WelTelOAKTREE: Text Messaging to Support Patients With HIV/AIDS in British Columbia

Purpose:

To provide a weekly text messaging intervention to 85 high risk HIV+ participants attending the Oak Tree HIV Clinic in order to improve medication adherence, attendance at appointments and subsequently, CD4 counts and HIV viral load values over a 1 year period.

Justification:

In Canada, around 65,000 people are living with HIV/AIDS, approximately 14,300 of whom are women. AntiRetroviral Therapy (ART) has led to enormous improvements in the health and survival of individuals with HIV. Moreover, by decreasing the amount of virus circulating in the body (viral load), HAART offers the possibility of treatment as a preventative measure. However, high levels of engagement in care, timely initiation of ARVs, and adherence to medication are required to maximize the benefits of HAART in order to prevent resistance, progression to AIDS, transmission or mortality. Unfortunately, engagement in ongoing HIV care can be poor, with one study from the United States (US) showing only 52% retention in care over 1 year. Further, adherence among high-risk populations is low, with women being less adherent partly due to their role as care providers for children and partners, potential abuse in partner relationships, fear of stigma, homelessness, and concerns regarding side effects. Conversely, active drug use (especially cocaine), lack of social supports, and depression are just a few of the variables that affect both men and women alike. Current methods of engagement in care have failed to overcome these barriers to adherence, which makes finding an effective adherence intervention critically important. Mobile health (mHealth), the use of mobile phone technology to deliver health care, is an emerging area of disease management that can assist in patient adherence to prolonged chronic treatment regimens and monitoring of care. A randomized controlled trial (WelTelKenya1), conducted by Dr. Richard Lester et. al, tested the clinical effectiveness of text message support for HIV treatment adherence in Kenya. WelTelKenya1, of which 67% were women, showed that patients receiving text message support had significantly higher rates of treatment adherence and viral suppression than patients who received standard care alone. In Canada, cell phone penetration exceeds 70% and is expected to reach 100% within the next decade. The WelTel system offers a clinical management model that can be carried out using standard services offered by cellular network providers with minimal additional infrastructure and is both flexible and scalable.

The investigators have completed one of the first studies of text messaging support for HIV care in Canada. The pilot study called WelTelBC1 involved 25 individuals from five patient groups 1) Non-suppressed (CD4 <200, VL >250); 2) Youth (ages 14-24); 3) Mature (Age ≥50); 4) English as a second language; and 5) Distance (those residing 3+ hours travel time from the clinic), who receive a weekly text message asking them "How are you?" Participants were then instructed to respond with "OK" or to let the investigators know if they had a problem. Participants who responded that they are "not okay" or did not respond were then followed up by a clinic nurse. The pilot study was designed to look at feasibility and acceptability of the weekly text messaging intervention in a Canadian HIV+ population, and resulted showed that the intervention was been perceived as beneficial among participants. In regards to acceptability, the pilot study has been extremely informative, and has enabled investigators to engage participants previously only seen sporadically; overcoming gaps that prevent optimal care and follow-up. In addition, we have seen from our pilot project, that to reach those in most need of a link to care we need to be prepared to provide cell phones and phone plan support to those without one (only 50% of those enrolled in our intervention owned a cell phone, and only 40% had unlimited text messaging - anecdotally these are the patients with whom engagement has most improved during the intervention). It is now critical to expand this program to all individuals at Oak Tree Clinic who could benefit and to study the efficacy of this intervention in engaging patients in care and improving adherence to HAART.

Research Method:

Participant Selection and Recruitment: A list of patients with a CD4 count ≤500 or previous prescription for antiretroviral therapy (other than for pregnancy) prior to the control year, as well as a detectable viral load (≥200) in the control year was assembled. The list was reviewed by the clinic physicians, nurse, pharmacist, dietician, counselor, outreach worker and social worker to determine which patients would benefit most from participating in the WelTel text messaging program (i.e. poor engagement in care, difficult to contact, poor or non-adherence to ARV therapy, advanced HIV infection/AIDS, vulnerable or socially isolated patients). A consensus based approach was used for patient selection. In addition, all 25 of the Oak Tree pilot study participants (WelTelBC1), were invited into the current study. Once nominated, when patients attended clinic for a clinical visit, they were introduced to the WelTel intervention concept. Those interested were approached by research staff for a full explanation. The intervention protocol was the same as that used for the WelTelBC1 pilot study at Oak Tree, developed through use of patient questionnaires as well as patient and health care worker focus groups/interviews at Oak Tree, and yet very similar to the intervention used in the initial WelTelKenya1 intervention. Following fully informed consenting with completion of consent forms, participants were provided with a cell phone with unlimited text messaging if they did not have one, or if they had their own cell phone, had their plan topped up to include unlimited text messaging service. Baseline clinical data including historical CD4 counts, and HIV viral loads were abstracted from patient charts.

Additionally, study participants were asked to complete a Quality of Life Assessment (QOL) questionnaire (the SF-12 questionnaire) at study entry (0 months), mid-way through study (6 months +/- 6 weeks) and at study exit (12 months +/- 6 weeks). The questionnaire consisted of 12 questions; was self-administered or interviewer administered; and took between 10 -15 minutes to complete.

Intervention:

The intervention protocol was the same as that used for the WelTelBC1 pilot study. This was modeled on the WelTelKenya1 intervention but adapted to the Oak Tree Clinic patient population through the use of patient questionnaires as well as patient and health care worker focus group and individual interviews. Each Monday, patients received a text message from a number not traceable to the clinic stating simply "How are you?" Patients were instructed to respond to the message if they are "OK" or to state that they have a problem. Messages were reviewed and triaged daily by our program research team member and in all cases of a negative or complex response other than OK, participants were contacted by the program nurse (patients were instructed that this is NOT an emergency service). Non-responders received a second text message on Wednesday at 12:00 pm (48 hours after the initial text sent out), and if there was no response, were contacted by the program nurse Wednesday afternoon or Thursday morning for follow-up.

Data Collection:

Participants wiwere asked their ethnicity at study enrolment. Participants were also asked to complete a 12 question QOL assessment questionnaire at study entry, mid-way through the study and at study end. Frequency of attendance in care was assessed from the outpatient clinic electronic booking system. Chart abstraction of clinical health status included: participant age (in years), housing status, current illicit drug use and postal code, CD4 counts and percentages, HIV viral loads, antiretroviral drug (ARV) regimen (including date of initiation or discontinuation), and degree of medication adherence (as determined from timing of ARV refills, and self-report), which was collected at baseline then at each clinical visit for the following one year. All available like data for one year prior (up to 2 years if in the Pilot Study) to enrollment in the WelTel program was alsocollected such that participants served as their own controls in the intervention (repeated measures study), and for the year following the intervention to assess the longevity of the intervention's impacts. Staff costs / savings were calculated by looking at Pharmacy, Nursing and Outreach worker time used both prior to and throughout the intervention. The planned duration of the intervention was one year, and of the study, 18 months. At that time data was analyzed and the program evaluated.