‘The phone reminder is important, but will others get to know about my illness?’ Patient perceptions of an mHealth antiretroviral treatment support intervention in the HIVIND trial in South India

‘The phone eminde is impo an , bu will o he s ge o know abou my illness?’Pa ien pe cep ions o an mHeal h an i e o i al ea men suppo in e en ion in he HIVIND ial in Sou h India Rashmi Rod igues, 1,2 S Poongulali, 3 Ka i ha Balaji, 3 Salla A kins, 1 Pe Asho n, 4 Ayesha De Cos a 1 To ci e: Rod igues R, Poongulali S, Balaji K, e al. ‘The phone eminde is impo an , bu will o he s ge o know abou my illness?’ Pa ien pe cep ions o an mHeal h an i e o i al ea men suppo in e en ion in he HIVIND ial in Sou h India. BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015- 007574 ▸P epublica ion his o y o his pape is a ailable online. To iew hese iles please isi he jou nal online (h p://dx.doi.o g/10.1136/ bmjopen-2015-007574). Recei ed 2 Janua y 2015 Re ised 1 July 2015 Accep ed 18 Augus 2015 Fo numbe ed a ilia ions see end o a icle. Co espondence o D Rashmi Rod igues; [email p o ec ed] ABSTRACT Objec i es: The ecen explosion o mHeal h applica ions in he a ea o HIV ca e has led o he de elopmen o mHeal h in e en ions o suppo an i e o i al ea men adhe ence. Se e al o hese in e en ions ha e been es ed o e ec i eness, bu ew s udies ha e explo ed pa ien pe spec i es o such in e en ions. Explo ing pa ien pe spec i es enhances he unde s anding o how an in e en ion wo ks o why i does no . We he e o e s udied pe cep ions ega ding an mHeal h adhe ence in e en ion wi hin he HIVIND ial in Sou h India. Me hods: The s udy was conduc ed a h ee clinics in Sou h India. The in e en ion comp ised an au oma ed in e ac i e oice esponse (IVR) call and a pic o ial sho messaging se ice (SMS), each deli e ed weekly. Six een pu posi ely selec ed pa icipan s om he in e en ion a m in he HIVIND ial we e in e iewed. All pa icipan s had comple ed a leas 84 weeks since en ollmen in he ial. Pe cep ions on he use ulness and pe cei ed bene i s and isks o ecei ing he in e en ion we e sough . The in e iews we e ansc ibed and analysed using he amewo k app oach o quali a i e da a analysis. Resul s: Despi e a ying pe cep ions o he in e en ion, mos pa icipan s ound i use ul. The in e en ion was pe cei ed as a sign o ‘ca e’ om he clinic. The IVR call was p e e ed o he SMS eminde . Two-way communica ion was p e e ed o au oma ed calls. Pa icipan s also pe cei ed a isk o unin en ional disclosu e o hei HIV s a us and s igma he eo ia he in e en ion and ook ini ia i es o mi iga e his isk. Ta ge ing eminde s a hose wi h poo adhe ence and hose in need o social suppo was sugges ed. Conclusions: mHeal h adhe ence in e en ions go beyond hei in ended ole o p o ide a sense o ca e and suppo o he ecipien . Al hough au oma ed in e en ions a e impe sonal, hey could be a solu ion o scale up. In e en ions ha engage bo h he ecipien and he heal hca e p o ide ha e g ea e po en ial o success. Pe sonalising mHeal h in e en ions could mi iga e he isk o s igma and p omo e hei up ake. T ial egis a ion numbe : ISRCTN79261738. INTRODUCTION The scope o mobile phones in suppo o heal hca e (mHeal h) has g own in ecen yea s as e idenced by he expanding body o li e a u e in his field. 12 One o he popula uses o mHeal h has been o suppo he con inuum o HIV ca e and p e en ion. 34 The echnology has been used ex ensi ely o sp ead awa eness ega ding HIV in ec ion and i s p e en ion, 1 including he p e en ion o mo he - o-child ansmission (PMTCT). 5 S eng hs and limi a ions o his s udy ▪This is he i s quali a i e s udy in he Indian con ex designed wi hin a andomised con olled ial ha in es iga es pa icipan pe cep ions ega ding an mHeal h adhe ence suppo in e - en ion o an i e o i al he apy. ▪Quali a i e esea ch alongside clinical ials con- ex ualises in e en ions and acili a es he de el- opmen o in e en ions ha a e e ec i e. ▪The amewo k app oach used in he analysis is poised be ween induc i e and deduc i e app oaches and encou ages he analysis o he esponden s’pe cep ions, in ligh o he esea che s’expe iences. ▪As he HIVIND ial was ongoing a he ime o he in e iews, he adhe ence s a us o he pa i- cipan s was unknown o he in e iewe . Hence, we we e unable o ela e he expe iences o he pa icipan s o hei adhe ence wi hin he ial. Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 1 Open Access Resea ch g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om Mo eo e , mHeal h has been ound accep able o com- munica ing labo a o y esul s o HIV-in ec ed indi i- duals, despi e sensi i i y su ounding he disease. 6 I is also being s udied o HIV counselling and e en ion 78 in ea men . Suppo ing adhe ence o an i e o i al ea men (ART) has been one o he popula uses o he echnology, wi h a numbe o s udies 49–12 showing mixed esul s. 91013 Adhe ence o ea men is a complex 14 ye essen ial phenomenon o posi i e heal h ou comes, i espec i e o he disease. I is key o he success o ART, as op imal adhe ence de e s ea men ailu e and dea h in HIV-in ec ed indi iduals. One o he ba ie s o adhe - ence is o ge ulness. Se e al adhe ence suppo in e - en ions ha e been de eloped o a ge o ge ulness such as pillboxes, elec onic eminde s 15 16 and mobile phone eminde s. These in e en ions a e posi ed wi hin he Beha io Lea ning Theo y (BLT) amewo k whe e hey a e conside ed as ex e nal cues ha suppo medica ion adhe ence. 16 The ecen ly concluded HIVIND andomised con- olled ial (RCT) in Sou h India was he fi s RCT ha explo ed he e ec o an mHeal h in e en ion on ART adhe ence in he Indian con ex . 17 The in e en ion es ed in he HIVIND ial comp ised an au oma ed in e ac i e oice esponse (IVR) call and a neu al pic u e sho messaging se ice (SMS). Quali a i e s udies can aid he unde s anding o he e ec s o an in e en ion and he possible pa hways h ough which hese e ec s (o lack o ) may be media ed. Such esea ch alongside RCTs can con ex ual- ise in e en ions and acili a e he de elopmen o e ec - i e in e en ions. 18 19 Howe e , quali a i e s udies alongside RCTs, especially in HIV in ec ion, a e ew. We explo ed pa icipan pe cep ions o he in e en- ion, including conce ns abou s igma and in usion o p i acy, and compa ed he IVR call wi h he SMS eminde in he Indian con ex . METHODS S udy con ex This was a quali a i e s udy wi hin he HIVIND ial. The HIVIND ial was a pa allel g oup open-label ial ha s udied he e ec o mobile phone eminde s on adhe ence o ART and ime o fi s -line ART ailu e in ea men naï e people li ing wi h HIV (PLHIV) in Sou h India. The RCT had h ee ec ui men si es in Sou h India, ha is, (1) S . John’s Medical College Hospi al, Bangalo e, a p i a e e ia y-le el non-p ofi heal hca e acili y, p o iding ca e and ea men o app oxima ely 3000 PLHIV/AIDS h ough a public– p i a e pa ne ship in Ka na aka S a e; (2) KR Hospi al, Myso e, a public e ia y-le el eaching heal hca e acil- i y, p o iding ca e and ea men o app oxima ely 10 000 PLHIV in Ka na aka S a e and (3) YRG Ca e, Chennai, a p i a e heal hca e acili y, p o iding ca e and ea men o app oxima ely 18 000 PLHIV in Tamil Nadu S a e. HIV ca e and ea men in India a e p o ided a no cos o minimal cos o PLHIV h ough a ne wo k o public, p i a e and public–p i a e heal hca e acili ies. All PLHIV egis e ed wi hin his HIV ca e ne wo k ecei e counselling suppo , basic in es iga ions such as CD4 coun , ea men and mon hly ollow-up ca e. T ea men is ini ia ed a CD4 coun s <350 cells/mm 3 . Fu he , PLHIV ecei e adhe ence counselling a ea - men ini ia ion, and a mon hly pill efill isi s. The ele- men s o his counselling (figu e 1) a e in line wi h he Social Cogni i e Theo y o beha iou change. Acco ding o he exis ing sys em, counsello s assess pa ien s’knowledge ega ding hei heal h and educa e hem based on hei knowledge gap. Counselling os e s sel -e ficacy by engende ing he belie ha one can inde- penden ly add ess adhe ence- ela ed issues and help iden i y ex e nal sys ems (an eceden s) sui able o suppo adhe ence. 20 The HIVIND T ial 19 : The ial es ed an mHeal h ART adhe ence suppo in e en ion in 631 ou pa ien s aged 18–60 yea s. Hal he pa icipan s in he ial (315) ecei ed he in e en ion, while he o he hal ecei ed s anda d ca e. All pa icipan s, i espec i e o ecei ing he in e en ion, we e ollowed up qua e ly o 2 yea s o un il ea men ailu e, whiche e was ea lie . O he pa icipan s in he ial, 502 (79%) we e li e a e, 273 (43%) we e emale, 398 (63%) we e employed and 286 (45%) we e om a u al backg ound. The ecen ly epo ed HIVIND ial esul s showed no e ec o he in e en ion on adhe ence. 21 The mHeal h in e en ion (figu e 2): The ial in e en ion was designed on he basis o pa ien p e e ences and pilo ed p io o he ial. 22 23 The in e en ion was au o- ma ed and comp ised an IVR call 24 and a pic o ial SMS eminde , each sen once a week o he in e en ion a m pa icipan s o 2 yea s. The IVR call said, “Hello, his is you good iend calling you! I you ha e aken all you medica ions yes e day please p ess ‘1’i no please p ess ‘2’.”I he pa ien missed he fi s call, h ee add- i ional calls we e a emp ed in he nex 24 h. The SMS eminde was a neu al pic o ial SMS ha depic ed a ‘lamp’. The SMS eminde was sen o he pa ien 3 days a e he IVR call. Pa icipan s could choose a day and ime o ecei e he IVR call and SMS eminde . Pa icipan s we e ained o ecei e and espond o he oice calls and access he SMS eminde . Da a collec ion We conduc ed in-dep h in e iews wi h 16 pa icipan s in he in e en ion a m o he ial (5 pa ien s each a he Myso e and Bangalo e si es and 6 a he Chennai si e). All pa icipan s had comple ed a leas 84 weeks since en ollmen in he ial. We achie ed da a sa u - a ion a ound he 16 h in e iew and he e o e decided o s op he in e iews, as no new in o ma ion was o h- coming. Pu posi e sampling included pa ien s om 2Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om di e en si es, sexes, occupa ions, and u al and u ban backg ounds. The fi s au ho , who wo ked wi h he HIVIND ial and he in e en ion, bu no wi h ou ine pa ien ca e, conduc ed he in-dep h in e iews. All in e iews we e conduc ed in a quie oom a each o he s udy si es and audio- eco ded wi h he pa icipan s’consen . Each in e iew was conduc ed in he local language, ha is, Kannada o Tamil, and ook app oxima ely 20 min. We used an in e iew guide ha was de eloped and pilo es ed o explo e pa icipan pe cep ions o he in e en- ion unde specific domains, ha is, help ulness o he in e en ion, equency, ease o use, disclosu e o HIV s a us and s igma he eo , p e e ence be ween he IVR call and he SMS eminde . Da a analysis A na i e local language expe , also p oficien in English, ansc ibed and ansla ed he in e iews in o English. The fi s au ho subsequen ly spo checked he ansc ip ions o ensu e consis ency wi h he eco dings. The hema ic ‘F amewo k App oach’ 25 was used in he analysis. Fo his, he fi s au ho amilia ised he sel wi h he ansc ip s and iden ified codes. The codes om di e en ansc ip s we e compa ed o simila i y and g ouped unde a single ca ego y. Codes and Figu e 1 Adhe ence counselling o suppo an i e o i al ea men p o ided o pa ien s in line wi h Social Cogni i e Theo y. Figu e 2 The in e en ion. Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 3 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om ca ego ies o med he p elimina y amewo k o he analysis (box 1). Discussions be ween he au ho s enabled he eo ganisa ion o codes o ma ch he amewo k. The codes we e hen applied o he ansc ip s, ha is, he ansc ip s we e indexed and he da a cha ed unde he ele an codes in he amewo k. Addi ional codes we e gene a ed as he analysis p og essed. Subsequen ly, he da a we e summa ised and mapped o iden i y con- nec ions be ween summa ies and o de i e sub hemes. The sub hemes we e collapsed unde b oade hemes desc ibing he da a ( able 1). E hics s a emen P io o en olmen , pa ien s ga e hei w i en consen o pa icipa e. The consen was also ead o all pa ici- pan s in he local language o ensu e ha hey unde - s ood he pu pose o he s udy and he s udy p ocedu es i espec i e o hei le el o li e acy. FINDINGS Pa icipan cha ac e is ics: Six een pa icipan s om he in e en ion a m o he HIVIND ial pa icipa ed in his s udy. They we e aged 25–56 yea s (median age: 38 yea s). O hese, se en we e women and nine we e men. O he women, h ee had pa - ime employmen ou side hei home. All he men we e employed. Six pa - icipan s had a u al backg ound. Themes: Th ee hemes eme ged om he in e iews. These we e (1) a ying use ulness o he in e en ion, (2) p e e ence o calls o e messages and (3) pe cei ed isk o unin en ional disclosu e o HIV s a us. Va ying use ulness o he in e en ion Some pa icipan s ound ha he mobile phone emin- de s aided es ablishmen o a ou ine o aking hei medica ion while o he s did no , ci ing cu en ly adequa e adhe ence ega dless o he in e en ion. Despi e being au oma ed, he in e en ion was pe cei ed o p o ide social suppo and eflec he conce n o he heal hca e p o ide . Di e en pe cep ions o help ulness as a eminde One pa icipan epo ed ha he in e en ion se ed as a eminde h oughou he day and week despi e i s bi-weekly equency. O he s epo ed ha he calls mini- mised o ge ulness, especially when hey we e busy wi h wo k o we e away om home. We will be busy wi h ou wo k. When we ge busy, I eel ha he eminde is e y impo an o me o ake he able s … o my heal h. The phone calls p o ided a cue o ake medica ions and helped some pa icipan s de elop a ou ine o aking medicines. One pa ien epo ed ha he emba - assmen o p o iding a nega i e esponse o he call ensu ed he adhe ence. In he beginning, I used o o ge . So, ini ially o se he ime, his became a good me hod. I used o miss he able s ini ially. Nowadays, I ake he able s a 10:30 in he mo ning and 10:30 a nigh . To ge in o his ou ine, his phone call has helped… Pa icipan s who conside ed hemsel es adequa ely adhe en o medica ion did no find he in e en ion use ul. They we e awa e o he impo ance o adhe ence o medica ion in o de o ensu e i al supp ession. They also epo ed conce n o hei own heal h and a desi e o be heal hy. No, (i is no ) only because o phone call ( ha ) I emem- be ed o ake able s. I is also because I wan o be heal hy ha I ( ake) able s egula ly. Pa icipan s sugges ed sending eminde s only o poo ly adhe en pa ien s ha he heal hca e p o ide iden ified a mon hly isi s when pill coun s we e done. They also sugges ed ha he equency o he eminde s could be based on he indi iduals’need. Fo hose i egula wi h medica ions you should make he calls e e y day, hose who a e egula weekly once is enough. Reminde s we e conside ed necessa y o hose esid- ing in u al a eas, hose equi ing social suppo , hose p eoccupied wi h wo k o hose wi h an illness ha a ec ed memo y. Box 1 Thema ic amewo k used in he s udy 1.1 Help ulness in e ac i e oice esponse (IVR) 1.2 Help ulness sho messaging se ice (SMS) 2. Pe cep ion o impo ance o he in e en ion 3. F equency o he in e en ion 4. Func ioning o he in e en ion 5. P i acy 6. In usion 7. S igma 7.1 S igma om in e en ion 7.2 Ways o a oid s igma 8. Discon inua ion o he in e en ion 9. Voice: Male/ emale p e e ences 10. Technical knowledge 11. P e e ence be ween IVR and SMS 12.1 Responding o he in e en ion 12.2 Reasons o non- esponse 12.3 Reasons o e o in esponse 13. Ba ie s o adhe ence 14. Communica ion 14.1 Two-way communica ion 14.2 Type o message 14.3 Sugges ions o deploymen 4Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om Table 1 A b ie desc ip ion o he amewo k used in da a analysis Thema ic amewo k componen s and quo es Codes Summa y Ca ego ies Sub hemes Theme* Help ulness o IVR Fo ge ulness due o wo k — eminde s help Phone calls minimised o ge ulness and enabled de elopmen o a medica ion ou ine, imp o ing adhe ence. Conce n om heal hca e p o ide pe cei ed Imp o es adhe ence Di e en pe cep ions ega ding he help ulness as a eminde 1 We will be busy wi h ou wo k, when we ge busy, I eel ha he eminde is e y impo an o me ake he able s When I ge a call a 8:00, I eel happy ha a call has come om hospi al. I eel people om hospi al ha e called Pe cei ed conce n o he heal hca e p o ide along wi h. Feels happy wi h he calls Heal hca e p o ide s conce n Di e en pe cep ions ega ding he help ulness as a eminde 1 Help ulness o SMS SMS no liked, he inconspicuous ale one makes he pa ien miss he SMS A gene al dislike o SMS. The SMS is missed on a i al because o i s inconspicuous ing one. Unin ended disclosu e o HIV s a us pe cei ed SMS disliked Di e en pe cep ions ega ding he help ulness as a eminde 1 I ne e liked SMS, I didn’ like i a all. I would be busy du ing he day. I would ne e hea i s sound a all…I ne e ge o know when SMS comes… I dele e i on he spo bu by chance i someone sees i , “LM a ogyam”†, hey will ques ion me Pe cei e he disclosu e o HIV s a us wi h he SMS Fea o disclosu e o HIV s a us P e en ing unin ended disclosu e o HIV s a us 3 Pe cep ion o impo ance o he in e en ion In e en ion does no ma e , pa ien adhe en The necessi y o calls is no seen as pa ien s claim o be adhe en wi hou he calls. Ex e nal cues o suppo adhe ence used Pe cep ion o being adhe en Di e en pe cep ions ega ding he help ulness as a eminde 1 Whe he I ge a phone call o SMS, i doesn’ ma e , I ha e aken able s egula ly. Tha is mo e impo an igh ? By 9.00 sha p, a e ha ing my b eak as and while aking he cash o my expenses…I will also ake he medicines Pa ien akes medicines daily a e b eak as Ex e nal cues o adhe ence used Di e en pe cep ions ega ding he help ulness as a eminde 1 P e e ence be ween IVR and SMS Fea o disclosu e o HIV s a us om he phone call A p e e ence o phone calls in compa ison o SMSs. Phone calls hough o ha e he po en ial o disclosu e o HIV s a us Fea o disclosu e o HIV s a us P e en ing unin ended disclosu e o HIV s a us 3 Yes, she has ead he message and asked who is he ‘a ogyam’, no one knows, bu i you a e calling, whoe e is a ending he call will come o know abou he p oblem The call is su icien , SMS is no necessa y…in he phone call hey speak, a leas o espec ( hem) we ake he call…I we ake he call we ha e o espond…We do no ha e o espond o he SMS… Calls a e conside ed su icien as hey a e in e ac i e, SMS conside ed passi e In e ac ion in he call p e e ed o passi i y o he SMS Engagemen : IVR s SMS 2 *Themes: 1. Pe cep ions o a ying use ulness o he in e en ion, 2. p e e ence o calls o e messages and 3. pe cei ed isk o unin en ional disclosu e o HIV s a us. †The name unde which he SMS is deli e ed, ‘a ogyam’means heal h. IVR, in e ac i e oice esponse; SMS, sho messaging se ice; LM, p e ix o he alpha nume ic sende iden i ie ‘a ogyam’L- code o he se ice p o ide , M- code o he se ice a ea. This p e ix ollows he egula ions o SMS sende iden i ica ion issued by he Telecom Regula o y Au ho i y o India on 10 h Decembe 2008, h p://www. ai.go .in/W i eReadDa a/Di ec ion/ Documen /di ec ion10dec08.pd . Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 5 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om Pe cei ed suppo : social suppo and conce n o he heal hca e p o ide The in e en ion made he pa icipan s eel he conce n and suppo om he heal hca e p o ide as a esul o he phone calls. I eel happy ha hospi al people ca e o my heal h, when I ecei e he call. The calls we e pe cei ed as being om a iend and no om a machine, despi e all pa icipan s being awa e ha he calls we e au oma ed. When I ge hese compu e ized phone calls asking me how my heal h is?? I eel con en ed …E en i I am eady o pay a hund ed housand upees, I don’ hink I will ge a p i ilege like his… P e e ence o calls o e messages Despi e au oma ing bo h he IVR call and he SMS eminde , he IVR call was p e e ed and conside ed mo e use ul when compa ed wi h he SMS eminde . Fu he , pa icipan s a ely iewed he SMS eminde . Un amilia i y wi h he concep o SMS combined wi h i s passi i y and he inconspicuous ale one led pa ien s o igno e he SMS eminde . Engagemen : IVR e sus SMS The phone call a ac ed a en ion and was unlikely o be missed because o he ‘noise’i made while he SMS was easily missed as i made a so e ‘noise’. Pa icipan s app ecia ed he in e ac i i y and echno- logical simplici y o he au oma ed call in compa ison o he SMS. The call is su ficien , ( he) SMS is no necessa y. I you compa e …in he phone call hey speak. A leas o espec wha hey speak we pick he call, a leas once … We do no ha e o espond o he SMSs, he SMS is no necessa y. Pa icipan s epo ed a desi e o speak o he pe son calling hem. They also wished o discuss hei heal h p oblems and sha e hei eelings, hei disease p og es- sion, ad ice ega ding medica ion and i s side e ec s, nu i ion, in e ac i e beha iou and any ecen ad ances in HIV ea men and cu e wi h he calle i gi en an oppo uni y The p oblem wi h his phone call is ha you canno speak any hing in o i . Only he pe son on he o he side (speaks) …We canno cla i y any hing …I ha op ion we e he e, i would be use ul …I would make me eel elie ed… I am a pa ien , I could ha e asked de ailed in o ma- ion abou ha abou aking able s, abou ea ing ood, how o mingle wi h e e yone, how o wo k …i may help. Only one pa icipan epo ed ha wo-way communi- ca ion would be unaccep able, as he pe son calling would know his disease s a us. (A) compu e ised message is ( he) bes way. Why should my in o ma ion be known o he pe son who is calling? Un amilia i y wi h using SMS in he se ing Many pa icipan s epo ed di ficul ies in accessing he SMS despi e being augh . One pa icipan epo ed ha nei he he no his s anda d 10 h educa ed spouse knew how o iew o send an SMS. Phone call is simple. The SMS is oo complica ed…I don’ unde s and SMS well… Pe cei ed isk o unin en ional disclosu e om he in e en ion Pa icipan s ound i impo an o p e en he disclosu e o hei HIV s a us, and made e o s o conceal he sou ce o he in e en ion (ie, he HIV clinic) om o he s. P e en ing unin ended disclosu e o HIV s a us Fea ul o o he s a ending he IVR call, pa icipan s a ely le he phone una ended on he day o he call. Pe sonalising he ime and day o he call enabled pa i- cipan s o choose imings ha a o ded some p i acy and secu i y while answe ing he call. No one has picked up he call. Wednesday (I ecei e) he phone call and (on) Sunday he SMS. On hese wo days I will ne e lea e he phone anywhe e. E en i he e is a p oblem, he phone will be kep wi h me. When iends and amily enqui ed abou he calls, some pa icipan s made a emp s o conceal he sou ce o he calls (ie, he HIV clinic). (When) people a ound me wan ed o know whe e he call comes om, I ell hem ha his call is om he Ai cel/mobile company and y o escape ( he) si ua ion. Pa icipan s conside ed ha equen calls migh inc ease he isk o unin en ional disclosu e o hei HIV s a us as he likelihood o someone else ecei ing he call while he phone is una ended cumula es wi h he numbe o calls ecei ed. I I ge he call epea edly, I am sca ed o he s will ecei e he call and will ge o know abou (my) illness. Pa icipan s epo ed ha hey kep he phone o ‘ hei ea s only’when in he p esence o o he s. One pa icipan epo ed going ou o he house o a end he call, while ano he pe sonalised he call o ecei e i a e he child en had le he house. Pa icipan s dele ed he SMS eminde s om hei phone inbox in an a emp o p e en o he s om iewing hem. One pa icipan ea ed ha i any PLHIV 6Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om ecei ing he same SMS eminde iewed his SMSs, his HIV s a us would be unin en ionally disclosed, as he o he pa ien would ecognise he SMS. Al hough in e en ion was no conside ed in usi e, pa icipan s ei he silenced o swi ched o he phone o p e en he calls om a ac ing a en ion in public places. DISCUSSION The esul s o ou s udy indica e ha pa icipan s had a ied iews o he in e en ion’s help ulness. Some pa - icipan s pe cei ed he conce n o he heal hca e p o- ide along wi h social and emo ional suppo . The IVR call was p e e ed o e he SMS eminde , and pa ici- pan s el hey needed a mo e engaging call. Pa icipan s ea ed HIV- ela ed s igma, and he e o e ook s eps o educe he isk o unin ended disclosu e o hei HIV s a us, ia he in e en ion. mHeal h in e en ions a e designed o imp o e adhe - ence by a ge ing o ge ulness, a common ba ie o ea men adhe ence. 26 Hence, ou in e en ion may be placed wi hin he con ex o he beha iou lea ning heo y (BLT) amewo k (figu e 3). 16 The pa icipan s in ou s udy desi ed o s ay heal hy. This need, an in e nal an eceden in he BLT amewo k, p obably encou aged hem o ake medicines. Fu he , he emin- de s p o ided ia mobile phones se ed as an ex e nal an eceden . Howe e , s udies ha e shown ha ex e nal eminde s used in isola ion may no be able o imp o e adhe ence. 27 28 E ec i e adhe ence suppo in e en- ions need amily o pee suppo , counselling and daily ea men suppo eflec ing hei complexi y. 28 Pa icipan s also epo ed eeling he ‘conce n’o hei heal hca e p o ide s owa ds hei heal h, mo e so wi h he oice calls han wi h he SMSs. Repo s o mHeal h in e en ions p o iding suppo and making he ecipien eel alued we e also obse ed in Kenya. 5929 These in e en ions, howe e , in ol ed a componen o pe sonal in e ac ion ia he phone unlike ou au oma ed in e en ion. A s udy om B i ish Columbia ha used ex messages in a ube culosis clinic epo ed ha pa icipan s el suppo ed and ca ed o by hei heal hca e p o ide . 30 Ano he s udy om Pe u- ela ed mobile phone eminde s o an ‘angel’and a‘ iend’gi ing he in e en ion an an h opomo phic cha ac e . 31 Simila epo s a e a ailable om an mHeal h smoking cessa ion ial in he UK. 32 Despi e eeling he conce n o he heal hca e p o- ide , he pa icipan s exp essed he need o wo-way communica ion. Engaging beneficia ies in communica- ion imp o es he e ficacy o mHeal h in e en ions. 33 Two-way communica ion in ol ing ex messages was ound o open communica ion channels and add ess unme needs o PLHIV in he Came oon Mobile Phone SMS (CAMPS) ial. Exp essing g a i ude, eques ing counselling, financial suppo and ad ice ega ding medica ion side e ec s— easons o using wo-way com- munica ion in he CAMPS ial—we e also mi o ed in ou s udy. 34 Al hough he con en o communica ion was iden ified du ing he design phase in h ee mHeal h ials om sub-Saha an A ica (SSA), none explo ed he need o in e ac ion. 91013 Fu he , hough he WelTel and CAMPS ials p o ided an oppo uni y o pe sonal in e ac ion wi h he heal hca e p o ide , only he WelTel ial was success ul. 910 The o he Kenyan ial was suc- cess ul e en hough he in e en ion was comple ely au oma ed. 10 The lack o in e ac ion is he e o e unlikely o be a majo eason o he nega i e ial esul s in ou s udy. Conside ing he p e alen li e acy le els in ou se ing, we chose o inco po a e a neu al pic u e SMS in he in e en ion. Elsewhe e, ex messages we e p e e ed o o he o ms o communica ion as hey we e conside ed o p o ide g ea e confiden iali y. 11 35 S udies om SSA epo high accep abili y o ex messaging o adhe - ence suppo , gene a ing awa eness ega ding HIV in ec- ion and communica ing labo a o y esul s in Figu e 3 The mHeal h an i e o i al ea men (ART) suppo in e en ion in he con ex o he Beha io Lea ning Theo y. Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 7 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om HIV-in ec ed indi iduals. 6913 None o hese s udies compa ed he accep abili y o ex messages o oice calls. Howe e , ano he s udy om Sou h A ica epo ed nea ly equal p e e ence o oice calls and ex messages as a mode o communica ion by a heal hca e p o ide . 36 A s udy in ol ing PMTCT indica ed ha ex messages we e p e e ed o b ie communica ions and oice calls o longe , mo e in e ac i e con e sa ions. 5 The lowe English language li e acy in ou con ex in compa ison o some o hese s udy con ex s could ha e esul ed in he p e e ence o oice calls o e ex mes- sages. I would be o alue o iden i y he p e e ed mode o communica ion and inco po a e i in o he in e en ion design. Ve bal communica ion has he po en ial o o e come he li e acy ba ie o ex messa- ging, and could be used in low-middle income con ex s such as ou s. The combined e ec o educa ion and ex messaging in imp o ing adhe ence, and he li e acy ba ie o he use o ex messages, eflec he need o al e na i es such as oice calls and pic u e messages o adhe ence suppo . 37 The in e en ion in ou s udy was designed subse- quen o a su ey ha used a semis uc u ed in e iew schedule. One in h ee pa icipan s in he su ey indi- ca ed a s ong p e e ence o oice calls. Howe e , he su ey did no explo e he p e e ed con en o commu- nica ion o he ex en o i s in e ac i eness. Quali a i e explo a ion o pa icipan pe cep ions while pilo ing he HIVIND ial in e en ion could ha e helped iden i y and add ess issues be o e he ial commenced. Also, hose who pa icipa ed in ou s udy we e exposed o he in e en ion o app oxima ely 2 yea s. Fi s -hand expe i- ence wi h he in e en ion p obably enabled pa icipan s o iden i y mo e ‘nega i es’ ega ding he SMS eminde han we e pe cei ed by po en ial ial pa icipan s. A ea o s igma om an unin ended disclosu e o hei HIV s a us om he in e en ion was el despi e he in e en ion’s neu al con en , pe sonalised iming and non- e bal in e ac i e communica ion. Non- e bal com- munica ions wi hin he IVR call p obably engende ed cu iosi y in hose wi nessing he call and a ea o HIV s a us disclosu e in he pa icipan s. S igma has been documen ed as a significan p oblem o pe sons li ing wi h HIV in he Indian se ing. 38 This ea -induced pa i- cipan s o keep hei phones wi h hem on he days ha hey ecei ed he call. They also p e e ed he eminde s weekly as hey could keep he phone wi h hemsel es o he day. As in o he s udies, pa icipan s in ou s udy also p e e ed no ecei ing he calls in public. 36 Li e a u e on s igma associa ed wi h mHeal h in e en- ions in HIV in ec ion is la gely in he con ex o ex messages. Reques s o code con en ha p e en ed dis- closu e o HIV s a us and minimised s igma we e obse ed in he de elopmen phase o he CAMPS in e - en ion. 39 S igma as a ba ie o mHeal h adhe ence suppo was iden ified in he design phase o he WelTel ial. 40 In ou s udy, he ea o s igma was less p o- nounced wi h he SMS eminde s han wi h he IVR calls, p obably because he SMS eminde s we e less popula and wen unno iced mo e o en. As in o he s udies, pa icipan s in ou s udy also eso ed o dele - ing o ex messages due o ea o unin ended disclos- u e o HIV s a us. 6 The pe cei ed isk o unin ended disclosu e o HIV s a us due o o he s iewing pa ici- pan s’ ex messages was also epo ed om China, SSA and Pe u. 13 29 31 36 41 Resea che s should weigh he ben- efi s agains he isk o engende ing he ea o s igma om mHeal h in e en ions in sensi i e heal h condi- ions like HIV in ec ion. Educa ing pa ien s and adequa e counselling suppo could o e come he s igma associa ed wi h he in e en ion. 34 Conside a ions o he u u e design o mHeal h in e - en ions: In e en ions pe sonalised o he beneficia ies’ need should be de eloped. 32 Mul icomponen in e en- ions could be designed such ha wo-way communica- ion wi h physicians/counsello s, along wi h in o ma ion on nu i ion, medica ion side e ec s and ad ances in HIV ca e, is inco po a ed. Al hough in e ac i e commu- nica ion may engage pa ien s o longe du a ions, i needs o be balanced agains he esou ces needed o in e en ion scale up. Sensi i i y o he disease and sociocul u al con ex s, gi en he possibili y o s igma i sensi i e in o ma ion we e in e cep ed, is needed. Ta ge ing hose wi h poo adhe ence, as sugges ed by pa icipan s in ou s udy, could imp o e he e ficacy o mHeal h in e en ions. 40 Fu he mo e, quali a i e assessmen s p io o and du ing a ial can help de elop and con ex ualise such in e en- ions 32 These conside a ions could be ex apola ed on o ch onic non-communicable disease and ube culosis. 42 Me hodological issues: The s udy sample was pu posi e and ep esen a i e o he pa icipan s in he HIVIND ial, enabling he ans e abili y o esul s. Fu he , a hick desc ip ion o he s udy con ex is p o ided in he me hods sec ion o his manusc ip , enabling he eade o judge he applicabili y o he esul s in hei se ing. To ensu e c edibili y o he findings, he esea che s discussed wi h each o he he in e iew guide, which was pilo es ed and modified sui ably o ensu e ha i accu - a ely assessed pa icipan pe cep ions. Simila ly, a con- sensus was a i ed a ega ding he amewo k, codes, hemes and sub hemes a e discussions wi h he coau- ho s. Dependabili y was suppo ed by desc ibing all p o- cedu es unde aken in de ail in he me hods sec ion o his manusc ip . Reflexi i y: The fi s au ho ’s unde s anding o he pos- i ion as a physician, a public heal h pe son and a esea che li ing and wo king in he se ing enabled he o e ec i ely con ex ualise he esponses o he pa ici- pan s. The mul idisciplina y esea ch eam in ol ed in he s udy con ibu ed o a holis ic in e p e a ion o pa - icipan pe cep ions ha we explo ed. The amewo k app oach is ideal o analyse heal h sys em ela ed da a as i d aws on many di e en adi- ions o answe a specific esea ch ques ion. Poised be ween he induc i e and deduc i e app oaches, i 8Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om encou ages he analysis o he esponden ’s in e p e - a ion o issues, in he ligh o he esea che s’expe i- ences, and suppo s he con ex ualisa ion o esea ch findings. 25 As he HIVIND ial was ongoing a he ime o he in e iews, he adhe ence s a us o he pa icipan s was unknown o he in e iewe . Hence, we we e unable o ela e he expe iences o he pa icipan s o hei adhe - ence wi hin he ial. CONCLUSION mHeal h adhe ence in e en ions go beyond hei in ended ole o p o ide a sense o ca e and suppo o he ecipien . Al hough au oma ed in e en ions a e impe sonal, hey could be he solu ion o scale up. The u ili y o conse ing spa se heal hca e esou ces by au o- ma ing in e en ions should be weighed agains he e ec i eness o a mo e engaging in e en ion in ol ing wo-way communica ion. Mul icomponen in e en ions pe sonalised on he basis o iming, con en , du a ion and adhe ence le els could be mo e engaging and accep able o beneficia ies, and mi iga e he isk o s igma. Mos o all, mHeal h in e en ions should be de eloped h ough ex ensi e beha iou al esea ch and es ed o e ec i eness p io o inco po a ing hem wi hin heal hca e sys ems. Au ho a ilia ions 1 Ka olinska Ins i u e , S ockholm, Sweden 2 Depa men o Communi y Heal h, S . John’s Medical College, Bangalo e, Ka na aka, India 3 YRG Ca e Cen e , Chennai, Tamil Nadu, India 4 Depa men o In e na ional Heal h, Uni e si y o Tampe e School o Medicine, Tampe e, Finland Acknowledgemen s The au ho s would like o hank he medical o ice s and s a a he h ee s udy si es and hei ART cen es, ha is, S . John’s Medical College, Bangalo e; KR Hospi al, Myso e; YRG Ca e, Chennai o suppo . The au ho s acknowledge he suppo om he p incipal in es iga o s o he HIVIND ial, D Kuma asamy N, Chennai and D Ani a She , Bangalo e. They also hank D Swa oop N, Senio esea ch o ice , HIVIND s udy a KR Hospi al, Myso e o o ganising he da a collec ion a he Myso e si e. They also acknowledge he con ibu ion o D BB Riwa i, D Sunil Kuma Dodde i and D Su esh Shas h i, along wi h he Na ional AIDS Con ol O ganisa ion, o enabling he es ablishmen o he s udy si e a Myso e and suppo ing he s udy. Con ibu o s RR pa icipa ed in concep , design, da a collec ion, analysis and d a ing he manusc ip . SP and KB we e esponsible o da a collec ion, analysis and c i ical e iew. SA was in ol ed in analysis and d a ing he manusc ip . PA was esponsible o analysis and c i ical e iew. ADC was in ol ed in concep , design, analysis and c i ical e iew. Funding The unding suppo om he Eu opean Union 7 h F amewo k Commission G an o he HIVIND ial 2009–2014, G an numbe 222946. Compe ing in e es s None decla ed. E hics app o al E hical clea ance o he s udy was ob ained om he espec i e e hics commi ees a he h ee si es unde he HIVIND s udy: ((1) S . John’s Medical College Hospi al, Bangalo e: IERB 1/369/08-92/2008; (2) Myso e Medical College and Resea ch Ins i u e: NO/PS/173/2010; (3) YRG Ca e Medical Cen e , Chennai: IRB Ap il18/2009). P o enance and pee e iew No commissioned; ex e nally pee e iewed. Da a sha ing s a emen No addi ional da a a e a ailable. Open Access This is an Open Access a icle dis ibu ed in acco dance wi h he e ms o he C ea i e Commons A ibu ion (CC BY 4.0) license, which pe mi s o he s o dis ibu e, emix, adap and build upon his wo k, o comme cial use, p o ided he o iginal wo k is p ope ly ci ed. See: h p:// c ea i ecommons.o g/licenses/by/4.0/ REFERENCES 1. Wo ld Heal h O ganiza ion. mHeal h: New ho izons o heal h h ough mobile echnologies: second global su ey on eheal h. Global Obse a o y o eHeal h se ies—Volume 3. 2011. WHO P ess, Wo ld Heal h O ganiza ion, 20 A enue Appia, 1211 Gene a 27, Swi ze land ISBN 978 92 4 156425 0. h p://apps.who.in /i is/ bi s eam/10665/44607/1/9789241564250_eng.pd 2. F ee C, Phillips G, Galli L, e al. The e ec i eness o mobile-heal h echnology-based heal h beha iou change o disease managemen in e en ions o heal h ca e consume s: a sys ema ic e iew. PLoS Med 2013;10:e1001362. 3. Muessig KE, Pike EC, Leg and S, e al. Mobile phone applica ions o he ca e and p e en ion o HIV and o he sexually ansmi ed diseases: a e iew. J Med In e ne Res 2013;15:e1. 4. Ho a h T, Azman H, Kennedy GE, e al. Mobile phone ex messaging o p omo ing adhe ence o an i e o i al he apy in pa ien s wi h HIV in ec ion. Coch ane Da abase Sys Re 2012;3: CD009756. 5. Jennings L, Ong’ech J, Simiyu R, e al. Explo ing he use o mobile phone echnology o he enhancemen o he p e en ion o mo he - o-child ansmission o HIV p og am in Nyanza, Kenya: a quali a i e s udy. BMC Public Heal h 2013;13:1131. 6. Siedne MJ, Habe e JE, Bwana MB, e al. High accep abili y o cell phone ex messages o imp o e communica ion o labo a o y esul s wi h HIV-in ec ed pa ien s in u al Uganda: a c oss-sec ional su ey s udy. BMC Med In o m Decis Mak 2012;12:56. 7. Kalichman SC, Kalichman MO, Che y C, e al. B ie beha io al sel - egula ion counseling o HIV ea men adhe ence deli e ed by cell phone: an ini ial es o concep ial. AIDS Pa ien Ca e STDS 2011;25:303–10. 8. Smillie K, Van Bo ek N, Abaki J, e al. A quali a i e s udy in es iga ing he use o a mobile phone sho message se ice designed o imp o e hi adhe ence and e en ion in ca e in Canada (WelTel BC1). J Assoc Nu ses AIDS Ca e 2014;25:614–25. 9. Les e RT, Ri o P, Mills EJ, e al. E ec s o a mobile phone sho message se ice on an i e o i al ea men adhe ence in Kenya (WelTel Kenya1): a andomised ial. Lance 2010;376:1838–45. 10. Pop-Eleches C, Thi umu hy H, Habya imana JP, e al. Mobile phone echnologies imp o e adhe ence o an i e o i al ea men in a esou ce-limi ed se ing: a andomized con olled ial o ex message eminde s. AIDS 2011;25:825–34. 11. Mille CW, Himelhoch S. Accep abili y o mobile phone echnology o medica ion adhe ence in e en ions among HIV-posi i e pa ien s a an u ban clinic. AIDS Res T ea 2013;2013:670525. 12. Puccio JA, Belze M, Olson J, e al. The use o cell phone eminde calls o assis ing HIV-in ec ed adolescen s and young adul s o adhe e o highly ac i e an i e o i al he apy: a pilo s udy. AIDS Pa ien Ca e STDS 2006;20:438–44. 13. Mbuagbaw L, Thabane L, Ongolo-Zogo P, e al. The Came oon Mobile Phone SMS (CAMPS) ial: a andomized ial o ex messaging e sus usual ca e o adhe ence o an i e o i al he apy. PLoS ONE 2012;7:e46909. 14. Saba é E. WHO Adhe ence o Long Te m The apies P ojec , Global Adhe ence In e disciplina y Ne wo k, and Wo ld Heal h O ganiza ion. Dep . o Managemen o Noncommunicable Diseases, Adhe ence o long- e m he apies: e idence o ac ion. Gene a: Wo ld Heal h O ganiza ion, 2003:196. 15. Ve loe M, Linn AJ, an Wee JC, e al. The e ec i eness o in e en ions using elec onic eminde s o imp o e adhe ence o ch onic medica ion: a sys ema ic e iew o he li e a u e. J Am Med In o m Assoc 2012;19:696–704. 16. Mun o S, Lewin S, Swa T, e al. A e iew o heal h beha iou heo ies: how use ul a e hese o de eloping in e en ions o p omo e long- e m medica ion adhe ence o TB and HIV/AIDS? BMC Public Heal h 2007;7:104. 17. De Cos a A, She A, Kuma asamy N, e al. Design o a andomized ial o e alua e he in luence o mobile phone eminde s on adhe ence o i s line an i e o i al ea men in Sou h India— he HIVIND s udy p o ocol. BMC Med Res Me hodol 2010;10:25. Rod igues R, e al.BMJ Open 2015;5:e007574. doi:10.1136/bmjopen-2015-007574 9 Open Access g oup.bmj.com on May 18, 2016 - Published by h p://bmjopen.bmj.com/Downloaded om