data basics - society for clinical data management (scdm) · electronic case report form (ecrf),...

DATA BASICSDATA BASICS To advanceexcellence in themanagement ofclinical data

A PUBLICATION SUPPORTED BY AND FOR THE MEMBERS OF THE SOCIETY FOR CLINICAL DATA MANAGEMENT, INC.

Letter from the Editors This Issue1

Letter from the Editors

1Data Managers’ Workflow with EDC –

The Real World, The Ideal World

3Letter from the Chair

5Database Quality Review: Starting With

the End in Mind

7Partnering with a CRO to Meet the

Challenges of the Changing Clinical DataManagement Function

9Diversity, Challenge, and Opportunity:The Role of the Clinical Data Manager

12Organizational Planning: Collaborationbetween the Clinical Data Management

Lead and the Programmer

15Ensuring Data Quality Through Process

Improvement Initiatives

17Integrating External Clinical Data into

Clinical Research Databases

Volume 11Number 42005 Winter

Continued on page 4

The SCDM Annual Fall Conference was a rousing success again this year, with 546participants and over 48 presentations and workshops. In this issue of Data Basics you’ll find

articles from several of the conference presenters dealing with the perennial topics of data quality,database integration and collaborative partnerships in data management. If you are transitioning into anEDC system, you’ll want to check out the article Data Manager’s Workflow with EDC – The Real World,The Ideal World starting below.

If you presented at the conference but haven’t yet submitted an article, don’t think you’re off the hook!We’re still eagerly anticipating your articles and will be including them in future issues of Data Basics asthe year progresses. Whether you’ve presented at the conference or not, we invite you to share yourexpertise and challenges in a Data Basics article. Upcoming themes:

Spring 2006 Traditional EDCSummer 2006 Data Privacy/EU DirectiveFall 2006 Non Traditional EDC/Data CaptureWinter 2006 Vendor Interaction/Topics

On behalf of myself and the Publications Committee, I’d also like to congratulate Lynda Hunter, DataBasics Co-Editor, on her acceptance of the Chair position for the committee. Many thanks also to KitHoward, outgoing Chair, for her unflagging support of the Publications Committee and SCDM as awhole. Kit will continue to provide editorial guidance in her role as Senior Editor of Data Basics.

Wishing you all a safe and happy new year,Your Data Basics Co-Editors,Chandra Wooten and Lynda Hunter

Data Managers’ Workflow with EDC – The RealWorld, The Ideal WorldAnne Zielinski, Director, ICTI

A fundamental concept driving Electronic DataCapture (EDC) adoption is the reduction in theeffort needed to conduct a clinical trial. Butexactly whose job gets easier? With the wide rangeof roles that EDC impacts, it’s not always obviouswho will benefit. Clearly, what’s important toSCDM members is the Data Manager’s role, sothat’s the focus of this article.

The Real WorldLooking at several real world examples of howEDC gets used by trial sponsors and ContractResearch Organization (CRO), I found there were

some areas of commonality and several areas wheredifferences were the norm.

In all EDC trials, data management is involved indetermining specifications. Data elements,Electronic Case Report Form (eCRF), visitstructure, edit rules, data imports, and dataexports must all be defined up front. The goodnews is that if the sponsor and/or CRO has, andadheres to, data and eCRF standards, there is agood deal of reusability from one trial to the next.The bad news is that, particularly for the first trial

PUBLICATION OF THE SCDM To advance excellence in the management of clinical data WINTER 20052

DATA BASICSDATA BASICS Submission Requirements

EDITORIAL

BOARD(also known as PublicationsCommittee)

Lynda L. Hunter, Chair &Liaison to BoardPRA InternationalPhone: (913) 577-2972Email:[email protected]

Chandra Wooten, Co-editorCorus Pharma, Inc.Phone: (206) 832-2016Email:[email protected]

Kim Tiefenbach, Co-editorOmniComm Systems, Inc.Phone: (610) 326-3249Email:[email protected]

Rehana BluntAmgenPhone: (805) 447-5729Email: [email protected]

Felicia Ford-RiceAcambis, Inc.Phone: (781) 302-3007Email:[email protected]

Katherine HowardKestrel ConsultantsPhone: (734) 576-3031Email:[email protected]

Crystal Lewis, CCDMKyowa Pharmaceutical, Inc.Phone: (609) 424-0869Email: [email protected]

Pilar Moore, CCDMKyowa Pharmaceutical, Inc.Phone: (609) 580-7394Email: [email protected]

Virginia Nido, CCDMGenentech, Inc.Phone: (650) 225-3805Email: [email protected]

Mary OlshewskiKelly ServicesPhone: (248) 244-4772Email:[email protected]

SUBMISSION DEADLINES (Articles and Advertising Art Work)

Our quarterly publication schedule for the next four issues requires the following input deadlines:

Volume 12, #1 (Spring) 9 January 2006

Volume 12, #2 (Summer) 10 April 2006

Volume 12, #3 (Fall) 10 July 2006

Volume 12, #4 (Winter) 9 October 2006

PUBLICATION POLICYWe welcome submission of materials for publication in Data Basics. Materials should preferably be submitted in electronic form (MSWord). Acceptance of materials for publication will be at the sole discretion of the Editorial Board. The decision will be based primarilyupon professional merit and suitability (i.e. publication may be edited at the discretion of the Editorial Board.)

Neither SCDM nor the Data Basics Editorial Board endorses any commercial vendors or systems mentioned or discussed in anymaterials published in Data Basics.

ADVERTISING POLICYAD RATES** x1 x2 x3 x4

FULL Page $675 each $640 each($1280) $610 each ($1830) $575 each($2300)

HALF Page $470 each $445 each($890) $425 each($1275) $400 each($1600)

QTR Page $285 each $270 each($540) $255 each($765) $240 each($960)

**Ads are net, non-commissionable.

Advertisers purchasing multiple ad packages will have the option of placing those ads anytime within the 12-monthperiod following receipt of payment by SCDM.

Quarter Page = (3 5/8 inches x 4 7/8 inches) Half Page-vertical = (3 5/8 inches x 10 inches)

Half Page-horizontal = (7 ½ inches x 4 7/8 inches) Full page = (7 ½ inches X 10 inches)

MECHANICAL REQUIREMENTS: Do not send logo/photos/images from word processing software, presentation software,or Web sites. Files should be saved in the native application/file format in which they were created. Photos/images should be highresolution and received in the file size you wish to have it printed.

600 dpi or higher for black and white, 300 dpi or higher for color/grayscale

Acceptable file formats include AI, EPS, and high resolution PDF, PSD, JPEG, and/or TIFF

PAYMENT: Payment must be received with advertising. Space reservations cannot be made by telephone. There is NO agencydiscount. All ads must be paid in full.

CANCELLATIONS: Cancellations or changes in advertising requests by the advertiser or its agency 5 days or later after thesubmission deadline will not be accepted.

GENERAL INFORMATION: All ads must be pre-paid. Publisher is not liable for advertisement printed from faulty admaterials. Advertiser agrees to hold SCDM harmless from any and all claims or suits arising out of publication on any of his/heradvertising. SCDM assumes no liability, including but not limited to compensatory or consequential damages, for any errors oromissions in connection with any ad. The SCDM does not guarantee placement in specific locations or in a given issue. SCDM reservesthe right to refuse or pull ads for space or content.

Please submit all forms, artwork, and payments to:Society for Clinical Data Management Phone: 1 (414) 226-0362555 E. Wells St. Fax: 1 (414) 276-3349Suite 1100 E-mail: [email protected], WI 53202 Web site: www. scdm.org

WINTER 2005 PUBLICATION OF THE SCDM To advance excellence in the management of clinical data 3

2006 SCDMCommittees

The following are currentlyactive Committees within theSociety for Clinical DataManagement, Inc.

Certification CommitteeChair: Colleen Cox, CCDMPhone: (617) 868-2020E-mail: [email protected]

Education CommitteeChair: Mary Foy, CCDMPhone: (201) 427-8094E-mail: [email protected]

Marketing CommitteeChair: Audra McRaePhone: (919) 465-6080E-mail:[email protected]

GCDMP CommitteeChair: Carol GarveyPhone: (650) 225-6317E-mail: [email protected]

Membership CommitteeChair: Deb ColePhone: (513) 579-9911E-mail: [email protected]

Publications CommitteeChair: Lynda HunterPhone: (913) 577-2972E-mail:[email protected]

Web Site CommitteeChair: David BorbasE-mail:[email protected]

Web Sites to Check OutACDM - www.acdm.org.ukCDISC - www.cdisc.orgFDA - www.fda.govICH - www.ich.org

There are more links to befound on our web site!SCDM - www.scdm.org

Please let the Web SiteCommittee know about anyother “hot” web sites that youfeel would be of interest to theSCDM membership.

Letter from the ChairBy: Sharon Miller, Business Consultant E-data Management Program, Eli Lilly and Company

Dear Colleagues:

It’s hard to believe that yet another year hasflown by. As my term as Chair of the Board ofTrustees for SCDM comes to an end, I wouldlike to take this opportunity to thank all of youfor making this yet another outstanding year forthe Society. Through your efforts, SCDM hasagain been able to amass an impressive list ofaccomplishments in 2005. Some of the manyaccomplishments this year included:

· Refreshing the long-term strategic plan for theSociety

· Introducing revised Mission, Vision, andValues statements

· Entering into a collaborative agreement withDuke Clinical Research Institute to provideeducational offerings to support candidates inpreparation for CDM certification

· Launching a series of educational webinarsrelated to the Good Clinical Data Manage-ment Practices (GCDMP) document

· Initiating a process to review and providecollective input into regulatory guidance andissues of interest to members of the CDMprofession

· Certifying 16 new Clinical Data Managers· Implementing multiple member option for

discounted & flexible group memberships· Launching version 4.0 of GCDMPs· Launching a new and improved version of

Data Connections electronic newsletter· Continuing to expand and evolve Data

Basics to include more technical andsubstantive articles

· Giving the SCDM website a new look

It is only through you, the members, thatSCDM can reach its goals and continue toexpand the opportunities & resources availableto support all who are engaged in the manage-ment of clinical research data. The intellectualcapital and collective experience of our memberstruly are our greatest assets. We rely on andembrace the participation and contribution of ourmembers and volunteers.

I look forward to continuing to work with all ofyou to make 2006 another stellar year forSCDM.

All the Best,

Members of the 2005 SCDM Board of Trustees (left to right) Lisa Freeman, Charlene Dark, JonathanAndrus, Jill Vath, Lynda Hunter, Sharon Miller, Jane Hiatt and Anthony Costello. (Missing from photo:Audra Mc Rae, Armelde Pitre, Judy Pyke and Mariann Plaunt.)


2005 SCDM BoardSharon A. Miller, ChairEli Lilly and [email protected]

Jill Vath, Vice-Chair/TrusteeGenentech, [email protected]

Judy Pyke, SecretaryKendle International [email protected]

Audra McRae, TreasurerCharles River Laboratories,Clinical [email protected]

Jonathan Andrus, TrusteePhoenix Data [email protected]

Anthony Costello, TrusteeNextrials, [email protected]

Charlene Dark, TrusteeOmnicare Clinical [email protected]

Lisa Freeman, TrusteeCorus Pharma, [email protected]

Jane Hiatt, TrusteeSchering Plough [email protected]

Lynda Hunter, TrusteePRA [email protected]

Armelde Pitre, TrusteePfizer, [email protected]

Marianne R. Plaunt, PastChairI3Statprobe, [email protected]

2006 SCDM ElectedTrusteesColleen Cox, CCDM, TrusteePROMETRIKA, [email protected]

Mary Foy, CCDM, TrusteeForest Research [email protected]

Derek Perrin, TrusteePfizer Global Research [email protected]

Data Managers’ Workflow with EDC – The Real World, The Ideal WorldContinued from cover

with an EDC system, defining these parameterscan take a lot of a Data Manager’s time.

Once a trial is running, some differences in datamanagement’s role become apparent.

Issuing Data Clarification QueriesClinical trial monitors always issue data clarifica-tion queries, as do data managers. However,sometimes data management issues queriesthrough a Clinical Data Management (CDM),system while monitors use the EDC system. Thereis questionable value in issuing queries from bothEDC and a CDM system. First, it is cleaner tohave a complete audit trail of queries in one place,and only one place. Secondly, the only reason touse a CDM system is if the EDC system cannotissue batch edit checks, or if the CDM system isbeing used as a data warehouse.

Locking Data (Revoking Data EntryRights)At some sponsors and CROs, data managementalways locks data. In others, if or when a monitordetermines that data is clean, he or she can lockthe data.

The Timing of Data Management’sReviewHow quickly data is reviewed by data manage-ment varies, largely based on how quickly data isentered. The best way to make certain that data isentered in a timely manner is to tie investigatorsite payments to data being entered in the EDCsystem. That way, the sponsor’s or CRO’s view isthat if the data isn’t entered, the visit didn’thappen. Even without the ability to tie paymentsto data entry, some EDC systems provide reportsthat show the lag between visit date and data entrydate, a handy way to keep track of sites’ timeliness.

The Ideal WorldSome trial sponsors have eliminated the questions ofwho issues queries and who locks data by meldingthe role of data manager and clinical monitor, anintelligent outcome of EDC’s ability to minimizethe mindless work that results from using paperCRFs. Individuals in these roles focus on issues thatrequire thought and expertise, rather than spendingtime doing largely clerical data clean-up. Even in anenvironment where this melding of roles cannot bedone, it points out one of the easiest opportunitiesto streamline the clinical review process: empowerthe first person who determines that an eCRF iscomplete and clean to lock it. Since EDC allowsincremental locking of data, eCRFs can be locked

as soon as they are determined to be clean, so that atthe end of a trial the only eCRFs that need to belocked and reviewed are the eCRFs for the lastpatient’s last visit.

Another area in which EDC can optimizeworkflow and that affects the timing of datareview revolves around the concept of eSource. Toestablish the fundamentals, Good Clinical Practice(GCP) define source documentation as the firsttime an observation is recorded to durable media,and require that the investigator must retaincontrol of the source data. In most EDC trials,observations are recorded initially on paper – amedical record, a blank paper copy of the eCRF,or some sort of a worksheet, typically devised bythe investigator site based on their workflow.

EDC systems that provide an on-line/off-lineoption, in which data is stored on the clientcomputer, allow users to have eSource, in whichthe first time an observation is recorded to durablemedia is when it is typed into a computer and thedata is stored locally under the investigator’scontrol. eSource eliminates the need for sourcedata verification, since data entered into EDC isthe source. But how does this make the DataManager’s life easier? First, the data is cleaner bydefinition, because transcription is reduced. Recallthat the sole purpose of double data entry (keyverification) in the world of paper CRFs is toensure that data is transcribed correctly! Thesecond major difference is that use of eSourceeliminates waiting for source document verifica-tion to be completed. The sooner data is available,the sooner it can be reviewed and locked. And weall know about that clichéd million dollars a daythat our efficiencies can save!

Getting From Real to IdealBack in the 90’s when I was first involved inspecifying EDC system capabilities, we realizedthat EDC needed to accommodate a variety ofworkflows, some that were logical and efficientand some that, frankly, were not. While the goalshould always be to use a system as efficiently aspossible, it’s unrealistic to expect to get to the idealworld in the first step. Rather, with the end inmind, it is reasonable and practical to start with aworkflow that’s good enough, and then take stepsto get to ideal.

Once final piece of advice: if you’re choosing anEDC system, make certain that both your startingworkflow and your ideal workflow are easilyaccommodated by the system.


The pharmaceutical industry has experienced increasing scrutiny ofclinical trial data utilized to support claims of product safety andefficacy. Database quality reviews for clinical trial reporting data-bases are an important component of the clinical trial process toinsure that the database meets rigorous quality standards and willprovide data that are accurate and reliable. It is implicit that anaccurate reporting database is the basis of obtaining meaningfulconclusions from a clinical study.

At Lilly, a Database Quality Review (DBQR) is a comparison of casereport form data and ancillary data (if applicable) to the correspondingdata elements in the electronic (clinical) database. The purpose of theDBQR is to estimate the accuracy of data processing and ensure thatobserved data match the raw data in an electronic database. The keyattributes of the database quality review process are the sampling plan,the error rate, and the error type. The sampling plan is a joint effortbetween the clinical data management coordinator (CDMC) and thestatistician. The CDMC provides the statistician with an estimatednumber of data points and the statistician creates a randomly selectedsubject list for the sampling plan. At Lilly, the parameters of thesampling plan are a minimum of 5 subjects; 10% of total data pointsup to a maximum of approximately 10,000 data points. The error rateis defined as the total number of errors divided by the total numberdata points reviewed. A successful DBQR will have an error rate lessthan or equal to 0.1%. There are three error types: data entry, datavalidation (i.e., data inconsistent with provided documentation) andsystemic (i.e., programming errors or recurring validation errors).Historically, the data modules that represent the greatest percentage oferrors are adverse events, diaries, scales and other (study specificmodules) (Fig 1).

Numerous issues have been identified throughout the pharmaceuti-cal industry with database quality reviews. DBQR processes differacross companies; for example trial selection, sampling plandetermination, error definition, acceptable error rate and criticalvariable reviews all differ throughout the industry. There is a generallack of guidance within the industry relative to acceptable qualityspecifications or processes. DBQRs typically consume significantresources and time. The end-of-trial review is focused on erroridentification, but offers no opportunity to prevent errors or takecorrective actions on quality issues during the trial. A failed DBQRat end-of-trial adversely impacts the database lock timeline.

An initiative has recently been undertaken at Eli Lilly to investigateopportunities to improve the DBQR process. The design and scopeof the initiative were to explore the current DBQR process, whichmeasures the quality of data processing and verifies that observeddata matches the data output, and to discover a means by which toinsure the overall quality of the data.

A lean Six-sigma approach was employed to investigate the DBQRprocess. [Editor’s note: Six-sigma is a specific methodology for eliminat-ing defects in a process.] Initially, a process map that reflects thecurrent process was constructed (Fig 2). This map was used todevelop a Failure Modes and Effects Analysis (FMEA). FMEA is adisciplined approach used to identify possible failures of a processand then determine the frequency and impact of the failure. It is alsoused to rank and prioritize the possible causes of failures as well asdevelop and implement preventative actions. The results of theFMEA pinpointed two steps of the DBQR process for furtherinvestigation and potential improvement.

The first step, waiting for 90% of the data to come in house and bedeclared clean, was identified as too late in the course of the trial to

Fig. 1 Historical Metrics

2000 2001 2002 2003

Datalocks 24 124 105 110

DBQRs Done 16 52 46 33

% DBQRs 67% 42% 44% 30%

% Miss. Data 53% 20% 0% 18%

MeanError Rates .055% .028% .048% .060%

Fig. 2 DBQR Process Map

Database Quality Review: Starting With the End in MindPeter H. Zervos, Teamleader, Global Clinical Data Management, Diabetes Product Group, Eli Lilly and Co.

Continued on page 8


Partnering with a CRO to Meet the Challenges of the Changing Clinical DataManagement FunctionTracy Mayer, MBA, BS, Senior Clinical Data Manager, Kendle

Research conducted at Tufts University in 2003 found that the costof research and development efforts to bring a new drug to marketaverages an estimated US$802 million.1 The average time associatedwith bringing a drug from lab to medicine chest is approximately10-12 years with roughly six years spent in Phase I-III clinical trials.2

Therefore, based on this data, each additional month in drugdevelopment will cost a biopharmaceutical company US$5.6million.1,2

Given the obvious cost associated with the clinical developmentlifecycle, biopharmaceutical companies are looking increasingly fordata management outsourcing partners that can decrease the timespent in clinical trials and increase their return on investment morequickly. In increasing numbers, biopharmaceutical companies areturning to CROs to minimize the length of time spent in clinicaldevelopment, to provide flexible and strategic trial data managementsolutions, and to provide an experienced team to meet the challengesof the changing clinical data management function. In 2004, 24percent of companies outsourced drug development activitiescompared to just 4 percent in the 1990s.3 It is clear that outsourcinghas become a growing trend in the clinical research industry and inthe data management area.

When considering the outsourcing of data management to a CRO,biopharmaceutical companies generally select one of two outsourcingmodels – the project model or the functional model. In the projectmodel, a single trial is outsourced, whereas with the functional model,an entire functional operation (e.g., data management) is outsourced.Each model has its advantages as well as its challenges, so carefulconsideration should be taken when choosing which model toimplement. Regardless of the model, following a few best practices willhelp companies work more efficiently with their CROs.

Choosing the Right ModelWhen choosing to outsource data management to a CRO, sponsorsmust first decide on the model. Project model outsourcing is bestfor single projects where there is a temporary lack of resourcecapacity at the biopharmaceutical company. Functional modeloutsourcing is better for companies that want to invest in a longer-term strategy, lack the operational function all together, or havedecided that data management is not one of their internal corecompetencies. Efficiencies and returns are generally greater on workoutsourced on a functional basis.

The Project ModelWhen a company has short-term data management needs, theproject model may be an attractive option. It can provide a highlyexperienced team composed of top professionals on a temporarybasis. A biopharmaceutical company gains the expertise of theseprofessionals without the long-term expense of keeping them onstaff.

There are challenges associated with the project outsourcing model,but they may be overcome with a little advanced planning. In somecases there may be a perceived loss of project ownership at the

biopharmaceutical company. As a project is outsourced to an off-sitelocation, it may be difficult for the sponsor’s team members to feelconnected with the work performed by the CRO. However,designating a sponsor liaison who provides project oversight andworks directly with the CRO team is an effective way forbiopharmaceutical companies to maintain close contact and thefeeling of ownership in any given project. To facilitate ownership onthe part of the liaison, he/she should be held accountable for thesame deliverables and metrics as the CRO team, as well as share inthe rewards of a successful project.

Another common challenge is that training a small pool of resourcesat the start of a project can be inefficient, especially if there is a needfor more resources over time. By employing the “train-the-trainer”concept, much of this inefficiency can be eliminated. This involvestraining one or two personnel on the CRO team who in turn canprovide project training on an ongoing basis whenever required atthe CRO.

Often more than one project from a biopharmaceutical company isoutsourced to the same CRO vendor, and if processes and proce-dures are not communicated across the different project teams, theresult may be a lack of consistency in how methods are applied. Thiscan result in the need to re-do work that has already been completedsuch as the development of edit libraries or case report forms(CRFs). The liaison can help here too, ensuring consistency acrossmultiple trials.

Another challenge associated with the project model is the ability tomanage costs. Even in the short term, tracking metrics weekly toidentify potential efficiencies and problem areas can alleviate someof the cost associated with short-term project outsourcing. Forexample, tracking the frequency of queries per CRF page allows theteam to identify CRF completion issues that can be communicatedto the sites, thus decreasing the volume of queries to issue andresolve.

A final consideration for lowering the cost of outsourcing is toestablish a preferred provider agreement. This decreases the newbusiness development costs associated with individual contracts. Ofcourse, the best way to lower individual trial outsourcing expenses isto consider employing the functional model for a full range of trials.

The Functional ModelFunctional model outsourcing is preferable in many ways to theproject model. The functional model is often able to provide costsavings through shared efficiencies and scalability that a projectmodel cannot. Under this arrangement there typically is a largegroup of people dedicated to serving multiple trials for an individualbiopharmaceutical customer, which ensures consistency across alltrials. This does, however, present the challenge of training a largegroup consistently over time. In this case, implementing web-basedor CD-ROM training or the “train-the-trainer” approach cansuccessfully alleviate this issue. Another challenge is the lack of

Continued on page 8


project-specific knowledge, as team members are trained on thesystem rather than individual project protocol. This can be solved byappointing project leaders at the CRO to provide project-specificoversight and to communicate any necessary training information tothe team.

A final challenge of this model is the need to customize proceduresfor specific projects. Most processes and procedures are standardizedacross all projects to allow for scalability and efficiencies. However,by appointing project leaders who are trained in the overall modelyet dedicated to the individual trials, a CRO can offer abiopharmaceutical company the customization they require whileadhering to a system that still provides efficiencies and cost savings.

Choosing the Right CROOnce the model is chosen, the biopharmaceutical company mustchoose the right CRO to implement the plan. The technology of theCRO and the compatibility with the needs of the biopharmaceuticalcompany should be considered in the selection process. Ideally, itshould allow for user connectivity and parallel workflow to allow formaximum performance and further increase efficiencies. The CROshould have a system architecture which enhances the processthrough reliability, flexibility and scalability.

Global capability is also an important factor in selecting a CRO.Ideally a CRO will have locations in the same regions as the studysites as well as the biopharmaceutical company’s locations to allowfor communication in real time. Another point to consider is thesponsor’s historical experience with the CRO. Knowledge gainedthrough historical experience can be applied to current and futureprojects to further gain efficiencies.

A final point to consider is the CRO management team. Themanagement team at the CRO should possess the technical andbusiness acumen to bridge the gap between the two and fullyunderstand the needs of the biopharmaceutical customer. CROmanagement must exhibit their commitment to the endeavor andthe success of the partnership.

ConclusionImplementing a successful outsourcing model results in manybenefits to the biopharmaceutical company including decreasedcycle times for database set-up and cleaning, customized applica-tions tailored to its needs, a large group of experienced professionalsfocused on a targeted data cleaning strategy, and a cost reductionover time associated with efficiencies and scalability. Developing apartnership enables both the biopharmaceutical company and theCRO to evaluate resourcing and project needs on an ongoing basis,resulting in a more cost and time effective outcome.

Tracy Mayer is a Senior Clinical Data Manager for Kendle, a globalCRO providing Clinical Development, Regulatory Affairs, Biometricsand Late Phase solutions for the biopharmaceutical industry. In thisrole, Ms. Mayer is responsible for managing all aspects of the clinicaldata management process, including data cleaning; building, testingand validating of clinical databases; project team training; budgeting;and reporting.

References1. DiMasi, JA, Hansen, RW, Grabowski, HG. 2003. The price of innovation: new

estimates of drug development costs, Journal of Health Economics.2. www.pharmiweb.com3. www.mindbranch.com

Partnering with a CRO to Meet the Challenges of the Changing Clinical Data Management FunctionContinued from page 7

perform DBQR and may interfere with the critical path database locktimeline. The potential effect of failure is that finding errors in thedata at the end of the trial may render the data useless and unchange-able if a site has closed, resulting in compromise to the timeline fordatabase lock and possibly delay in submission of the data. There is noopportunity to assess the cause of errors and implement correctiveactions during the course of the study. The proposed change to thecurrent process that is being evaluated is the use of a progressiveDBQR, setting reviews at specific milestones, whether a number ofpatients or percent of data or a given time frame. The specificmilestones would be recommended by study the team.

The second step to explore was the methodology used in counting of dataissues or database errors. Placing equal value on modules that arecompleted as well as modules with the ‘none’ box ticked or inclusion ofhard-coded data fields contributes to misleading database field counts.An error in Comments, perhaps due to a misspelling, which could havelittle or no significance to the data, may be treated equally as compared toan error in a critical data field. The potential effects of failure are a

positive DBQR for a database containing significant errors or a failedDBQR due to repetitive errors in spelling found in Comments, whichhave no significance. Potential causes of failure are: one data entry errorin the database shows up in more than one data field; validation rules orsteps are not followed; no verification of data in the collection database;and data appears to have been changed without appropriate correspond-ing documentation found in the collection database. The proposedchange to the current process that is being assessed is that not all data hasthe same significance or importance in a study and that it is rational tomatch up a module’s importance in relation to a protocol objective todetermine a weight value for error rate calculation.

The two steps identified in the FEMA of the DBQR process arecurrently being investigated by two subgroups to the DBQR initiativeat Lilly. Recommendations for improvements in these two steps of theDBQR process will be forthcoming in the coming months.

Peter Zervos has been with Eli Lilly for 17 years. His pharmaceuticalindustry career spans discovery research and clinical development. He isa certified Six Sigma Black Belt.

Database Quality Review: Starting With the End in MindContinued from page 5


Continued on page 11

Diversity, Challenge, and Opportunity: The Role of the Clinical Data ManagerNina M. Trocky, RN, MSN, CNA, Project Director, Harris Technical Services Corporation

Clinical data management is a process; a set ofsequential and parallel activities; a coordinatedand regulated series of actions that togethersupport the conduct of clinical research trials.If the same question is posed to five differentresearch professionals asking them to explainwhat a Clinical Data Manager does, you willprobably get five different responses. Why isthis? The replies are different because there isno one answer. The role of the Clinical DataManager comprises many different activities soeach person’s view may be very different.

Clinical Data Manager is a title or a jobclassification. The job title and associatedposition description are specific to a particularorganization or work place. Therefore, it canbe a challenge for an organization to train theincumbent to meet the unique needs of thedata management activities. On-the-job training and focused skillsdevelopment might vary since prior work experiences would likely bedifferent. But are the required skills consistent across organizations orworkplaces? Table 1 shows a sample of key responsibilities detailed inrecent classified ads.

These key responsibilities were taken from eight different classified ads,all advertising a position titled Clinical Data Manager. The variabilityreflects the unique activities of the research site or workplace. No twoclassified ads were the same. The inconsistency among the variousclassified ads is expected because the incumbents in the Clinical DataManager position are carrying out specific actions or are accomplishingdistinct tasks unique to that particular work place. The Clinical DataManager role is one of many roles within a variety of staff, teamedtogether to perform all aspects of clinical research.

Variability within the classified ads is noted regarding required educationand past experience. For example, the Bachelor’s degree seemed to be theminimal level of formal education, but the areas of focus seemed to vary.Table 2 describes requirements regarding education and experience in theeight classified ads.

Table 1

Sample of Key Clinical Data Manager Responsibilities in Recent Classified Ads

1. Perform medical records abstraction and/or data abstraction

2. Perform data entry and data reporting per institutional procedures

3. Development and design case report forms

4. Development of the Data Management Plan

5. Execute data validation routines and initiate queries on potential discrepant data

6. Assist in reconciling clinical database adverse events with serious adverse eventreporting systems according to project-specific guidelines

7. Perform databases and transfer file quality control procedures

8. Format and organize data to support clinical review and statistical analysis

DiscussionUnlike a degreed position, such as nursing, in which the corecompetencies and training curriculum are standard andaccredited by a collegiate or regulatory body, the Clinical DataManager role is defined by workplace requirements. Clinical

Data Managers, as noted, come from variousbackgrounds, some with skills complemen-tary to the specific role. One interestingobservation was that none of the classifiedads stated that professional certificationswere either desirable or required (e.g.,Certified Clinical Data Management fromthe Society of Clinical Data Management,Certified Clinical Research Professional fromSociety of Clinical Professionals, CertifiedClinical Research Associate or CertifiedResearch Coordinator from Association ofClinical Research Professionals). Certifica-tion is adjunct to a degreed program and isused to support a specialized level ofknowledge within the various roles of clinical

research. This diversity in practice, expertise, and workplacespecialization presents those of us in clinical research manage-ment a challenge on several fronts, none of which is withouttremendous benefit and opportunity. It will be interesting to seeif the requirements regarding education and past experiencebegin to add specialized certifications as a desirable or evenmandatory requirement. The following are some specificrecommendations to develop Clinical Data Managers within anew workplace.

RecommendationsClearly define the staffing mix and their associated roles andresponsibilities based on the type of operations or the business

Table 2

Sample of Key Clinical Data Manager Experiences in Recent Classified Ads

• Associates degree (AA) preferably in Life Sciences and 1 year of related experience

• Bachelor’s of Arts/Bachelor’s of Science in Biology or Health Science with at least 1year experience in clinical database management

• Bachelor’s degree or equivalent in Natural Science or Computer Science and 3 plusyears of experience in data management

• Bachelor’s in Biology, Chemistry, Pharmacy, Nursing and 2-3 years industryexperience

• Bachelor’s degree preferably in a Life Science and at least 2 years of clinical datamanagement experience in a pharmaceutical company or CRO

• High school diploma, Bachelor’s degree preferred with 2-3 years relevant experience

• Appropriate experience to successfully perform the essential job-specificresponsibilities


model of the organization. This will alreadybe set when working in a pharmaceuticalcompany or contract research organizationbut maybe missing in a new facility such as aphysician’s community practice.

Develop and establish a set of core competen-cies for the Clinical Data Manager role at theresearch site. Next, develop a competency-based training program within the work placethat meets the specific and unique demandsof each position performing key activities tosupport the conduct of clinical research. Theorganization needs to develop an orientationand training program that clearly defines the core competencies andthe associated skills that must be accurately demonstrated.

Training aspects, such as minimal standards, best practices, andspecific strategies are discussed in the Good Clinical Data Manage-ment Practices, Version 3. This document also includes core topicsand details needed to ensure regulatory compliance and a role-specific training curriculum. These elements may ensure individualperformance and proficiency at a minimal level, which may be

further improved by a role–specific, competency-based trainingprogram.

The choice to participate actively in professional development restswith both the Clinical Data Manager and the management team atthe work place. There are a significant number of training programsand resources specifically directed toward the various activitiesconducted throughout the clinical research process. For example,numerous professional organizations offer computer-based training,newsletters, and local chapter memberships. Various organizationsoffer focused competency-based research and data managementfocused certifications. Several universities offer Master’s Degreeprograms. Table 3 lists a sample of universities offering accrediteddegree programs. Each program offers a broad yet focused curricu-lum that will support an individual’s advancement within the fieldof clinical research. Additionally, certifications and degrees willbetter prepare the CDM for progressively challenging positionswithin pharmaceutical and related industries.

Finally, Clinical Data Managers should not narrowly define theirpresent roles. They should maintain a constant vigilance, observingand participating in the technological and regulatory advanceswithin the entire field of clinical research. The acquisition ofknowledge, skills and competencies will enhance the incumbent’spursuits in any number of career paths whose roots are all intercon-nected in clinical data management processes and activities.

Nina M. Trocky manages the clinical data management contract for theNational Cancer Institute Center for Cancer Research (NCI/CCR) inBethesda, Maryland. In that position, she is responsible for the supportof approximately 250+ Phase I, II, and III oncology and AIDSintramural clinical trials.

ReferencesGood Clinical Data Management Practices, Version 3, Society for Clinical Data

Management, September 2003, page 103-111.Sullivan R. 1995. The Competency Based Approach to Training. JHPIEGO Strategy

paper. ReproLine The Reading Room. (1995) http://www.reproline.jhu.edu/english/6read/6training/cbt/cbt.htm#CBT.

Table 3

University Program Degree

University of Maryland Clinical Research Management Master’sBaltimore, MD

Boston University Clinical Investigation Master’sBoston, MA

George Washington University Clinical Research Administration Bachelor’sWashington, DC

Northwestern University Clinical Investigation Master’sChicago, IL

Diversity, Challenge, and Opportunity: The Role of the Clinical Data ManagerContinued from page 9


Organizational Planning: Collaboration between the Clinical DataManagement Lead and the ProgrammerJodi L Carey RN BSN, Principal Team Leader, GlaxoSmithKline; Co-Author, Robyn L Eichenbaum, Sr. Programmer Analyst, GlaxoSmithKline

A frequently debated discussion among our industry is: what experi-ence and skill sets are required to be an effective data manager? Whatattributes facilitate optimal data quality and appropriate resourcemanagement? One philosophy is to start with a strong clinicalbackground and train the technical competencies. This positionimplies that understanding the medical science and focusing on thepatient results in enhanced data quality. The opposing view is to startwith a strong technical background and teach the clinical aspects ofData Management. This position boasts that understanding underly-ing data structures and system functionality results in enhanced dataquality. Our solution is a balance of both philosophies and collabora-tion between clinical and technical staff.

Based on our experience, the roles of the past tend to be an exercise inworking in isolation with a series of handoffs. Historically, the datamanager’s main function was that of a data cleaner, which includedreview and validation of data, reconciliation of coding dictionaries,and safety and lab data clean up; much of this was done manually.Our past roles also included minimal input into study start-upactivities, which include protocol review, database and case reportform (CRF) development. This resulted in less direct involvementwith the Study Team and Programmers. There was a separate role forCRF design and another group performed the transfer of external datasuch as labs and ECGs. The programmer transferred data for analysisafter confirmation from the data manager that the database was fit forrelease. The programmer was not aware of or routinely involved in anyof the data management activities between database setup and release.Edit check development consisted of creating non-technical versionsof data validation specifications and handing them off to a program-mer, usually in another department, thus creating two distinct workefforts. For the programmer, the main function was to program datavalidation checks, which comprised 80% of their workload.

So, why change our roles or work habits? There is rapid movementtoward electronic clinical (eClinical) trials. Our company has set anaggressive goal to have 90% of all new and ongoing trials in usingelectronic data capture (EDC) by the end of 2006. Other reasons forchange include reduction in data collection cycle times and thedevelopment and management of standards. Clinical trials aregrowing larger and more complex, often combining novel conceptsand Phase II and III approaches. There is a definite need for quality,real time, in-stream data to facilitate go/no-go decisions, safety boardreviews and independent reviews. There are multiple data sourcesoutside the CRF and electronic CRF (eCRF), many which captureprimary endpoints. Data aggregation, data mining and the ability tomix and match software and systems also impact our need to reevalu-ate our roles. It is crucial to work with the end in mind and under-stand the deliverable and its impact beginning with study set-up.

So, where are we now and what do we need for the future? Datamanagers need to enhance their technical aptitude and understand-ing of programming and database design language. It is importantto be an active participant in creating seamless data transfers from

eCRF/CRF development through reporting dataset delivery. Datamanagers also need to continue to play an active role in the editcheck development process by utilizing standard edit check libraries,creating detailed specifications and keeping an open dialogue withtheir programming colleagues.

For the programmer, current and future needs continue to includethe ability to learn new tools and systems effectively and efficiently.They also need to enhance clinical aptitude by increasing study teamparticipation and accountability and gaining a greater understand-ing of the protocol, the disease under study and the drug develop-ment process. Combining these skills with the understanding ofprimary and secondary clinical trial endpoints and the rationale fordelivering data fit for analysis is also valuable.

We put our philosophy into practice by merging the roles of the datamanager and programmer. Together, we shared the project vision andserved as active participants in all data management activities fromstudy set up through database freeze. The two roles worked jointly inteam and study meetings, process and data standard development,team mentoring and training. We created a standards library toshorten eCRF and edit check development timelines and using thislibrary enabled us to produce final eCRF screens in three days. As ourroles are different, it was important to develop communication plansand create a common language. We identified areas that could causemisinterpretation and addressed them directly with visual cues andhands-on demonstrations. Working more closely together improvedwork efforts and resource planning by joining different approachesand providing quicker resolution of issues utilizing technical solutions.Co-developing Lessons Learned meetings and documentation alsoaided the project by addressing what went right and wrong, applyingthat knowledge to future studies and decreasing the potential to repeatmissteps multiple times.

Through this collaboration or ‘marriage of the skill sets,’ we wereable to demonstrate a seamless interaction, which enhanced teamperformance and decreased redundancy. In an 18 month period, wesimultaneously started and released seven studies. Several of thesestudies had rapid enrollment (940 subjects in 35 days and anotherwith 314 subjects in 17 days) and relatively short study durations of8-14 weeks. Cross-functional mentoring, coaching and trainingenhanced knowledge sharing and decreased skill set gaps. There wasan increased ownership and accountability by joining our roles.Enhanced communication and language blending was notedthrough the life of the studies. Another major benefit of thismarriage of skill sets was improved issue resolution management.This ability to go directly to the right person for the right issue in atimely manner ultimately increased customer satisfaction.

In essence, we have remained relatively specialized within our roles buthave created a marriage of the philosophies. This collaborationincreases communication, ownership and performance. Data managersare enhancing their technical skills; programmers are enhancing theirclinical skills; together, we are delivering timely quality clinical data.


Ensuring Data Quality Through Process Improvement InitiativesBy: Kristan Gallitano, Associate Director Data Management, Serono Laboratories, Inc. and Sherri DeGiorgio, Senior Data Analyst, SeronoLaboratories, Inc.

The clinical trial data we manage is used for analysis, reporting andultimately submissions; therefore, ensuring high quality is of criticalimportance. Determining best practices to meet the goal of qualitydata can take many avenues. This article will discuss Serono’sapproach to creating best practices in achieving quality data throughtwo process improvement initiatives we undertook in the past yearand a half.

The first initiative, known as the Policies, Operating Procedures,Instructions (POPI) Team, began in 2004. Within the ClinicalDevelopment Organization, this team was created to take a closerlook at the organization’s Standard Operating Procedures (SOPs)and supporting documents. The formation of the team stems fromthe belief that quality has its origins in ethical principles and thatprocesses are developed to ensure a high level of quality. The goal ofthis team was to develop a series of regulated documents to improveour ability to meet our needs for conducting clinical trials incompliance with Serono policies and SOPs, including regulatoryrequirements and specific work instructions reflecting departmentalprocesses.

One of the main issues identified with the existing structure was thelarge volume of clinical SOPs with many of these SOPs in excess of20 pages. This volume made it difficult to confirm that the SOPswere read, understood and/or followed. Another issue identified wasthat it was taking an average of nine months to review and approve anew SOP. In addition, access to departmental working conventionswas limited to the authoring department. For cross-functional workinstructions, this presented a problem.

Principal participants of the POPI team included representativesboth within Clinical Development (CD) and outside CD, thusmaking this a global project with worldwide effort. The teamapproached quality through the development of new clinicalpolicies, followed by a rewrite of high level SOPs. In parallel with theSOPs, detailed work instructions, templates and forms were created.

Specifically, the policies were written to be in line with subject andbusiness needs in addition to regulatory and customer requirements.The policies are 1-2 pages in length and reflect a statement thatpresents the guiding principle behind our approach to the topic asan organization.

The SOPs are 5-8 pages in length and contain a more detailedpresentation of our approach to the relevant topic in order tocomply with the Declaration of Helsinki, ICH Guidelines and keyregulations. SOPs include the requirements for what we do with aconcise description of high-level activities, roles and responsibilitiesand the end result of the process.

Departmental Work Instructions (WIs) were created from existingSOP and Working Convention content and contain a description ofthe process/activity flow, including details of the relevant roles andresponsibilities within the process and a description of supportingforms and tools.

All documents created underwent review cycles within each depart-ment in addition to a global cross-functional review. An additional

tool was developed, an organizational tree, which allows for easyidentification of how each SOP, WI and supporting tool linktogether.

Results to date include the release and publication of clinicalpolicies, SOPs, WIs and supporting tools in one database accessibleto all CD personnel. Policy training was conducted through globalLunch and Learn presentations. An e-Learning tool was used forassessing all CD personnel on the SOPs. Departmental and cross-functional WI training occurred via traditional classroom trainingsessions within Data Management (DM).

The second initiative was the creation of a DM/Clinical TrialManagement (CTM) Efficiency Team, which is comprised ofrepresentatives from both DM and CTM Departments within CD.This was a global initiative with representatives from both theUnited States (US) and Switzerland. This team was created based ondiscussions during monthly meetings between the DM and CTMmanagement teams. As the POPI Team progressed, issues were raisedduring these meetings regarding potential duplication of effortbetween the two departments. Thus, the DM/CTM Efficiency Teamwas launched in early 2005 to evaluate any duplication between thetwo departments. The goal of this team is to develop and implementan action plan that looks at ways to increase quality and efficiency inthe handling of clinical trial data and decrease duplication of effortsbetween DM and CTM, resulting in data fit for purpose/reporting.

Biweekly meetings were held and processes from trial planningthrough trial conduct were reviewed from both departments’perspectives. The team focused primarily on the electronic datacapture (EDC) process versus the paper process since the majority ofSerono trials are managed using EDC. After four months ofcollaboration, an Action and Implementation Plan was developedand approved.

Some of the processes identified as areas in which greater efficiencycould be achieved included: inconsistent trial team communication;the need to optimize query text and resolution; overlap between theData Validation Plan (DVP) and Clinical Monitoring Conventions(CMC); the need for a method to analyze queries and providegeneral feedback to the Data Standards Team (DST) regarding casereport form (CRF) design, the content of the DVP and CMCtemplates.

Actions currently in progress include the addition of a Question andAnswer log to all EDC trial home pages. A query writing andresolution training module is under development. The goal is topresent this training in a joint DM/CTM session with the focusbeing ‘best practices’ and ‘tips and tricks’ on effective query writing.This training will focus on EDC query process. A combined DVPand CMC template was created and is currently being piloted in atrial; feedback is expected by year-end. Additionally, query metricsreports were created to analyze the query/requery rates, to identifyedit checks firing/not firing and to access the frequency and natureof queries by CRF module and item.

Continued on page 16


Based on the initiatives, resulting action items will include feedbackand suggestions given to the DST regarding CRF design modifica-tions and changes to the standard DVP, including modification ofedit checks and the production of a combined template for theDVP/CMCs.

Within CD, the Efficiency Team updates occur via EDC newsletter,which highlights changes such as the home page enhancement.Another avenue used is the Lunch and Learn sessions, which occurin both US and Switzerland and allow for new processes to bepresented to all of CD, including field-based staff via video. Inaddition, when field-based Clinical Research Associates are in-housefor meetings, DM and CTM hold joint sessions to discuss relevanttopics.

As the results of these initiatives are implemented, the teams willbegin to monitor the progress and the impact. Ensuring data qualityis an ongoing process that begins with initiatives such as the twodescribed above.

Ensuring Data Quality Through ProcessImprovement InitiativesContinued from page 15

SCDM Professional CertificationThe Society for Clinical Data Management (SCDM) wouldlike to congratulate the following individuals for receivingtheir Certified Clinical Data Management designation!!

Curtis Campbell, CCDM Dawn Harris, CCDMAngel Lazarov, CCDM Nimita Limaye, CCDM

Call for Articles – Future Issues of Data BasicsSCDM is looking for articles for future issues of Data Basics. Formore information regarding article submission, please contactSamantha Bordeaux at [email protected].

Themes for the 2006 issues of Data Basics include:Spring 2006 Traditional EDCSummer 2006 Data Privacy/EU DirectiveFall 2006 Non-traditional EDC/Data CaptureWinter 2006 Vendor Interaction


Integrating External Clinical Data into Clinical Research DatabasesCynthia A. Brandt, MD, MPH, Assistant Professor, Center for Medical Informatics, Yale University School of Medicine

Large-scale clinical study database management systems (CSDMS)must provide facilities to incorporate data from other systems orelectronic sources for centralized data monitoring and reporting. Inmany cases, users need to import data previously collected in aretrospective study, or data collected in other computer systems such aslaboratory data. Also, data may be entered off-line (not through theCSDMS web data entry) and this data should easily be imported intothe CSDMS. All data imports should be accomplished withoutdifficult manipulation and data transformation.

It is a challenge to provide seamless integration and linkage of data.The process of importing data into a large-scale CSDMS is fre-quently complicated by the fact that these databases may bedesigned to store information of hundreds of clinical studies withseveral thousand clinical parameters. Similar to data stored inelectronic patient record systems (EPRS), clinical studies containdata that is highly variable in form and content. CSDMS datadiffers from EPRS clinical data by the strict structure enforced in acontrolled trial or research study. In the controlled study, a writtendocument (protocol) specifies when data or tests are to be performed(at pre-determined and scheduled events). Importing data into aCSDMS must make use of this study information (metadata) tostructure the import process as well as the data item specificationsthat are present in study specific data dictionaries.

Various informatics solutions and approaches to integration havebeen applied in the biomedical research domain. Solutions includeapproaches that involve the use of standards and clinical vocabulariesfor inter-operation and data mapping. Data mapping is a processwhereby semantically identical attributes in separate sources of dataare linked to each other. Both manual and automatic linkingmethods are possible through the use of extensive system metadata,including clinical study-specific data dictionaries and the use ofstandard vocabularies, such as Logical Observation IdentifiersNames and Codes (LOINC®)1 and the Health Level Seven, Inc(HL7)2 standard message format for laboratory data. There areapproaches to mapping data to standards or vocabularies that can beapplied and may include the following: 1) lexical mapping or exactword or phrase equivalence, 2) mapping terms to common concepts(e.g., chest pain = angina, heartburn etc), and/or 3) using the linksor relationships between concepts to look for similar terms.3

With the increased availability of data in electronic form and theincreased emphasis on collaborative research, the need to provideresearchers and data managers with the tools required for mapping dataand integration is even greater. The mapping must be valid to ensureaccurate integration and the semantics of the data that is being mappedare frequently complex. The importing process may be automatedthrough the use of this data and metadata mapping. Issues andproblems that can occur with mapping include: 1) differences in the

granularity of the data at the answer and the data question level, 2)differences in the wording or description of data/questions and answersand 3) differences in mapping (lose data or lose granularity). Because ofthese issues, it is important to test the results of the mapping. This caninclude qualitative or quantitative approaches to evaluate the mapping.

In summary, approaches that are used in a research context to mapdifferent questions and/or answers, or to map and merge differentclinical vocabularies, can be applied to production system dataintegration and mapping efforts. Although the individual mapping ofdata for data integration efforts is generally given little emphasis inintegration projects and presentations, if not performed correctly, thedata that is now “integrated” is likely to be invalid. It is important toprovide visual tools, and understand the effort that will frequently berequired when integrating and merging different data. Lastly, results ofintegration and mapping must be validated and tested.

References1). Logical Observation Identifiers Names and Codes, accessed 3/05;2). Health Level Seven, Inc, accessed 3/05; http://www.HL7.org.3). Dolin, R.H., et al., Evaluation of a “lexically assign, logically refine” strategy for semi-

automated integration of overlapping terminologies. Journal of the American MedicalInformatics Association, 1998. 5(2): p. 203-13.

Bring your expertise to the SCDM DiscussionForum, on the web at http://www.scdm.org/.


SCDM Committee Chairs meet for the first Cross Communication Commit-tee meeting at the 2005 Annual Fall Conference.

Future SCDM ConferenceDates and Locations

SAVE THE DATE! SCDM Spring Forum 2006Transforming the Role of the Clinical Data Manager…

Addressing the Elephant in the Center of the Room

We talk about it over the water cooler, over a cocktail and in ourstaff meetings… How do we respond to the rapid pace of changeoccurring within clinical data management (CDM)?

Forum participants will examine various aspects influencing thefuture of our profession. Brainstorming discussions will address thefollowing areas:

~ Off Shoring ~ Technology ~ Standardization ~~ Redefinition of traditional roles and responsibilities ~

March 19 – 21, 2006 • Washington Duke Inn and Golf ClubDurham, NC

Please go to www.scdm.org for more information!

Fall Conference Dates2006 Fall ConferenceOctober 8-11, 2006

Buena Vista Palace in the Walt Disney ResortLake Buena Vista, Florida

2007 Fall ConferenceSeptember 16-19, 2007

Hyatt Regency Chicago on the RiverwalkChicago, IL


Please contact [email protected] if you have questions about registration forupcoming meetings, advertising, renewal of membership, or if you need toprovide updated mailing/contact information.

Society for Clinical Data Management, Inc.555 E. Wells StreetSuite 1100Milwaukee, WI 53202-3823Phone: 414-226-0362Fax: 414-276-3349E-mail: [email protected]

@Info

scdm.org

Visit us atwww.scdm.org

Society for Clinical Data Management, Inc.555 E. Wells Street

Suite 1100Milwaukee, WI 53202-3823

FIRST CLASS PRSRTU.S. POSTAGE

PAIDPERMIT NO. 2066

EAU CLARIE, WI 54701

92-1205-004

data basics - society for clinical data management (scdm) · electronic case report form (ecrf),...

Documents