labrique global health v4
TRANSCRIPT
Alain B. Labrique, PhD, MHS, MS
DirectorJHU Global mHealth Initiative (JHU-GmI)
Associate ProfessorProgram in Global Disease Epidemiology and Control
Dept. of International Health & Dept. of Epidemiology (jt)
Johns Hopkins Bloomberg School of Public Health
JHU School of Nursing
JHU School of Medicine (Health Informatics)
Maternal Mortality 2010, Worldmapper.org
Euclidean map of 10 million of the 850 million
Facebook users friend networks© Paul Butler, FB
Mobile – Social Networks : New Frontiers for Global Health
6.8 BILLION MOBILE-CELLULAR SUBSCRIPTIONS
Untethered, yet connected: Diverse applications of ubiquitous
wireless and mobile technologies
designed to improve and
enhance health research, health services
delivery and health outcomes
mHealth
mHealth:The Four C’s
Harnessing ubiquitous information
and communication technology to
collect data, connect individuals to
each other and to information,
compress time and create
opportunities to intervene.
Global “mHealth” is a complex, diverse
development space, and is not homogenous.
jhumhealth.org
133 mHealth Projects at JHU, as of September 2014
“JiVitA” Maternal and Child Health Research Project
(WWW.JIVITA.ORG)
Public Health, Maternal and Child Health
and Nutrition Efficacy Research
to
Improve Health and Save Lives in
Bangladesh, South Asia and Globally.
RANGPUR
Rural families use mobile phones
during severe pregnancy crisesN=11,451 (2007-2010)
Source: Labrique, mHealth Summit, Washington DC, 2011
168,231 Woman Survey –Gaibandha, Bangladesh
(January-March 2012)
• 71% Households own phones
• 20% Used a phone in past 30 days for
emergency health purpose
• Phone owners 2.8 times more likely to
use phone for health call
• ONLY 23% Electricity in home!
Labrique et al., Unpublished data, mHealth Summit 2012
0.2
.4.6
.81
Pro
ba
bili
ty o
f M
ob
ile P
ho
ne
Ow
ne
rship
2008 2009 2010 2011 2012
Year
Lowest Quartile WI (n=17,176) Low Quartile WI (n=19,789)
High Quartile WI (n=6,472) Highest Quartile WI (n=1,032)
Mobile Phone Ownership by WI over Time
Household Mobile Phone Ownership over time in rural Bangladesh, by “Wealth Index” (n=44,469)
Labrique, Tran et al, 2013 (in press)
Pro
po
rtio
n o
f H
H r
ep
ort
ing “
Mo
bile
Ph
on
e O
wn
ers
hip
”
Challenges in averting neonatal mortality –
being at the right place, at the right time…
•1st Day – 50% of deaths
•1st Week – 75% of deaths
Source: Lawn JE et al Lancet 2005, Based on analysis of 47 DHS datasets (1995-2003), 10,048 neonatal deaths)
“Hot Zone”
m-Labor
Notification
System
Pilot Study
Source: Gernand, JiVitA Data 2011
(Unpublished)
306 (88.9%)
Births Attended
Tremendous time and effort is invested in manual data collection, aggregation and reporting.
Example: Bangladesh CHW’s 19 ledgers contain 473
unique data fields.
Only 60 fields are unique, required for a digital system
to process the same information.
Census Enumeration
Smart Scheduling of Daily Activities, by Priority
Assessing pregnancy status
2.5 minutes saved for a SINGLE task resulted in ~13 FTEs over a district.
X X /60=
mCARE: Integrated Community-Health Worker System to ImproveAntenatal & Postnatal Care and Increase Client Demand
Allow clients to report data to the system
Try it:
Text / SMS “birth” to
1(443) 393-2228
25.2
65.8
74.8
34.2
Non-interventiongroup (n=135)
Intervention group(n=193)
June 2014 - Preliminary Results:Antenatal Care Utilization
Received Not-received
mTikka: Virtual Vaccine Registry and Immunization Improvement System
Partners:
! " # $ %&' ( ) $ * %+ $
, $
- $
. , $
. - $
/ , $
/ - $
0 , $
0 - $
1 , $
1 - $
A
B
C DE
F
G
Pregnancy Surveillance
Pregnancy registration & survey of vaccination beliefs
Birth notification Reminder for upcoming vaccination
EPI camp open notification
Up-to-date vaccination record
Timely availability of performance indicators
! "#$$%&%'(#)*+%,-'#. */ )%0*#( / -'! 12#,/ 3451. / '2%6/ 7'89! ! +. #:%*#1. '; / 01)7<'*1'
#! 4)1( / '(%00#. %*#1. )%*/ 6#. ')+)%,'=%. >,%7/ 65
?)@'A,%#. 'B%2)#C+/
9. D%. *6#. )/ 61+)0/ 411) 6/ **#. >6
E9'F
G10+6F
E)1H/ 0* I ( / )(#/ J F! " #$$%#6 %. / ,/ 0*)1. #0 0,1+73
9. D%. *6'#. ')/ 61+)0/ 3411)'6/ **#. >6@
KL'! 1. *56'MLNKL'3LNKOP
; +)%,'=%. >,%7/ 65
G10+6'F
E)1H/ 0*'E/ )#17'F
B10%*#1. 'F
E)1H/ 0* I ( / )(#/ J F! " #$$%#6 %. / ,/ 0*)1. #0- 0,1+7
2%6/ 7 6Q6*/ ! D1) #. D%. * )/ >#6*)%*#1. - (%00#. %*#1.
)/ 01)7$/ / 4#. >- #. 0/ . *#(#: %*#1. - %. 7 6+)(/ Q 1D
(%00#. %*#1. 2/ ,#/ D6@! " #$$%J#,, D10+6 1. *5/ 411)/ 6*
%. 7 5%)7/ 6* *1 )/ %05 6/ >! / . *6 1D*5/ 414+,%*#1. -
5/ ,4#. > *1 #7/ . *#DQ M#. )/ %,3*#! / P )/ >#1. 6 J5/ )/
(%00#. / 01(/ )%>/ #6 ,#! #*/ 7- %. 7 *1 4/ )! #*
01! ! +. #*Q32%6/ 7- *%)>/ */ 7 #. */ )( / . *#1. 6 *5%* J#,,
D#. 0)/ %6/ *5/ )/ %05 1D #! ! +. #: %*#1. 6@R%00#. /
J1)$/ )6 MRS 6P J#,, +6/ *5/ ! " #$$% %44,#0%*#1.
,1%7/ 7 1. ,1J3016* 6! %)* 451. / 6 *1 )/ >#6*/ ) %.
#. D%. * %* 2#)*5@! " #$$% J#,, *5/ . >/ . / )%*/ %.
%+*1! %*/ 7 (%00#. %*#1. %. 7 )/ ! #. 7/ ) 605/ 7+,/ -
0+6*1! #: / 7 *1 *5/ #. D%. *T6 2#)*5 7%*/ - *1 / . 6+)/
*#! / ,Q(%00#. %*#1. @?+/ *1 #*6 ,10%*#1. 1. *5/ 0,1+7
7%*%2%6/ *5/ #. D%. *T6 (%00#. %*#1. 6*%*+6 0%. 2/I 2H/ 0*#( / 6 F" 5/ >1%, 1D! " #$$%#6 *1 / . 6+)/ *#! / ,Q
(%00#. %*#1. %! 1. > #. D%. *6 #. )+)%, =%. >,%7/ 65 %. 77%*%2%6/ - *5/ #. D%. *6 (%00#. %*#1. 6*%*+6 0%. 2/
%00/ 66/ 7 %. Q*#! / %. 7 %. QJ5/ )/ - 2Q 60%. . #. > %
+. #C+/ 2%)017/ %66#>. / 7 *1 / %05 #. D%. *@E%)/ . *6
J51 1J. 0/ ,, 451. / 6 J#,, 2/ %2,/ *1 ! 1. #*1) %. 7
)/ 0/ #(/ )/ ! #. 7/ )6 %21+* *5/ (%00#. %*#1. 6*%*+6 1D
*5/ #) #. D%. *6 %. 7 J#,, )/ 0/ #(/ 0%65 #. 0/ . *#(/ 6 D1) *5/
*#! / ,Q 01! 4,/ *#1. 1D *5/ 4)/ 60)#2/ 7 (%00#. %*#1.
605/ 7+,/ 2Q *5/ #. D%. *T6 D#)6* 2#)*57%Q@" 5/ 6! %)*
D D
(%00#. %*#1. %! 1. > #. D%. *6 #. )+)%, =%. >,%7/ 65 %. 7
%(/ )* ! 1)2#7#*Q %. 7 ! 1)*%,#*Q 7+/ *1 4)/ (/ . *%2,/
7#6/ %6/ 6@! " #$$% J#,, %77)/ 66 D1+) 2%))#/ )6 *1
605/ 7+,/ 7 (%00#. %*#1. 01! 4,#%. 0/ &
K@9. %2#,#*Q*1 *)%0$%05#,7T6(%00#. %*#1. 5#6*1)Q-
L@B%0$1D*#! / ,Q)/ ! #. 7/ )6-
O@B%0$1D#. 0/ . *#(/ 6*1 4%)/ . *6-%. 7
U@B#! #*/ 7 %J%)/ . / 66%21+*(%00#. / / DD/ 0*#(/ . / 66451. / #. */ )D%0/ J#,, %,61 2/ ,/ ( / )%>/ 7 D1) 4)/ 6/ . *#. >
/ 7+0%*#1. %, ! 17+,/ 6 4/ )*%#. #. > *1 (%00#. /
/ DD/ 0*#(/ . / 66*1 4%)/ . *6@
V1. */ W*F9. LNKK- *5/ (%00#. %*#1. 01(/ )%>/ )%*/ 6 D1)
D+,,Q#! ! +. #: / 7 =%. >,%7/ 65#05#,7)/ . 2/ *J / / . KL *1
LO! 1. *561D%>/ J / )/ XYZ @[ 1J/ (/ )- #. )+)%, 5%)73
*13)/ %05 %)/ %6 / (#7/ . 0/ 6+>>/ 6*6 *5%* 01! 4,/ */
32Q+6#. > %0,1+732%6/ 7 #. D%. * )/ >#6*)%*#1. 6Q6*/ ! -
\ ] \ 32%6/ 7 )/ ! #. 7/ )6 *1 4%)/ . *6 %. 7 (%00#. /
J1)$/ )6- 0%65 #. 0/ . *#(/ 6 *1 4%)/ . *6 J516/ 05#,7)/ .
%)/ (%00#. %*/ 7- %. 7 5%). / 66#. > *5/ ! 12#,/ #. */ )D%0/
D1) 5/ %,*5 / 7+0%*#1. @
^W4/ 0*/ 7 I +*01! / 6F
• A 0,1+732%6/ 7 #. D%. * )/ >#6*)%*#1. 6Q6*/ ! +6#. >*1 )/ %05 %)/ %6 / (#7/ . 0/ 6+>>/ 6*6 *5%* 01! 4,/ */
#! ! +. #: %*#1. )%*/ 6 %)/ 6#>. #D#0%. *,Q,1J / ) %. 7 )%. >/
D)1! UUZ *1 YNZ @\ *+7#/ 6 651J *5%*01(/ )%>/ )%*/ 6
D+)*5/ ) 7/ 0,#. / D1) D1,,1J3+4 716/ 6@] %H1) 126*%0,/ 6
*1 %05#/ (#. > 5#>5 #! ! +. #: %*#1. 01(/ )%>/ #.
=%. >,%7/ 65 #. 0,+7/ F#))/ >+,%) ^E9 605/ 7+,/ 6- D/ J
6*)%*/ >#/ 6 *1 %00/ 66 411) )/ 61+)0/ 6/ **#. >6- %. 7
,#! #*/ 7 01! ! +. #*Q/ . >%>/ ! / . *J#*5 *5/ ^E9@
• A 0,1+732%6/ 7 #. D%. * )/ >#6*)%*#1. 6Q6*/ ! +6#. >
%. 7)1#7 451. / 6D1) 7%*%/ . *)Q@
• " #! / ,Q(%00#. %*#1. 1D#. D%. *6 *5)1+>5 \ ] \ 32%6/ 7
)/ ! #. 7/ )6*1 4%)/ . *6%. 7 (%00#. / J1)$/ )6@
• ^! 41J / )#. > 4%)/ . *6 *1 %00/ 66 *5/ #) 05#,7T6
(%00#. %*#1. )/ 01)76@
• 97/ . *#D#0%*#1. 1D2%))#/ )6- %. 7 #! 4,/ ! / . *%*#1. 1D
*%)>/ */ 7- 01! ! +. #*Q32%6/ 7 #. */ )( / . *#1. 6 *1
G+. 7#. >F_)%. 7 V5%,,/ . >/ 6 ^W4,1)%*#1. _)%. * - =#,,
%. 7 ] / ,#. 7%_%*/ 6G1+. 7%*#1.
#. 0)/ %6/ (%00#. / 01(/ )%>/ @
GoB National Health
Information System
Emerging “Lessons”
• User-centered / User-engaged design
• Extensive formative research & workflow mapping
• Iterative technical deployment and stabilization
• Early government and community engagement
• Mixed-methods evaluation
• Plan for technical failures / build-in system
redundancy
• “Control” systems to prevent & monitor misuse
mHealth doesn’t work in a Vacuum
PROVIDERHEALTHSYSTEM
PATIENTAccess to information
Behavior changeActivity MonitoringSelf-reported Data
Workflow managementDecision Support
Surveillance and TrackingRemuneration / Incentives
Workforce monitoringReal-time Data Streams
Supply-chain management
Providing families access to timely information
“If you have any bleeding during this month, seek medical attention right away”
Expectant women/ new mothers sign
up for service
Users receive health-related
messages weekly
“Freemium” model to drive coverage
“Your baby needs an immunization this week
to stay healthy: Available free at all
EPI clinics”
Photo: Text to Change
Healthcare Worker Communication and Training
• Data collection and communication tools
• Multimedia courses and lectures• mLearning on Demand• Interactive Quizzes
www.emocha.org
Project Mwana: SMS to reduce Infant HIV PCR
Turnaround Time (46%)
Amader Gram (Our Village) Breast Care
• Educate• Identify• Refer• Track
SmartRegister.org
Emphasis on user-focused design to facilitate FHW utilization and feedback.
Nutrition (6) >
Integrate workforce and client training as part of the exposures
New frontiers!
• US FDA Approved• 2-lead ECG
New frontiers!
Remote, Point-of-care Diagnostic tools
Breslauer D., et al. 2009 Mobile Phone Based Clinical Microscopy for Global Health Applications. PLoS ONE 4(7): e6320
Mobile-based Flow Cytometry
Ozcan Research Group (Nano-Bio Photonics / UCLA): Optical imaging techniques for point-of-care diagnosticsHongying Zhu , Serhan O. Isikman , Onur Mudanyali , Alon Greenbaum and Aydogan Ozcan Lab Chip, 2012, Advance Article
62
Agriculture
Health
Money
Research
m
Information
Social Networks
Entertainment
New paradigms for health data collection
Blood chemistry
Urinalysis
+
Medication adherence
Vital signs
Movement, activity
ECG
Body weight, mass
The Gartner “Hype” Cycle
Fenn J, Maskino M: When to Leap on the Hype Cycle. Gartner Group 2008.
“Pilotitis”
Healthy mSkepticism
The Bellagio eHealth Evaluation Declaration 2011
“Rigorous evaluation of e- & m-Health is necessary to generate useful evidence and
promote the appropriate integration of technologies to
improve health and reduce inequalities.”
Bellagio Call to Action 2011
If used improperly, eHealth may divert
valuable resources and even cause
harm… implementation must be
guided by evidence…
“mHealth tools and interventions must be backed up by rigorous scientific development, evaluation, and evidence generation to enhance meaningful innovation and best practices, and to validate tools and methods for health professionals, consumers, payers, governments, and industry.”
Why “Evidence” ?
1. Health investments in global health are driven by more than market forces
2. Limited resources = Need for stringent, cost-effectiveness based planning
3. Two decades of Emphasis on EBD !
4. Donors: Increased transparency / scrutiny
5. Population-side demand for improved quality
6. e-Health / ICT induced political fatigue
Evidence for whom ?
Is there evidence ? Who is asking the question ?
Improving the Evidence for Mobile Health, 2011
Evidence of what ?
“Maturity” of the mHealth Project
Am
ou
nt
of
Info
rmat
ion
(R
ED)
Threshold of “Information”
Stability Functionality Useability Efficacy Effectiveness
Methodology
Systems Engineering Qualitative Quantitative Mixed Q/Q / M&E
“Evidence” Across The mHealth Maturity Lifecycle
OF WHAT ?
MEASURED HOW ?
mHealth Technical Evidence Review Group for RMNCH
“m-TERG”
“Providing governments and implementing agencies
objective, evidence-based guidance for the
selection and scale of mHealth strategies
across the reproductive, maternal,
newborn and child health continuum”
INTERVENTIONOF KNOWN
EFFICACY
EFFECTIVECOVERAGE
mHEALTH: A Health Systems Catalyst
Jo Y, Labrique AB et al. PLOS One 2014
Shift focus from “Does mHealth work?” to “Does mHealth optimize what we know works ?”
Need for Structure
Step 1: Develop a common vocabulary
Help us as innovators, researchers, funders talk about mHealth…
A Taxonomy for mHealth
What is the problem we’re trying to solve ?
AVAILABILITY
4.2.1 Supply of commodities
4.2.2 Supply of services
4.2.3 Supply of equipment
4.2.4 Diversity of treatment options
INFORMATION
4.1.1 Lack of population
enumeration
4.1.2 Delayed reporting of
events
4.1.3 Quality/unreliability of
data
4.1.4 Communication
roadblocks
4.1.5 Access to information or
data
COST
4.7.1 Expenses related to
commodity production
4.7.2 Expenses related to
commodity supply
4.7.3 Expenses related to
commodity disbursement
4.7.4 Expenses related to service
delivery
4.7.5 Client-side expenses
UTILIZATION
4.5.4 Loss to follow up
4.5.1 Demand for services
4.5.2 Geographic inaccessibility
4.5.3 Low adherence to treatments
ACCEPTABILITY
4.4.3 Stigma
4.4.1 Alignment with local norms
4.4.2 Addressing individual beliefs
and practices
EFFICIENCY
4.6.1 Workflow management
4.6.2 Effective resource allocation
4.6.5 Timeliness of care
4.6.3 Unnecessary referrals/
transportation
4.6.4 Planning and coordination
QUALITY
4.3.1 Quality of care
4.3.3 Quality of Commodity
4.3.4 Health worker
motivation
4.3.2 Health worker
competence
4.3.6 Supportive supervision
4.3.5 Continuity of care
mHealth Strategy Intermediate Outcome Outcome / Impact
Provider Competence,
Accountability,
Effectiveness.
Client Knowledge
and Self-Efficacy
Improved
Health Outcomes
Improved
Quality
of Care
Improved
Health
Behaviors
Disease Surveillance
Electronic Medical Records
Remote Monitoring
Logistics monitoring and tracking
Decision Support Systems
Point-of-care Diagnostics
Appointment Scheduling
Client reporting of quality / performance
On-Demand Training / Assessment
Client Education
On-demand Information / Helplines
Supply Chain Integrity
Accuracy of Information
Continuity of Care
Affordability of Care
Financing (Banking, Insurance)
Enhanced Counseling
Improved
Efficiency /
Coverage
Vital Statistics ReportingImproved
Population
Health
Real-time Data Access / PHRC
LIE
NT
PR
OV
IDE
R
H
EA
LT
H S
YS
TE
M
Remote Consultation
Improved Dem. / Hlth. Data
Appropriate Resource Alloc.
Policy Adjustments
Workflow Management Systems
Responsive
Health System
Is your “mHealth” the same as my “mHealth” ?
Why a mHealth and ICT Framework for RMNCH?
•Allows focus on health systems strategy of the mHealth innovation, not just the technology.
•Provides projects with a communication tool when talking with different stakeholders, including governments about what mHealth offers.
•Allows identification of uniqueness, commonalities and gaps across multiple mHealth projects through the use of a consistent and health systems-focused vocabulary.
12 Common mHealth Applications
RMNCH Continuum:
Known Interventions
mHealth Strategy: …overcoming
these constraints:
Touching these
“actors” in the
system:
Labrique, Mehl, Vasudevan et al. 2013 (MS in Review)
Labrique, Mehl, Vasudevan et al. 2013 (MS in Review)
Step 2: Develop repositories of m-evidence and m-activities
Help to identify, collate and grade the quality of information on mHealth
strategies
What do we know ? What has been tried ?
mHealthEvidence.org / mHealthKnowledge.org
Helping to Consolidate efforts Globally
And other partners…
MREGISTRY.ORGA Global mHealth Project Registry
Step 3: Facilitate the review and synthesis of evidence
Help to understand when sufficient information exists to recommend
mHealth as part of the standard of care
What kind of evidence ?
mTERG Criteria for Grading mHealth Information Quality
Step 4: Create tools to help with structured evaluation, common
indicators moving forward
Develop Common Indicators and Measurement Standards for mHealth Projects
Agarwal et al. mHS 2013
Evidence Prioritization Summary
mHealth strategies likely to demonstrate:
• improved client access to information• enhanced traditional methods of counseling and BCC• bolstered client adherence to medication, and attendance to
scheduled appointments• shortened turnaround time for performance data submission• improved workforce scheduling, monitoring and accountability• improved workforce training and continued education• supported caregivers through decision support tools• strengthened commodities supply chains and reduce risk of
stockouts• created shorter feedback loops for systemic response
“mHealth Extends REACH, Creates CONVENIENCE, Shortens INFORMATION lag, and Facilitates TARGETTED CARE when and where its needed.”
mTERG
Where can we have the most impact ?
Mehl G, Labrique AB. Science Sept. 2014.
An Ecosystem of mTools for
Cross-Sectoral Development exists!
“m” – spans Health, Agriculture, Education,
Politics, Finance, Data Collection
Eras of mHealth
I
Innovation and Experimentation
II
Discordant Proliferation
III
Scrutiny and Consolidation
IV
Integration and Scale
Degree to which the mHealth strategy changes the status quo
INCREMENTAL CHANGE DISRUPTIVE INNOVATION
DIF
FIC
ULT
Y O
F SC
ALI
NG
CO
MP
LEX
ITY
OF
ENG
AG
ED E
CO
SYST
EM
INST
ITU
TIO
NA
L /
HEA
LTH
SYS
TEM
INER
TIA
Challenges- Tentative funding for pilots and
demonstrations, limited investment in
scale
- Rapidly growing, complex ecosystem
with new non-health actors
- Duplicative efforts, lack of
interoperability
- Siloes of innovation, without clear
pathways to integration
- Economic evaluations of mHealth
interventions are lacking
• For scale-up / Mainstreaming of mHealth, we need to:
• …Reach BEYOND the “converted”Speak the language of HEALTH decision-makers
• …STOP taking shortcuts – measuring attributable impact or cost is not an afterthought, an inexpensive or easy task.
• …SUPPORT a high threshold of information quality, establishing new methods where appropriate, but aligning claims with data.
Two last thoughts
A phone… as a phone !
From this…
To this ?
More data is not better
information.
Draw inspiration from Botswana and Bangladesh to Brussels and Baltimore to
understand what is m…… POSSIBLE
Thank you.
http://tinyurl.com/mpossible-video
Follow a robust process
USERS
• Identify Users
•Define Target Population
ROLES
•Define Roles
•Map Workflow / Scheduling rules
DATA
•Map Data “Universe”
•Deconstruct data elements
OPTIMIZE
•Assess Data Efficiency
• Identify opportunities for Optimization
DESIGN
•End User Engagement
•User-Acceptability / Functionality
BUILD
•Program, Deploy, Test
•Evaluate
Other Tools: Balsamiq.com
Other Tools: Captricity