Pilot study of an e-intervention for symptoms of depression among university students in South Africa

Gericke, Franco (2020-03)

Thesis (MA)--Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: Background. Major Depressive Disorder (MDD) is associated with a range of adverse outcomes among university students, including academic failure, suicidal thoughts and behaviours, and severe role impairment. Despite the variety of effective treatments available, most university students do not seek or receive help. Evidence suggests that internet-based interventions (e-interventions) might be as effective as existing treatments available to treat MDD. Therefore, e-interventions have been suggested to address the treatment gap among university students. However, e-interventions have not been extensively used or evaluated in low- and middle-income countries (LMIC) like South Africa (SA). Therefore, I set out to pilot the use of an e-intervention, namely iCare, aimed at addressing depressive symptoms among SA university students. Aims. This randomised external pilot study had two primary aims: (1) to assess the key feasibility aspects of iCare, namely (a) recruitment, (b) randomisation, and (c) implementation (utilisation, retention and follow-up rates, and the assessment of the outcome measures); and (2) to investigate students’ experiences of using iCare and to document their suggestions on improving iCare and making it more culturally appropriate for use in the SA context. Methods. All first-year university students (n=5094) were invited to complete a voluntary mental health survey. A total of 138 participants met the inclusion criteria and were randomised (according to a 2:1 ratio) to iCare (n=91) and treatment-as-usual (TAU) (n=47). iCare participants received one-month and three-month post-intervention follow-up assessments to monitor symptom change. Each iCare participant also received an email inviting them to participate in an individual in-depth post-intervention interview. The key feasibility aspects were quantified and assessed using various statistical analyses. A large proportion of the participants did not complete he follow-up assessments. The outcome measures were, therefore, assessed using three iii approaches commonly used to handle missing data in clinical trials: (1) complete case analysis (CCA), (2) intention-to-treat analysis using multiple imputations (ITTA-MI), and (3) per-protocol analysis (PPA). Sample size estimations to scale this study to a randomised control trial were conducted based on the ITTA-MI, using (1) safeguard power analysis, and (2) the minimal clinically significant effect. The interviews were transcribed and analysed through thematic analysis following a data-driven approach. Findings. Only 31.87% (n=29) of participants indicated their willingness to use iCare, of which 24.18% (n=22) started with iCare’s first session. A substantial number (72.73%, n=16) of participants dropped out during treatment. Low follow-up rates were observed at the one-month (65.9%, n=60) and three-month (71.4%, n=65) post-intervention follow-up assessments. However, the ITTA-MI indicated a significant small to moderate decrease in iCare participants’ depressive symptoms at these assessments. The PPA indicated that a large proportion (50%, n=3) of iCare completers achieved a successful treatment outcome. I identified seven superordinate themes and several sub-themes. The majority of participants felt the anonymity and accessibility of iCareen-abled them to overcome the barriers they faced to traditional face-to-face therapy; others found the lack of direct human contact to be problematic, expecting iCare to mimic the responsiveness and reflectiveness of face-to-face therapy. The perceived time-consuming nature of iCare was put forward by some participants for their discontinuation of iCare; while others discontinued due to symptom improvement. The majority of participants indicated the need for more engaging and interactive content and suggested that iCare should be used in addition to, rather than replacing, traditional face-to-face therapy.Conclusion. The findings of this study indicate that there is a proportion of students for whom e-interventions may provide the first step to accessing mental health services, although nota replacement for traditional face-to-face therapy. A randomised control trial is needed to determine the efficacy of iCare. I also offer recommendations for addressing the key feasibility aspects that were identified, and for future research on e-interventions among university students in SA.

AFRIKAANSE OPSOMMING: Agtergrond. Major Depressieversteuring word onder universiteitstudente geassosieer met ’n verskeidenheid nadelige effekte, insluitendakademiese mislukking, selfdoodgedagtes en -gedrag, en ernstige funksioneringsprobleme. Ten spyte van die verskeie effektiewe behandelings wat beskikbaar is, soek en ontvang die meeste universiteitstudente nie hulp nie. Navorsing bewys dat internet-gebaseerde intervensies (e-intervensies) net so effektief kan wees soos die bestaande depressiebehandelings. E-intervensies word dus aanbeveel om die gebrek aan behandeling onder universiteitstudente aan te spreek. In lae- en middelinkomstelande soos Suid-Afrika (SA) is e-intervensies egter nog nie in diepte gebruik of geëvalueer nie. Ek het dus onderneem om die gebruik van ’n e-intervensie genaamd iCare uit te toets, wat daarop gemik is om depressiesimptome onder SA universiteitstudente aan te spreek.Doelstellings. Hierdie ewekansige eksterne toetsstudie het twee primêre doelstellings gehad: (1) om die lewensvatbaarheid van iCare te assesseer in terme van (a) werwing, (b) ewekansige verdeling, en (c) implementering (benutting, retensie- en opvolgkoerse, en die assessering van die uitkomste); en (2) om die studente se ervaring van iCare te ondersoek en hul voorstelle te dokumenteer oor hoe om iCare te verbeter en dit meer kultureel toepaslik vir die SA konteks te maak. Metodologie. Alle eerstejaarstudente (n=5094) is uitgenooi om ’n vrywillige geestesgesondheidsvraelys in te vul. ’n Totaal van 138 deelnemers het aan die insluitingskriteria voldoen en is ewekansig verdeel (in die verhouding 2:1) tussen iCare (n=91) en gewone behandeling (n=47). Die iCare-deelnemers het een maand asook drie maande ná die intervensie opvolg-assesserings ontvang om simptoomveranderinge te monitor. Elke iCare-deelnemer het ook ’n e-pos ontvang wat hom/haar uitnooi om aan ’n individuele in-d iepte post-intervensie onderhoud deel te neem. Die lewensvatbaarheidsaspekte van só ’n intervensie is gekwantifiseer en geassesseer deur die gebruik van verskeie statistiese ontledings. ’n Groot proporsie van die deelnemers het nie die opvolg-assesserings voltooi nie, dus is die uitkomste geassesseer met behulp van drie benaderings wat algemeen gebruik word om vermiste data in kliniese toetse te hanteer: (1) volledige gevalsontleding (CCA), (2) veelvuldige imputasies (ITTA-MI), en (3) per-protokol-ontleding (PPA). Om te bepaal wat nodig sal wees om hierdie studie na ’n ewekansige kontroletoets uit te brei, is steekproefskattings op grond van die ITTA-MI bereken met behulp van (1) ontleding met magsbeskerming, en (2) die minimaal klinies beduidende effek. Die onderhoude is getranskribeer en ontleed met tematiese analise, in ooreenkoms met ’n datagedrewe benadering. Bevindinge. Slegs 31.87% (n=29) van die deelnemers het hul bereidwilligheid aangedui om iCare te gebruik. Hiervan het 24.18% (n=22) iCare se eerste sessie begin. ’n Aansienlike aantal deelnemers (72.73%, n=16) het tydens behandeling onttrek. Lae opvolgkoerse is een maand (65.9%, n=60) en drie maande (71.4%, n=65) ná die intervensie tydens opvolg-assesserings waargeneem. Die ITTA-MI het egter tydens hierdie assesserings ’n beduidende klein tot middelmatige afname in die iCare-deelnemers se depressiesimptome getoon. Die PPA het getoon dat ’n groot proporsie (50%, n=3) van die iCare-voltooiers (n=6) ’n suksesvolle behandelingsuitkoms bereik het. Ek het sewe hooftemas en verskeie subtemas geïdentifiseer. Die meerderheid deelnemers het gevoel dat die anonimiteit en toeganklikheid van iCare hulle in staat stel om die hindernisse te oorkom wat hulle verhoed om tradisionele een-tot-een-terapie te gebruik. Ander het die gebrek aan direkte menslike kontak problematies gevind omdat hulle verwag het dat iCare die responsiwiteit en refleksiwiteit van een-tot-een-terapie sou deel. Party deelnemers het nie met iCare voortgegaan nie omdat hulle dit as tydrowend gesien het, terwyl ander as gevolg van simptoomverbetering onttrek het. Die meerderheid deelnemers het daarop gewys dat meer boeiende en interaktiewe inhoud nodig is, en voorgestel dat iCare aanvullend tot tradisionele terapie gebruik word, eerder as in plaas daarvan. Gevolgtrekking. Die bevindinge van hierdie studie dui daarop dat daar ’n proporsie studente is vir wie e-intervensies moontlik ’n eerste stap kan bied om ’n geestesgesondheidsdiens te begin gebruik, alhoewel dit nie ’n plaasvervanger vir tradisionele terapie is nie. ’n Ewekansige kontroletoets sal nodig wees om die doeltreffendheid van iCare te bepaal. Ek maak aanbevelings oor hoe om die lewensvatbaarheidsaspekte wat in hierdie studie geïdentifiseer word, aan te spreek, asook vir toekomstige navorsing oor e-intervensies onder universiteitstudente in SA.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/107758
This item appears in the following collections: