Date Published: January 15, 2019
Publisher: Public Library of Science
Author(s): Parya Saberi, Kristin Ming, Dominique Legnitto, Torsten B. Neilands, Monica Gandhi, Mallory O. Johnson, Ethan Moitra.
Due to marked reductions in morbidity and mortality, antiretroviral (ARV) adherence monitoring is of high interest. Researchers and clinicians often resort to the most feasible and cost-effective adherence methods possible, which may result in biased or inaccurate estimates and require the physical presence of a participant at a research or clinic site. The objective of our study was to evaluate the feasibility and acceptability of three objective, innovative, and remote methods to estimate ARV adherence which may be conducted with less time and financial resources in a wide range of clinic and research settings. These three methods included: (a) text-messaged photographs of pharmacy refill dates to measure refill-based adherence, (b) text-messaged photographs of ARV medications to estimate pill-count-based adherence, and (c) home-collected hair samples for the measurement of ARV concentration to determine pharmacologic-based adherence. We conducted a pilot study from March through October 2017 to examine the feasibility and acceptability of these three adherence measures and the remotely conducted study procedures in 93 adults living with HIV nationwide. From our diverse national sample of participants, 95.7% were retained until the end of the study, 89.9% sent all text messages, and 84.3% sent all hair samples. Approximately 74.2% of participants reported excellent overall experience with the study, 60.2% were very or extremely satisfied with participating in a hair collection study, and 76.3% noted extremely high likelihood of participating in a similar study including text messaging pictures of medications or refill dates. We noted high levels of feasibility and acceptability with the remote study methodology, collection of photographed and text messaged pharmacy refill dates and pill counts, and home-collected hair samples. Here we describe the feasibility and acceptability metrics, results from the exit qualitative interviews with the participants, and lessons learned. These adherence measures represent innovative approaches to expand monitoring tools for HIV treatment and prevention adherence in future research.
Adherence to antiretroviral (ARV) medications is the primary determinant of virologic suppression, leading to marked benefits on morbidity and mortality from HIV[1, 2], but long-term adherence is difficult to maintain. Adherence monitoring is thereby of increasing interest; however, there is no gold standard for estimating adherence, as all methods offer opportunity for bias or inaccuracy. Current methods include the use of patient self-report[4, 5], pill counts, pharmacy refill dates[6, 7], electronic drug monitoring (such as MEMS caps[4, 8] or Wisepill), and pharmacologic measures (such as concentrations in blood or hair). Each method has varying advantages and disadvantages and the choice of which to use often depends on feasibility and cost-effectiveness. For example, self-report is inexpensive, but subject to a host of reporting biases that can result in an overestimation of adherence. Electronic medication monitoring systems, such as MEMS or Wisepill, can provide granular, daily data reflecting detailed adherence patterns. However, the devices measure medication bottle/device opening only (not actual ingestion) and are expensive. Furthermore, some individuals may prefer not to carry them for their HIV medications, as the bottles/devices are often large and conspicuous, thus potentially inviting attention and activating HIV-related stigma regarding the purpose of the medication. Feasible, acceptable, cost-effective, and objective measures of ARV medication adherence are needed to minimize biased or inaccurate estimates.
We screened 166 people nationwide and enrolled 93. Seventy three individuals were ineligible due to not being interested (N = 34); not able or willing to provide hair samples due to baldness, very short hair, or not wanting to cut hair (N = 9); receiving automated ARV refills (N = 9); not taking an ARV regimen from the list of study ARVs (N = 7); having recently changed ARV medications or planning to switch soon (N = 7); HIV-negative serostatus (N = 3); not having access to the internet (N = 2); participating in a blinded clinical trial (N = 1); or being mono-lingual Spanish speaker (N = 1).
We examined the feasibility and acceptability of conducting an entirely remote study using three novel measures of ARV adherence among a diverse sample of participants nationwide. We noted high levels of feasibility and acceptability with the overall study methodology, remote data collection, collection of adherence data using photographed and text messaged pharmacy refill dates and pill counts, and home-collected hair samples. Despite some individuals requiring additional assistance with study activities and reporting that they missed the personal connection, the majority of participants noted that the study was simple, convenient, not time-consuming, and private. While many reported an improvement in their ARV adherence, several individuals noted no change due to an already high level of adherence. To our knowledge, this is the first study among PLWH that has been conducted in a completely remote manner and examined remote collection of refill dates, pill counts, and hair samples.