Background
Methods
Search strategies and selection criteria
Source | Variations of the following search terms were used | |
---|---|---|
PubMed, MEDLINE, EMBASE, CINAHL, Web of Science, Cochrane Library, WHOLIS, World Bank e-Library, BHIVA, International AIDS Society, International AIDS Conference, The Australasian HIV/AIDS conference | HIV infection | HIV, AIDS |
Technology intervention | mHealth, mobile health, cellular phone, cell phone, mobile phone, handphone, smartphone, personal digital assistant, portable media player, handheld video-game consoles, computer, personal computer, handheld and ultra-portable computer, desktop, laptop, palm pilot, netbook, mobile application, SMS, MMS, text messaging, reminder systems, email, instant messaging, chat room, live chat, multimedia, blogging, podcast, social media, Facebook, Twitter, MySpace, YouTube, social networking, internet forums, wireless technology, wi-fi, world wide web, website, internet, online, eHealth, telehealth, telemedicine | |
HIV treatment and care cascade | Testing, screen, diagnose, retention, linkage, care, cascade, follow up, counselling, treatment, suppression, PrEP, PEP, ART, medication adherence | |
Priority population | MSM, sex workers, people who inject drugs, transgender |
Screening and data extraction
Quality assessment
Results
Study | Country | Study design | Study Population Sample size (N=) | Technology mode | Purpose of study | Study description | Key outcomes of interest in this review | Results/Authors’ conclusions |
---|---|---|---|---|---|---|---|---|
SMS reminders to increase HIV testing and re-testing
| ||||||||
Bourne et al. 2011 | Australia | Cross-sectional | MSM I = 714 C = 1084 PI = 1753 | SMS | Evaluate impact of SMS reminder system on HIV/STI re-testing rates | SMS testing reminders were sent 3–6 monthly for MSM considered high-risk based on self-reported sexual behaviour. Comparison of HIV/STI re-testing rates among 3 groups. | HIV re-testing rates | Significant increase HIV re-testing rates within 9 months, I (64%) vs. C (30%) (p < 0.001) and I (64%) vs. PI (31%) (p < 0.001). SMS group was 4.4 times more likely to re-test than Control (95% CI 3.5 to 5.5, p < 0.001); SMS group was 3.1 times more likely to re-test than PI group (95% CI 2.5 to 3.8, p < 0.001). |
Instant messaging to promote HIV testing
| ||||||||
Zou et al. 2013 | China | Cross-sectional | MSM N = 429 | Instant messaging | Explore the feasibility of using internet outreach to encourage MSM to get tested for HIV | Two trained MSM volunteers promoted VCT using active (instant messaging, chat rooms, mobile phone, email) and passive (website banner ads) methods. Those who came for testing completed a survey and HIV/syphilis tests. | HIV testing uptake Motivation for seeking HIV testing Response to internet outreach | Instant messaging was the most effective mode for HIV testing promotion (1:4 men). The email was the least effective (1:140 men). Active internet outreach recruited younger MSM (X2 = 11.400, p = 0.001), never tested for HIV (X2 = 4.281, p = 0.039), tested less often (X2 = 5.638, p = 0.018). Note: Internet effective in encouraging testing but a confounding factor of financial reward. |
Social media campaigns to increase HIV testing in MSM
| ||||||||
Cheng et al. 2016 | China | Cross-sectional | MSM Primary outcome N = 22,282 Secondary outcomes N = 999 | Internet Social media | Evaluate impact of integrated service including internet based prevention services, online-to-offline line service linkage and offline one-stop shop service | Three project components: (1) internet-based prevention services to facilitate HIV prevention; (2) online-to-offline service linkage-online appointments for MSM for HIV testing; and (3) offline one-stop shop service-HIV testing and linkage to care for PLHIV | Primary outcome: HIV testing uptake Secondary outcomes: linkage to and retention in care | Six years of project implementation, the project accounted for 80% of total HIV tests (22,282/26,884) and new HIV diagnoses (999/1218) among MSM in Guangzhou. Of the 999 HIV-positive diagnoses, 948 (95%) linked to care services, while 891 (94%) of those linked were successfully retained in care. |
Ko et al. 2013 | Taiwan | Cross-sectional | MSM N = 1037 | Social media | Evaluate the effectiveness of iPOL in disseminating information about HIV, increasing the frequency of HIV testing, and reducing risky behaviours | The iPOLs actively disseminated HIV-related information via the Facebook social networking website and discussed and responded to questions from Internet-using MSM. MSM who visited the intervention or control website were surveyed after 6 months. | Frequency of online discussion or accessing information about HIV Incidence of HIV testing and condom use | MSM who visited intervention website were more likely to have HIV tests within 6 months (43.89% vs. 22.31%, p < 0.001); consistently use condoms during anal sex with online sex partners (34.15% vs. 26.19%, p = 0.004); receive HIV-related information (25.49% vs.10.47%, p < 0.001); discuss HIV issues with others (41.88% vs. 23.79%, p < 0.001); review articles about HIV (90.58% vs.79.73%, p < 0.001); and be asked about or discuss HIV-related questions (51.11% vs. 31.78%, p < 0.001) than those on the control website. |
Web-based health promotion to increase HIV testing
| ||||||||
Minas et al. 2012 | Australia | Cross-sectional | MSM Various sample size as multiple methods of evaluation | Website | Evaluation of a communication strategy to improve the awareness and appropriate use of nPEP | The communication strategy included: • development of nPEP information pamphlet for distribution through gay websites • nPEP resources (aimed at people at risk of acquiring HIV) and the Western Australia nPEP guidelines aimed at HCWs available on Western Australia AIDS Council website • 24 h nPEP phone line |
Immediate outcomes:
• Access and use of nPEP information • Awareness of nPEP among MSM and HCWs
Ultimate outcomes:
• nPEP treatment practice and follow up testing | Significant increase in the proportion of clients tested for HIV at 3 to 4 months after the initial visit (38.8% in 2002–2005 to 51.9% in 2008–2010, p = 0.023). No increase in clients tested at six to seven month after the initial visit. |
Pedrana et al. 2012 | Australia | Cross-sectional | MSM N = 295 | Social marketing campaign | Evaluate the impact of a social marketing campaign in 2008–2009 aimed to increase health-seeking behaviour and STI testing and enhance HIV/STI knowledge in gay men | Impact evaluation of “Drama Downunder” health promotion campaign, by surveying online sample of gay men and analysing HIV and other STI testing data from high case load clinics before, during and after the campaign. | HIV/STI testing | Compared with the pre-campaign period, 17% increase in HIV testing rate (p < 0.01) were observed during the initial campaign period and 27% increase (p < 0.01) during the continued campaign period. |
Wilkinson et al. 2016 | Australia | Cohort | MSM
N = 242
| Social marketing campaign | Evaluate impact of campaign on HIV sexual health testing | Impact evaluation of social marketing campaign, by surveying online sample of MSM and analysing HIV and other STI testing data from Victorian Primary Care Network for Sentinel Surveillance HIV network before, during and after the campaign. | HIV/STI testing among MSM | Though increasing HIV/STI testing trends were observed for MSM pre and post marketing period, there was insufficient evidence to significantly attribute impact exclusively to the campaign. |
Study | Country | Study design | Study Population Sample size (N=) | Technology mode | Purpose of study | Study description | Study duration | Key outcome assessed of interest in this review | Results/Authors’ conclusions |
---|---|---|---|---|---|---|---|---|---|
Mobile phone calls or SMS reminders to increase adherence
| |||||||||
Huang et al. 2013 | China | RCT | PLHIV on ART N = 172 | Mobile phone call | Investigate the effect of a phone call intervention to promote adherence to ART and QoL among PLHIV | I=Usual care combined with bi-weekly 3-min reminder phone calls made by trained registered nurse or other health workers C=Usual care Both groups included HIV treatment-naive and treatment-experienced HIV patients. | 12 weeks | Self-reported adherencea QoL | No significant improvements in adherences rates in the intervention group. Significant improvements in QoL in the intervention groups for treatment-naïve HIV patients (physical health p = 0.003; level of independence p = 0.018; environment p = 0.002; and spirituality/religion/personal beliefs p = 0.021) at 3 months. |
Shet et al. 2014 | India | RCT | PLHIV initiating ART N = 631 | Mobile phone call and SMS reminders | Assess whether customised mobile phone reminders would improve adherence | I=Standard care and weekly customised, interactive, automated voice reminders, and a pictorial message sent weekly to the participants’ mobile phones C=Standard care | 96 weeks | Time to virological failure ART adherence measured by pill count | No significant effect of the mobile phone intervention on either time to virological failure or ART adherence at the end of two years of therapy. |
Swendeman et al. 2015 | India | Cohort | PLHIV N = 44 | IVR system using mobile phones | To design, pilot and refine IVR intervention to support ART adherence | All subjects received two IVR calls daily, timed to dosing schedules with brief messages on strategies for medical, mental health and nutrition and hygiene. | 4 weeks | Self reported adherence at baseline and 1 month | Self reported missed doses decreased from 39 to 18% at one month (p = 0.005). |
Tran et al. 2013 | Vietnam | Cross sectional | PLHIV N = 1016 | Mobile phone reminders | Assess ART adherence and its determinants among PLHIV | Multi-site cross-sectional survey: Inpatients and outpatients adult PLHIV were interviewed using structured questionnaires | N/A | Self-reported medication adherence questions Questions about medication adherence self-efficacy, reasons for missing doses and adherence aids | The main devices used for adherence supports were mobile phone alarms (62.2%). In multivariate analysis, the use of reminder strategies, such as mobile phone alarms was associated with ≥95% optimal adherence (Coefficient 0.89, 95% CI 0.02 to 1.99, p < 0.05). |
Uzma et al. 2011 | Pakistan | RCT | PLHIV initiating ART N = 76 | Mobile phone call reminders | Assess the efficacy of interventions for improving adherence to ART regimens | I = Routine counselling and weekly phone/mobile phone call reminders C = Routine counselling | 8 weeks | Self-reported adherence Pill identification test; defined as ≥95% CD4 counts Viral load | Those in the intervention condition had significantly better self-reported adherence (p < .001) and significantly lower viral load (p = .012). |
Biofeedback to improve adherence
| |||||||||
Sabin et al. 2010 | China | RCT | PLHIV on ART N = 64 | EDM | Determine whether EDM feedback improved ART adherence | I=Counselling using EDM feedback C=No EDM feedback Both groups included participants assessed 6 months after initiation of treatment as either ‘low adherers’ or ‘high adherers’ | 12 weeks pre-interventi-on 12 weeks interventi-on | Adherence measured by EDM; defined as ≥95% Markers of disease progression | At month 12 intervention, mean adherence had risen significantly (p = 0.003) among intervention subjects to 96.5% but remained unchanged in controls. The mean CD4 count rose by 90 cells/μl and declined by 9 cells/μl among intervention and control subjects, respectively. |
Sabin et al. 2015 | China | RCT | PLHIV N = 119 | EDM including real time wireless medication communicator | Determine whether EDM feedback improved ART adherence | Subjects with optimal and suboptimal adherence randomised to intervention or control arms. I=Individualised SMS mobile phone reminders triggered by late dose taking, and data-informed counselling. C=No reminders, standard adherence counselling | 12 weeks pre-interventi-on 24 weeks interventi-on | Adherence measured by EDM; defined as ≥95% Markers of disease progression | At last intervention month, the proportion of optimal adheres was significantly higher in I group 87.3% vs. 51.8% (RR for optimal adherence in month 9, I vs. C, 1.69; CI: 1.29 to 2.21, p < 0.001). The mean adherence during intervention period was significantly higher in I group (I vs. C: 96.3% vs. 88.9%, p < 0.001). Post intervention clinical outcomes not significant. |
Perera et al. 2014 | New Zealand | RCT | PLHIV on ART N = 28 | Smartphone app | Examine the efficacy of a smartphone application incorporating personalised health-related visual imagery to improve adherence to ART. | I = 24-h medication clock and augmented version of the smartphone app which comprised a daily real-time graphical representations of the current estimated plasma concentrations of antiretroviral drugs and simulation of protection against HIV C = standard version of the smartphone app which comprised a 24-h medication clock displaying daily ART dosing schedule and allowed participants to record when they had taken their medications each day | 12 weeks | Viral load Self-reported adherence Pharmacy dispensing records | Participants in the intervention group showed a significantly higher level of self-reported adherence to ART at 3 months (p = 0.03) and decreased viral load (p = 0.023). Greater usage of the extra components of the augmented application was associated with greater perceived understanding of HIV infection and increased perceived necessity for ART. |
Study | Country | Study design | Study Population Sample size (N=) | Technology mode | Purpose of study | Study description | Study duration | Key outcome assessed of interest in this review | Results/Authors’ conclusions |
---|---|---|---|---|---|---|---|---|---|
Internet to increase HIV testing and treatment uptake
| |||||||||
Mi et al. 2015 | China | Quasi-RCT | HIV positive MSM N = 202 | Internet | Investigate efficacy of web-based intervention in increasing HIV testing and treatment uptake among MSM | I = participants were given access to online program including an information exchanges website, a bulletin board system, an individualised online counselling with trained peer educators, and an animation game C = standard of care | 24 weeks | Early initiation of ART Motivating partners to receive ART | The intervention group had significant increases in motivating partners to accept HIV testing (42.3% vs 25.5%, p = 0.0156) compared with the control group. There were no between-group differences on receiving ART (p = 0.368). |
Characteristics of the included studies
STUDY | Assigned design | COMPONENT RATINGS | GLOBAL RATING | |||||
---|---|---|---|---|---|---|---|---|
Selection bias | Study design | Confounders | Blinding | Data collection method | Withdrawals and dropouts | |||
eHealth and HIV testing and linkage to care
| ||||||||
Bourne et al | Cross-sectional | Moderate | Moderate | Strong | Moderate | Weak | Weak | Moderate |
Zou et al | Cross-sectional | Weak | Moderate | Weak | Moderate | Weak | Weak | Weak |
Cheng et al | Cross-sectional | Weak | Weak | Weak | Moderate | Weak | Weak | Weak |
Ko et al | Cross-sectional | Moderate | Moderate | Strong | Moderate | Weak | Weak | Moderate |
Minas et al | Cross-sectional | Weak | Moderate | Weak | Moderate | Weak | Weak | Weak |
Pedrana et al | Cross-sectional | Moderate | Moderate | Weak | Moderate | Weak | Weak | Weak |
Wilkinson et al | Cohort | Weak | Weak | Weak | Moderate | Weak | Weak | Weak |
eHealth and HIV treatment adherence
| ||||||||
Huang et al | RCT | Moderate | Strong | Strong | Moderate | Strong | Strong | Strong |
Shet et al | RCT | Moderate | Strong | Strong | Moderate | Strong | Strong | Strong |
Swendeman et al | Cohort | Weak | Moderate | Weak | Moderate | Weak | Strong | Weak |
Tran et al | Cross-sectional | Moderate | Weak | Weak | Moderate | Weak | Weak | Weak |
Uzma et al | RCT | Moderate | Strong | Strong | Moderate | Strong | Strong | Strong |
Sabin et al. 2010 | RCT | Moderate | Strong | Strong | Weak | Strong | Strong | Moderate |
Sabin et al. 2015 | RCT | Moderate | Strong | Strong | Moderate | Strong | Strong | Strong |
Perera et al | RCT | Moderate | Strong | Strong | Moderate | Strong | Strong | Strong |
eHealth and HIV testing, linkage to care and treatment adherence
| ||||||||
Mi et al | RCT | Moderate | Strong | Strong | Moderate | Weak | Strong | Moderate |