch, MD, MBA (Leonard Davis Institute ofHealth Economics, University of Pennsylvania) and David C. Herman, MD(Clinical Practice Committee, Mayo Clinic) for helpful comments on thismanuscript. We also thank Larry D. Albrecht (Department of Systems andProcedures, Mayo Clinic) for assistance with data collection and Jaysen P.Ness, Julie A. Lisowski, and Tina K. Sass (Rochester Financial Analysis,Mayo Clinic) for ongoing analytical support.

From the Department of Internal Medicine (LHL, CAG, DLW), the Department ofRadiology (SJS), and Rochester Financial Analysis (RRM), Mayo Clinic and Foundation, 200First Street SW, Rochester, Minn.

Address correspondence to: Lawrence H. Lee, MD, MBA, Department of InternalMedicine, Siebens 660, Mayo Clinic, 200 First Street SW, Rochester, MN 55905. E-mail:lee.lawrence@mayo.edu.

1. Maitino AJ, Levin DC, Parker L, Rao VM, Sunshine JH. 

Nationwide trends inrates of utilization of noninvasive diagnostic imaging among the Medicare populationbetween 1992 and 1999. 

2. Hahn PF, Gervais DA, O'Neill MJ, Mueller PR. 

Nonvascular interventional procedures:analysis of a 10-year database containing more than 21 000 cases.

Hospitals to boost spending, a move that may hurtprofits. 

5. Cai Q, Bruno CJ, Hagedorn CH, Desbiens NA. 

Temporal trends over 10 yearsin formal inpatient gastroenterology consultations at an inner city hospital. 

6. Varnava AM, Sedgwick JE, Deaner A, Ranjadayalan K, Timmis AD. 

Restrictedweekend service inappropriately delays discharge after acute myocardial infarction.

Restricted weekend services result in delays in discharges from hospital.

Articles

MANAGERIAL

Optimizing Weekend Availability for Sophisticated Tests and Procedures in a Large Hospital

To evaluate the safety and effectiveness of a simvastatin-to-lovastatin therapeutic conversion program.

Observational database study of a therapeuticconversion in members of the Northern and Southern Californiaregions of Kaiser Permanente, using a pretest/posttest design.

All patients actively converted from simvastatin tolovastatin between April 1, 2002, and March 31, 2003, were identifiedfor inclusion in the analysis. The conversion from simvastatinto lovastatin was based on an equipotent dose ratio of 1 mg of simvastatinto 2 mg of lovastatin. Electronic prescription record andlaboratory data were collected for converted patients beginning365 days before changing therapy through June 30, 2003. The primaryeffectiveness end point was a comparison of the preconversionand postconversion low-density lipoprotein cholesterol(LDL-C) levels. Safety end points included an analysis of preconversionand postconversion alanine aminotransferase (ALT) testsand creatine kinase values.

: A total of 33 318 converted patients met criteria forinclusion in the analysis. The mean LDL-C was lowered from 110.9to 108.4 mg/dL (

< .001) following the conversion to lovastatin.The percentage of patients with serum ALT levels greater than 3times the upper limit of normal (ULN) was similar before (0.7%)and after (0.6%) conversion from simvastatin to lovastatin.Creatine kinase elevations greater then 10 times the ULN occurredat similar rates before and after the conversion.

Overall, patients had an improvement in theirlipid profile without evidence of hepatic or muscle enzyme elevations.Appropriately selected patients can be safely and effectivelyconverted from simvastatin to lovastatin.

It is well established that the 3-hydroxy-3-methylglutarylcoenzyme A (HMG-CoA) reductase inhibitors(statins) reduce the risks of cardiovasculardisease. Results from large, randomized clinical trialshave established the safety and efficacy of statins withdemonstrated reductions in total and coronary heartdisease (CHD) mortality, acute coronary syndromes,revascularization procedures, and stroke.

Currently,the National Cholesterol Education Program's AdultTreatment Panel III (ATP III) recommends statins asfirst-line agents when drugs that lower low-densitylipoprotein cholesterol (LDL-C) are indicated to achievetreatment goals.

Selection of a statin(s) for inclusion on a formulary isreasonably made on the basis of comparative efficacy,clinical benefit, safety, and relative cost. In December2001, generic versions of lovastatin were approved bythe Food and Drug Administration. With the availabilityof a generic formulation, lovastatin became the mostcost-effective statin for patients requiring mild to moderateLDL-C reductions within Kaiser Permanente (KP).

As the first statin approved for clinical use, lovastatinhas undergone extensive study evaluating safety andefficacy. Dose-ranging studies with lovastatin demonstratedLDL-C reductions of up to 41% with lovastatin80 mg daily, along with favorable effects on high-densitylipoprotein cholesterol (HDL-C) and triglyceride(TG) levels.

Lovastatin has been shown to reducecardiovascular events in a large primary preventionstudy.

In this study lovastatin was shown to reduce therisk of a first coronary event in men and women withaverage or mildly elevated LDL-C and TG levels andbelow-average HDL-C.

Angiographic studies havedemonstrated beneficial effects of lovastatin with areduction in the progression of coronary athersclerosis.

In 2 of these studies, lovastatin also demonstratedclinical benefit in a secondary-prevention population.

At the time generic lovastatin was approved, brandedsimvastatin (Zocor) was commonly prescribed withinKP. Compared with simvastatin, the LDL-C-loweringpotency of lovastatin is approximately half on a milligram-to-milligram basis (10-40 mg of simvastatinbeing approximately equipotent to 20-80 mg of lovastatin).

Relative safety appears to be similarwhen equipotent doses of simvastatin and lovastatin arecompared.

Based on the evidence of comparable efficacy andsafety, a conversion program to switch eligible patientsfrom simvastatin to lovastatin was undertaken in theNorthern and Southern California Regions of KP. In thispaper, we report our experience in converting a largepopulation of patients from branded simvastatin togeneric lovastatin.

KP is a large, national, not-for-profit, group-modelHMO providing integrated healthcare services to itsmembers. Between April 2002 and March 2003, a campaignwas undertaken within KP California to convertpatients taking simvastatin to an equipotent dose oflovastatin. The conversion from the formulary brandedsimvastatin product to generic lovastatin in selectedpatients (see criteria under "Patients") was deemedappropriate by medical staff committees involved withformulary decisions (regional pharmacy & therapeuticscommittees), cholesterol management, and specialistcommittees. Primary care providers were informed whentheir patients were potential candidates for the conversion.Providers either approved or disapproved the conversionfor each patient.

Patients approved for conversion were informed by aletter from the prescribing physician/provider that theconversion was going to occur. Conversion occurred atthe pharmacy when the patients came in to pick up theirprescription. Patients were counseled by the pharmacistabout the conversion and given written materials withtheir lovastatin prescription. Converted patients alsowere provided with instructions and a lab slip for a follow-up fasting LDL-C and serum alanine aminotransferase(ALT) test to be performed 6 to 8 weeks afterstarting lovastatin. This study (evaluating the conversion)was approved by the regional KP institutionalreview boards in both Northern and Southern California.

The study was designed as a pre-post observationalanalysis of patients converted from simvastatin tolovastatin. The conversion from simvastatin to lovastatinwas based on an equipotent dose ratio of 1 mg ofsimvastatin to 2 mg of lovastatin (eg, patients receivingsimvastatin 10 mg/day were converted to lovastatin 20mg/day). The patient's healthcare provider had discretionover the conversion dose of lovastatin, includingreducing the amount a patient received based on clinicalcircumstances.

Electronic data were collected on all patients activelyconverted between April 1, 2002, and March 31,2003. Patient prescription records for the year beforeconversion (simvastatin therapy) and the months afterconversion (while on lovastatin therapy) were recorded.The daily doses of simvastatin and lovastatin weredetermined based on the tablet strength and the directionsfor use. Laboratory data were collected beginning365 days before the conversion through June 30, 2003,for each patient. This allowed for a minimum of 3months of follow-up for all patients. Laboratory monitoringconsisted of baseline and postconversion lipidresults (LDL-C, HDL-C, and TG) and all serum transaminasetests (ALT and aspartate aminotransferase[AST]). Creatine kinase (CK) tests were not required bythe conversion protocol, but were captured if orderedby the patient's provider as part of usual care.

If a patient's baseline LDL-C was more than 10%above the identified goals for primary and secondaryprevention, <130 mg/dL and <100 mg/dL, respectively,then a dosage adjustment was suggested as part of theconversion. A threshold of 10% above goal was basedupon the prevailing opinion within the medical groupsin early 2001 that slight elevations in LDL-C (eg, anLDL-C of 103 mg/dL for secondary prevention) did notnecessarily warrant a dosage adjustment. Therefore, thethresholds for dose escalation recommendations were143 mg/dL or greater for primary prevention and 110mg/dL or greater for secondary prevention. For example,if a primary prevention patient had a baseline LDL-Cof 150 mg/dL while taking simvastatin 10 mg/day, therecommended lovastatin conversion dose would be 40mg/day instead of an equipotent lovastatin dose of 20mg/day. The maximum lovastatin dose allowed on anyconversion was 80 mg/day. If a patient was not at goalon simvastatin 40 mg daily, the recommendation was toadjust the dose to simvastatin 80 mg daily. In responseto the postconversion LDL-C values, the decision toadjust a patient's lovastatin dosage was left up to the primarycare provider. For purposes of this study we comparedthe final lovastatin dosage with the preconversionsimvastatin dosage.

All patients age &#8805;18 years receiving simvastatin atdoses up to 40 mg daily were potential candidates forthe conversion. To be eligible for the analysis, patientswere required to have a baseline and a postconversionLDL-C lab test.

Exclusions from the conversion were based on safetyconsiderations and the presence of potentially interactingdrugs. Therefore, patients at increased risk of muscletoxicity were not converted. These included patientswith decreased renal function (women with a serumcreatinine [SCr] concentration of &#8805;2.0 mg/dL and menwith a SCr concentration of &#8805;2.5 mg/dL), patients takingdrugs known to interfere with lovastatin metabolism(cyclosporine, macrolide antibiotics, azole antifungals,verapamil, amiodarone, protease inhibitors, and nefazodone),patients currently taking niacin or fibrates,patients with known HIV infection, and patients receivingmore than 40 mg of simvastatin per day.

Classification of Diseases, Ninth Revision

Patients were divided into primary-prevention andsecondary-prevention categories. Estimating a patient's10-year CHD risk was not possible because family historyof heart disease, tobacco use, and blood pressuremeasurements were not available in our electronicdatabases. Therefore, secondary-prevention patientswere defined as those meeting criteria for inclusioninto 1 of 2 KP disease management registries at thetime of conversion (Diabetes Registry and/or CoronaryArtery Disease [CAD] Registry). Inclusion in theDiabetes Registry is based on diagnosis coding (

] code 250 -diabetes) and prescription records(oral antidiabetic agents and/or insulin). Inclusion inthe CAD Registry is based on diagnosis coding (

codes 410, 411, 412 and 414—ischemic heart disease,440—atherosclerosis, 443.9—peripheral vascular disease),procedure codes (Current Procedural Terminology36.0x—percutaneous transluminal coronaryangioplasty, 36.1x—coronary artery bypass graft), andprescription records (oral and topical nitrates).Patients not meeting inclusion criteria for 1 of theseregistries were categorized in the primary-preventiongroup.

Conversion effectiveness was determined by comparingbaseline and postconversion LDL-C results.Baseline LDL-C tests had to be done within 365 days ofthe conversion date while the patient was on simvastatin.If a patient had multiple LDL-C tests within the365-day window, then the lab value closest to the conversiondate was used for the analysis. PostconversionLDL-C lab tests needed to be done at a minimum of 4weeks after the conversion date. For patients with multipleLDL-C tests performed postconversion, the lastLDL-C test result on lovastatin therapy was the primaryend point used in the analysis. This result reflected anydosage titrations done subsequent to the initial conversion.We also analyzed the first postconversion LDL-Ctest results.

Although HDL-C and TG laboratory tests were recommendedfor all converted patients, this was not arequirement for inclusion in the analysis. Baseline andpostconversion HDL-C and TG laboratory tests used thesame timing criteria established for LDL-C testing.

The primary safety end points were a comparison ofALT and CK elevations postconversion versus preconversion.All ALT and CK tests done within 365 daysbefore the conversion date (baseline) were collected.Postconversion ALT and CK tests were those done atany point after the conversion date through the end oftherapy or June 30, 2003. If a patient had multiple ALTor CK tests, then the highest (maximum) test result wasused for comparison. Therefore, the maximum resultwas not necessarily the test done immediately beforeconversion. CK increases associated with a myocardialinfarction were excluded from the safety analysis. Forpurposes of this analysis, a myocardial infarction wasdefined as a Troponin I result of &#8805;0.3 ng/mL, or a combinationof a CK-MB percentage of &#8805;3 plus a total CKMBof &#8805;8.1 mg/dL.

Patients served as their own controls with the prepostdesign. A chi-square test and McNemar's test wereused for categorical outcome variables and frequencyresults. A paired 

test and the Wilcoxon signed ranktest were used for continuous variables, with a decisionbased on the distribution of the results (normal vs nonnormaldistribution). Multivariate logistic regressionwas used to evaluate patient risk factors associated withelevated CK laboratory results after conversion. Giventhe large population of patients converted from simvastatinto lovastatin, a 2-sided 

< .01 was defined as thelevel necessary to achieve statistical significance.

A total of 33 318 patients met criteria for inclusion inthe analysis. Patient demographics are listed in 

Patients converted to lovastatin were elderly and predominantlymale. Within our population, more womenwere being treated for primary prevention than for secondaryprevention, while the split was more even withmale patients. The average duration of lovastatin followupwas 11 months, and the mean time between conversionand the last LDL-C laboratory test was more than6 months. On initial conversion, the majority ofpatients (82.7%) received an equipotent dose of lovastatin,with a small percentage (5.6%) receiving a doseincrease. We observed further dosage titration duringthe follow-up period, and an additional 5.1% of patientshad their lovastatin dose increased.

The frequency of follow-up LDL-C testing varied, with47.9% of patients having only a single test performed and6.9% having 4 or more tests done. Based on eachpatient's last or only test result during follow-up (primary end point), LDL-C decreased by an average of 2.5mg/dL, HDL-C increased by 2.2 mg/dL, and TG decreasedby 1.6 mg/dL following the conversion (

). Results were slightly different when the firstpostconversion lab tests were analyzed: average LDL-Cdecreased by 1.8 mg/dL, HDL-C decreased by 1.0 mg/dL,and TG increased by 1.0 mg/dL. Improvements in lipidtest results were seen in both primary-prevention andsecondary-prevention patients. All lipid changes werehighly significant (

< .001), except for TG changes insecondary-prevention patients. This level of statisticalsignificance is likely due to the large size of the studypopulation.

When grouped by postconversion dosage adjustment,all 3 groups of patients (dose increase, dose decrease,and equipotent dose) had lower LDL-C values after conversion,although differences were not always statisticallysignificant. In patients who received an equipotentdose of lovastatin, the change in LDL-C was not statisticallysignificant (

The percentage of patients at target LDL-C values forsecondary prevention (<100 mg/dL) changed from55.7% preconversion to 56.4% postconversion (

= .12,not significant). For primary-preventionpatients, the percentage ofpatients at goal postconversion wassignificantly higher than the percentageat goal preconversion (

Only 72.6% of the patients whomet inclusion criteria had both abaseline and a postconversion ALTmeasurement done (

). Anequal number of patients experiencedmild elevations in ALT beforeand after converting to lovastatin.Marked elevations in ALT, greaterthan 3 times the upper limit of normal(ULN), occurred in 0.7% of thepatients before and 0.6% of patientsafter the conversion. Compared withpatients who had normal ALT valuesat baseline, patients with marked elevationsin ALT before converting weremore likely to have marked elevationsafter conversion (

As CK tests were ordered at thediscretion of the primary careproviders, the total number ofpatients with CK tests at baseline andpostconversion was different. Ofthose tested, marked CK increases(>10 ULN) occurred at the same ratebefore and after the conversion(Table 3). No patient with a marked elevation in CKbefore conversion experienced a similar increase onlovastatin. Male patients (odds ratio [OR] = 1.8, 95%confidence interval [CI] = 1.1, 3.0) and those with elevatedCK results (>500 U/L) before conversion (OR =16.1, 95% CI = 9.7, 26.3) were more likely to have elevatedCK tests (>500 U/L) postconversion. Lovastatindosage and age were not associated with increased oddsof elevated CK test results postconversion.

Our results in 33 318 patients support the effectivenessand safety of converting patients from simvastatinto lovastatin. Postconversion lipid, ALT, and CK levelswere similar to the preconversion levels. In a comparisonof preconversion and postconversion average lipidtest results, LDL-C decreased, HDL-C increased, andTG decreased. Rates of ALT and CK elevations werenot significantly different before and after the conversion.These data confirm our a priori hypothesisthat lovastatin is an effective and safe alternative formanaging hypercholesterolemia.

Our results are consistent with previous statin conversionstudies.

Three of these studies included relativelysmall numbers of patients (approximately 100) intheir analysis.

The investigators in these studiesfound no difference in the preconversionand postconversion lipid levelsand concluded that the conversionwas safe and effective. Safety andeffectiveness, however, are difficult toassess with sample sizes of only 100patients. Ito et al conducted a largestudy involving 1032 VeteransAdministration patients who completeda conversion from pravastatinto simvastatin.

Individual simvastatindoses were titrated to achievepredefined LDL-C results. In thisstudy, LDL-C improved from 116mg/dL to 99 mg/dL, and the percentageof patients at goal improved from44% to 69% postconversion. Liverfunction tests and CK results weresimilar at baseline and postconversion.The authors concluded thatconversion was effective and costsaving.

One strength of our study is thelarge population of patients we wereable to observe. This is importantwhen evaluating both the effectivenessand the safety of a therapeuticconversion. In particular, our largesample provided the power necessary to evaluaterates of hepatic and muscle enzyme elevationsassociated with these 2 statins. Increasesin ALT greater than 3 times the ULN andincreases of CK greater than 10 times the ULNoccurred at similar rates with simvastatin andlovastatin.

Our rate of marked hepatic enzyme elevationwas comparatively low (0.6-0.7%).

A possibleexplanation for the low rate observed inour population was that patients were beingsuccessfully treated with simvastatin and weretherefore self-selected. Patients intolerant tosimvastatin and not taking the drug were notinvolved in the conversion. We also specificallyexcluded certain populations who werepotentially at higher risk for toxicity based oncoexisting conditions and potentially interactingdrugs.

There was no difference in the rate of increasedCK values between simvastatin (preconversion)and lovastatin (postconversion). However, the rate ofmarked CK elevations was higher in our population than isusually reported in the literature.

An explanation forthe higher rates in our study is that our rates reflect onlythose patients who had CKtests ordered; most of thepublished incidence ratesreport CK elevations as apercentage of the total numberof patients on statinprescriptions.

The slightly improvedlipid results we observed atequipotent doses mayreflect external patient factorsassociated with theconversion. Patients mayhave become more compliantwith their new regimendue to increased contactwith providers. Patients alsoknew that laboratory followupwas a necessary part ofthe conversion, and the factthat outreach efforts weremade to those patients whofailed to get laboratory tests done could also have affectedthe results.

A limitation of this study is that it relied on electronicdata capture and did not include direct patient contactor chart reviews by the investigators. This can leadto misclassification errors, especially when separatingpatients into primary-prevention and secondary-preventioncategories based on registry data, or when evaluatingpostconversion dose-adjustment groupings. Oneanomalous finding was that patients with effective postconversiondosage reductions also achieved lower postconversionLDL-C values. Information not capturedelectronically but available to the provider may explainthese results. For example, providers may have decidedto reduce the dose based on clinical circumstances outsideof LDL-C results and CK or ALT values.

Another limitation to this study is that clinical outcomeswere not measured; lipid laboratory results wereused as surrogate effectiveness markers and the primaryend point for this analysis. The follow-up period also wasshort and variable, ranging from 3 to 15 months.Therefore, we were unable to assess patients' persistence(long-term adherence) with their new lovastatin regimen.

CLINICAL

Successful Conversion of Patients With Hypercholesterolemia From a Brand Name to a Generic Cholesterol-Lowering Drug

To determine the relationship between provider type(eg, resident, mid-level practitioner, attending physician) andattainment of clinical goals with respect to the treatment of dyslipidemia,diabetes mellitus, and hypertension.

Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) 

Using electronic medical records, we identifiedall patients seen in the Veterans Affairs Connecticut Health CareSystem in a 6-month period with an 

codecorresponding to a diagnosis of coronary artery disease, diabetesmellitus, or hypertension.

We recorded the most recent low-density lipoproteincholesterol level for patients with diabetes or coronary arterydisease, glycosylated hemoglobin level for patients with diabetes,and blood pressure for patients with hypertension. We thenexamined the relationship between these measures and providertype.

After controlling for patient age and practice site, nosignificant differences were noted between attending physiciansand residents in attaining treatment goals for dyslipidemia (oddsratio [OR] 1.15, 95% confidence interval [CI] 0.96-1.38) or diabetes(OR 1.05, 95% CI 0.82-1.33). However, attending physicianswere significantly more likely to attain blood pressure goals thanwere residents (59% vs 54%, 

= .002; OR 1.23, 95% CI 1.08-1.39). Controlling for additional patient characteristics did not alterthe findings.

Internal medicine residents may quickly developexpertise in managing dyslipidemia and diabetes mellitus.Residents in our sample, however, were less likely than attendingphysicians to reach goal blood pressure for patients with hypertension.Educational efforts aimed at house staff to improve thetreatment of hypertension are warranted.

Cardiovascular disease (CVD) is the leading causeof death in the United States.

Estimates are thatnearly 50 million Americans have high bloodpressure, 42 million have elevated blood cholesterol levels,and almost 11 million have diabetes mellitus.

Notonly does CVD account for substantial morbidity andmortality, it also exerts a substantial financial burdenon the healthcare system. In 1999, more than $26 billionwas paid to Medicare beneficiaries for hospitalexpenses due to cardiovascular problems.

Becauseseveral of the modifiable risk factors for CVD are wellknown and prevalent (eg, hypertension, dyslipidemia,and diabetes mellitus), efforts aimed at targetingthese conditions have been promulgated by majororganizations.

The Veterans Health Administration in theDepartment of Veterans Affairs (VA) is the largest fullyintegrated healthcare system in the United States, providingcare to nearly 3.5 million persons annually.

Within the VA system, primary care is provided byattending physicians, mid-level practitioners (eg, nursepractitioners and physician assistants), and internalmedicine house staff (residents).

Although residents have been shown to be deficientin their knowledge and screening practices with respectto colon cancer,

they are more likely to prescribe narrow-spectrum antibiotics than attending physicians forsinusitis.

In a study conducted in a large urban publichospital, diabetic patients of residents had similar ratesof goal attainment for glycosylated hemoglobin (A1C)and cholesterol when compared to diabetic patients offaculty.

Little is known about the relative performance ofthese provider types with respect to the treatment ofcommon outpatient conditions within the VA, such asdiabetes mellitus, dyslipidemia, and hypertension. Wesought to determine whether internal medicine residentsin training were less likely to be successful inattaining recommended goals for their patients thantheir more experienced attending physicians or midlevelproviders.

The VA Connecticut Health Care System cares formore than 44 000 veteran patients at 2 major academichealthcare centers (the West Haven and Newington VAMedical Centers) and 6 community-based clinics.Patients are randomly assigned a primary care provider.These primary care providers consist of internal medicineresidents, who practice under the supervision ofattending physicians; mid-level practitioners, andattending physicians. Fifty-six internal medicine residents(40 affiliated with Yale University School ofMedicine and 16 with the University of ConnecticutSchool of Medicine) have their longitudinal generalmedicine clinic within the VA Connecticut Health CareSystem and provide primary care for 8.4% of patientsenrolled in primary care. Each resident is supervised bya designated general internal medicine faculty member.Each attending physician supervises 3 to 4 residents atany 1 time and must cosign all resident notes. Residentsare expected to present each patient to the attendingphysician, during which time the patient's problems arediscussed and a plan for care is made conjointly. Whena resident completes residency training, the resident'spanel of patients is transferred to that of another incomingresident.

In the VA Connecticut Health Care System, 18 mid-levelpractitioners (13 nurse practitioners and 5 physicianassistants) practice independently, although theyare assigned designated attending physicians shouldclinical questions arise. All attending physicians forresidents or mid-level practitioners are full-or part-timestaff physicians in primary care, see their ownpanel of primary care patients, and hold academicappointments at the University of Connecticut Schoolof Medicine or at Yale University School of Medicine.When attending physicians supervise residents in primarycare, however, they do not see their ownpatients. Within the VA Connecticut Health CareSystem 39 attending physicians are engaged in thepractice of primary care.

The VA uses an electronic medical records system,which allowed us to compile patient lists and correspondinglaboratory values. Within the electronic medicalrecord, a patient's primary care provider is clearlyidentified, thus allowing us to identify provider types.Blood pressure measurements are obtained at all primarycare visits, and blood tests are ordered at the discretionof the providers. Blood pressure values areentered into the electronic medical record, as are theresults of all blood tests performed at the VA.

International Classification of Disease, 9th Edition,

Each patient visit is electronically assigned an

) code correspondingto the diagnoses for which the patient was treated. Weidentified patients with coronary artery disease (CAD)using the codes 410 to 414.9, diabetes mellitus using thecodes 250.00 to 250.93, and hypertension using thecodes 401.0 to 401.9. Psychiatric comorbidity, that is,alcohol disorder, drug disorder, and severe mental illness(bipolar disorder, major depressive disorder, andschizophrenia), were identified using the codes 305.00,305.01, 305.02, 305.03, 303.xx, 291.xx, 305.2x, 305.99,292.xx, 304.xx, 295.xx, and 296.xx. Patients who wereseen at VA Connecticut between April 1, 2003, andSeptember 30, 2003, were identified for our study bythese codes.

codes relating to eitherdiabetes mellitus or CAD, we recorded the most recentlow-density lipoprotein (LDL) measurement obtainedwithin the last 25 months. For patients with diabetesmellitus, we also recorded the most recent A1C measurementobtained within the past 13 months. Forpatients with hypertension, we noted the most recentblood pressure obtained during the previous 13 months.These time frames were chosen because VA standardsrecommend obtaining LDL cholesterol level measurementsat least biannually for patients with diabetes mellitusor CAD, A1C at least yearly for patients withdiabetes, and blood pressures at least yearly for allpatients. Computer-generated prompts remindproviders of these standards.

Target goals for LDL, A1C, and blood pressure wereless than 100 mg/dL,

less than 7.5%, and less than140/90 mm Hg,

respectively. Because the AmericanDiabetes Association recommends that target A1C beless than 1% above the upper limit of normal,

we useda target A1C of less than 7.5% since the upper limit ofnormal for our laboratory is 6.5%.

Patients without blood pressure or laboratory testvalues or without an identified provider were excludedfrom analysis. Because female veterans were treatedalmost entirely by residents and at a separate site, theywere also excluded from analysis. The study protocolwas approved by the institutional review board at theWest Haven Veterans Affairs Medical Center.

Descriptive statistics on the sample characteristicsincluded frequencies, proportions, and ranges.Indicator variables were created for the site of care, the3 conditions of interest (hypertension, diabetes, andCAD), and for each comorbid psychiatric condition (alcoholdisorders, drug disorders, and severe mental illness).

Chi-square and Fisher's exact test, in cases in whichcell sizes were small, were used to examine the relationshipbetween provider type and dichotomous patientcharacteristics. Logistic regression was used to adjust forpotential confounding factors in the association betweenprovider type and the patients' attainment of treatmentgoals. Patient age and site of care were controlled for inall multivariate analyses. Additional analyses controlledfor comorbid psychiatric disorders and comorbid medicalconditions. Separate analyses were performed foreach goal of interest (LDL, A1C, and blood pressure).Each analysis utilized data from patients diagnosed withthe condition of interest for which a particular goal wasto be met (eg, only patients with diabetes were includedin the analysis on attaining A1C < 7.5%).

Odds ratios (ORs) and 95% confidence intervals (CIs)were calculated using the coefficients and standarderrors of the respective variable in the logistic regressionanalysis. Odds ratios from the logistic regressionswere calculated using resident as the referent group.Statistical significance was defined as a 2-tailed 

< .05.All statistical analyses were performed using SAS version8 (SAS Institute, Inc, Cary, North Carolina).

A total of 20 453 unique patients with a diagnosis ofhypertension, diabetes mellitus, or CAD were seen duringthe observation period. From this group, 294 wereexcluded because they had no identifiable primary careprovider and an additional 499 were excluded becausethey were female. Afterexcluding these 793 patients,19 660 unique patients metentry criteria and wereincluded in the analyses.

Attending physicians saw12 682 unique patients, mid-levelpractitioners saw 5572patients, and residents saw1406 patients. A total of15 893 patients had a diagnosisof hypertension, 6292a diagnosis of CAD, and 5830a diagnosis of diabetes mellitus;7277 patients had 2 ormore of these diagnoses.Patient characteristics, diagnoses,and comorbidity arelisted in 

A significant differencewas noted in the age of thepatients by the providertypes (

< .001). Residentsand attending physicianswere more likely to seeolder patients than midlevelpractitioners. Significant differences were notedin the proportion of patients with comorbid psychiatricdisorders by provider type: mid-level practitionerssaw more patients with alcohol disorders (

= .017); residents saw a smaller proportion ofpatients with these comorbidities than either attendingphysicians or mid-level providers.

A significant difference was noted in the proportion ofpatients with hypertension, diabetes, or CAD by providertype: mid-level practitioners were more likely to seepatients with hypertension compared to attending physicians;residents were less likely to see patients with CADcompared to attending physicians; and mid-level practitionerswere less likely to see patients with diabetes mellituscompared to attending physicians (Table 1). Whenresidents were compared with mid-level practitionersand attending physicians combined, residents saw significantlyolder patients and were less likely to see patientswith CAD or comorbidities (data not shown).

For patients with CAD, 70% of attending physicians'patients had met the treatment goal of LDL less than100 mg/dL, compared with 68% of patients cared for bymid-level practitioners and 72% cared for by residents(

). No significant differences werenoted between the provider types on attainment of LDLgoals.

For diabetic patients, 71% of attending physicians'patients had A1C levels less than 7.5%, compared with74% of mid-level practitioners' patients and 71% ofresidents' patients (

= .190). No statistically significantdifferences were noted in A1C attainment by providertype.

For patients with hypertension, the target bloodpressure of less than 140/90 mm Hg was attained by 59%of attending physicians' patients, compared with 57% ofmid-level practitioners' patients and 54% of residents'patients (

= .001) (Table 2). Significant differences werenoted in blood pressure control between residents' andmid-level practitioners' patients (

= .018) and betweenresident and attending physicians' patients (

After controlling for patient age and practice location,attending physicians were significantly more likelyto attain blood pressure goals compared to internalmedicine residents (OR 1.23, 95% CI 1.08-1.39). No significantdifference was noted in blood pressure controlbetween mid-level practitioners and residents (OR 1.12,95% CI 0.98-1.28) (

No significant difference was noted for patientattainment of LDL cholesterol goals between attendingphysicians and residents (OR 1.15, 95% CI 0.96-1.38)or between mid-level practitioners and residents (OR1.03, 95% CI 0.85-1.26). Mid-level practitioners weresignificantly more likely to attain A1C goals comparedto residents (OR 1.30, 95% CI 1.01-1.70); however,attending physicians were not more likely to attainA1C goals than residents (OR 1.05, 95% CI 0.82-1.33)(Table 3).

In separate logistic regression models controlling forthe additional effects of co-occurring medical and psychiatriccomorbidities, the pattern of ORs did notchange significantly.

Given the complexity of healthcare, use of qualitymeasures is essential to ensure the promotion of healthand prevention of disease complications. This is especiallytrue in academic medical centers, where traineesare rendering care to patients. Substandard care hasnegative implications not only to the training programbut, more importantly, to the health of individualpatients as well.

With respect to the control of 3 common outpatientproblems, we found that although attending physicianswere not more successful than residents in obtainingtreatment goals for patients with dyslipidemia and diabetesmellitus, they were more successful in attainingblood pressure goals for hypertensive patients.

Several theories may explain why the rates of goalattainment were not markedly different betweenattending physicians and residents for the treatment ofdyslipidemia or diabetes mellitus. Because these conditionsare so prevalent, perhaps residents in trainingquickly develop expertise in these common clinicalentities. Additionally, familiarity with well-promulgatedguidelines by major organizations may further improvethe quality of care.

At our facilities, residents practice in the same locationas their attending physicians and have the sameaccess to ancillary support, such as clinical pharmacists,nurses, and health technicians. All practitionersreceive feedback from 

regarding their compliance with these guidelines, apractice that has been shown to improve the performanceof residents.

It is also possible that attendingphysician input improved the performanceof the residents, and the"resident" group in our study shouldactually be thought of as "residentswith attending supervision."Similarly, some degree of crossovermay have occurred: Because residentsare in primary care clinic only1 to 2 afternoons per week and mid-levelpractitioners and attendingphysicians often have full schedules,opportunities for intervention mighthave occurred during urgent visitsthat might have been handled byother provider types in the clinic.

MEDICAL EDUCATION & MANAGED CARE

Relationship Between Provider Type and the Attainment of Treatment Goals in Primary Care

To determine the effect on medical residents' perceptionsresulting from attending a medical policy meeting.

Pre- and postsurveys administered to residentsattending BlueCross BlueShield of Tennessee's (BCBST) MedicalPolicy Subcommittee meeting. This group is a subcommittee of theMedical Policy Review Committee (MPRC), which is responsiblefor the final approval of medical policies at BCBST.

Residents were invited to attend BCBST's MedicalPolicy Subcommittee meeting to determine what impact observingthis process would have on their attitudes toward health insurers'use of scientific evidence. This subcommittee, which consists ofboth BCBST medical directors and network physicians, plays a keyrole in determining whether new medical technologies or new usesof established technologies are medically necessary, cosmetic, notmedically necessary, or investigational. Residents were given apre- and postsurvey to assess their attitudes regarding the methodsinsurance companies use to make coverage decisions.

Medical residents rated BCBST's use of scientific evidencehigher after attending the subcommittee meeting. Only 4 ofthe 10 residents surveyed prior to the subcommittee meetingagreed with the statement, "Insurance companies base their coveragedecisions for medical treatment on scientific evidence." In thepostsurvey, all 10 residents indicated that they agreed with thestatement. Other questions showed similar positive changes in residents'opinions of the health plans' decision making.

Many medical residents likely have negative perceptionsof health insurers and may be prone to believe that medicalpolicy decisions are based predominantly on financial impact.Such attitudes can easily persist as they move into private practice.Although drawing firm conclusions from this small study is difficult,exposing residents to actual decision-making processes withininsurance companies may be one means of improving theiropinions of how such decisions are made.

The public often has a negative opinion of insurancecompanies, particularly of managed careorganizations.

Such negative opinions are alsoheld by some practicing physicians. These physiciansbelieve that insurance companies are concerned mainlywith finances and are not interested in scientific evidencewhen important coverage determinations mustbe made.

People working within the medical policy decisionprocess at health plans know that scientific evidence iscarefully considered when determining what servicesare reimbursable for members. Evidence-based medicine,defined as the "conscientious, explicit, and judicioususe of current best evidence in making decisionsabout the care of individual patients" is a tool used bymany insurance companies.

The BlueCross BlueShieldAssociation's Technology Evaluation Center, for example,uses the results of scientific studies to assess newand established technologies. Many health insuranceplans use such sources to assess medical technologies.BlueCross BlueShield of Tennessee (BCBST) and otherhealth plans often balance the assessments publishedby these organizations with practitioner experience.

Medical policies at BCBST are developed by a staff oftrained researchers and reviewed through a 2-step committeeapproval process. The first step involves a presentationof a drafted policy and supporting evidence toour Medical Technology Assessment Subcommittee(MTAS), the voting members of which are BCBST staffphysicians and community physicians. These physiciansconsider medical policies from a scientific andclinical standpoint. Two of the community physiciansare also affiliated with the University of TennesseeCollege of Medicine, Erlanger Unit. The second step ofthe process is consideration by the Medical PolicyReview Committee. This committee consists of bothadministrative and medical staff and is charged withfinal approval of medical policy proposals. This committeegenerally accepts the policy proposals approvedby MTAS.

The goal of the small study presented in this articlewas to determine whether allowing residents from theUniversity of Tennessee College of Medicine, ErlangerUnit, to attend MTAS meetings would positively impacttheir views of health insurance decision making.Influencing physicians early in their careers can positivelyimpact their long-term perceptions of insurancecompanies. Clearly, skepticism and negativity are barriers to a productive dialogue among the stakeholders inthe healthcare system. Time wasted dealing with falseimpressions could be better spent focusing on providingthe best care for members.

Encouraging services based on the best available scientificknowledge and discouraging services that areunlikely to benefit patients was 1 aim of the Institute ofMedicine's influential book, 

The report encouraged all of the constituentsinvolved in healthcare, including providers and payers,to work together to achieve this goal. The first step inachieving this goal is for health insurers and cliniciansto achieve a greater understanding of how each groupapproaches decision making about clinical care. Thisstudy offers one way in which such understanding canbe achieved.

Residents were invited to attend an MTAS meetingby 1 of the subcommittee members who was a facultymember at the University of Tennessee College ofMedicine, Erlanger Unit. The faculty member typicallybrought 1 resident with him per month; MTAS meetingsoccur on a bimonthly basis. During the course of 15months, 10 residents attended. Each meeting typicallylasted 60 to 90 minutes. Drafted medical policy addressingnew medical technologies and new uses of establishedtechnologies were reviewed and discussed ateach meeting. Examples of medical policies consideredinclude plasmapheresis/plasma exchange and genetictesting for colorectal cancer.

Residents were given the same brief questionnairebefore and after the meeting. The questionnaire consistedof questions in an agree/disagree/don't know format(

All 10 medical residents attending MTAS meetingscompleted the pre-and post-MTAS questionnaires.

lists the results of the questionnaire with thestrongly agree/agree and strongly disagree/disagreeanswers combined into 2 categories. The pre-MTASquestionnaire shows a marked negativity toward insurancecompany decision making with respect to newtechnologies. Only 4 of the 10 residents surveyedthought that insurance companies considered scientificevidence when making coverage decisions about medicaltreatment. Only 2 of the 10 responded that insurancecompanies valued clinical/practitioner expertisewhen making coverage determinations. Most thoughtthat insurance companies were primarily concernedwith cost containment in making coverage decisionsabout medical technologies, and only 4 thought thatinsurance companies considered patients' needs whenmaking determinations.

The post-MTAS questionnaire results were significantlydifferent. All 10 of the respondents agreed thatcoverage decisions were based on scientific evidence.Nine of the 10 responded that insurance companies valuedclinical expertise in making such decisions. Only 3responded that coverage decisions were based primarilyon cost, and 7 responded that coverage decisionswere based on patients' needs.

Managed care organizations and practicing physiciansboth strive to practice evidence-based medicine.Managed care organizations want members to receiveproven treatments that may improve health. Physiciansalso want to use treatments proven to be effective inimproving health. Unfortunately, differencesof opinion often arise between the2 groups as each becomes suspicious ofthe other. For physicians, this suspicioncan manifest itself as skepticism aboutthe intents of health insurers and theopinion that health insurers are concernedprimarily with containing costs.

Missing from this dialogue is theview that health insurers often serve asdefault stewards of health insuranceresources. As with any resource, healthinsurance is not limited, but ratherfinite. Health insurance plans need tobe able to price their product to appealto employers and individuals. Theyneed to minimize healthcare spendingon unproven and potentially harmfultechnologies, which drive up the cost ofcare, because spiraling costs for healthinsurance makes it more difficult for employers andindividuals to purchase healthcare coverage.

Evidence-based medicine is one tool that can beused to stem the tide of increased costs due tounproven or unnecessary care. Because evidencebasedmedicine is also an important part of the practiceof medicine, it is an area of common ground betweenhealth insurance plans and healthcare practitioners.

If physicians see health insurance plans basing theirdecisions on scientific evidence, then their views ofhealth plans might be less negative and skeptical. Thiscould foster an atmosphere of collaboration betweenthese 2 groups. One way to cultivate this partnership isto allow physicians to see how health plans make decisionsabout new medical technologies. Encouragingresidents to attend medical policy meetings at healthinsurance plans can promote an important alliancebetween plans and physicians.

Results of this admittedly small study have shownthat attendance of physicians at such meetings might bean important tool in improving their perceptions ofmanaged care. At best, the results of this study are preliminaryand further studies are needed to confirm thefindings. The number of survey respondents was notlarge enough to calculate statistical significance.Additionally, the fact that residents attended with a facultymember may have introduced bias into the resultsas far as influencing the willingness of some residents toattend the meeting and their responses to the survey.The administration of the survey by BCBST staff mayhave been a source of potential conflict of interest.Finally, there is a question of whether attending 1 meetingallowed the residents enough exposure to formulateinformed opinions. Future studies of this ongoing endeavorwill help address such issues.

We are hopeful that the results from this small studywill encourage similar efforts and studies by other plans.

From BlueCross BlueShield of Tennessee, Chattanooga, Tenn.

Address correspondence to Richard S. Mathis, PhD, Senior Manager, Medical PolicyResearch and Development, BlueCross BlueShield of Tennessee, 801 Pine Street, 2E,Chattanooga, TN 37402-2555. E-mail: richard_mathis@bcbst.com.

Who do you trust? Americans' perspectives on health care,1997-2001. 

Hamilton county physicians' experiencewith managed care. 

3. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. 

Evidencebasedmedicine: what it is and what it isn't [editorial]. 

September 2005

Optimizing Weekend Availability for Sophisticated Tests and Procedures in a Large Hospital

Successful Conversion of Patients With Hypercholesterolemia From a Brand Name to a Generic Cholesterol-Lowering Drug

Relationship Between Provider Type and the Attainment of Treatment Goals in Primary Care

Improving Medical Resident Perceptions of Health Plan Decision Making

Teaching Evidence-based Medicine in a Managed Care Setting: From Didactic Exercise to Pharmacopolicy Development Tool

Use of Administrative Data to Identify Health Plan Members With Unrecognized Bipolar Disorder: A Retrospective Cohort Study

Correspondence regarding, "A Systematic Review of Diabetes Disease Management Programs"

A Group Randomized Trial to Improve Safe Use of Nonsteroidal Anti-inflammatory Drugs