Data Management Issues: University Affiliation and Transitioning Data Managers

1
Data Management IssuesUniversity Affiliation
and Transitioning Data Managers

• Presented By Sara Becker Christina
  SanchezPrevious / Current Data Managers Duke
  University Medical CenterDurham, North Carolina

2
Duke University Medical Center
3
Drug Abuse Treatment for Adolescents (DATA) Study

• Brief, outpatient treatment (MET/CBT-5) for
  adolescents age 13-21
• Clinical staff
• 1 Principal Investigator
• 1 Clinical Supervisor
• 1 Medical Director
• 4 MET / CBT -5 counselors
• Research staff
• 1 Evaluator
• 1 Data Manager / Study Coordinator
• 1 Undergraduate Research Assistant

4
Site-Specific Research Elements

• Different IRB requirements for participants age
  13-17 vs. age 18-21
• Compensation of 25 per assessment
• Several local instruments at each assessment
• Child Behavior Checklist (age 13-18)
• Personal Experience Screening Questionnaire
  (PESQ age 13-18) and PESQ-Adult Version (age
  19-21)
• Quality of Life Profile - Adolescent Version
• Parental Monitoring - Adolescent and Parent
  Version
• Parental Communication - Adolescent and Parent
  Version
• Peer Substance Use and Peer Tolerance of
  Substance Use

5
Site-Specific Clinical Elements

• 2 optional family sessions
• Parental Monitoring
• Enforcing Consequences
• 1 final interpretive session
• Share urine screen results and patient progress
  with participants guardian
• Provide continuing care recommendations

6
Data Manager / Study Coordinator Role

• Oversees data collection and management
• Reviews assessment schedule and tracking
• Enters client data into the GRL
• Responds to Gain Edits and submits data to CHS
• Enters all GPRA discharge interviews
• Monitors GAIN and GPRA follow-up rates
• Coordinates financial incentive process
• Prepares agenda and minutes for team meetings
• ABS Co-Administrator
• Certified GAIN Interviewer

7
Evaluator Role

• Leads study recruitment efforts
• Screens all potential clients to determine
  eligibility
• Conducts majority of intake interviews
• Enters GAIN and GPRA data for completed
  interviews
• ABS Co-administrator
• Certified local GAIN trainer and interviewer
• Works closely with DM to QA the data

8
How do we make the evaluation system work?

• THREE key elements
• ) Tight COORDINATON
• ) Consistent PROCESSES
• ) Site-Specific MANUAL

9
TIGHT COORDINATION between research and clinical
staff

• Weekly executive committee meeting (PI, Clinical
  Supervisor, 1 Clinician, Evaluator, DM)
• Meeting objectives
• Determine caseness of potential participants
• Monitor intake and follow-up rates
• Assess ongoing recruitment and retention efforts
• Evaluate status of all active cases
• Discuss ongoing DM issues

10
TIGHT COORDINATION between research and clinical
staff

• Weekly clinical meeting (Full Project Team)
• Meeting objectives
• Research staff provides relevant information from
  assessments to clinical staff
• Clinical staff provides updates on attendance and
  treatment recommendations for GRL
• Clinical staff submits completed research
  materials (TxSat, GPRA discharge, clinical
  binders) to DM
• Clinical and research staff work together to
  locate and contact slippery clients for
  follow-up

11
Creation of clear, CONSISTENT PROCESSES

• Research binders prepared in advance for 3 age
  groups (13-17, 18, 19-21)
• Binders contain all materials needed for intake,
  3-, 6- and 12-months (consent,
  demographics, GAIN, local instruments, etc)
• Clinical binders prepared in advance for all
  subjects
• Binders contain materials for all 5 MET/CBT
  sessions with envelopes for items to be given
  to DM (TxSat, GPRA discharge)
• Research staff conducts intake, adds reports
  (PFR, ICP, GRRS) and gives binder to clinician
• Clinicians submit materials to DM at weekly
  clinical meetings
• GOAL 5 of each binder ready to go at all times!
  

12
Creation of SITE-SPECIFIC MANUAL

• Our Duke Manual consists of local data
  tracking, reference and QA forms
• Group Tracking Form - spreadsheet to track status
  of cases
• Data Entry Shell - spreadsheet to enter all of
  local data
• GAIN-I QA Checklist - QA form of common Gain-I
  issues
• M-90 QA Checklist - QA form of common M-90 issues
• Study Specific Codes - document of all the user
  and site IDs used by Duke
• Team Reference Sheet - cheat sheet of all of
  the websites, user IDs, and passwords needed by
  the DM

13
So, we got the system to work How did we train
someone new?

• FOUR key steps
• ) Informal INTEGRATION
• ) Formal TRAINING
• 3 ) QUALITY ASSURANCE
• 4 ) Continuing COLLABORATION

14
Step One INFORMAL INTEGRATION

• Prior to the transition, new DM was already
  integrated into the system
• Attended weekly team meetings
• Earned GAIN certification
• Demonstrated competence using ABS on the computer
• Entered GAIN and GPRA data, printed relevant
  reports
• Although the new DM had not been formally
  trained, she had participated in a number of key
  processes.

15
Step Two OFFICIAL TRAINING

• Old DM offered a two day training session
• Created Study Coordinator Reference Sheet and DM
  Manual
• Reviewed responsibilities described in reference
  sheet
• Reviewed materials in the Duke Manual
• Chestnut Health System (CHS) offered telephone
  training
• Reviewed any questions about GAIN, GRL, GAIN
  edits, and data submission process

16
Step Three QUALITY ASSURANCES

• Old and new DM submitted data together the month
  of the switch
• Entered any new GAIN-I, M-90 or TxSat records
• Completed GAIN edits
• Checked new records using QA forms
• Updated GRL
• Exported GAIN-I, M-90, and TxSat data
• Co-submission assured that all of the old DMs
  strategies were shared with the new DM!

17
Step Four Continuing COLLABORATION

• New and old DM maintain an open dialogue
• Review and evaluate old methods on an as needed
  basis (example process of binder creation, GPRA
  discharge)
• Share and learn new techniques
• New DM frequently consults with the Evaluator
• The Evaluator is an integral part of the system
• Establishing a relationship between the Evaluator
  and new DM helped the new DM adjust to the system
  
• It also prepared the Evaluator for potential
  bumps in the road

18
Step Four Continuing COLLABORATION

• New DM works closely with a research assistant
  who
• Previously worked with the old DM
• Manages the creation of binders
• Oversees entry of local data
• Reminds new DM of emerging issues!
• ALL team members share a common goal ?
  ACCURATE AND CONSISTENT DATA!
• In situations when data are ambiguous,
  consultation facilitates consistency.

19
Sounds easy enough What were the hardest parts?

• THREE issues
• 1) DOCUMENTATION
• ROLE CONFUSION
• PATIENCE (!!)

20
Issue 1 DOCUMENTATION

• Because even when you think your manual has it
  all, something might surprise you!
• We had a glitch in the system with some
  un-entered GPRA data.
• The processes had been documented in a way that
  made sense to the OLD DM, but were not clear to
  the NEW DM
• If this happens to you, flexibility and
  collaboration are critical!
• Our new and old DM worked together for 3 hours to
  address the problem and document the resolution!

21
Issue 2 ROLE CONFUSION

• The Co-DM role during the switch had pros and
  cons
• PROS
• Old DM available for questions about data issues
• Facilitates on the job learning
• Provides QA on the data
• CONS
• Creates role confusion for the team which DM
  should be approached with data issues? (example
  Clinical Meetings)
• New DM may not feel a sense of ownership
• Old DM may feel burdened with extra
  responsibilities

22
Issue 3 PATIENCE!

• Such a major shift required a LOT of patience!
• Old DM needed to stay patient and calm
• It was hard to believe shed ever cover it all
  but she did!
• New DM needed to stay patient and hopeful
• It was hard to believe shed ever learn it all
  but she did!
• Both DMs needed to support one another
• We got through the transition by acknowledging
  how tough it was for BOTH of us!

23
So, what would we recommend to other sites
training new DMs?

• Our Top 10 Tips for a
• Seamless Transition!

24
Our Top 10 Transition Tips
10) Create a system that works for your site (it
wont work unless it works for you) 9) Fill
the system with consistent processes (consistency
facilitates accuracy!) 8) Create a manual
that ANYONE can follow (seriously, would your
grandmother be able to follow it?) 7) Document
everything and anything (because youll never
regret having too much information) 6)
Communicate early and often (early detection
fosters early resolution!)
25
Our Top 10 Transition Tips
5) Practice makes perfect! (nothing prepares
you for submitting data better than actually
submitting data!) 4) Integrate the new DMs
ideas ASAP (remember that consistency is a
work in progress, integration promotes
credibility!) 3) Collaborate with other key
team members (data management is an integrative
process collaboration is required to bring all
of the pieces together)
26
And the Top 2 Transition Tips.

• 2) Old DM stay calm!
• You will cover it all eventually!
• Normalize the process share stories about when
  you started
• Dont take it personally if the new DM changes
  your system what works for you might not work
  for someone else.
• 1) New DM stay hopeful!
• You will get it all eventually!
• Dont be afraid to ask questions and share
  successes
• DM is an ongoing process. Make the system fit
  you!

27
And an extra tip for good measure Feel Free to
Contact Us!

• Sara Becker (Previous DM)
• Sara.becker_at_duke.edu
• [phone]
• Christina Sanchez (Current DM)
• Ccs8_at_duke.edu
• [phone]
• Cindy Jones (Evaluator)
• jones106_at_mc.duke.edu
• [phone]