Increasing Access to Research Data for Secondary Data Analysis in - - PowerPoint PPT Presentation

increasing access to research data for secondary data
SMART_READER_LITE
LIVE PREVIEW

Increasing Access to Research Data for Secondary Data Analysis in - - PowerPoint PPT Presentation

Increasing Access to Research Data for Secondary Data Analysis in the Rehabilitation & Disability Fields AcademyHealth Annual Conference 2017 New Orleans, LA Disclosures NIH grant support: P2C HD065702. K01 HD086290. We have no


slide-1
SLIDE 1

Increasing Access to Research Data for Secondary Data Analysis in the Rehabilitation & Disability Fields

AcademyHealth Annual Conference 2017 New Orleans, LA

slide-2
SLIDE 2

Disclosures

  • NIH grant support:

P2C HD065702. K01 HD086290.

  • We have no financial interests to disclose.
slide-3
SLIDE 3

Presenters

  • Module 1: Amol M. Karmarkar, Assistant Professor, Co-

Director of Techniques Development, Center for Large Data Research and Data Sharing in Rehabilitation, University of Texas Medical Branch, amkarmar@utmb.edu

  • Module 2: Alison M. Stroud, Archive Manager, Archive of Data
  • n Disability to Enable Policy and research, National Archive
  • f Data on Arts and Culture, ICPSR, alistrou@umich.edu.
  • Module 3: Chih-Ying (Cynthia) Li, Postdoctoral Fellow, Division
  • f Rehabilitation Sciences, University of Texas Medical

Branch, chili@utmb.edu

slide-4
SLIDE 4

Presentation Outline

  • Overview
  • Module 1: Motivation behind data archiving & sharing

activities, overview of the existing ‘data repositories’ supported by the National Institutes of Health (NIH), data directory & ADDEP.

  • Module 2: ICPSR Data Curation process & workflow, Open

ICPSR, ADDEP

  • Module 3: Example of Accessing & Analyzing Archived Data
  • Discussion & QA
slide-5
SLIDE 5

Center for Large Data Research & Data Sharing in Rehabilitation

The CLDR includes a consortium devoted to building rehabilitation research capacity by increasing the quantity and quality of outcomes research using large data.

The CLDR also supports data sharing and archiving of completed rehabilitation research studies.

Visit us at: http://rehabsciences.utmb.edu/cldr

Workshops, on-line seminars and training modules to develop skills in large data research Listing of available datasets including purpose, variables, access and contact information Collaborative projects with CLDR mentors / investigators using large data relate to rehabilitation & recovery Collaborate with CLDR mentors using large rehabilitation

  • datasets. Support

for up to six months Support for linking / merging and archiving data from completed rehabilitation studies to promote secondary data analyses Supported by the National Institutes of Health – the NICHD/NCMRR, the NINDS, and the NIBIB # P2C-HD065702

Education & Training Data Directory Pilot Projects Visiting Scholars Data Archiving

Activities supported by the CLDR Medical Rehabilitation Research Infrastructure Network (MRRIN)

slide-6
SLIDE 6

https://ncmrr.org/about/

slide-7
SLIDE 7

Key Players

slide-8
SLIDE 8

Key Players

http://annals.org/aim/article/2630766/data-sharing-statements-clinical-trials-requirement-international-committee-medical-journal http://www.nejm.org/doi/pdf/10.1056/NEJMe1705439
slide-9
SLIDE 9

Data Repositories Supported by the NIH

slide-10
SLIDE 10

Center for Large Data Research & Data Sharing in Rehabilitation

slide-11
SLIDE 11

Data Access Data Archive

http://www.disabilitystatistics.org/sources-rehab.cfm http://www.icpsr.umich.edu/icpsrweb/content/addep/index.html

slide-12
SLIDE 12

Module 2

Alison M. Stroud, Archive Manager, Archive of Data on Disability to Enable Policy and research, National Archive of Data on Arts and Culture, ICPSR, alistrou@umich.edu.

Sharing & Using Data at ADDEP/ICPSR

slide-13
SLIDE 13

Lots of Data Sharing!

  • Over 1,800 research

data repositories listed

  • n re3data.org
  • Lots of open access to

research data out there – but are the data usable for the purposes

  • f further research?

Will it increase research impact?

slide-14
SLIDE 14

Some researchers still reluctant to share data...

  • Protection of confidentiality of study participants
  • Concerns about getting “scooped”
  • Concerns about errors in data and data

documentation

  • Limited resources for preparing data &

documentation

Don’t ¡worry. ¡ICPSR’s ¡got ¡your ¡back! ¡

slide-15
SLIDE 15

Has anyone in the audience had similar concerns or experiences?

…Or maybe just didn’t know where to start?

slide-16
SLIDE 16

This is where ICPSR comes in...

ICPSR - International Leader in Data Stewardship and Data Analysis Services

  • Over 10,000 studies archived at ICPSR
  • 4.82 Million variables tagged
  • Over 70,000 citations in the searchable bibliography
  • 750+ Member Institutions
  • ~1,000 Summer training participants annually
slide-17
SLIDE 17

Archive of Data on Disability to Enable Policy and research (ADDEP)

slide-18
SLIDE 18

Curation Services

  • Processing Planning
  • Disclosure Review
  • Missing Data

Standardization

  • Outliers, Wild Codes
  • Variable-Level Metadata
  • ICPSR Codebook
  • Study Description
  • Quality Checks
  • Dissemination in SAS,

SPSS, Stata, R, ASCII

Ahh! These services are divine!!

slide-19
SLIDE 19

ICPSR’s Curation Process is rigorous

slide-20
SLIDE 20

Metadata Ensures Use and Impact

  • Summary ¡of ¡the ¡data ¡collec0on ¡
  • Access ¡condi0ons ¡
  • Methodology ¡
  • Data ¡cita0on ¡and ¡bibliography ¡
  • Scope ¡
slide-21
SLIDE 21

Impact of data and materials

  • Views, Downloads
  • Publications
  • Inform design of new studies
  • Training, Student Use
  • Grant Applications (funding opportunities!)
  • Altmetrics

Use ¡of ¡Study ¡(Children ¡with ¡Disabili@es) ¡-­‑ ¡released ¡in ¡Oct ¡2016 ¡

slide-22
SLIDE 22

https://www.utmb.edu/cldr/pilot-projects/data-sharing-studies

Example of funding opportunity

slide-23
SLIDE 23

Data Citation

Markides, Kyriakos, Nai-Wei Chen, Ronald Angel, Raymond Palmer, and James Graham. Hispanic Established Populations for the Epidemiologic Study of the Elderly (HEPESE) Wave 7, 2010-2011 [Arizona, California, Colorado, New Mexico, and Texas]. ICPSR36537-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2016-12-05. http://doi.org/10.3886/ICPSR36537.v2

Authors Persistent Identifer

Ensuring Researcher Credit for Data

My study is famous!!!

slide-24
SLIDE 24

Yes, we have self-publishing options

slide-25
SLIDE 25
  • penICPSR v. ADDEP/ICPSR

Cura@on ¡Services ¡

< ¡

For ¡long ¡term ¡data ¡re-­‑usability, ¡growing ¡publica0ons, ¡etc…. ¡ Full ¡cura0on ¡is ¡usually ¡best. ¡

slide-26
SLIDE 26

Outreach and spreading awareness

ADDEP and ICPSR also offer:

  • Webinars
  • Social media
  • Connections to member institutions
  • Summer Program courses (workshop on June

29-30)

slide-27
SLIDE 27

Let’s take a look at the website

Also, openICPSR if time.

slide-28
SLIDE 28

Chih-Ying (Cynthia) Li, PhD Postdoctoral Fellow, Division of Rehabilitation Sciences, University of Texas Medical Branch, chili@utmb.edu

Example of Accessing & Analyzing Archived Data

Module 3

slide-29
SLIDE 29

https://www.icpsr.umich.edu/icpsrweb/

ICPSR Where to Start?

Finding Data Accessing Data Using Data Lessons Learned

slide-30
SLIDE 30

Finding Data Accessing Data Using Data Lessons Learned

ICPSR Webpage

slide-31
SLIDE 31

Finding Data Accessing Data Using Data Lessons Learned

slide-32
SLIDE 32

Finding Data Accessing Data Using Data Lessons Learned

  • Study Relevance
  • Title A-Z
  • Released/Updated
  • Time Period
  • Most Cited in ICPSR Bib.
  • Variable Relevance
slide-33
SLIDE 33

Finding Data Accessing Data Using Data Lessons Learned

slide-34
SLIDE 34

Finding Data Accessing Data Using Data Lessons Learned

§ PI’s § Summary § Funding § Related publications § Access (publically available vs any restrictions) § Scope (units, geographic coverage, time period) § Methodology (purpose, design, description of variables, response rates) § List of Variables § Datasets

Study Information

slide-35
SLIDE 35

Finding Data Accessing Data Using Data Lessons Learned

Search Variables You Want

slide-36
SLIDE 36

Finding Data Accessing Data Using Data Lessons Learned

Search/Compare Variables

slide-37
SLIDE 37

Finding Data Accessing Data Using Data Lessons Learned

Search/Compare Variables

slide-38
SLIDE 38

Finding Data Accessing Data Using Data Lessons Learned

slide-39
SLIDE 39

Finding Data Accessing Data Using Data Lessons Learned

ICPSR Membership List

slide-40
SLIDE 40

Finding Data Accessing Data Using Data Lessons Learned

slide-41
SLIDE 41

Finding Data Accessing Data Using Data Lessons Learned

slide-42
SLIDE 42

Finding Data Accessing Data Using Data Lessons Learned

slide-43
SLIDE 43

Finding Data Accessing Data Using Data Lessons Learned

slide-44
SLIDE 44

Finding Data Accessing Data Using Data Lessons Learned

slide-45
SLIDE 45

Finding Data Accessing Data Using Data Lessons Learned

Download/Upload Data

FREE

slide-46
SLIDE 46

Finding Data Accessing Data Using Data Lessons Learned

§ Research Question: Is depression at discharge from inpatient rehabilitation associated with functional decline over the following year for the underserved individuals with stroke? § Sample: 840 individuals with stroke with complete data at inpatient rehabilitation discharge and 1-year follow-up

  • Average age: 68.3 (SD: 13.0) years
  • Sex: 52.1% female
  • Race/Ethnicity: 76.2% white, 17.3% black, 4.5% Hispanic
slide-47
SLIDE 47

Finding Data Accessing Data Using Data Lessons Learned

0% 5% 10% 15% 20% 25% 10.1% 12.8% 17.2% 22.5%

No Depressive Symptomology Depressive Symptomology

Self-care Decline Mobility Decline

  • Figure. Rates from a multilevel model adjusted for patients’ age, sex, race/ethnicity, comorbidity count, level of social

support, and whether or not they received any additional therapy over 3 months after inpatient rehabilitation discharge.

slide-48
SLIDE 48

Finding Data Accessing Data Using Data Lessons Learned

Lessons Learned

§ Potential Challenges:

  • Searching available datasets
  • Identifying novel research question
  • Understanding data

§ Potential Benefits:

  • Free, accessible data
  • Publications
  • Evidence to support grant applications
slide-49
SLIDE 49

Department Name Goes Here

slide-50
SLIDE 50

Department Name Goes Here

slide-51
SLIDE 51

Thank you!

  • Amol M. Karmarkar, Assistant Professor, Co-Director of

Techniques Development, Center for Large Data Research and Data Sharing in Rehabilitation, University of Texas Medical Branch, amkarmar@utmb.edu

  • Alison M. Stroud, Archive Manager, Archive of Data on

Disability to Enable Policy and research, National Archive of Data on Arts and Culture, ICPSR, alistrou@umich.edu.

  • Chih-Ying (Cynthia) Li, Postdoctoral Fellow, Division of

Rehabilitation Sciences, University of Texas Medical Branch, chili@utmb.edu