J A C C : C A R D I O V A S C U L A R I M A G I N G VO L . 1 1 , N O . 4 , 2 0 1 8

ª 2 0 1 8 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

A Randomized Trial ofPocket-Echocardiography IntegratedMobile Health Device Assessments inModern Structural Heart Disease Clinics

Sanjeev P. Bhavnani, MD,a Srikanth Sola, MD,b David Adams, RCS, RDCS,c Ashwin Venkateshvaran, PHD,b

P.K. Dash, MD,b Partho P. Sengupta, MD, DM,d for the ASEF-VALUES Investigators

ABSTRACT

ISS

Fro

Wh

He

Am

Qu

We

Dr

Ro

the

Ap

Ma

OBJECTIVES This study sought to determine whether mobile health (mHealth) device assessments used as clinical

decision support tools at the point-of-care can reduce the time to treatment and improve long-term outcomes among

patients with rheumatic and structural heart diseases (SHD).

BACKGROUND Newly developed smartphone-connected mHealth devices represent promising methods to diagnose

common diseases in resource-limited areas; however, the impact of technology-based care on long-term outcomes has

not been rigorously evaluated.

METHODS A total of 253 patients with SHD were randomized to an initial diagnostic assessment with wireless devices in

mHealth clinics (n ¼ 139) or to standard-care (n ¼ 114) in India. mHealth clinics were equipped with point-of-care devices

including pocket-echocardiography, smartphone-connected-electrocardiogram blood pressure and oxygen measure-

ments, activity monitoring, and portable brain natriuretic peptide laboratory testing. All individuals underwent

comprehensive transthoracic echocardiography to assess the severity of SHD. The primary endpoint was the time to

referral for therapy with percutaneous valvuloplasty or surgical valve replacement. Secondary endpoints included the

probability of a cardiovascular hospitalization and/or death over 1 year.

RESULTS An initial mHealth assessment was associated with a shorter time to referral for valvuloplasty and/or valve

replacement (83 � 79 days vs. 180 � 101 days; p <0.001) and was associated with an increased probability for

valvuloplasty/valve replacement compared to standard-care (34% vs. 32%; adjusted hazard ratio: 1.54; 95% CI: 0.96 to

2.47; p ¼ 0.07). Patients randomized to mHealth were associated with a lower risk of a hospitalization and/or death on

follow-up (15% vs. 28%, adjusted hazard ratio: 0.41; 95% CI: 0.21 to 0.83; p ¼ 0.013).

CONCLUSIONS An initial mHealth diagnostic strategy was associated with a shorter time to definitive therapy among

patients with SHD in a resource-limited area and was associated with improved outcomes. (A Randomized Trial of

Pocket-Echocardiography Integrated Mobile Health Device Assessments in Modern Structural Heart Disease Clinics;

NCT02881398) (J Am Coll Cardiol Img 2018;11:546–57) © 2018 by the American College of Cardiology Foundation.

N 1936-878X/$36.00 http://dx.doi.org/10.1016/j.jcmg.2017.06.019

m the aScripps Clinic and Research Foundation, San Diego, California; bSri Sathya Sai Institute of Higher Medical Sciences,

itefield, Bangalore, India; cDuke University School of Medicine, Durham, North Carolina; and the dWest Virginia University

art and Vascular Institute at West Virginia University School of Medicine, Morgantown, West Virginia. Supported by The

erican Society of Echocardiography Foundation. Dr. Bhavnani has received an educational and research grant from the

alcomm Foundation to Scripps Health; is a consultant to Proteus Digital; and is an advisory board member to iVEDIX,

llSeek, and Misceo. Dr. Sola has received a research grant from General Electric Healthcare (outside of this investigation).

. Sengupta has received research grants from Heart Test Labs and Echo Sense Ltd.; and is a consultant to TeleHealth

botics, Intel, Hitachi Aloka, and Heart Test Labs. All other authors have reported that they have no relationships relevant to

contents of this paper to disclose. A complete list of investigators in the ASEF-VALUES study is provided in the Online

pendix. Maurice Enriquez-Sarano, MD, served as Guest Editor for this paper.

nuscript received March 30, 2017; revised manuscript received June 19, 2017, accepted June 19, 2017.

AB BR E V I A T I O N S

AND ACRONYM S

ASEF = American Society of

Echocardiography Foundation

BNP = brain natriuretic peptide

iECG = iPhone-

electrocardiography

mHealth = mobile health

POC = point-of-care

SHD = structural heart disease

SSSIHMS = Sri Sathya Sai

Institute of Higher Medical

Sciences

TTE = transthoracic

cardiogram

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8 Bhavnani et al.A P R I L 2 0 1 8 : 5 4 6 – 5 7 mHealth in Structural Heart Disease

547

T he transformative potential for cellular tech-nologies, expanding internet connectivity,and the development of innovative mobile

health (mHealth) devices to improve health care de-livery in resource-limited areas is promising (1,2). Asthese areas begin to leverage new digital infrastruc-tures for health care, several key factors haveemerged. These include: 1) to identify the heuristicfactors and evaluative methods that lead to appro-priate use of new technologies; 2) to determine theintegration of device-based findings into existinginformational systems and health records; 3) todemonstrate the patterns of effective use at thepoint-of-care; and 4) to identify those patterns thatlead to earlier diagnostic and treatment decisions(3,4). In the aggregate, an emphasis on the determin-istic approaches of mHealth must include pragmaticdevice use and outcomes-based assessments as newtechnology-based health care initiatives are orga-nized (5,6).

SEE PAGE 558

Recent shifts in the global burden of cardiovasculardiseases have led to an increasing prevalence inresource-limited areas with more than 25 milliondeaths in these regions predicted by 2030 (7). Thisproblem is further compounded with resource-limitedareas receiving a disproportionately low allocation ofglobal resources ranging from the availability ofappropriate diagnostic tests to sufficiently trainedhealth care professionals (8). Portability, lower cost,and simple-to-use form factors are among the designfeatures of mHealth that may be well suited to bridgethese inequalities, and to mobilize care from hospital-and clinic-based encounters to the practitioner at thepoint-of-care and in remote locations (9). Althoughattractive from a technological perspective, the impactof mHealth used as a practitioner-based clinical-deci-sion support tool on long-term outcomes has not beenrigorously evaluated (10).

Therefore, the objective of the present study was tocompare the outcomes of mHealth with smartphone-connected devices and pocket-echocardiography onmedical decision making among patients with rheu-matic and structural heart disease (SHD) in a healthcare system of a resource-limited area.

METHODS

STUDY DESIGN. The study was performed underthe ASEF-VALUES (American Society of Echocardio-graphy Foundation–Valvular Assessment Leading toUnexplored Echocardiographic Stratagems) program—

a philanthropic and educational initiative to explore

health care solutions for patients with SHDusing new technologies. Within the programwas a nested, single-site, randomized trialconducted at the Sri Satya Sai Institute ofHigher Medical Sciences (SSSIHMS), a chari-table, free-of-charge, tertiary-care, andteaching institution in Bangalore, India thatexclusively provides care to the underserved,sees more than 20,000 SHD patients per year,and performs 1,100 percutaneous valvulo-plasties and 1,000 valve replacements on anannual basis.

The primary study sponsors were the ASEFand SSSIHMS. General Electric Healthcare(Bangalore, India) provided local instruments

and logistical support. Additional device support wasprovided by CoreSound Imaging (Raleigh-Durham,North Carolina) and iHealth (San Francisco, Califor-nia). Five cardiologists and 12 sonographers from 12academic medical centers across the United States, 15cardiologists and cardiothoracic surgeons fromSSSIHMS, and 30 cardiologists from across Indiaparticipated in the study.

PARTICIPANTS. The study participants were out-patients with a new or an established diagnosis of SHD.The definition of SHD included valvular disease, left/right ventricular failure and congenital heart defects,and included adult, pediatric, and pregnant patients.We decided a priori to include SHD patients with aprior valvuloplasty or valve replacement. Exclusionsincluded neonatal patients and those with an unstablehemodynamic status. All subjects provided writteninformed consent in their native language.

TRIAL ORGANIZATION, RANDOMIZATION, AND

MASKING. Consecutive subjects were randomlyassigned to an initial evaluation with mHealth or tostandard care. Study subjects were evaluated in either1 of 10 (5 mHealth, or 5 standard care) clinical sites alllocated at SSSIHMS. Each site used for a patientencounter after randomization was an individualclinic. We decided to create mHealth and standard-care sites within 1 hospital to minimize variabilitywith the initial clinical encounter (mHealth orstandard care) after randomization, and separatedthese clinics to reduce any potential bias introduced byusing mHealth devices. All mHealth clinics wereequipped with the same devices and used theinstitution’s electronic medical record to standardizeworkflow the data generated during the trialencounter. The standard-care clinics were designedwith pragmatic intention and to mimic usualcare practice patterns from a group of physiciansacross India. To minimize confounding resulting

echo

Bhavnani et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8

mHealth in Structural Heart Disease A P R I L 2 0 1 8 : 5 4 6 – 5 7

548

from the same physician participating in the clinicalassessments as well as conducting procedures, physi-cians performing mHealth or standard care assess-ments did not participate in procedural interventionsassessments and vice versa.

Following the clinical encounter, with mHealth orstandard-care, all study participants underwent acomprehensive transthoracic echocardiogram (TTE)for the severity of SHD. The TTE was performed on thesame day and was interpreted either on the sameday or within 24 h in both randomized arms (OnlineFigure 1). The decision to complete TTE on the sameday as randomization was to: 1) eliminate the time todiagnostic testing with TTE as a variable that iscommonly observed in resource-limited areas; 2)minimize the time to TTE as a factor resulting intreatment delays; 3) determine the yield of diagnosticinformation provided by mHealth or standard care onthe referral rate for treatment (valvuloplasty or valvereplacement) and; 4) assess the impact of mHealth andstandard care on medical decision making at the timeof enrollment and on follow-up. A randomizationschedule was created (based on hospital outpatientestimates) that on a given day outpatient numberswould not exceed a maximum 400 patients, and wasformulated using a simple randomization scheme(random number generator SPSS version 23.0 [IBMCorporation, Armonk, New York]).

To ensure concealment of allocation that wouldotherwise introduce selection bias despite randomi-zation, randomization was performed by study staffnot involved in the study and was concealed until theprimary and secondary endpoints were analyzed.Given the diagnostic and treatment procedures in thepresent study, it was important to differentiate thosecardiologists who performed the initial diagnosticassessment (mHealth or standard care) from thosecardiologists and surgeons who performed valvulo-plasty or valve replacement. The purpose of theinitial assessment was to allow participating physi-cians to make clinical decisions based on mHealth orstandard-care findings. The treatment plan andreferral for intervention was generated by these car-diologists conducting the initial assessment and wasformulated through the aggregate of diagnostic in-formation available at the time of enrollment. FormHealth it was history and the findings on activitymonitoring, pocket ultrasound, smartphone electro-cardiograph (ECG) and point-of-care brain natriureticpeptide (BNP) (the latter if applicable), and for stan-dard care the usual physical examination findingsand diagnostic tests on follow up. Subsequently,operating interventional cardiologists and surgeons(different than those performing the initial

assessment) were blinded to a study subject’s groupallocation; however, they reviewed the findings onTTE for diagnostic accuracy at the time of plannedpercutaneous intervention or surgical procedures.

STUDY PROCEDURES. Initial mHealth assessments. EachmHealth clinic was equipped with wireless mHealthdevices that were selected to assess functional andstructural abnormalities at the point-of-care (OnlineFigure 2) including: 1) pocket echocardiography(VScan, General Electric, Whitefield, India); 2) vitalsigns with smartphone-connected oximetry and bloodpressure monitors (iHealth, San Francisco, California);3) 6-min walk test with a trial-axial activity monitor(Ozeri, San Diego, California); 4) cardiac rhythm ab-normalities were classified by a smartphone-connected-iECG (AliveCor, San Francisco, California);and 5) point-of-care testing with fingerstick B-typenatriuretic peptide (Alere Triage, Gurgoan, India).Pocket echocardiography. All participating physiciansand study staff received training on the use of mHealthdevices before initiation of the program. Point-of-careechocardiographic examinations were performedusing the VScan, a pocket-sized device. Scans wereperformed by local physicians to execute a protocolconsisting of 11 standard views including color-flowDoppler images of all valves (11), and were trainedby ASE sonographers similar to the trainingmethodology used in the ASEF-VISION (Value ofInteractive Scanning for Improving Outcomes of NewLearners) study (6). By design, ASE sonographers didnot participate in pocket-echocardiographic imageacquisition and local physicians were independentwhen using pocket-echocardiography in the mHealthassessment and for clinical decisions. The VScan is ahandheld-sized device (135 � 73 � 28 mm) thatweighs 400 g and has an 8.9-cm (diagonal) displaywith a resolution of 240 � 320 pixels. The device usesa phased-array transducer (1.7 MHz to 3.8 MHz) anddisplays gray scale images with a sector width of 75�

and color Doppler images with a fixed sector width of30�. Current-generation devices do not have thecapabilities of spectral Doppler or M-mode imaging.Qualitative assessments (mild, moderate, or severe)of chamber size, volumetric estimations and severityof valvular stenosis, regurgitation, and left and rightventricular dysfunction were interpreted at the timeof the examination. Left ventricular ejection fractionwas stratified into normal $55% or low if it was <55%by visual estimation and the presence of valvularabnormalities (regurgitant or stenotic) and severity(mild, moderate, or severe), or mitral stenosis(progressive, severe, very severe) were recordedaccording to ASE recommended definitions. Theseverity of regurgitant lesions was based on

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8 Bhavnani et al.A P R I L 2 0 1 8 : 5 4 6 – 5 7 mHealth in Structural Heart Disease

549

2-dimensional findings (atrial or ventricularenlargement, hyperdynamic left ventricle) andqualitative color Doppler findings (width of venacontracta and jet area), whereas the severity ofstenotic lesions was based on 2-dimensional findingsof valve opening and leaflet mobility, thickness, andcalcification.Activity monitoring. An assessment of functional ca-pacity was quantified by a tri-axial activity monitorthat calculates the number of steps, duration ofactivity (minutes), distance (meters) and gait speed(miles per hour) to quantify the level of activity on a6-min walk test (6MWT). The 6MWT was conductedby nurses trained in device use, data acquisition, andwas performed along a fixed distance of 50 m markedat 10-m intervals (12). Internal quality control ofactivity measurements to fixed distances wereperformed daily with study staff as volunteers. Error>10% required removal and replacement of thedevice (none required in the study). To quantify theNew York Heart Association (NYHA) functional class,the following calculations were applied: gait speed(mph) was automatically calculated from the distancewalked in meters over the duration of activityachieved. NYHA functional class was stratified by gaitspeed into the following categories: NYHA functionalclass I $2.2 mph, NYHA functional class II ¼ 1.5 to 2.2mph, NYHA functional class III #1.5 mph, and NYHAfunctional class IV symptoms at rest (13).Vital signs. Smartphone-connected oxygen and bloodpressure monitors provided assessments of oxygensaturation, and blood pressure at rest and with exer-tion. Oxygen saturation was measured by photo-plethysmography and blood pressure withoscillometric measurements and automated inflation.The resulting measurements are displayed on a tabletapplication.Cardiac rhythm. A smartphone-connected iPhone-ECG (iECG) was used to determine the heart rate andcardiac rhythm (2). In general, the iECG produces asingle-lead ECG when held with the left and rightfingers (lead I) or placed directly on the chest wall fora precordial lead. The ECG recording was transmittedto the tablet using frequency modulation of theelectrical signal to ultrasound. Capture of this soundsignal on the tablet’s microphone produces a real-time cardiac rhythm on the tablet’s display.Diagnostic findings were classified as an atrialarrhythmia (atrial fibrillation or atrial flutter, supra-ventricular tachycardia, or ventricular arrhythmias)or bradyarrhythmia (second- or third-degree atrioven-tricular block, or sinus node dysfunction).Natriuretic peptide levels. Fingerstick B-type natri-uretic peptide (BNP) was performed among select

patients with findings of severe mitral stenosis,nonequivocal symptoms on functional assessments.The test is a rapid, point-of-care fluorescenceimmunoassay used to measure BNP in K2

Ethylenediaminetetraacetic acid anticoagulatedwhole blood droplets. Results are displayed within15 min on a miniaturized, portable, and batterypowered device (14).Initial standard-care assessments. The standard-careclinics used available resources including a 12-leadECG, radiographs, and laboratory testing asrequired. Participating cardiologists who have expe-rience treating patients with SHD conducted theinitial clinical examination, interpreted all point-of-care assessments, made preliminary medical andsurgical treatment decisions, and determined thefrequency of follow-up visitations.Transthoracic echocardiography and cloud-based andpaperless reporting. After enrollment, all subjects un-derwent a comprehensive transthoracic echocardio-graphic examination (General Electric Vivid-E9,Philips Healthcare-ie33) performed by onsite ASEsonographers according to ASE guidelines (15,16).Local and ASE cardiologists interpreted all echocar-diographic studies by using a paperless and cloud-based system as previously reported (5) (OnlineAppendix). mHealth devices, with the exception ofpocket-echocardiography and point-of-care BNP,were connected via Bluetooth to a tablet computer(Samsung Galaxy Ta, Samsung Electronics, Seoul,South Korea). The mHealth clinics were designed tooperate without WiFi and used only local powersupply. The institutional electronic medical record(Enterprise Manage, Computer Science Corporation,Bangalore, India), a standards-based, modular appli-cation running on Oracle11G database, was modifiedwith templates to input all mHealth and standard-care findings to facilitate paperless data collection.

OUTCOMES

The primary outcome was the time to treatment withvalvuloplasty or valve replacement over 12-monthsafter the initial mHealth or standard-care assess-ment. Secondary outcomes included the occurrenceof a cardiovascular hospitalization and/or death onfollow-up. The primary investigators at SSSIHMSadjudicated all clinical endpoints and determined thenecessity for percutaneous or surgical treatment.Outcomes were obtained at the time of a procedure orwere determined by telephone, text message, or bycommunity health worker visitation to the home.

STATISTICAL ANALYSIS. Our hypothesis was that aninitial assessment with mHealth would result in a

FIGURE 1 CONSORT Diagram

254 consecutive patients withSHD

Randomized (n = 253)

139 Were assigned to an initialassessment with mHealth

At 12-month follow-up129 (93%) completed study0 withdrew consent10 (7%) were lost to follow-up

139 (100%) Were included in the analysis47 (34%) Underwent valvuloplasty and/orvalve replacement21 (15%) Experienced a cardiovascularhospitalization and/or death

114 (100%) Were included in the analysis38 (32%) Underwent valvuloplasty and/orvalve replacement32 (28%) Experienced a cardiovascularhospitalization and/or death

At 12-month follow-up105 (92%) completed study0 withdrew consent9 (8%) Were lost to follow-up

114 Were assigned to an initialassessment with standard-care

Enrollment

Excluded (n = 1)Neonate (n = 1)Declined to participate (n = 0)

Allocation

Follow-Up

Analysis

Enrollment, randomization, and follow-up of the study population. SHD ¼ structural heart disease.

Bhavnani et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8

mHealth in Structural Heart Disease A P R I L 2 0 1 8 : 5 4 6 – 5 7

550

shorter time to treatment. We performed a post hocpower calculation derived from the number of par-ticipants enrolled and time to primary endpoint.Based on the observed means and SDs across trialarms, our sample size had a power of 80% even if thetype I error rate was as small as 5.22 � 10-14. The fullrelationship between type I and type II error rates isshown in Online Figure 3. All analyses wereintention-to-treat based on randomized treatmentallocation. Descriptive analyses of continuous vari-ables are described as means and SDs and categoricalvariables as frequencies and percentages, and werecompared using the Student’s t test or the Mann-Whitney test, or the chi square or Fisher exact testswhere appropriate, respectively. Statistical compari-sons of the randomized groups were based on atime-to-first-event (primary outcome of the rate ofvalvuloplasty/valve replacement and secondaryoutcome of a cardiovascular hospitalization/death)that was reported with mean differences and the Cox

proportional hazard model. All outcomes wereadjusted for the presence of moderate or severemitral and/or aortic valve disease, a history of SHD,and a prior valvuloplasty/valve replacement. Relativerisks were expressed as means and adjusted hazardratios (AHRs). Additionally, the outcome of a hospi-talization and/or death was also computed usingvalvuloplasty/valve replacement as a time-dependentcovariate. Cumulative event rates were calculated foreach randomized group as a function of time fromrandomization with the use of the Kaplan-Meiermethod. Pre-specified subgroup analyses were per-formed according to relevant demographic variables:age (younger than or older than 40 years), sex, pres-ence of symptoms, history of SHD using the Coxmodel, and the interaction of individual mHealthdevices on outcomes using chi-square test associa-tions. All analyses reported 95% confidence intervals(CIs) where appropriate. The protocol received ethicsand Institutional Review Board committee approval

TABLE 1 Baseline Demographic Characteristics and Major Transthoracic

Echocardiographic Findings

mHealth Clinics(n ¼ 139)

Standard-Care Clinics*(n ¼ 114)

Clinical characteristics

Age, yrs 39 � 13 36 � 12

Female 59 (42) 48 (42)

NYHA functional class

I 72 (52) 58 (51)

II to III 67 (48) 56 (49)

IV 0 (0) 0 (0)

History of SHD 22 (16) 23 (20)

History of SHD with prior mitral valvuloplastyor valve replacement

47 (34) 26 (23)

History of mitral valve prolapse 1 (0.1) 0 (0)

History of atrial or ventricular septal defect 2 (0.1) 0 (0)

Coronary artery disease 8 (6) 6 (5)

Diabetes 4 (3) 3 (3)

Hypertension 9 (7) 3 (3)

Symptoms

Chest pain 60 (43) 44 (39)

Dyspnea 96 (69) 72 (64)

Syncope 12 (9) 8 (7)

Palpitations 53 (38) 51 (45)

Peripheral edema 6 (5) 10 (7)

Transthoracic echocardiographic findings

Left ventricular ejection fraction, % 56 � 14 54 � 12

Mitral valvular disease

Mitral stenosis

Progressive 27 (19) 21 (18)

Severe 32 (23) 29 (24)

Very severe 17 (12) 19 (16)

Mean transmitral pressure gradient, mm Hg 7.1 � 4.4 8.2 � 5.1

Mean mitral valve area, cm2 1.6 � 1.0 1.5 � 0.9

Mitral regurgitation

Mild 32 (23) 27 (23)

Moderate 14 (10) 14 (12)

Severe 10 (7) 9 (8)

Aortic valvular disease

Aortic stenosis

Mild 22 (16) 12 (10)

Moderate 17 (12) 5 (4)

Severe 1 (0.7) 2 (1)

Aortic valve area, cm2 1.8 � 0.8 1.8 � 0.7

Mean transvalvular pressure gradient, mm Hg 11.0 � 11.7 14.4 � 17.0

Aortic regurgitation

Mild 23 (17) 20 (22)

Moderate 23 (17) 16 (13)

Severe 7 (5) 9 (8)

Mixed mitral and aortic valve disease 31 (22) 26 (23)

Atrial or ventricular septal defect 8 (5) 7 (6)

Pericardial effusion 2 (1) 1 (1)

Values are mean � SD or n (%). All images were acquired and interpreted according to American Society ofEchocardiography guidelines. *No statistical significant differences were observed between randomizedstudy arms.

NYHA ¼ New York Heart Association; SHD ¼ structural heart disease.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8 Bhavnani et al.A P R I L 2 0 1 8 : 5 4 6 – 5 7 mHealth in Structural Heart Disease

551

and was registered with ISRCTN61479659 andNCT02881398.

RESULTS

STUDY POPULATION. All 253 study subjects wererecruited and randomized over 3 days betweenAugust 10, 2014, and August 12, 2014. Follow-up wascompleted in October 2015 and data analysis per-formed in January 2016 (Figure 1). A comprehensivemHealth assessment was performed in 139 subjectsand a standard-care assessment in 114. Data wasavailable from all devices with the exception of 5activity assessments that were deleted on the activitymonitoring devices.

CHARACTERISTICS AT BASELINE. The mean age ofthe study population was 39 � 14 years and 42% (107 of253) of participants were women. The mean follow-uptime was 337 � 116 days. The study population had asubstantial burden of disease with an establisheddiagnosis of SHD with or without prior valvuloplasty/valve replacement observed in 46% (118 of 253). Nearlyone-half of the study population were symptomaticwith 46% (116 of 253) exhibiting NYHA functional classII to III symptoms. Pertinent clinical characteristicswere well balanced with no statistically significantdifferences between randomized arms (Table 1).Transthoracic echocardiographic findings are listed inOnline Table 1. The presence of SHD with mitral ste-nosis, mitral regurgitation, aortic stenosis, or aorticregurgitation were frequently observed in 57%, 42%,23%, and 32% of transthoracic echocardiographicstudies, respectively. The prevalence of severe mitralor aortic valve disease was similar between random-ized groups.

INITIAL TESTING. Pertinent findings on mHealthdevices are listed in Table 2. Activity parameters weresignificantly lower among study subjects with anactivity duration of <6 min (n ¼ 40) compared tosubjects able to complete 6 min of activity (n ¼ 94)(Online Figure 4). The iECG was interpretable withhigh diagnostic certainty in 98% of rhythm strips.Ninety-six percent of pocket-echocardiographicstudies were graded to have good/excellent imagequality and showed adequate diagnostic correlationto transthoracic echocardiography for moderate/se-vere valvular stenosis/regurgitation (areas under thecurve of 0.74 and 0.79, respectively) (OnlineFigure 5). An example of an initial mHealth assess-ment is shown in Figure 2.

PRIMARY ENDPOINTS. Overall, 34% (85 of 253) ofthe study population underwent treatment withvalvuloplasty or valve replacement on follow-up.

TABLE 2 mHealth Device and Major Pocket Echocardiographic

Findings

Activity monitor

6-min walk test

Duration of activity, min 5.3 � 1.3

Number of steps 515 � 153

Distance, m 257 � 77

Walking speed, mph 1.8 � 0.3

Wireless oxygen and heart rate monitor

Resting oxygen saturation, % 88 � 12

Oxygen saturation with exercise, % 86 � 15

Resting heart rate, beats/min 81 � 16

Peak heart rate with exercise, beats/min 91 � 18

Wireless blood pressure monitor

Systolic blood pressure, mm Hg 122 � 23

Diastolic blood pressure, mm Hg 75 � 14

iECG

Normal findings 114

Baseline heart rate 79 � 19

Abnormal findings 25 (18)

Atrial arrhythmia 20 (14)

Bradyarrhythmia 5 (3)

Continued in the next column

TABLE 2 Continued

Point-of-care BNP 82 � 31

Major pocket-echocardiographic findings*

Left ventricular ejection fraction

Normal 118 (85)

Mildly reduced 15 (11)

Moderately reduced 4 (3)

Severely reduced 2 (1)

Mitral stenosis

None 56 (40)

Mild 29 (21)

Moderate 35 (25)

Severe 19 (14)

Mitral valve prolapse 5 (4)

Mitral regurgitation

None-trace 57 (41)

Mild 44 (32)

Moderate 23 (17)

Severe 14 (10)

Aortic stenosis

None 96 (87)

Mild 5 (4)

Moderate 8 (7)

Severe 2 (2)

Aortic regurgitation

None-trace 75 (54)

Mild 33 (24)

Moderate 28 (20)

Severe 3 (2)

Tricuspid regurgitation

None-trace 64 (46)

Mild 42 (30)

Moderate 22 (16)

Severe 11 (8)

Atrial or ventricular septal defect 8 (7)

Pericardial effusion (present) 2 (2)

Values are mean � SD or n (%). *Pocket echocardiographic findings were acquiredand quantified according to American Society of Echocardiography FocusedUltrasound Recommendations.

BNP ¼ brain natriuretic peptide; iECG ¼ iPhone-electrocardiography.

Bhavnani et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8

mHealth in Structural Heart Disease A P R I L 2 0 1 8 : 5 4 6 – 5 7

552

The total number of procedures was 119 comprised of26 valvuloplasties, 27 aortic and 53 mitral valvereplacements and 13 dual aortic and mitral valvereplacements with 8 patients receiving both valvu-loplasty and valve replacement. The occurrence ofthe individual types of procedures within thecomposite primary outcome was similar betweenrandomized groups (Figure 3).

At 12 months, a similar treatment rate wasobserved between the mHealth and standard-caregroups (34% [95% CI: 26% to 42%] vs. 32% [95% CI:25% to 42%], mean difference 0.5% [95% CI fordifference between study arms of �11% to þ12%;p ¼ 0.52]). Compared to standard care, a shorterduration from enrollment to primary outcome wasobserved with mHealth (83 � 79 days vs. 180 � 101days, mean difference �96% [95% CI: �136 daysto �56 days]; p <0.001) with twice as many partici-pants randomized to mHealth undergoing treatmentat 90 days (20% vs. 10%). On follow-up, 51 subjects(20%) experienced a cardiovascular hospitalizationand 3 died (1%). The occurrence of a hospitalizationand/or death was lower in the mHealth than in thestandard-care arm (15% [95% CI: 9% to 21%] vs. 28%[95% CI: 2% to 36%], mean difference �13% [95% CIof �23% to �3%; p ¼ 0.012]).

Study subjects randomized to mHealth were morelikely to undergo treatment with valvuloplasty and/or valve replacement (AHR: 1.54 [95% CI: 0.96 to2$47; p ¼ 0.07]) (Figure 4A) compared to standardcare and was associated with a lower hazard of hos-pitalization and/or death on follow-up (AHR: 0.41

[95% CI: 0.21 to 0.83; p ¼ 0.013]) (Figure 4B).Modeling the occurrence of the primary outcome as atime-dependent covariate, the probability of a car-diovascular hospitalization and/or death was lower inthe mHealth arm as compared to standard-care (AHR:0.31 [95% CI: 0.15 to 0.63]; p ¼ 0.001).

SUBGROUP ANALYSES. Pertinent subgroup analysesof relevant demographic cohorts can be found inFigure 5. Within the mHealth study arm, an incre-mental correlation to the primary outcome wasobserved with a X2 association of 2.7 (p ¼ 0.23), 11(p <0.001), and 32 (p ¼ 0.001), resulting from thepresence of an abnormal finding on the iECG, <6 minof walking observed on activity monitoring, andsevere valvular disease shown on pocket echocardi-ography, respectively.

FIGURE 2 Example of a Comprehensive mHealth Assessment in a 29-Year-Old Study Subject

ClinicalHistory

Duration2.3 minutesSymptomsDyspneaActivity116 steps87 meters1.4 mph

Resting71 beats/min96% O2

Peak Exercise94 beats/min92% O2

NYHA reclassifiedto functional class III

29 year oldwoman

No medicalhistory

Mildexertionaldyspnea

NYHA functional class I-II

6 Minute WalkTest

Mobile Blood Pressure Pocket-Echocardiogram iECG Point-of-Care BNP

126

mHealth devices showed rheumatic mitral valve disease with severe mitral stenosis, severe regurgitation, normal ejection fraction,

hypotension, sinus rhythm, and a normal BNP. Activity monitoring reclassified the functional capacity to class III from class I to II. The patient

was subsequently referred for surgical valve replacement. BNP ¼ brain natriuretic peptide; iECG ¼ iPhone-electrocardiography; DYS ¼ dia-

stolic; NYHA ¼ New York Heart Association; SYS ¼ systolic.

FIGURE 3 Individual Outcomes of the Primary Composite Outcome

ValvuloplastyValve

Replacement

p = 0.522 p = 0.533 p = 0.554 p = 0.454 p = 0.444 p = 0.355

ValveReplacement

MVR AVR MVR + AVR

6%4%

Valvuloplasty

40%35%30%25%20%15%10%5%0%

11% 10%

20%22%

29%32%

10%10%

32%34%

Standard-Care mHealth

The occurrence of the individual types of procedures—valvuloplasty and valve

replacement—within the composite primary outcome. AVR ¼ aortic valve replacement;

MVR ¼ mitral valve replacement.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8 Bhavnani et al.A P R I L 2 0 1 8 : 5 4 6 – 5 7 mHealth in Structural Heart Disease

553

DISCUSSION

The principle findings from ASE-VALUES are thefollowing: 1) compared to standard care, an initialdiagnostic strategy with mHealth was associated witha shorter time to referral for valvular interventionsand a lower probability of a hospitalization or deathamong a community cohort of SHD patients; 2) for asimilar severity of SHD, an incremental effecton outcomes was observed with mHealth; and 3)point-of-care mHealth devices used to assess theseverity of symptoms, structural, and functionalabnormalities can be used at the point of care asclinical decision support tools.

The ASE digital global health programs have aimedto maximize the yield of pocket echocardiography.The ASEF-REWARD (Remote Echocardiography withWeb Based Assessments for Referrals at a Distance)study (5) was the first investigation to determine thefeasibility of cloud-computing– and internet-basedechocardiographic image transfer. In a remote re-gion of India, 1,000 individuals with symptoms ofSHD were imaged with pocket echocardiographywithin 48 h. The digitized studies were uploadedwithin 4 min and interpreted by a global consortiumof 75 cardiologists within 12 h. Results of complexSHD were delivered back to local physicians for clin-ical decisions providing a novel mechanism foraccessing expert consultation at the point of care.To standardize image reporting, the follow-up

ASEF-VISION (Value of Interactive Scanning forImproving Outcomes of New Learners) study (6) wasperformed in which sonographers in the United Statestrained clinicians in India using a web-based imagingand educational platform resulting in improvementsin image acquisition and interpretation. ASEF-VALUES aimed to advance these findings and todetermine the impact pocket echocardiography andmHealth on outcomes in SHD.

The design of the present investigation requiredreferral of patients with SHD; therefore, we observed a

FIGURE 4 Primary and Secondary Outcomes

.0 100.0 200.0

Standard Care (n = 114) mHealth Clinics (n = 139)

Time (Days)

Cum

ulat

ive

Prob

abili

ty o

f Val

vulo

plas

ty a

nd/o

rVa

lve

Repl

acem

ent

300.0 400.0

138 108 96 91 90114

mHealthNumber at risk

Standard-Care 102 91 80 75

0.5

0.4

0.3

AHR 1.54 (95% CI 0.96 – 2.47), p = 0.07

0.2

0.1

0.0

.0 100.0 200.0Time (Days)

Cum

ulat

ive

Prob

abili

ty o

f Hos

pita

lizat

ion

and/

orDe

ath

300.0 400.0

138 133 125 105 84114

mHealthNumber at risk

Standard-Care 98 93 74 63

0.5

0.4

0.3

AHR 0.41 (95% CI 0.21 – 0.83), p = 0.013

0.2

0.1

0.0

(A) Outcome of treatment with percutaneous valvuloplasty and/or valve replacement. (B) Outcomes of hospitalization and/or death stratified by

randomized study groups. Both survival analyses were adjusted for the presence of moderate or severe mitral and/or aortic valve disease on transthoracic

echocardiography and a history of SHD. AHR ¼ adjusted hazard ratio; CI ¼ confidence interval; other abbreviation as in Figure 1.

Bhavnani et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8

mHealth in Structural Heart Disease A P R I L 2 0 1 8 : 5 4 6 – 5 7

554

high incidence of complex valve disease. Of our par-ticipants,>50% exhibited severe valvular disease with1 of 4 patients with combined mitral and aortic valvepathologies. For example, 38% of the populationexhibited severe rheumatic-mitral stenosis (meanmitral valve area and transmitral pressure gradient of1.1 � 0.4 cm2 and 9.2 � 4.7 mm Hg, respectively) andwere identified as a high-risk cohort at risk of hospi-talization and death. These findings are representa-tive of a cross-sectional sample of patients commonlyseen in endemic regions where patients often presentlate in the disease process and with advanced symp-toms (5,17,18). Within India, a 6- to 24-month waitingperiod for cardiac procedures and cardiothoracic sur-gery is commonly observed in nonprofit andgovernment-run hospitals (19). In these health sys-tems, determination of severity of disease and priori-tization of sicker patients is of paramount importance.Similar to that observed in the standard-care arm ofthe present study, longer waiting periods for treat-ment or interventions commonly result in increasedmorbidity and mortality. Various socioeconomic fac-tors also contribute to long treatment delays includingpatient refusal for surgery, poor health literacy, and alack of social support as additive factors and has beenshown in 50% of patients with SHD in rural India (20).

Compared to standard care, an initial assessmentwith mHealth was associated with a shorter time totreatment of 3 months (89 days vs. 180 days) withtwice as many participants in the mHealth arm un-dergoing treatment at 90 days (20% vs. 10%). Overall,our treatment rate of 34% at 12 months is dramaticallyshorter than the national average (19). Upon multi-variate and time-dependent analysis, an initialmHealth assessment was associated with improvedoutcomes within both the overall mHealth cohort andamong those mHealth subjects who underwent val-vuloplasty or valve replacement on follow-up. Thelatter suggests that earlier treatment is associatedwithimproved outcomes. We postulate the following rea-sons for these observations. The first is due to animproved characterization of SHD with mHealth thatprovided a comprehensive assessment of the severityof valvular abnormalities at the time of enrollment.Such point-of-care diagnostics likely facilitated timelyclinical decisions and referral for treatment amongthose patients with severe disease (18,21). The yield ofpocket echocardiography showed moderate to highdiagnostic accuracy compared to comprehensiveechocardiography for qualitative determinationof SHD lesions, and is consistent with the diagnosticaccuracy observed in the ASEF-REWARD study (5).

FIGURE 5 Subgroup Analysis

Women

Men

≤ 40 years old

> 40 years old

Presence of Symptoms

Asymptomatic

History of Rh-SHD with priorvalvuloplasty/valve replacementHistory of Rh-SHD without priorvalvuloplasty/valve replacement

2.11 (1.10–4.05)

1.10 (0.50–2.03)

1.43 (0.74–2.75)

1.78 (0.87–3.63)

1.65 (1.02–2.67)

1.09 (0.39–3.31)

0.66 (0.22–1.95)

1.79 (1.09–2.95)

0.1 1Adjusted Hazard Ratio

10

0.02

AHR (95% CI) p Value

0.98

0.29

0.11

0.04

0.89

0.45

0.02

Hazard ratios were adjusted for the presence of moderate/severe mitral and aortic valve disease, and a history of SHD. AHR ¼ adjusted hazard

ratio; CI ¼ confidence interval; Rh ¼ rheumatic; SHD ¼ structural heart disease.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8 Bhavnani et al.A P R I L 2 0 1 8 : 5 4 6 – 5 7 mHealth in Structural Heart Disease

555

The second is due to a quantified approached ofsymptoms and functional limitations that facilitatedmedical decision making. Within the mHealth arm,activity monitoring reclassified 37% of participants toa different NYHA functional class with 30% reclassi-fied to a higher and 6% to a lower NYHA functionalclass, respectively. Third, a shorter notification rate—the rapid availability of diagnostic information fromthe point of care and input into electronic medicalrecords—may have improved care coordination, facil-itated medical decisions such as early initiation ofdiuretics for heart failure or rate control and anti-coagulation for atrial fibrillation, and may haveresulted in shorter follow-up for higher-risk patients(22,23). Taken together, point-of-care diagnostics,functional assessments, and the availability of digi-tally acquired patient data likely influenced clinicaldecisions and referral of those patients that were morelikely to benefit with earlier interventions (18).

Because of sheer numbers of individuals at risk,mHealth utilization in resource-limited areas must bedesigned in pragmatic and cost-effective approaches.Recent progress with digital health in these regionshas included using text messaging to improve

medication compliance among patients with HIV andto encourage lifestyle changes among individuals atrisk for diabetes (24,25), genome sequencing todiagnose tuberculosis at the point of care (26), andelectronic intensive care units to streamline care foran acute coronary syndrome (27). In contrast to mostchronic diseases, the diagnosis of SHD requiresspecialized diagnostics including echocardiographyand trained users to accurately perform and interpretthe severity of structural abnormalities. Traininghealth care workers to perform portable echocardio-graphic studies and to shift imaging to community-based screening programs are potentially scalablemethods to address the shortage of technically profi-cient health care personnel (28,29). In this context,Engelman et al. (30), Mirabel et al. (31), and Ploutzet al. (32) have provided seminal results and address“task-shifting” with trained nurses proficient in theacquisition of echocardiographic images to diagnoseSHD. When compared to transthoracic echocardiog-raphy, nurse-based screening with portable echocar-diographic devices showed a diagnostic accuracy of85% for the detection of SHD in more than 4,000 at-risk children. Such training may emerge as a practical

PERSPECTIVES

COMPETENCY IN MEDICAL KNOWLEDGE:

Newly developed mHealth and smartphone-

connected technologies have emerged as potentially

transformative innovations to improve health care

delivery and for public and global health benefit.

COMPETENCY IN PATIENT CARE AND PROCE-

DURAL SKILLS: Despite significant progress in pre-

vention and screening, rheumatic and SHD remain

leading causes of morbidity and mortality in resource-

limited areas. In such areas, the intersection of

expanding digital infrastructures with near ubiquitous

cellular phone use and internet connectivity, com-

bined with new devices such as portable and wireless

mHealth devices, handheld-ultrasound, and lab-on-a-

chip technologies may bridge common health care

disparities by providing the necessary diagnostic in-

formation for health care practitioners to formulate

clinical decisions at the point of care.

TRANSLATIONAL OUTLOOK 1: This is the first

study to compare the impact mHealth device assess-

ments such as pocket echocardiography and smart-

phone electrocardiography to the standard of care on

treatment rates and outcomes among patients with

advanced SHD in the health system of a developing

nation. In doing so, this study identifies the capacity

of mHealth to identify high-risk patients and those

patients who may derive benefit with earlier medical

therapies and surgical interventions.

TRANSLATIONAL OUTLOOK 2: The present

investigation provides an understanding for how

technology-enabled care in such regions can be

delivered. Integrating mHealth findings into existing

health information technology systems and using

diagnostic information provided by mHealth to

improve risk stratification are among the important

heuristic factors that can improve patient outcomes.

New integration methods should remain a focus of

future studies evaluating the outcomes of

technology-enabled care in resource-limited areas.

Bhavnani et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8

mHealth in Structural Heart Disease A P R I L 2 0 1 8 : 5 4 6 – 5 7

556

and potentially transformative method to increase theyield of portable imaging in endemic areas.

STUDY LIMITATIONS. We observed unequal samplesizes for 2 main reasons: 1) using an a priorirandomization schedule used for daily enrollmentversus over the enrollment period; and 2) a simple (orunrestricted) rather than a restricted (i.e., permutedblock) randomization method. The simple randomi-zation method allows for random variation in samplesizes important for pragmatic trials and to minimizebias particularly in non–double-blinded studies. Insuch designs, equal randomization is not necessarilyrequired (33,34). Despite this finding, baseline de-mographics were well balanced and the overalltreatment rates at 12 months were equal betweenrandomized groups, suggesting that bias was mini-mized when analyzing the effectiveness and safety ofmHealth. A multiple-arm trial and blinded assess-ment of pocket echocardiography compared to TTEwas not performed as this would be ethically unac-ceptable because Doppler measurements cannot beperformed on pocket devices for accurate hemody-namic assessment required for interventional/surgi-cal referral of cases. Although our study was small insize, its randomized comparison of initial testingstrategies, use of available technologies, broad rep-resentation of a real-world community cohort, andthe use of hard clinical events as outcome measuresenhances internal and external validity, and mayrepresent potentially reproducible methods formHealth use in other underserved areas.

CONCLUSIONS

Compared to standard care, an initial testing strategywith mHealth was associated with a shorter referraltime for treatment among symptomatic patients withadvanced SHD, and was associated with improvedhealth outcomes in an endemic area with a highburden of disease. These data have important impli-cations for the use of pocket echocardiography andsmartphone-connected mHealth devices at the pointof care as clinical decision support tools in the healthcare system of resource-limited areas.

ACKNOWLEDGMENTS The authors thank theAmerican Society of Echocardiography Foundationfor program organization, strategic planning, andfunding; General Electric Healthcare, iHealth, andCoreSound Imaging for their generous contributionsof devices and information technology resources thatwere integral to the execution of this investigation;and Hemant Kulkarni, MD, for statistical support.They also thank the participants of this study and fortheir willingness to contribute towards these efforts.

ADDRESS FOR CORRESPONDENCE: Dr. Partho P.Sengupta, West Virginia University Heart and VascularInstitute, West Virginia University School of Medicine,1 Medical Center Drive, Morgantown, West Vir-ginia 26506. E-mail: partho.sengupta@hsc.wvu.edu.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 4 , 2 0 1 8 Bhavnani et al.A P R I L 2 0 1 8 : 5 4 6 – 5 7 mHealth in Structural Heart Disease

557

RE F E RENCE S

1. Chib A, van Velthoven MH, Car J. mHealthadoption in low-resource environments: a reviewof the use of mobile healthcare in developingcountries. J Health Commun 2015;20:4–34.

2. Bhavnani SP, Narula J, Sengupta PP. Mobiletechnology and the digitization of healthcare. EurHeart J 2016;37:1428–38.

3. Agarwal S, LeFevre AE, Lee J, et al., WHOmHealth Technical Evidence Review Group.Guidelines for reporting of health interventionsusing mobile phones: mobile health (mHealth)evidence reporting and assessment (mERA)checklist. BMJ 2016;352:i1174.

4. Piette JD, List J, Rana GK, et al. Mobile healthdevices as tools for worldwide cardiovascular riskreduction and disease management. Circulation2015;132:2012–27.

5. Singh S, Bansal M, Maheshwari P, et al., ASE-REWARD Study Investigators. American Society ofEchocardiography: Remote Echocardiography withWeb-Based Assessments for Referrals at a Dis-tance (ASE-REWARD) study. J Am Soc Echo-cardiogr 2013;26:221–33.

6. Bansal M, Singh S, Maheshwari P, et al.,VISION-in-Tele-Echo Study Investigators. Valueof interactive scanning for improving theoutcome of new-learners in transcontinental tele-echocardiography (VISION-in-Tele-Echo) study.J Am Soc Echocardiogr 2015;28:75–87.

7. Yusuf S, Teo RK, Li IW, et al. Cardiovascular riskand events in 17 low-, middle-, and high-incomecountries. N Engl J Med 2014;371:818–27.

8. Crisp N, Chen L. Global supply of health pro-fessionals. N Engl J Med 2014;370:950–7.

9. Agarwal S, Perry HB, Long LA, et al. Evidenceon feasibility and effective use of mHealth stra-tegies by frontline health workers in developingcountries: systematic review. Trop Med Int Health2015;20:1003–14.

10. Eysenbach G, CONSORT-EHEALTH Group.CONSORT-EHEALTH: improving and standardizingevaluation reports of web-based and mobilehealth interventions. J Med Internet Res 2011;13:e126.

11. Spencer KT, Kimura BJ, Korcarz CE, et al.Focused cardiac ultrasound: recommendationsfrom the American Society of Echocardiography.J Am Soc Echocardiogr 2013;26:567–81.

12. Peters DM, Fritz SL, Krotish DE.Assessing the reliability and validity of ashorter walk test compared with the 10-meterealk test for measurements of gait speed inhealthy, older adults. J Geriatr Phys Ther 2013;36:24–30.

13. Pulignano G, Del Sindaco D, Di Lenarda A, et al.Incremental value of gait speed in predictingprognosis of older adults with heart failure: in-sights from the IMAGE-HF study. J Am Coll CardiolHF 2016;4:289–98.

14. Maisel A, Barnard D, Jaski B, et al. Primaryresults of the HABIT trial. J Am Coll Cardiol 2013;61:1726–35.

15. Lang RM, Badano LP, Mor-Avi V, et al. Rec-ommendations for cardiac chamber quantificationby echocardiography in adults: an update from theAmerican Society of Echocardiography and theEuropean Association of Cardiovascular Imaging.J Am Soc Echocardiogr 2015;28:1–39.

16. Nishimura RA, Otto CM, Bonow RO, et al.,American College of Cardiology/American HeartAssociation Task Force on Practice Guidelines.2014 AHA/ACC guideline for the management ofpatients with valvular heart disease: executivesummary: a report of the American College ofCardiology/American Heart Association Task Forceon Practice Guidelines. J Am Coll Cardiol 2014;63:2438–88.

17. Zühlke L, Karthikeyan G, Engel ME, et al.Clinical outcomes in 3343 children and adults withrheumatic heart disease from 14 low and middleincome countries: 2-year follow-up of the globalrheumatic heart disease registry (the REMEDYstudy). Circulation 2016;134:1456–66.

18. Evangelista A, Galuppo V, Méndez J, et al.Hand-held cardiac ultrasound screening per-formed by family doctors with remote expertsupport interpretation. Heart 2016;102:376–82.

19. Agarwal A. Standard operating procedures forhospitals in India: complete with stationary for-mats used in various department hospitals. NewDelhi, India: Atlantic Publishers & Distributors (P)Ltd; 2007; p 473.

20. Meenakshisundaram R, Thirumalaikolundu-subramanian P. Valvular heart disease in Indiansubcontinent: social issues. Indian J CommunityMed 2009;34:57–8.

21. Bansal M, Sengupta PP. Setting global stan-dards in adult echocardiography: where are we?Indian Heart J 2015;67:298–301.

22. Drain PK, Hyle EP, Noubary F, et al. Diagnosticpoint-of-care tests in resource-limited settings.Lancet Infect Dis 2014;14:239–49.

23. Were MC, Emenyonu N, Achieng M, et al.Evaluating a scalable model for implementingelectronic health records in resource-limited set-tings. J Am Med Inform Assoc 2010;17:237–44.

24. Lester RT, Ritvo P, Mills EJ, et al. Effects of amobile phone short message service on antire-troviral treatment adherence in Kenya (WelTel

Kenya1): a randomised trial. Lancet 2010;376:1838–45.

25. Ramachandran A, Snehalatha C, Ram J, et al.Effectiveness of mobile phone messaging in pre-vention of type 2 diabetes by lifestyle modificationin men in India: a prospective, parallel-group,randomised controlled trial. Lancet DiabetesEndocrinol 2013;1:191–8.

26. Theron G, Zijenah L, Chanda D, et al. Feasi-bility, accuracy, and clinical effect of point-of-care Xpert MTB/RIF testing for tuberculosis inprimary-care settings in Africa: a multicentre,randomised, controlled trial. Lancet 2014;383:424–35.

27. Gupta S, Dewan S, Kaushal A, et al. eICU re-duces mortality in STEMI patients in resource-limited areas. Glob Heart 2014;9:425–7.

28. Saxena A. Task shifting rheumatic heart dis-ease screening to non-experts. Lancet Glob Health2016;4:e349–50.

29. Marijon E, Ou P, Celermajer DS, et al. Preva-lence of rheumatic heart disease detected byechocardiographic screening. N Engl J Med 2007;357:470–6.

30. Engelman D, Kado JH, Reményi B, et al.Focused cardiac ultrasound screening for rheu-matic heart disease by briefly trained healthworkers: a study of diagnostic accuracy. LancetGlob Health 2016;4:e386–94.

31. Mirabel M, Bacquelin R, Tafflet M, et al.Screening for rheumatic heart disease: evaluationof a focused cardiac ultrasound approach. CircCardiovasc Imaging 2015;8:e002324.

32. Ploutz M, Lu JC, Scheel J, et al. Beaton A.Handheld echocardiographic screening for rheu-matic heart disease by non-experts. Heart 2016;102:35–9.

33. Bacchetti P, Deeks SG, McCune JM. Breakingfree of sample size dogma to perform innovativetranslational research. Sci Transl Med 2011;3:87ps24.

34. Schulz KF, Grimes DA. Unequal group sizes inrandomised trials: guarding against guessing.Lancet 2002;359:966–7.

KEY WORDS clinical trial, devices, mobilehealth, outcomes, pocket-echocardiography,structural heart diseases

APPENDIX Foracomplete listof investigators,a description of training procedures for mHealthdevices, supplemental figures and tables, pleasesee the online version of this paper.