System Dynamics (SD) Model of the HIV Care Continuum

Margaret Weeks; David W. Lounsbury; Jianghong Li; H. Danielle Green; Marcie Berman; Lucy Rohena; Rosely Gonzalez; Heather I. Mosher; Gary Hirsch

Sep 21, 2020

Version 3

System Dynamics (SD) Model of the HIV Care Continuum V.3

DOI

dx.doi.org/10.17504/protocols.io.bmj2k4qe

Margaret R Weeks¹,
David W. Lounsbury²,
Jianghong Li¹,
H. Danielle Green¹,
Marcie Berman¹,
Lucy Rohena¹,
Rosely Gonzalez¹,
Heather I. Mosher¹,
Gary Hirsch³

¹Institute for Community Research, Hartford, CT;
²Albert Einstein College of Medicine, Montefiore, NY, NY;
³Creator of Learning Environments, Wayland, MA

Margaret R Weeks

Institute for Community Research

DOI: dx.doi.org/10.17504/protocols.io.bmj2k4qe

External link: https://doi.org/10.1371/journal.pone.0230568

Protocol Citation: Margaret R Weeks, David W. Lounsbury, Jianghong Li, H. Danielle Green, Marcie Berman, Lucy Rohena, Rosely Gonzalez, Heather I. Mosher, Gary Hirsch 2020. System Dynamics (SD) Model of the HIV Care Continuum. protocols.io https://dx.doi.org/10.17504/protocols.io.bmj2k4qeVersion created by Margaret R Weeks

Manuscript citation:

Weeks, M.R., Lounsbury, D.W., Li, J., Hirsch, G., Berman, M., Green, H.D., Rohena, L., Gonzalez, R., Montezuma-Rusca, J.M., Jackson, S. (2020). Simulating system dynamics of the HIV care continuum to achieve treatment as prevention. PLOS ONE. 2020 Mar 19;15(3):e0230568. doi: 10.1371/journal.pone.0230568. eCollection 2020. PMID: 32191771 [PubMed - in process] PMCID: PMC7082036.

License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Protocol status: Working

We use this protocol and it's working

Created: September 21, 2020

Last Modified: September 21, 2020

Protocol Integer ID: 42330

Keywords: system dynamics model, HIV care continuum, group model building,

Disclaimer

Development of this simulation model was supported by the National Institute of Mental Health of the U.S. National Institutes of Health under award numbers R01MH103176 and R21MH110335. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Abstract

This protocol describes the development, design, and calibration of a system dynamics (SD) model of the HIV Care Continuum (CC) in one geographic region. Community stakeholders can use this SD model to better understand structural and behavioral characteristics of the HIV CC system and to inform strategic planning to determine the most promising strategies to prevent new HIV infections while also improving care for those infected and affected. This SD model can be used to learn about the dynamics of an HIV CC system by using the “base case scenario” to simulate system results. Model users can also test different “hypotheses” or “what-if scenarios” to see which combination of resources and actions generate the best improvements to the system. The model also can potentially be recalibrated to simulate the HIV CC in other geographic regions. 

To build this SD model, our research team collaborated with a coalition of 25 community stakeholders called the System Dynamics Modeling Task Force. This group included doctors, nurses, community clinic support staff, HIV/AIDS service organizations director, front line case managers and outreach workers, people living with HIV (PLWH), community advocates and activists, and representatives of the city and state departments of health. Using SD “group model building,” we conducted 16 iterative systems modeling workshops over an 18-month period (Jan. 2017 – July 2018) to diagram, calibrate, and simulate the regional HIV CC system.

The computational model was designed using system dynamics graphical software Stella Architect1.9.4®. It was parameterized using primary and secondary data, related literature, and stakeholder best estimates. These include CT Department of Public Health HIV surveillance data for 2015-2017 and regional Ryan White health services data for the same period. The resulting “base case scenario” reproduces the epidemic in the initially selected catchment area, namely, the Hartford TGA (Transitional Grant Area), a HRSA-designated Ryan White (RW) funded area that includes Hartford, Middlesex, and Tolland Counties in Connecticut, USA. Model parameters (initial values) and equations were calibrated to simulate the base case scenario over a time horizon of 60 months (2018-2023). To demonstrate effective SD model performance, we compared base case simulated output for the prior calendar year (2017) with reported trends over the same period in the catchment area. The model was rigorously validated with tests for structural and behavioral continuity and sensitivity analysis of some key parameters that have no or limited certainty.

Nine “modules” make up the full HIV CC SD model. These include: 
A. HIV Infection and Treatment as Prevention; 
B. HIV Testing and Prevention Services; 
C. Medical Care Services for PLWH; 
D. Ryan White Case Management Services; 
E. Housing, Substance Use Treatment, and Mental Health Services for PLWH; 
F. Peer Outreach to Promote HIV Testing; 
G. Peer Advocacy to Support PLWH; 
H. Expanded HIV Testing and Comprehensive Sexual Health Screening in Primary Care; 
I. Mobilizing Community Initiatives to Support PLWH. 

The “Action Strategies” (Modules F-I) are designed to improve delivery and effectiveness of the “Basic Services” (Modules B-E). Similarly, the “Basic Services” are designed to improve the effectiveness of the central HIV Infection and Treatment as Prevention (Module A) to reduce community-level viral load. 

This protocol and the resulting SD model were part of two studies funded by the National Institute of Mental Health, U.S. National Institutes of Health, award numbers R01MH103176 and R21MH110335. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the protocol. The authors have declared that no competing interests exist. All protocols for the protection of human subjects in research were followed during this study. The studies were reviewed and approved by the Institutional Review Board of the Institute for Community Research (IRB #2014-01 and IRB #2016-02).