An official website of the United States government
Abstract AimsGLP‐1 and GIP‐GLP‐1 agonists have emerged as potent weight‐loss medications. These incretin mimetics often have low patient adherence, and as with any medication, clinically meaningful efficacy requires adequate adherence. But what constitutes “adequate” adherence for incretin mimetics? The purpose of this paper is to address this question. Materials and MethodsWe use mathematical modelling and stochastic simulation to investigate the weight loss efficacy of incretin mimetics under imperfect adherence. We use validated pharmacokinetic and pharmacodynamic models of semaglutide and tirzepatide and assume that simulated patients randomly miss doses. ResultsWe find that semaglutide and tirzepatide forgive nonadherence, meaning that strong weight loss efficacy persists despite missed doses. For example, taking 80% of the prescribed doses yields around 90% of the weight loss achieved under perfect adherence. Taking only 50% of the prescribed doses yields nearly 70% of the weight loss of perfect adherence. Furthermore, such nonadherence causes only small fluctuations in body weight, assuming that patients do not typically miss more than several consecutive doses. ConclusionIncretin mimetics are powerful tools for combating obesity, perhaps even if patients can consistently take only half of their prescribed doses. The common assumption that significant weight loss requires at least 80% adherence needs revision.
more »
« less
An Activity Recognition System for Taking Medicine Using In-The-Wild Data to Promote Medication Adherence
Nearly half of people prescribed medication to treat chronic or short-term conditions do not take their medicine as prescribed. This leads to worse treatment outcomes, higher hospital admission rates, increased healthcare costs, and increased morbidity and mortality rates. While some instances of medication non-adherence are a result of problems with the treatment plan or barriers caused by the health care provider, many are instances caused by patient-related factors such as forgetting, running out of medication, and not understanding the required dosages. This presents a clear need for patient-centered systems that can reliably increase medication adherence. To that end, in this work we describe an activity recognition system capable of recognizing when individuals take medication in an unconstrained, real-world environment. Our methodology uses a modified version of the Bagging ensemble method to suit unbalanced data and a classifier trained on the prediction probabilities of the Bagging classifier to identify when individuals took medication during a full-day study. Using this methodology we are able to recognize when individuals took medication with an F-measure of 0.77. Our system is a first step towards developing personal health interfaces that are capable of providing personalized medication adherence interventions.
more »
« less
- Award ID(s):
- 1952236
- PAR ID:
- 10294629
- Date Published:
- Journal Name:
- IUI '21: 26th International Conference on Intelligent User Interfaces
- Page Range / eLocation ID:
- 575 to 584
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The prevalence of medication non-adherence poses a critical challenge with far-reaching health implications and substantial economic consequences. The PERACTIV project addresses this issue with a wrist-worn device that meticulously tracks hand and finger movements, offering real-time insights into medication intake routines. In the context of smart cities and healthcare innovation, this project provides a groundbreaking solution to the persistent problem of medication adherence. This paper introduces the integration of PERACTIV wearable technology with an automated annotation pipeline utilizing Force Sensing Resistor (FSR) technology. Placed strategically on the index finger, FSR records force during pivotal pill-related moments, eliminating the need for time-consuming manual data annotation. Together, the PERACTIV wearable and FSR-powered pipeline form a promising solution for medication adherence, particularly beneficial for individuals living independently and those with memory impairments. The research methodology involved focus groups to explore the intricacies of medication adherence challenges. Section 2 discusses the findings derived from these groups, providing a foundational understanding. Section 3 introduces the FSR-powered automated annotation pipeline, comprehensively analyzing micro-activities during pill-taking. These micro-activities encompass diverse elements, contributing to a holistic understanding crucial for effective analysis. This paper underscores the transformative potential of technology and innovation in healthcare research, setting the stage for seamless integration of medication adherence into daily life. The future promises a more connected and informed healthcare landscape driven by the synergy of PERACTIV wearable and FSR automation.more » « less
-
Abstract BackgroundCancer-care complexity heightens communication challenges between health care providers and patients, impacting their treatment adherence. This is especially evident upon hospital discharge in patients undergoing surgical procedures. Digital health tools offer potential solutions to address communication challenges seen in current discharge protocols. We aim to explore the usability and acceptability of an interactive health platform among discharged patients who underwent oncology-related procedures. MethodsA 4-week exploratory cohort study was conducted. Following hospital discharge, a tablet equipped with an integrated Personal Virtual Assistant (PVA) system was provided to patients who underwent oncology-related procedures. The PVA encompasses automated features that provide personalized care plans, developed through collaboration among clinicians, researchers, and engineers from various disciplines. These plans include guidance on daily specific assignments that were divided into 4 categories: medication intake, exercise,symptom surveys, and postprocedural specific tasks. The aim was to explore the acceptability of the PVA by quantification of dropout rate and assessing adherence to each care plan category throughout the study duration. The secondary aim assessed acceptability of the PVA through a technology acceptance model (TAM) questionnaire that examined ease of use, usefulness, attitude toward use, and privacy concerns. ResultsIn total, 17 patients were enrolled. However, 1 (5.8%) patient dropped out from the study after 3 days due to health deterioration, leaving 16/17 (94.2%) completing the study (mean age 54.5, SD 12.7, years; n=9, 52% Caucasian; n=14, 82% with a gynecological disease; n=3, 18% with a hepatobiliary disease). At the study end point, adherence to care plan categories were 78% (SD 25%) for medications, 81% (SD 24%) for exercises, 61% (SD 30%) for surveys, and 58% (SD 44%) for specific tasks such as following step-by step wound care instructions, managing drains, administering injectable medications independently, and performing pelvic baths as instructed. There was an 80% patient endorsement (strongly agree or agree) across all TAM categories. ConclusionThis study suggests the potential acceptability of the PVA among patients discharged after oncology-related procedures, with a dropout rate of less than 6% and fair-to-good adherence to tasks such as medication intake and exercise. However, these findings are preliminary due to the small sample size and highlight the need for further research with larger cohorts to validate and refine the system.more » « less
-
Abstract Lithium is a drug widely employed for the treatment of bipolar disorder owing to its high efficacy in mood management and suicide prevention. However, this efficacy is often undermined by misdosing and nonadherence, and diligent drug monitoring is required during treatment. Standard lithium monitoring involves invasive blood collections and laboratory analysis with low time granularity. Recent advances in sensor technology have enabled the development of personalized drug‐monitoring devices that analyze biomarker information noninvasively. Herein, based on the fact that the analyte partition onto the fingertip with a high flux, a touch‐based noninvasive monitoring modality for managing lithium pharmacotherapy is devised. The system is built based on a thin organohydrogel‐mounted lithium ion‐selective electrode (TOH‐ISE). The TOH coating provides a stable environment for sensing. Through the utilization of a water/glycerol bi‐solvent matrix, the gel exhibits dehydration‐resist properties, rendering a controlled micro‐environment for ISE conditioning, and subsequently minimizing signal drift. To illustrate the clinical application of the solution, the system is tested on a subject prescribed lithium. The system successfully detected the increase in circulating drug levels following medication intake. Collectively, the results indicate the devised solution is capable to facilitate lithium adherence monitoring and has broader potential for optimizing lithium pharmacotherapy.more » « less
-
Medication adherence is a major problem in the healthcare industry: it has a major impact on an individual’s health and is a major expense on the healthcare system. We note that much of human activity involves using our hands, often in conjunction with objects. Camera-based wearables for tracking human activities have sparked a lot of attention in the past few years. These technologies have the potential to track human behavior anytime, any place. This paper proposes a paradigm for medication adherence employing innovative wrist-worn camera technology. We discuss how the device was built, various experiments to demonstrate feasibility and how the device could be deployed to detect the micro-activities involved in pill taking so as to ensure medication adherence.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3397481.3450673
