Mobile phone reminders and peer counseling improve adherence and treatment outcomes of patients on ART in Malaysia: A randomized clinical trial

Background Adherence to treatment remains the cornerstone of long term viral suppression and successful treatment outcomes among patients receiving Antiretroviral Therapy (ART). Objective(s) Evaluate the effectiveness of mobile phone reminders and peer counseling in improving adherence and treatment outcomes among HIV positive patients on ART in Malaysia. Methods A single-blind, parallel group RCT conducted in Hospital Sungai Buloh, Malaysia in which 242 adult Malaysian patients were randomized to intervention or control groups. Intervention consisted of a reminder module delivered through SMS and telephone call reminders by trained research assistants for 24 consecutive weeks (starting from date of ART initiation), in addition to adherence counseling at every clinic visit. The length of intended follow up for each patient was 6 months. Data on adherence behavior of patients was collected using specialized, pre-validated Adult AIDS Clinical Trial Group (AACTG) adherence questionnaires. Data on weight, clinical symptoms, CD4 count and viral load tests were also collected. Data was analyzed using SPSS version 22 and R software. Repeated measures ANOVA, Friedman’s ANOVA and Multivariate regression models were used to evaluate efficacy of the intervention. Results The response rate after 6 months follow up was 93%. There were no significant differences at baseline in gender, employment status, income distribution and residential location of respondents between the intervention and control group. After 6 months follow up, the mean adherence was significantly higher in the intervention group (95.7; 95% CI: 94.39–96.97) as compared to the control group (87.5; 95% CI: 86.14–88.81). The proportion of respondents who had Good (>95%) adherence was significantly higher in the intervention group (92.2%) compared to the control group (54.6%). A significantly lower frequency in missed appointments (14.0% vs 35.5%) (p = 0.001), lower viral load (p = 0.001), higher rise in CD4 count (p = 0.017), lower incidence of tuberculosis (p = 0.001) and OIs (p = 0.001) at 6 months follow up, was observed among patients in the intervention group. Conclusion Mobile phone reminders (SMS and telephone call reminders) and peer counseling are effective in improving adherence and treatment outcomes among HIV positive patients on ART in Malaysia. These findings may be of potential benefit for collaborative adherence planning between patients and health care providers at ART commencement.


Modes of transmission:
HIV is present in blood, semen, and other body fluids such as breast milk and can be transmitted through heterosexual route (>75%) and from mother to child (5-10%) with wide racial and ethnic variations among countries" subpopulations as a result of immigrants influx from high prevalence countries. The incidence in injecting drug users also vary from country to country with UK having <1% and >50% in Eastern Europe, Vietnam, India, China and Malaysia.
Clinical history and features: Primary infection is characterized by sero-conversion illness in 70-80% of patients owing to high levels of circulating HIV 1. Common symptoms range from fever (80%), malaise (68%), arthralgia (54%), maculapapular rash (51%), myalgia, oral ulcers and pharygitis: neurological complications such as GBS and Bell"s palsy occasionally occur. Most symptoms generally resolve in 7-10 days though the illness is more severe in a few patients as a result of opportunistic infections (eg PCP, esophageal candidiasis) reflecting a more profound CD4 cells depletion often to as low as <200cells/mm3 (Penny Lewthwaite, Ed Wilkins, 2009). Patients with severe symptoms are more likely to have a rapid disease progression, though the illness is mild in many patients and only recognized through retrospective enquiry. Symptomatic recovery parallels a rise in CD4 count and viral load suppression. Diagnosis is made by detection of HIV RNA in serum or immunoblot assays (antibody detection). The appearance of specific anti-HIV antibodies in serum (seroconversion) takes place later at 2-12 weeks (median 8weeks) (Penny Lewthwaite, Ed Wilkins, 2009). In asymptomatic infection, the CD4 count increases again usually below its pre-infection level and viral load stabilizes at a particular set point for several years. Patients with CD4 count of 350-800 cells/mm3 are usually well. The period between infection and development of symptomatic or late-stage disease varies. Infected individuals may be completely asymptomatic or have generalized lymphadenopathy (CDC classification category disease A, WHO stage 1). Without treatment, the CD4 count eventually declines and at CD4 count <350, the individual becomes increasingly susceptible to a wide range of opportunistic infections ranging from TB, Pneumonia, Herpes zoster, recurrent oral and vaginal candidiasis and oral hairy leukoplakia. This is observed at CD4 count 200-350 cells/mm3 and is typical of CDC classification category disease B or WHO stage 2 & 3. At CD4 count <200 cells/mm3 other opportunistic infection or HIV-related tumor may develop. AIDS (CDC classification category disease C or WHO stage 4) is defined by the appearance of other opportunistic infections and tumors such as PCP, cerebral toxoplasmosis, Kaposi"s sarcoma, oesophageal candidiasis and diarrhoea diseases.  Breckenridge, 2009). These drugs are often given in combinations in order to achieve maximum viral suppression and prevent resistance to any one of the drugs by the virus. These classes of drugs are characterized by unique pharmacodynamic and pharmacokinetic properties, adverse effects, drug interactions and cross resistance between and among groups.

EFFECTIVENESS OF MOBILE PHONE TECHNOLOGY IN IMPROVING ADHERENCE AND
Adherence to treatment: According to the WHO Adherence Project, 2003, adherence can be defined as "the extent to which a person"s behaviourtaking medications, following a diet, and/or executing lifestyle changes, corresponds with the agreed recommendations from a health care provider". Medication adherence may also be defined as the extent to which a patient takes his or her medications in the way intended by the health care provider (Machtinger, E.L., Bangsberg, D.R., 2005). Medication adherence in HIV/AIDS has been defined by Jani (2004:1) as "the ability of a person living with HIV/AIDS to be involved in choosing, starting, managing and maintaining a given therapeutic combination medication regimen to control viral replication and improve immune function". According to WHO (2005), minimum adherence levels of 95% are required for treatment success (Fairly et al., 2005;Jani, 2004;Paterson et al., 2000;Saple, 2005). Other studies have demonstrated that ARV medication adherence levels of 54 -95% is required to maintain prolonged viral load suppression depending on the allowable flexibility margins of each ART program. It is however generally accepted that those clients who adhere strictly to their medication achieve viral suppression, while those who are not adherent may not (Lee Preininger et al., 2011). Adherence is a primary determinant of the effectiveness of treatment because poor adherence attenuates optimum clinical benefit. Good adherence improves the effectiveness of interventions, aimed at promoting healthy lifestyles such as diet modification, improved physical activity, non-smoking and safe sexual behaviours, and of the pharmacologicalbased risk-reduction interventions. It also affects secondary prevention and disease treatment interventions (WHO, 2003). Treatment adherence also involves regular clinic attendance in line with schedules jointly determined by the patient and the health care provider, based on approved international protocols for standard care of a particular disease. The success of antiretroviral treatment has been widely correlated with strict medication adherence (adherence >95%), behavioural and lifestyle modifications, disclosure issues, nutrition, psychosocial support, issues of stigma and discrimination, employment status and family income, in different studies. However, most studies have identified strict medication adherence as the single most important predictor of immunologic, viral and clinical outcome of treatment in HIV clients. Adherence measurements can be grouped into measures based on a patients self-report of pill-taking behaviour and measures that are objective surrogates of pill-taking behaviour such as pill count or MEMS cap (Machtinger and Bangsberg, 2005).

B. Research methodology
Study Location: The study will be conducted in Hospital Sungai Buloh, Selangor, Malaysia.
Study Design: It will be an experimental (randomized Controlled Trial) study design, single-blinded among newly diagnosed HIV positive patients who are eligible to commence ART from December 2013 to August 2014. Following enrollment and allocation to study arm, baseline adherence, clinical symptoms, CD4 count, viral load, Weight, TB status and Opportunistic infection index will be assessed and measured for the 2 groups, prior to the commencement of ART. The intervention group will receive appointment and medication reminders by text-messaging and telephone calls, in addition to standard care. Based on the Theory of Planned Behaviour Model (Azjen, 1985), appointment reminders will be sent by Short Text Message (in Malay and English languages) three days prior to each individual client"s scheduled appointment and telephone call reminders a day prior to the scheduled clinic appointment visits. Weekly short text messages will also be sent to clients to remind them of taking their medications. The control group will only receive standard care and paper-based appointment during clinic visits. Both groups of clients will be followed up for a period of 6 months on ART. Repeated CD4 count, viral load, Weight, TB status and OI index assessment and measurements will be conducted at 3 and 6 months, respectively to determine treatment outcome. Medication adherence will be measured using structured self-report questionnaires. Clinic visits will be measured through medical records of clinic attendance and clinicians" notes.
Study Duration: Recruitment and data collection will be from December 2013 to February 2015.

Study population:
The study population is all HIV positive clients assessed and found to be eligible for ART commencement at Hospital Sungai Buloh, Selangor Inclusion criteria: Age range of 15 to 65 years eligible to commence ART, able to read text messages and have a mobile phone access. Sampling method: A simple random sampling method will be used to select eligible patients for the study.
Random allocation: Participants will be allocated to either of the intervention or control groups based on simple, complete randomization technique using their unique ID numbers from a random number table. An independent project biostatistician will generate 1:1 randomization numbers for study arm assignments using a random number generating program. Random allocation will consider and eliminate all forms of possible bias based on the socio-demographic characteristics of the clients.
Allocation concealment: Written allocation of assignment will be sealed in individual opaque envelopes marked with study identification numbers which will be available in the study clinic to allocate the target number of participants. After consenting to participate and meeting the inclusion criteria, screened subjects will be enrolled and immediately afterwards are assigned to a randomized study arm by the study coordinator opening the sealed envelopes to determine allocation. Age and gender will be assessed for balance of study arm allocation.
Blinding: Investigators will be blinded to intervention assignments. Z1-a/2 = standard error when a = 0.05 (95% Confidence Interval) = 1.96 Z1-ß = standard error associated with power = 0.842 (ß = 0.20) Power (1-ß) = 80% Expected CD4 increase within 6months of treatment with >95% adherence is 50 cells/ml n = 2 x (132)²[1.96+0.842]²/(550-500) n = 110 which is the minimum required sample size 110 + 10% of 110= 121 (attrition) 121 x 2 = 242 (for experimental and control groups) Hence, sample size of 121 per group will be used for the study Intervention Protocol: For the intervention group -a "reminder module" will be developed and delivered via SMS and telephone calls by 2 PLHIV (research assistants). This will include weekly SMS medication reminders, SMS reminders 3 days prior to scheduled clinic appointments and telephone call reminders a day prior to scheduled clinic appointment (in addition to standard careadherence counseling). To ensure confidentiality, typical medication reminder text messages will include a short slogan in Malay language "Apa khabar" "Ini untuk menberithau anda ubat" meaning "How are you?" "This is to remind you of your medications". Appointment reminder text message would be "Apa khabar" "Tolong ingat tarikh temu janji lusa" meaning "How are you?" "Remember your appointment day after tomorrow" and telephone conversation would be standardized and short, with the message "Apa khabar" "Tolong ingat tarikh temu janji besok" meaning "How are you?" "Remember your appointment tomorrow". The control group will receive standard care -routine adherence counseling only. Patients will not be required to provide any responses to the text messages. A call and SMS log will be recorded and kept. Two PLHIV that are stable and have been consistent and adherent to ARV medications for at least 2 years will be recruited as research assistants. Using standard training module developed and validated by a team of experts from Universiti Putra Malaysia and Hospital Sungai Buloh, a two-day onsite training of the research assistants on HIV prevention methods, care and treatment issues and how to respond to common questions related to medication adherence, including medication side effects, when asked by study participants, will be conducted. In addition to sending SMS and making telephone calls to patients, they will also be required to provide counseling support and provide guidance to patients (when required) in filling out medication adherence questionnaires during every clinic visits. Patients in the intervention group will be further divided into 2 sub-groups, each under the supervision of one research assistant, who will be responsible for providing them with counseling and psychosocial support, in addition to the care and management provided by the clinicians and nurses, throughout the duration of the study. Each patient in the intervention group would undergo a minimum of three (during clinic visits at month 1, month 3 and month 6) individual counselling sessions with the research assistants lasting an average of 15 minutes per encounter. An intervention manual containing the training modules and implementation guidelines will be provided to the research assistants as spiral-bound hard copy reference document, to guide their interactions with the patient and assure quality and uniformity in the implementation of the intervention.
Data collection: Research data will be collected using the below plan: • How: 2 types (Primary and Secondary) of data will be collected from 3 main sources; (a) Hospital medical records -Patients" socio-demographic and clinic attendance data will be collected from hospital records; pharmacy records will be utilized for measuring adherence to drug refill appointments, while clinicians" notes will be assessed for clinical information regarding weight, TB status, and OI index of clients (through standardized OI monitoring forms). Laboratory records will be accessed for baseline and follow-up CD4 count and viral load test results.
(b) Structured questionnaire and standardized medication adherence monitoring forms would be used to collect information from clients" medication adherence self-reports during interaction with clinician/nurses/pharmacists at every scheduled clinic visit/appointment.
• Who and When: Socio-demographic (at enrolment) and clinic visit information (at every visit) will be collected by clinic medical record officers, laboratory information by laboratory scientists (at baseline, Months 3 and 6) and clinical and medication adherence information will be collected by clinicians/nurses/pharmacists at every client visit.
A log book on phone calls and text messages will also be recorded.
• Duration: Data will be collected over a 13 months period. will be adapted with slight modifications, for the purpose of this study. Specialized questionnaires will be used at baseline and follow up. The questionnaire will be pre-tested on a population not included in the sample for validity and reliability.  • Data Storage: All data generated from this study will be stored for a maximum of 3 years after the completion of the study, after which they will be destroyed. However, during this period they will remain joint intellectual property of Universiti Putra Malaysia and Hospital Sungai Buloh.

Primary Outcomes:
The first primary outcome is (improved) patient adherence to ART at six months. This is assessed by self-report using a standardized questionnaire of the number of pills missed in the last 30 days and calculating percent adherence based on the expected number of doses taken and/or number of missed doses. Percent adherence will be analyzed as continuous variable and polytomous variable. Subjects will be considered as "Good adherence" when they achieve an optimal adherence cut of >95% of medications taken as directed with <3 missed doses in the last 30 days period. "Fair adherence" will be considered as 80-95% of medications taken as directed with 3-8 missed doses in the last 30 days period, while "Poor adherence" will be considered as <80% of medications taken as directed with >8 missed doses in the last 30 days period. Medication adherence will be assessed at each follow-up visit, particularly at months 3 and 6. The six month scheduled follow-up is the primary analysis but the three month assessment will allow for evaluating trends but is not the primary outcome. The second primary outcome is (regularity of) scheduled clinic attendance at six months. This will be assessed from records of scheduled clinic visits in the medical records unit and supported by record of scheduled drug pick-up in the pharmacy. Regularity of scheduled clinic visits will be calculated based on number of scheduled clinic visits attended and/or number of times defaulted. Number of missed appointments will be analyzed as a continuous variable and polytomous variable. Subjects will be considered as "Regular clinic attendee" if they never missed any scheduled clinic appointment. A "Defaulter" will be a patient who has missed one or more scheduled clinic appointment for any reason(s), while a patient will be considered "Lost-to-follow-up" if s/he refuses to show up for scheduled clinic visit for 3 consecutive months, after 3 consecutive attempts to track the client and bring them back on treatment. The six month scheduled follow-up is the primary analysis but the three month assessment will allow for evaluating trends. Secondary Outcomes: The first secondary outcome is (improved) clinical status of patients at six months. This will be assessed by measuring their weights, assessing their TB and opportunistic infection status at every clinic visits, particularly at three and six months and comparing these with the baseline status. Patients" weight is routinely measured and recorded by nurses at the study site, as part of standard patient management protocol. These records will be accessed and utilized in the analysis. Patients" TB status is also routinely assessed and recorded by clinicians/nurses/adherence counselors at every clinic visits. These records will be accessed and utilized to categorize patients into four groups, namely "No signs and symptoms of TB", "TB suspected and referred for evaluation", "Currently on INH prophylaxis" and "Currently on TB treatment". Opportunistic infection status will be assessed through WHO clinical staging at every clinic visits. The six month scheduled follow-up is the primary analysis but trend will be evaluated at three months.

EFFECTIVENESS OF MOBILE
The other secondary outcome is (improved) immunologic status/response of patients at six months. This will be assessed based on records of two (2) laboratory measurements -CD4 T cell count and viral load. Patients" CD4 count results at six months will be assessed and compared with baseline values to determine if there"s a consistent increase (or otherwise) as expected with strict medication adherence. If patients are adherent to ART regimens, absorb the drugs normally, and if their virus is not resistant to the ART drugs in the regimen, it is expected that their CD4 count should increase from baseline levels by at least 50 cells per microliter of blood within 6-12 months of initiating treatment. CD4 count value at six month scheduled follow-up is the primary analysis. Similarly, plasma HIV RNA load results at six months will be assessed and compared with baseline levels to determine if there is a significant suppression at six months. If patients are adherent to ART regimens, absorb the drugs normally, and if their virus is not resistant to the ART drugs in the regimen, it is expected that their plasma HIV load should be suppressed to undetectable levels (<400 copies/ml) by six months and remain suppressed thereafter. Analysis of "suppressed" (<400 copies/ml) versus "failure to suppress" (>400 copies/ml) at six months as a dichotomous variable is the primary analysis.
Data analysis: Data collected will be collated, checked, cleaned, entered into and analyzed using Statistical Package for Social Sciences software (SPSS) version 21 and AMOS software. Parametric tests (T-test, repeated measures ANOVA) and non-parametric tests will be conducted on the data. P value for test of significance of results will be set at 0.05, alpha level (Type 1 error) at 0.05 (Confidence Interval of 95%), Power at 80% to detect an impact of 15% in the fraction of patients with adherence of at least 90% (assuming a no intervention median adherence level of 90%) using a two-tailed test, Z=1.96, and strength by effect size analysis. Analysis will be by intention-to-treat and complete case analysis. Limitations: (1) Self-report of adherenceexaggeration of true adherence (Pharmacy refill records, CD4 count and viral load will provide a means of validating the responses) (2) Pill count will not be used (3) Structured questionnaires may not elicit in-depth information Ethical considerations: The study and materials will be presented to the UPM Ethics approval committee and the Malaysian Ministry of Health"s Institutional Review and Ethics Committee for review and approval. Thereafter, permission will be sought from the Medical Director of the hospital in writing to use their clients. The researcher will work in collaboration with attending clinicians, nurses and pharmacists in identification and monitoring of the clients. Expected outcomes: The expected outcomes of this research include: 1. A significant difference in treatment adherence (medication adherence and regular, scheduled clinic attendance) is expected in the intervention group compared to the control group. 2. Better clinical outcome in the intervention group compared to the control group. 3. Socio-demographic, psychosocial and other risk factors that determine differences in treatment adherence between the intervention and control groups will be identified. 4. Provide baseline information on the effectiveness of current standard of care (routine adherence counselling) on treatment adherence and clinical outcomes.