NETWORK ANALYSIS TO STUDY STAKEHOLDERS INTERESTS 2 AGENDA - - PowerPoint PPT Presentation

network analysis to study
SMART_READER_LITE
LIVE PREVIEW

NETWORK ANALYSIS TO STUDY STAKEHOLDERS INTERESTS 2 AGENDA - - PowerPoint PPT Presentation

ACADEMYHEALTH ANNUAL RESEARCH MEETING JUNE 3, 2019 GRACE WANG, STEPHEN HINES, TENNILLE DANIELS, MARIA ALVA, SIYING LIU FEASIBILITY OF TWEETS AND SOCIAL NETWORK ANALYSIS TO STUDY STAKEHOLDERS INTERESTS 2 AGENDA Background Methods


slide-1
SLIDE 1

ACADEMYHEALTH ANNUAL RESEARCH MEETING JUNE 3, 2019 GRACE WANG, STEPHEN HINES, TENNILLE DANIELS, MARIA ALVA, SIYING LIU

FEASIBILITY OF TWEETS AND SOCIAL NETWORK ANALYSIS TO STUDY STAKEHOLDERS’ INTERESTS

slide-2
SLIDE 2

AGENDA

▸ Background ▸ Methods ▸ Results ▸ Lessons learned and practice implications

2

slide-3
SLIDE 3

BACKGROUND

PATIENT PRIORITIES IN RESEARCH

3

slide-4
SLIDE 4

BACKGROUND

▸ Social media is a potentially useful

data source for identifying the priorities of patients, families, and

  • ther stakeholders

SOCIAL MEDIA

4

slide-5
SLIDE 5

BACKGROUND

STUDY OBJECTIVES

▸ To show the feasibility of using social media (i.e., tweets) and

social network analysis to:

▸ (1) identify stakeholder groups and their interests, and ▸ (2) determine whether patient, researcher, and other

stakeholder interests are aligned.

5

slide-6
SLIDE 6

METHODS

STEP 1. SEARCH TWITTER

6

Search

  • Software: NodeXL
  • Terms: #chronickidneydisease, #CKD, #ESRD
  • Timeframe: 10/9/18 to 11/8/18

Dataset

  • 2303 tweets (including retweets and mentions) by 1051 Twitter

users

slide-7
SLIDE 7

METHODS

STEP 2. ANALYZE DATA

7

Social network analysis

  • Applied Girvan Newman method in NodeXL
  • Identified 163 clusters of users

Content analysis

  • Reviewed user accounts within each cluster
  • Categorized cluster into a stakeholder groups (i.e., patient/family;

advocacy; researchers; physicians/care delivery; industry)

  • Analyzed tweets to determine the stakeholder’s priorities
slide-8
SLIDE 8

RESULTS

▸ Dots: 74 Twitter users ▸ Colors: 25 clusters

8

PATIENT & FAMILY NETWORK

slide-9
SLIDE 9

RESULTS

▸ One Twitter user in the

center

▸ About 20 followers

9

PATIENT & FAMILY NETWORK

slide-10
SLIDE 10

RESULTS 10

PATIENT & FAMILY NETWORK

slide-11
SLIDE 11

RESULTS 11

PATIENT & FAMILY NETWORK

slide-12
SLIDE 12

RESEARCHER NETWORK

12

slide-13
SLIDE 13

RESEARCHER NETWORK

13

slide-14
SLIDE 14

RESEARCHER NETWORK

14

slide-15
SLIDE 15

DISCUSSION

LIMITATIONS

▸ Only tweets and not other forms of social media (e.g., Facebook) ▸ Twitter search results ▸ Active Twitter users with public accounts

15

slide-16
SLIDE 16

DISCUSSION

LESSONS LEARNED AND PRACTICE IMPLICATIONS

▸ Feasible method ▸ Chronic kidney disease example ▸ Patients and families: how condition affects daily life (boredom, diet) ▸ Researchers: CIRC study and clinical guidelines ▸ Twitter as a supplementary information source for stakeholders to learn

more about one another

16

slide-17
SLIDE 17

SUMMARY

▸ It is feasible to use Twitter and social network analysis to identify

stakeholders and their interests

▸ Social media posts can be a supplementary information source to

learn about stakeholder interests

17

slide-18
SLIDE 18

QUESTIONS?

Grace Wang, PhD, MPH Principal Research Associate IMPAQ International 206-939-4980 gwang@impaqint.com

18