Technology-Enhanced Peer Navigation to Improve IDUs' Engagement in HIV Care

Research Question

Can real-time collection of patient-level data using smartphones be feasibly and effectively
utilized by peer navigators to deliver timely and individualized support interventions
outside of a clinical setting, thereby facilitating improved engagement in HIV care and
medication adherence for HIV-infected drug users?

Aims

Optimal benefit from antiretroviral therapy (ART) is contingent upon consistent engagement
in HIV care and reliable medication adherence. HIV-infected intravenous drug users (IDUs)
delay entrance into and under-utilize medical care. Those who access HIV care are frequently
lost to follow up or are intermittent users. Out-of-care IDUs are more likely to develop
AIDS, have reduced survival, engage in risky behavior and place others at risk for HIV
transmission; therefore, evidence-based tools to promote engagement of IDUs in care are
urgently needed.

Clinic-based, individualized support delivered by case managers or patient navigators
improves engagement in HIV care for many patients, but few settings have sufficient
resources to offer adaptive, individualized, long-term support to all who need it. Mobile
health (mHealth) technologies able to collect patient-level data in real time and to
facilitate responsive communication hold promise as an effective strategy to optimize HIV
care and medication adherence for IDUs.

The objective of this pilot feasibility trial planning grant is to obtain feasibility,
acceptability and trial optimization data to inform the design and implementation of a
randomized trial to fully evaluate the hypothesis. To accomplish this objective, specific
aims include:

Specific Aim 1. To evaluate the feasibility and participant acceptability of
mHealth-enhanced peer navigation to improve retention in care for out-of-care, ART-eligible
IDUs.

In the mobile phone-based intervention, participants will utilize a smartphone interface to
report perceived barriers to accessing care and adhering to prescribed therapies. Through
repeated quantitative and qualitative evaluations, investigators will assess feasibility,
participant acceptability and logistical challenges (e.g., retention, participant burden,
device loss, data quality, response fatigue, perceived support) of the intervention
sustained over one year of follow-up.

Specific Aim 2. To define trial elements required for optimization of a future randomized
clinical trial of mHealth-enhanced peer navigation.

The long-term effectiveness of mHealth interventions (i.e., maintenance of sustained HIV RNA
suppression over years), will require large and expensive clinical trials. In prudent
preparation for a future trial, investigators will collect preliminary efficacy data
comparing the peer-navigation/smartphone intervention to usual care for several alternative
endpoints, including time to ART initiation, missed visit frequency, and rate of virologic
suppression. Investigators will also define the optimal patient load and scope of work for
peer navigators and the intensity and frequency of electronic communication that facilitate
meaningful peer support.

Background

Patient navigators have been successfully employed to assist persons utilizing complex
medical services. In contrast to the medical case management paradigm, which tends to
involve delivery of specific, predefined services, patient navigation features responsive,
flexible problem-solving to overcome perceived barriers to care. The World Health
Organization (WHO) recommends psychosocial support for all HIV-infected persons and has
promoted the use of peers in addressing care and retention needs. In resource-constrained
settings, peers trained for specific health-related tasks represent a potentially
cost-efficient and sustainable workforce. Besides addiction and mental health co-morbidity,
HIV-infected IDUs often have limited income, education, insurance, and social support,
unstable housing and food insecurity. Poor health literacy, negative experiences with
medical institutions and negative health-related beliefs may pose additional barriers to
care.

The role of peer navigators in HIV care is to influence areas which impact engagement in
care, namely to: 1) Educate and counsel regarding acceptance of HIV diagnosis and
understanding of HIV prevention and treatment, 2) Assist in overcoming practical barriers to
care (e.g., accessing social services like drug treatment, housing or transportation
assistance, 3) Identify and reduce sources of instability, 4) Facilitate patient coping
through life skills training, and 5) Improve patient-provider communication. Data suggest
that peer-led interventions should be personalized and practical, addressing individual
concerns and barriers rather than group-level support.8 The frequency and intensity of
patient-peer interaction needed for this individualization and the extent to which mobile
technology can improve communication toward this end need clarification.

mHealth approaches to optimizing HIV care

Use of mobile cellular devices has grown rapidly worldwide, increasing from 2 billion
subscriptions in 2005 to an estimated 5.3 billion at the end of 2010.9 The promise of mobile
health (mHealth) technologies for strengthening health care delivery in resource-limited
settings has been acknowledged by the United Nations Joint Programme on HIV/AIDS (UNAIDS) in
development of its strategic plan. Short Message Service (SMS) text messages were recently
shown in a large randomized trial to be an effective intervention for smoking cessation.
Text messages have also been effectively utilized to improve attendance at clinic
appointments, although few data are available from HIV clinics. ART dose reminder messages
sent with pagers or mobile phones have been used in several settings to improve adherence.
Two recent randomized mHealth studies in sub-Saharan Africa have demonstrated benefit of
weekly text messages for improving adherence and rates of viral suppression. Additional
randomized trials of SMS adherence reminders are under development in India and Cameroon.

Rationale

The rationale for Peer Navigators (PNs) working with HIV-infected IDUs is that similar
backgrounds and shared experiences will strengthen relationships and provide PNs access and
credibility not easily achieved by medical professionals. HIV infection among IDUs is
transmitted through drug-using networks. However, these same peer networks have also served
as the vehicle for several peer-driven interventions with demonstrated efficacy in reducing
HIV risk behavior, often outperforming traditional approaches. With lower-income, poorly
educated, predominantly minority HIV-infected IDUs, investigators propose that having a
knowledgeable 'peer' provide counsel and support during entrance to care and ART initiation,
and remain a stable presence into the maintenance phase could directly addresses many of the
barriers encountered by this community.

Population Description

Study participants will be primarily recruited from the existing ALIVE cohort. Additional
participants will be recruited among patients who have received HIV care in the past at the
Johns Hopkins Moore Clinic.

Sample Size

The total sample size for this study will be 60, with 30 participants in each study arm.
Because the main objective of this study is to demonstrate feasibility and acceptance of a
mHealth intervention, analysis of the data generated will largely be exploratory and
descriptive, rather than driven by a priori hypotheses.

In November 2012, the ALIVE Study identified 54 participants who meet the criteria for both
being out of HIV care and recent drug use. The Moore Clinic clinical staff estimates that
around 180 patients could meet these criteria, providing a combined pool of roughly 230
individuals from which the study can begin recruitment.

Recruitment

Potential participants will be approached at their regular visit by their interviewer and
the Research Coordinator about participating in a new study. Invitation will be determined
by clinic interviewers or staff, who will notify the Research Coordinator if a participant
is eligible. Potential participants will be contacted at the ALIVE Study or the Moore
Clinic, based on recent data or records. If an ALIVE participant self-reports to a study
interviewer that they have never been in care or have been out of care for over a year, they
will be referred to the study coordinator for more information and screening. Eligible
individuals who confirm they have not received any HIV care in the past year will be
assisted in scheduling an appointment with a primary care provider at the Moore Clinic. Only
participants who attend their first HIV care visit (confirmed through the electronic medical
record) will undergo enrollment and randomization.

As part of Moore clinic efforts to identify out-of-care persons, they periodically review
their records and contact individuals directly to clarify if patients are seeing another
provider or to support them to re-engage in care at the Moore clinic. The clinic (namely
Jeanne Keruly and her staff) will share this information on out-of-care Moore clinic
patients with the mP2P study and will assist us in identifying when out-of-care persons are
scheduled for an appointment. At their subsequent visit, Moore Clinic patients will meet
with the coordinator (preferably that day) for more information and screening. The study
will use a recruitment script and screening tool.

Potential participants may also be referred to the study coordinator for eligibility
screening from Moore Clinic providers who encounter patients who have been lost to follow-up
(i.e. have not seen their primary HIV provider for over one year), but present to the clinic
for a new visit, because of an administrative issue or urgent care appointment. With
provider assent, the recruiter will approach these patients, explain the study and if
interested, and schedule them to attend an enrollment visit where consent will be obtained.

Informed Consent Process

The Research Coordinator will read the consent form aloud to each enrolling participant
before providing training and instruction in the use of all devices. The study participant
will sign then their name and date the form, as will the Coordinator. A copy of the consent
form will be provided to the client.

Individuals who are cognitively impaired to the extent that they either cannot give informed
consent or cannot give self-report will be excluded. Self-report is a central outcome
measure; including participants who cannot do it would invalidate the study.

Part of the evaluation of the intervention will include elicitation of feedback from the
peer navigator staff regarding their impressions about the effectiveness of the
intervention, unforeseen challenges and best practices. As part of monthly staff meetings
with the study coordinator and PN clinical supervisor, a 30-60 minute focus group will be
conducted with peer navigators. Because their conversations will be audio recorded and
analyzed, their participation in the focus groups constitutes research. Written informed
consent will be obtained from PNs by the study coordinator at each of these sessions.

Peer Health Navigation

After the first HIV care visit, participants will return to the Wood Clinic for a baseline
interview and randomization into one of the two study arms (PN or treatment as usual).
Investigators modeled the peer navigation intervention after several HIV prevention trials
that aimed to assist high-risk individuals in overcoming logistical barriers to receiving
social services and medical care. Peer Navigators will be part-time employees of the study,
hired through Johns Hopkins School of Public Health (JHSPH) Human Resources procedures, and
will work approximately 10-15 hours per week, meeting study participants in person or by
phone to discuss strategies to remain engaged in care and overcome logistical challenges
encountered in the health care system. Persons living with HIV who have successfully
navigated the health care system and who are receiving ongoing HIV care will be encouraged
to apply, but HIV-positive status will not be a requirement of the position.

The PN will log all PN-participant contacts daily using the eMOCHA system. This will allow
an assessment of the level of intensity of PN-participant contact that may be associated
with clinical benefit, as well as inform the appropriate caseload for PNs. The frequency,
duration, and nature of each PN/patient interaction as well as an assessment of action plan
goals and identified barriers will be compared to outcomes from the study aims, including
time to ART initiation, missed visit frequency, and rate of virologic suppression.

The study coordinator will serve as the operational supervisor for peer navigators, and will
be responsible for facilitating communications among PNs and research staff. The study
coordinator will provide technical assistance with the web-based patient tracking software
and assist with review of participant-specific data. The coordinator will be responsible for
matching PNs with eligible participants, ensuring that all data are collected according to
the schedule and specifications in the research protocol, organizing the training and hiring
of PNs, and for setting up regular operational supervision meetings.

A co-investigator (Dr. Hutton) trained in clinical psychology with extensive experience with
HIV-infected persons with substance abuse will serve as the clinical supervisor for peer
navigators. She will be responsible for organizing regular clinical supervision sessions to
review case studies, address issues of PN/counseling overlap and discuss issues arising from
the execution of the work. These sessions should be a mix of group and individual level
supervision. The clinical supervisor should help PNs deal with issues of setting boundaries
with participants and developing methods for dealing with urgent issues. Clinical
supervisors will ensure that clinical assistance is available for PNs when situations arise
in which a licensed mental health worker's assistance is needed. Additional support will be
provided by clinicians and HIV care providers who have been trained to address the specific
medical needs of IDUs.

Feedback about perceived logistical challenges related to PN/participant interaction and the
overall conduct of the study will be collected at monthly meetings by the operational
supervisor. The entire team of investigators will compile and review these comments at
quarterly meetings throughout the study. At 3 months and 9 months after the start of the
study, the monthly PN meeting will consist of a formal focus group, led by a co-investigator
with qualitative research methods expertise.

Ecologic Momentary Assessment (EMA)

EMA refers to using electronic handheld data collection devices to collect real-time
information about stress, drug cravings and drug use participants in the context of their
daily routine, rather than in a research study. In this study, participants will be given a
smartphone to carry for up to one year. The current phone model used is the Motorola Droid
X2, running applications developed using a software platform created by the Center for
Clinical Global Health Education (CCGHE) at the Johns Hopkins School of Medicine. The
application, named the electronic Mobile Open-source Comprehensive Health Application
(eMOCHA), has been used to support community health workers in resource poor areas heavily
impacted by HIV. It has been modified by the CCGHE programming team specifically for this
study. A staff member will train participants in how to use the eMOCHA interface and all
relevant features of the smartphone. Participants will be required to show they can run a
"demo" questionnaire before commencing use in the field.

Smartphone-based data collection from participants in the intervention arm will commence two
business days following randomization, to allow for troubleshooting of device problems and
participants' increased familiarity with the eMOCHA interface. At scheduled data-collection
times, the participant's phone will generate an audible alert and the eMOCHA application
will automatically launch. To protect confidentiality, every eMOCHA session will begin with
a password-entry screen containing no other information. Frequency and length of EMA
sessions will be dynamic over the course of the study in order to individualize the
intervention and collect process data related to Aim 2. By default, participants will
receive two prompts daily for the first three months of the study. Based on participant
feedback, rate of EMA response and level of engagement in HIV care, the protocol may be
converted to a "maintenance phase" of less frequent prompts after the initial three-month
in-person study visit.

Each eMOCHA session will feature multiple-choice and free-text question items related to
three domains: engagement in HIV care (retention and adherence), drug/alcohol use, and mood.
The drug use items will query participants regarding the intensity of drug cravings and the
occurrence of any drug use. The brief version of the Profile of Mood States (POMS) will be
used to assess moods such as anger, depression and anxiety. Within each domain, the eMOCHA
application will guide the participant through three steps that (1) assess the participant's
current experiences or symptoms, (2) assess barriers to engagement in care, and (3) provide
support messages or information. Peer navigators will monitor the responses of each
participant over time, and will contact the participant if he or she reports specific
problems, or appears to have increasing levels of self-reported negative mood states, drug
use, or poor medication adherence.

Twice each day, the smartphone will signal the participant to answer a questionnaire.
Participants will have 15 minutes to respond after notification. The alarms can occur
between 9:00 am and 9:00 pm. One alarm will go off at a random time every day and the other
alarm will always go off at 9:00 pm to recap what happened during your day. The questions
should take about five minutes every day. Also, if participants use drugs or have a craving
to use drugs, they should report this event using a questionnaire that is similar to the
random questionnaire, but tailored to contain more drug-specific content.

Global Positioning System (GPS) Data

As a continuation of previous efforts during the EXACT Study, investigators will compare
global positioning systems (GPS) data and neighborhood-level psychosocial stress indicators
with drug-related EMA behavioral data. Participants' location will be recorded every five
minutes using the smartphone's internal GPS. GPS data will be encrypted and transferred
wirelessly to a secure server daily. The GPS data on the phone will be deleted after
transfer. Participants' daily EMA data will be compared to weekly ACASI data and tagged with
GPS data to construct a temporal and spatial profile of drug use and psychosocial stress.

Biological Samples

Every three months, a trained phlebotomist will draw 10mL of whole blood to place in a
repository, using approved biosafety procedures. Current research is actively investigating
several promising biological markers of stress. If, as anticipated, a valid and reliable
biomarker of longer-term stress (on the order of weeks to months) is identified, blood
samples from the repository will be tested at an appropriate laboratory. Genetic studies may
include both host and viral genetics.

Follow-up and assessment of HIV care utilization

Participants in both study arms will return to the Wood clinic every three months to collect
reimbursement for participation and to complete a follow-up questionnaire assessing the same
domains evaluated in the baseline questionnaire. The study will also collect 10mL (2
teaspoons) of blood serum for storage and future testing. This study visit should take about
30 minutes.

This study will uniquely benefit from an existing data sharing agreement with the Johns
Hopkins HIV Clinical Cohort (JHHCC) study, which will allow highly accurate and timely
ascertainment of care utilization endpoints ART prescription and pharmacy fill dates will be
collected from this database to verify participant-reported dates of ART initiation.
Appointment data are electronically managed through the Moore Clinic Registration System
(Epic Enterprise Scheduling Software, Madison, WI). All ambulatory visits scheduled are
input into this system at the point of service or through incoming calls by staff trained in
the use of this database. When a patient registers on the day of the appointment, the visit
is keyed as a completed visit. JHHCC data are available for download within approximately
two weeks of being entered into the system. Importantly, assessing clinic attendance and
prescription of ART using this external database will allow assessment of the main endpoints
for all randomized participants even if some are not retained in this study.

Study staff will conduct in-depth, qualitative interviews among the 30 participants in the
intervention arm. Ten will be randomly selected to complete an interview after 3-6 months of
the intervention, in order to provide feedback that may be used to refine the intervention
during the subsequent trial period; the remaining 20 will be interviewed at the end of
follow-up to collect data about their overall experience with the intervention. In order to
start processing the data, memos will be developed by the interviewer as data are collected.
Interviews will be semi-structured; interview guides will have predefined main themes, but
these guides are meant to be dynamic and will allow new topics to emerge during the course
of the interview. Important topics will include: (1) satisfaction and challenges with eMOCHA
interface, (2) satisfaction with relationship to peer navigator (3), perceived intrusiveness
of the intervention, (4) ongoing challenges to engagement in HIV care, (5) ways the
smartphone may facilitate getting HIV care information/help (e.g. from informal support
networks, the internet).

Focus groups conducted with the peer navigator staff will allow the investigator team to get
a better understanding about the effectiveness of the intervention, unforeseen challenges
and best practices. These will occur as part of monthly staff meetings with the study
coordinator and PN clinical supervisor, last 30-60 minutes, and will be audio-recorded for
later thematic analysis.

Blinding

Neither investigators nor participants will be blinded to which of the study arms
individuals are randomized.

Rationale for use of a non-treatment group

In order to justify spending resources for new, technology-based interventions, it is
essential to demonstrate that they add value compared to currently available care. The Johns
Hopkins Moore Clinic employs a limited number of full-time social workers, nurses, and peer
advocates who share responsibility for coordinating outpatient HIV care and facilitating
referrals for ancillary services. While this level of staffing is inadequate to provide
comprehensive, individualized case management for every patient, patients who desire
assistance with an aspect of their care have same-day access to this support, which
represents the currently available standard of care. Nearly all new patients seen in the
Moore Clinic meet with a social worker after completing the initial intake visit with their
medical provider, and may schedule follow-up meetings as needed or be seen on a walk-in
basis. Control-group participants will work with clinic-based support staff for coordination
of their care, meaning peer navigators will not actively facilitate interaction between
control group participants and clinic providers after the initial HIV care visit.

Participant removal criteria

Study participants may terminate their participation in the study at any time. If the
smartphone is lost, stolen or damaged during use, the study will assign a new one at no
charge. Participants never have to pay to replace a device. However, if a second smartphone
is lost, stolen or damaged, they will be asked to leave the mP2P Study. This will not affect
their participation in the ALIVE Study or their clinical care with the Moore clinic.

Certificate of Confidentiality

The mP2P study will collect data from a population of persons who formerly or currently
engage in illegal reportable activities, namely drug use. As the mP2P study is designed as a
sub-study of the ALIVE cohort, modifications will be made to the existing Certificate of
Confidentiality issued by the National Institute on Drug Abuse (DA-87-042) to protect
sensitive information reported by participants.

Consent Process and Documentation

Informed consent will take place at Johns Hopkins University, and will be carried out by the
Project Coordinator. Additional study staff may be trained in consent procedures, to provide
backup to the study. The Coordinator will read the consent form aloud while the potential
participant reads along. The participant will be able to ask any questions during the
consent process and before signing the form.

Risks and Benefits

Minimal risk: The probability and magnitude of harm or discomfort anticipated in the
research are not greater than those ordinarily encountered in daily life or during routine
physical and psychological examinations or tests.

Smartphone (EMA): Participants may feel hassled or annoyed by the EMA prompts. Also, some of
the questions are personal and they might not feel comfortable answering those questions.
Participants who are unfamiliar with smartphones or other electronic devices could feel some
discomfort or frustration in dealing with these technologies.

Smartphone (GPS): The main risk from carrying the GPS unit is loss of privacy. Using a GPS
device will not significantly increase participants' risk burden more than they currently
experience. Many participants already use mobile phones daily, which contain a system by
which a user's location can be tracked. Encrypted data will be wirelessly uploaded to a
secure server and erased from the device daily. The risks and procedures regarding the
smartphone will be explained during both screening and consent.

Phlebotomy: Phlebotomy is performed by drawing blood from a vein in the forearm or
antecubital fossa. This may cause pain during the insertion of the needle or slight bruising
afterward.

ACASI: Some questions in the ACASI questionnaire are personal and participants might feel
uncomfortable or upset about answering them. Participants may refuse to answer any question.

Confidentiality: Although precautions will be taken to protect privacy, there is the chance
that confidentiality can be broken. All data collected outside the study clinic are
protected by the Certificates of Confidentiality for the ALIVE I and ALIVE II studies.
Participants will be reminded of the legal requirements for confidentiality within the ALIVE
Study. Confidentiality risks will be minimized by using study materials without personal
identifiers, using only unique study numbers, and by following strict procedures in data
collection, data entry, and data analysis.

Peer Navigators will know participants by name and may have contact information for
participants stored on their own mobile phones, but will not carry any information linking
participants' identities with study-related information outside the locked offices of the
Wood Clinic.

Description of Level of Research Burden

Volunteers in this project will have their normal daily activities interrupted twice per
day. However, each interruption for an EMA response should take only 3-5 minutes. One
purpose of this study is to find out if this is too high of a burden for this population.
The maximum total daily time devoted to the study should be a no more than 15 to 20 minutes,
even with event-generated responses. Each weekly visit should take about 30 minutes, not
including time spent in the waiting room.

Description of Potential Benefits

Participants may not benefit directly for being in this study. Benefits to society include
helping us to understand research techniques that can promote engagement in HIV care as well
as improve data collection. The research may help society understand the environment in
which people use drugs and the specific challenges IDUs may encounter when accessing HIV
care.

Reporting Unanticipated Problems/Adverse Events

No adverse events are anticipated as part of the study. However, participants will be
instructed to contact the Principal Investigator, Dr. Kirk, the Research Coordinator, Mr.
Genz, or the Johns Hopkins Institutional Review Board (IRB) if they feel they have been
harmed during the course of study procedures.

Safety monitoring

Participants will have 24-hour access to an on-call provider through the Moore Clinic. If
needed, this communication will be facilitated by the study smartphone, which has important
clinic and study-related numbers pre-programmed. Participants' will be formally evaluated
with questionnaires and in-person interviews every three months at the ALIVE clinic by
research staff, at which time they can report any unforeseen challenges related to their
participation. They will have the ability to leave non-urgent messages for the study
coordinator or peer navigator, which will be answered within one business day.

Inclusion Criteria:

- History of current or former injection drug use

- No clinic visits with an HIV provider in the preceding 12 months

- Not taking ART

- HIV RNA level of greater than 1,000 copies/mL

- Agree to attend at least one HIV care visit at the Johns Hopkins Moore clinic

Exclusion Criteria:

- Any medical or psychiatric condition that would interfere with the participant's
ability to comply with study procedures or make participation unsafe

- Current enrollment in another HIV retention-in-care study