TXT4HEALTH IN THE CRESCENT CITY
ACADEMY HEALTH
ANNUAL RESEARCH MEETING
BALTIMORE, MD
JUNE 24, 2013
David Kulick, MPH
Louisiana Public Health Institute
New Orleans, LA
Acknowledgments:
Nebeyou Abebe
Anjum Khurshid
Lisanne Brown
Snigdha Mukherjee
Tom Carton
2
Crescent City Beacon Txt4health mhealth
Program
Program Rationale:
• Diabetes Prevalence
• Louisiana has 2nd highest diabetes related mortality
• Diabetes cost Louisiana approximately $231M in 2012
• 60% of adult residents (18+) in New Orleans have at least
one risk factor for Type-2 diabetes
• Ubiquitous mobile phone usage
• 93% of adults have mobile phones1
• 80% of adults with mobile phones use text messaging1
1. Pew Internet and American Life Project, 2013
2. Louisiana Department of Health and Hospitals, Bureau of Primary Care and Rural Health, 2013
3. LAHIDD, 2010
3
Txt4health Overview
Target population:
• Adults between 18-44
• Undiagnosed or at risk for
Type-2 diabetes
• Live in greater New
Orleans
Campaign objectives:
• Raise awareness about
personal risk for diabetes
• Set personal weight loss
and physical activity goals
• Link individuals to care and
community resources
4
Engagement Strategy
• Community Advisory Group – Public and Private Partnerships
• TV and Radio PSAs
• Outdoor billboards
• Co-branded posters
• Outreach at community events, neighborhood associations,
•
•
•
•
conferences, universities, and faith based organizations
Outreach with providers including promotion through providers
Support through BCBS
Outreach at businesses and through employee events
Contests and incentive programs
5
Data Generated
Programmatic data
1.
•
•
•
•
Risk assessment questions
Demographic information
Activity
Weight goals
Enrollment trends
3. Population based survey
2.
•
•
Pre-campaign survey
Post-campaign survey
Participant satisfaction survey
5. Combinations of the data
4.
6
Programmatic Data
Age distribution of
participants who entered age
(N=1060)
Reported weight goal (N=1057)
30-44
Reported
weight goal
39%
61%
29%
BMI of participants who
entered weight and height
information (N=1395)
44%
29%
BMI < 25
Normal or
Underweight
BMI 25-30
Overweight
27%
Did not
report
weight goal
71%
Other
Reported active for at least 30 minutes a day (N=1431)
45%
50%
40%
25%
30%
20%
10%
BMI > 30
White
19%
45+
41%
African
American/Black
10%
18-29
30%
Race/Ethnicity of those
reporting (N=639)
5%
4%
0
1
8%
7%
2%
4%
6
7
0%
2
3
days
4
5
7
Programmatic Data: Activity Responses
Reported Days Active For At Least 30 Minutes
Over Last Week Throughout 14 Week Program
(log values)
3.5
3
2.5
2
1.5
1
0.5
0
Week 1Week 2Week 3Week 4Week 5Week 6Week 7Week 8Week 9 Week Week Week Week Week
10
11
12
13
14
No response
0 days
1 day
2 days
3 days
4 days
5 days
6 days
7 days
8
Enrollment Trends
Enrollment in txt4health Over Time
2000
1800
1600
Enrollment contest
1400
Enrollment
1200
1000
University outreach events
800
600
Community outreach event
400
Community outreach event
200
Launch event
0
1/20/12
3/10/12
4/29/12
6/18/12
8/7/12
9/26/12
Date
11/15/12
1/4/13
2/23/13
4/14/13
9
Population Based Survey:
Pre-Campaign Survey
Mixed method population
based survey N=701
internet panel and RDD
Overall Texting Capabilities in Household
100
4
24
9
1
6
9
2
4
36
80
6
Don’t Have, No One Does
12
60
9
Don't Have but, Someone Does
95
84
40
86
18
Have, Don't Use
Have, Use
58
20
36
Source: Pre-Campaign Survey, 2012
0
A.
Total
(N=701)
G.
18-29
(N=103)
H.
30-34
(N=72)
I.
35-44
(N=126)
J.
45 and Older
(N=393)
10
Population Based Survey:
Post-Campaign Survey
N=701
Diabetes in Family
Mixed method population
based survey N=701
internet panel and RDD
African
American/Black
18-29 Years Old
Pre
Post
Pre
Post
Pre
Post
Awareness of
“Txt4Health”
4%
19%
8%
29%
9%
29%
Awareness of
“Text ‘HEALTH’
to 300400”
5%
16%
5%
23%
4%
33%
= significantly greater than comparison group at 95% confidence level
Source: Post-Campaign Survey, 2012
11
Participant Satisfaction Survey
N=77
Overall, how
satisfied were
you with the
program?
Txt4health made
me conscious of
my risky
behaviors.
I will recommend
txt4health to my
friends and
family.
Strongly Disagree
1%
3%
2%
Disagree
16%
4%
7%
Agree
32%
34%
24%
Strongly Agree
51%
59%
67%
Total
100%
100%
100%
Online rolling biweekly satisfaction survey
prompted by a consent text message
followed by entry in the survey via URL
12
Data: value to the program
1. Outcome success
• Demonstrates our success in targeting the population of interest
2. Resource monitoring
• Utilization of resources
• Cost/enrollee and over time making most effective use of resources and
relationships
3. Effectiveness of social marketing campaign
• Demonstrating the value of partnerships and in-kind support
• Facilitates the sustaining and expanding relationships with stakeholders
4. Program scaling
• Realistic expectations for future scaling in terms of engagement and social
marketing success
13
Data: lessons learned and value to the
mhealth field
1. Scaling barriers from pilot to community
• Social marketing barriers: significant cost/enrollee
demonstrating the need for formative research to
understand the most effective use of resources
• Evaluation barriers: what to measure and the value of the
data
2. Engagement approach for mhealth is transferrable
for community chronic care management
• Demonstrates the importance of PPP for cost savings
14
Next Steps
1. Evaluating the data further and understanding
the profile of users to create most successful
engagement model
2. Programmatic expansion to state and
continued partnership development
15
Thank you
David Kulick
dkulick@lphi.org