Use of systems thinking and adapted group model building methods to understand patterns of technology use among older adults with type 1 diabetes: a preliminary process evaluation.

BACKGROUND: A growing number of older adults (ages 65+) live with Type 1 diabetes. Simultaneously, technologies such as continuous glucose monitoring (CGM) have become standard of care. There is thus a need to understand better the complex dynamics that promote use of CGM (and other care innovations) over time in this age group. Our aim was to adapt methods from systems thinking, specifically a participatory approach to system dynamics modeling called group model building (GMB), to model the complex experiences that may underlie different trajectories of CGM use among this population. Herein, we report on the feasibility, strengths, and limitations of this methodology. METHODS: We conducted a series of GMB workshops and validation interviews to collect data in the form of questionnaires, diagrams, and recordings of group discussion. Data were integrated into a conceptual diagram of the "system" of factors associated with uptake and use of CGM over time. We evaluate the feasibility of each aspect of the study, including the teaching of systems thinking to older adult participants. We collected participant feedback on positive aspects of their experiences and areas for improvement. RESULTS: We completed nine GMB workshops with older adults and their caregivers (N = 33). Each three-hour in-person workshop comprised: (1) questionnaires; (2) the GMB session, including both didactic components and structured activities; and (3) a brief focus group discussion. Within the GMB session, individual drawing activities proved to be the most challenging for participants, while group activities and discussion of relevant dynamics over time for illustrative (i.e., realistic but not real) patients yielded rich engagement and sufficient information for system diagramming. Study participants liked the opportunity to share experiences with peers, learning and enhancing their knowledge, peer support, age-specific discussions, the workshop pace and structure, and the systems thinking framework. Participants gave mixed feedback on the workshop duration. CONCLUSIONS: The study demonstrates preliminary feasibility, acceptability, and the value of GMB for engaging older adults about key determinants of complex health behaviors over time. To our knowledge, few studies have extended participatory systems science methods to older adult stakeholders. Future studies may utilize this methodology to inform novel approaches for supporting health across the lifespan.

Racial disparities in stroke incidence in the Women's Health Initiative: Exploring biological, behavioral, psychosocial, and social risk factors.

Background: - Disparities in incident stroke risk among women by race and ethnicity persist. Few studies report the distribution and association of stroke risk factors by age group among a diverse sample of women. Methods: - Data from the Women's Health Initiative (WHI) Observational Study collected between 1993 and 2010 were used to calculate cumulative stroke incidence and incidence rates among non-Hispanic African American (NHAA), non-Hispanic white (NHW), and Hispanic white or African American (HWAA) women by age group in participants aged ≥50 years at baseline (N = 77,247). Hazard ratios (HRs) and 95% CIs for biological, behavioral, psychosocial, and socioeconomic factors overall and by race or ethnicity were estimated using sequential Cox proportional hazard regression models. Results: - Average follow-up time was 11.52 (SD, 3.48) years. The incident stroke rate was higher among NHAA (306 per 100,000 person-years) compared to NHW (279/100,000py) and HWAA women (147/100,000py) overall and in each age group. The disparity was largest at ages >75 years. The association between stroke risk factors (e.g., smoking, BMI, physical activity) and incident stroke varied across race and ethnicity groups. Higher social support was significantly associated with decreased stroke risk overall (HR:0.84, 95% CI, 0.76, 0.93); the degree of protection varied across race and ethnicity groups. Socioeconomic factors did not contribute additional stroke risk beyond risk conferred by traditional and psychosocial factors. Conclusions: - The distribution and association of stroke risk factors differed between NHAA and NHW women. There is a clear need for stroke prevention strategies that address factors driving racial disparities in stroke risk.

The motivation and process for developing a consortium-wide time and motion study to estimate resource implications of innovations in the use of genome sequencing to inform patient care.

Costs of implementing genomic testing innovations extend beyond the cost of sequencing, affecting personnel and infrastructure for which little data are available. We developed a time and motion (T&M) study within the Clinical Sequencing Evidence-Generating Research (CSER) consortium to address this gap, and herein describe challenges of conducting T&M studies within a research consortium and the approaches we developed to overcome them. CSER investigators created a subgroup to carry out the T&M study (authors). We describe logistical and administrative challenges associated with resource use data collection across heterogeneous projects conducted in real-world clinical settings, and our solutions for completing this study and harmonizing data across projects. We delineate processes for feasible data collection on workflow, personnel, and resources required to deliver genetic testing innovations in each CSER project. A critical early step involved developing detailed project-specific process flow diagrams of innovation implementation in projects' clinical settings. Analyzing diagrams across sites, we identified common process-step themes, used to organize project-specific data collection and cross-project analysis. Given the heterogeneity of innovations, study design, and workflows, which affect resources required to deliver genetic testing innovations, flexibility was necessary to harmonize data collection. Despite its challenges, this heterogeneity provides rich insights about variation in clinical processes and resource implications for implementing genetic testing innovations.

Improving Prediction of Tobacco Use Over Time: Findings from Waves 1-4 of the Population Assessment of Tobacco and Health Study.

INTRODUCTION: First-order Markov models assume future tobacco use behavior is dependent on current tobacco use and are often used to characterize patterns of tobacco use over time. Higher-order Markov models that assume future behavior is dependent on current and prior tobacco use may better estimate patterns of tobacco use. AIMS AND METHODS: This study compared Markov models of different orders to examine whether incorporating information about tobacco use history improves model estimation of tobacco use and estimated tobacco use transition probabilities. We used data from four waves of the Population Assessment of Tobacco and Health Study. In each Wave, a participant was categorized into one of the following tobacco use states: never smoker, former smoker, menthol cigarette smoker, non-menthol cigarette smoker, or e-cigarette/dual user. We compared first-, second-, and third-order Markov models using multinomial logistic regression and estimated transition probabilities between tobacco use states. `RESULTS: The third-order model was the best fit for the data. The percentage of former smokers, menthol cigarette smokers, non-menthol cigarette smokers, and e-cigarette/dual users in Wave 3 that remained in the same tobacco use state in Wave 4 ranged from 63.4% to 97.2%, 29.2% to 89.8%, 34.8% to 89.7%, and 20.5% to 80.0%, respectively, dependent on tobacco use history. Individuals who were current tobacco users, but former smokers in the prior two years, were most likely to quit. CONCLUSIONS: Transition probabilities between tobacco use states varied widely depending on tobacco use history. Higher-order Markov models improve estimation of tobacco use over time and can inform understanding of trajectories of tobacco use behavior. IMPLICATIONS: Findings from this study suggest that transition probabilities between tobacco use states vary widely depending on tobacco use history. Tobacco product users (cigarette or e-cigarette/dual users) who were in the same tobacco use state in the prior two years were least likely to quit. Individuals who were current tobacco users, but former smokers in the prior two years, were most likely to quit. Quitting smoking for at least two years is an important milestone in the process of cessation.

A conceptual model of the continuous glucose monitoring integration process for older adults with diabetes developed using participatory systems science methods.

AIMS: Continuous glucose monitoring (CGM) use remains low in older adults. We aimed to develop a conceptual model of CGM integration among older adults with type 1 and type 2 diabetes. METHODS: We previously engaged older adults with type 1 diabetes using participatory system science methods to develop a model of the system of factors that shape CGM integration. To validate and expand the model, we conducted semi-structured interviews with 17 older adults with type 1 and type 2 diabetes and 3 caregivers. Vignettes representing each integration phase were used to elicit outcomes and strategies to support CGM use. Data were analyzed using team-based causal loop diagraming. RESULTS: The model includes six phases spanning (1) CGM uptake; (2) device set-up; acquisition of (3) belief in oneself to use CGM effectively; (4) belief that CGM is preferable to blood glucose monitoring; (5) belief in future CGM benefits CGM; and (6) development of a sense of reliance on CGM. Causal loop diagrams visualize factors and feedback loops shaping outcomes at each phase. Participants proposed support strategies spanning clinical, educational, and behavioral interventions. CONCLUSIONS: The model underscores the complex transition of learning new technology and provides opportunities for tailored support for older adults.

Implementation costs of sugary drink policies in the United States.

To support implementation of important public health policies, policymakers need information about implementation costs over time and across stakeholder groups. We assessed implementation costs of two federal sugar-sweetened beverage (SSB) policies of current policy interest and with evidence to support their effects: excise taxes and health warning labels. Our analysis encompassed the entire policy life cycle using the Exploration, Preparation, Implementation, and Sustainment framework. We identified implementation actions using key informant interviews and developed quantitative estimates of implementation costs using published literature and government documents. Results show that implementation costs vary over time and among stakeholders. Explicitly integrating implementation science theory and using mixed methods improved the comprehensiveness of our results. Although this work is specific to federal SSB policies, the process can inform how we understand the costs of many public health policies, providing crucial information for public health policy making.

Optimizing process flow diagrams to guide implementation of a colorectal cancer screening intervention in new settings.

PURPOSE: The goal of this study was to assess acceptability of using process flow diagrams (or process maps) depicting a previously implemented evidence-based intervention (EBI) to inform the implementation of similar interventions in new settings. METHODS: We developed three different versions of process maps, each visualizing the implementation of the same multicomponent colorectal cancer (CRC) screening EBI in community health centers but including varying levels of detail about how it was implemented. Interviews with community health professionals and practitioners at other sites not affiliated with this intervention were conducted. We assessed their preferences related to the map designs, their potential utility for guiding EBI implementation, and the feasibility of implementing a similar intervention in their local setting given the information available in the process maps. RESULTS: Eleven community health representatives were interviewed. Participants were able to understand how the intervention was implemented and engage in discussions around the feasibility of implementing this type of complex intervention in their local system. Potential uses of the maps for supporting implementation included staff training, role delineation, monitoring and quality control, and adapting the components and implementation activities of the existing intervention. CONCLUSION: Process maps can potentially support decision-making about the adoption, implementation, and adaptation of existing EBIs in new contexts. Given the complexities involved in deciding whether and how to implement EBIs, these diagrams serve as visual, easily understood tools to inform potential future adopters of the EBI about the activities, resources, and staffing needed for implementation.

Local Health Department Values and Organizational Authorities Guiding Cross-Sector Work During COVID-19.

OBJECTIVE: The COVID-19 pandemic highlighted the role that local health departments (LHDs) have in cross sector can address alone, including the work of value alignment and the strategic use of organizational authorities. The practices by which LHDs used their authorities to conduct cross-sector work during the pandemic need exploration. METHOD: We conducted semi-structured interviews with 19 public health leaders from metropolitan LHDs across the United States. Our interview guide assessed the values that LHD leadership prioritized in their cross-sector work as well as the range of organizational authorities they leveraged to influence external decision-making in other sectors. RESULTS: We found that LHDs approached cross-sector work by leaning on diverse values and authorities, each with unique implications for their work. The LHDs emphasized their approach to value alignment on a sector-by-sector basis, strategically using diverse organizational authorities-economic, political, moral, scientific, and logistical. While each authority and value we assessed was present across all interviewees, how each shaped action varied. Internally, LHDs emphasized certain authorities more than others to the degree that they more closely aligned with prioritized core values. CONCLUSION: Our findings highlight the ongoing need for LHD leadership to improve their ability to effectively communicate public health values and the unique authorities by which health-supporting work can be facilitated, including how this message must be adapted, depending on the specific sectors with which the LHD needs to partner and the governance arrangement in which the LHD is situated.

Effects of caregiver-child acculturation on child obesity among US Latino individuals.

OBJECTIVE: The goal of this study was to assess the association of Latino caregiver-child nativity status (US- and foreign-born) with child obesity using a nationally representative sample. METHODS: Using data from the National Health and Nutrition Examination Survey (NHANES 1999-2018), this study used generalized linear models to identify associations between caregiver-child nativity status (as a proxy for acculturation) and children's BMI. RESULTS: Compared with foreign-born caregiver-child dyads, US-born caregiver-child dyads had 2.35 times the risk of class 2 obesity (95% CI: 1.59-3.47) and 3.60 times the risk of class 3 obesity (95% CI: 1.86-6.96). Foreign-born caregiver and US-born child dyads had 2.01 times the risk of class 2 obesity (95% CI: 1.42-2.84) and 2.47 times the risk of class 3 obesity (95% CI: 1.38-4.44; p < 0.05 for class 2 and class 3). CONCLUSIONS: Compared with foreign-born Latino caregiver-child dyads, dyads with US-born caregivers and children and dyads with foreign-born caregivers and US-born children had significantly increased risk across the severe classes of obesity. Examining the influence and relationship of varying acculturation levels in an immigrant household will help guide more effective clinical and policy guidelines surrounding obesity and weight management in both pediatric and adult US Latino populations.

Engaging stakeholders in the use of an interactive simulation tool to support decision-making about the implementation of colorectal cancer screening interventions.

PURPOSE: We aimed to understand how an interactive, web-based simulation tool can be optimized to support decision-making about the implementation of evidence-based interventions (EBIs) for improving colorectal cancer (CRC) screening. METHODS: Interviews were conducted with decision-makers, including health administrators, advocates, and researchers, with a strong foundation in CRC prevention. Following a demonstration of the microsimulation modeling tool, participants reflected on the tool's potential impact for informing the selection and implementation of strategies for improving CRC screening and outcomes. The interviews assessed participants' preferences regarding the tool's design and content, comprehension of the model results, and recommendations for improving the tool. RESULTS: Seventeen decision-makers completed interviews. Themes regarding the tool's utility included building a case for EBI implementation, selecting EBIs to adopt, setting implementation goals, and understanding the evidence base. Reported barriers to guiding EBI implementation included the tool being too research-focused, contextual differences between the simulated and local contexts, and lack of specificity regarding the design of simulated EBIs. Recommendations to address these challenges included making the data more actionable, allowing users to enter their own model inputs, and providing a how-to guide for implementing the simulated EBIs. CONCLUSION: Diverse decision-makers found the simulation tool to be most useful for supporting early implementation phases, especially deciding which EBI(s) to implement. To increase the tool's utility, providing detailed guidance on how to implement the selected EBIs, and the extent to which users can expect similar CRC screening gains in their contexts, should be prioritized.

Supporting reproductive health among birthing persons with chronic conditions in the United States: A qualitative multilevel study using systems thinking to inform action.

OBJECTIVE (STUDY QUESTION): To use systems thinking with diverse system actors to (a) characterize current problems at the intersection of chronic conditions (CCs) and reproductive health (RH) care and their determinants, (b) determine necessary system actors for change, and (c) document cross-system actions that can improve identified problems in the United States. DATA SOURCES/STUDY SETTING: Data were collected from six groups of system actors via online focus groups. STUDY DESIGN: This is a qualitative multilevel study using the iceberg systems thinking framework. DATA COLLECTION/EXTRACTION METHODS: Data were collected by note-taking and recording six focus groups; analysis incorporated perspective triangulation using the systems thinking iceberg and system mapping to visualize interconnected system challenges, actors, and action ideas. PRINCIPAL FINDINGS: Participants described eight necessary system actors: health care institutions, medical leaders, medical providers, patient advocates and foundations, patients and families, payors, policy makers, and research funders. Forty pain points were identified, spread across each of the four levels of the systems thinking iceberg: undesirable outcomes (6), concerning trends (9), system structure flaws (15), and problematic mental models (10). In response to these pain points, a set of 46 action ideas was generated by participants and mapped into nine action themes: (1) adjust QI metrics, incentives, and reimbursement, (2) bolster RH medical education and training, (3) break down medical silos, (4) enrich patient education, (5) expand the health care team, (6) improve holistic health care, (7) modify research and programmatic funding to prioritize RH and CC, (8) spur innovation for patient visits, and (9) support professional champions and leaders. CONCLUSIONS: By embracing system complexity, creating visual maps, and pushing participants to identify actionable strategies for improvement, this study generates a set of specific actions that can be used to address pain points across the multiple system levels that make improving reproductive care for people with CCs so challenging.

Using systems science to advance health equity in tobacco control: a causal loop diagram of smoking.

OBJECTIVES: Develop and use a causal loop diagram (CLD) of smoking among racial/ethnic minority and lower-income groups to anticipate the intended and unintended effects of tobacco control policies. METHODS: We developed a CLD to elucidate connections between individual, environmental and structural causes of racial/ethnic and socioeconomic disparities in smoking. The CLD was informed by a review of conceptual and empirical models of smoking, fundamental cause and social stress theories and 19 qualitative interviews with tobacco control stakeholders. The CLD was then used to examine the potential impacts of three tobacco control policies. RESULTS: The CLD includes 24 constructs encompassing individual (eg, risk perceptions), environmental (eg, marketing) and structural (eg, systemic racism) factors associated with smoking. Evaluations of tobacco control policies using the CLD identified potential unintended consequences that may maintain smoking disparities. For example, the intent of a smoke-free policy for public housing is to reduce smoking among residents. Our CLD suggests that the policy may reduce smoking among residents by reducing smoking among family/friends, which subsequently reduces pro-smoking norms and perceptions of tobacco use as low risk. On the other hand, some residents who smoke may violate the policy. Policy violations may result in financial strain and/or housing instability, which increases stress and reduces feelings of control, thus having the unintended consequence of increasing smoking. CONCLUSIONS: The CLD may be used to support stakeholder engagement in action planning and to identify non-traditional partners and approaches for tobacco control.

"Nothing is linear": Characterizing the determinants and dynamics of CGM use in older adults with type 1 diabetes.

AIMS: Continuous glucose monitoring (CGM) can reduce hypoglycemia in older adults with type 1 diabetes (T1D). We aimed to characterize factors that influence effective use in this age group. METHODS: Older adults with type T1D (age ≥ 65) and their caregivers participated in one of a series of parallel group model building workshops, a participatory approach to system dynamics involving drawing and scripted group activities. Data were synthesized in a qualitative model of the hypothesized system of factors producing distinct patterns of CGM use in older adults. The model was validated through virtual follow-up interviews. RESULTS: Data were collected from 33 participants (four patient-caregiver dyads, mean age 73.8 ± 4.4 years [range 66-85 years]; 16 % non-CGM users, 79 % pump users). The system model delineates drivers of CGM uptake, drivers of ongoing CGM use, and feedback loops that either reinforce or counteract future CGM use. Participants emphasized the importance of different sets of feedback loops at different points in the duration of CGM use. CONCLUSIONS: The holistic system model underscores that factors and feedback loops driving effective CGM use in older adults are both individualized and dynamic (e.g., changing over time), suggesting opportunities for staged and tailored age-specific education and support.

The Burden of COVID-19 on Caregivers of Children with Suspected Genetic Conditions: A Therapeutic Odyssey.

AIMS: Children with disabilities and rare or undiagnosed conditions and their families have faced numerous hardships of living during the COVID-19 pandemic. For those with undiagnosed conditions, the diagnostic odyssey can be long, expensive, and marked by uncertainty. We, therefore, sought to understand whether and how COVID-19 impacted the trajectory of children's care. METHODS: We conducted semi-structured qualitative interviews with 25 caregivers who, prior to the pandemic, were on a diagnostic odyssey for their children. RESULTS: Most caregivers did not report any interruptions to their child's diagnostic odyssey. The greatest impact was access to therapy services, including the suspension or loss of their child's in-person therapeutic care and difficulties with virtual therapies. This therapy gap caused caregivers to fear that their children were not making progress. CONCLUSION: Although much has been written about the challenges of diagnostic odysseys for children and their families, this study illustrates the importance of expanding the focus of these studies to include therapeutic odysseys. Because therapeutic odysseys continue regardless of whether diagnoses are made, future research should investigate how to support caregivers through children's therapies within and outside of the COVID-19 context.

Modeling the Impact of Nonpharmaceutical Interventions on COVID-19 Transmission in K-12 Schools.

Background. The novel coronavirus SARS-CoV-2 spread across the world causing many waves of COVID-19. Children were at high risk of being exposed to the disease because they were not eligible for vaccination during the first 20 mo of the pandemic in the United States. While children 5 y and older are now eligible to receive a COVID-19 vaccine in the United States, vaccination rates remain low despite most schools returning to in-person instruction. Nonpharmaceutical interventions (NPIs) are important for controlling the spread of COVID-19 in K-12 schools. US school districts used varied and layered mitigation strategies during the pandemic. The goal of this article is to analyze the impact of different NPIs on COVID-19 transmission within K-12 schools. Methods. We developed a deterministic stratified SEIR model that captures the role of social contacts between cohorts in disease transmission to estimate COVID-19 incidence under different NPIs including masks, random screening, contact reduction, school closures, and test-to-stay. We designed contact matrices to simulate the contact patterns between students and teachers within schools. We estimated the proportion of susceptible infected associated with each intervention over 1 semester under the Omicron variant. Results. We find that masks and reducing contacts can greatly reduce new infections among students. Weekly screening tests also have a positive impact on disease mitigation. While self-quarantining symptomatic infections and school closures are effective measures for decreasing semester-end infections, they increase absenteeism. Conclusion. The model provides a useful tool for evaluating the impact of a variety of NPIs on disease transmission in K-12 schools. While the model is tested under Omicron variant parameters in US K-12 schools, it can be adapted to study other populations under different disease settings. Highlights: A stratified SEIR model was developed that captures the role of social contacts in K-12 schools to estimate COVID-19 transmission under different nonpharmaceutical interventions.While masks, random screening, contact reduction, school closures, and test-to-stay are all beneficial interventions, masks and contact reduction resulted in the greatest reduction in new infections among students from the tested scenarios.Layered interventions provide more benefits than implementing interventions independently.

Cross-sector decision landscape in response to COVID-19: A qualitative network mapping analysis of North Carolina decision-makers.

Introduction: The COVID-19 pandemic response has demonstrated the interconnectedness of individuals, organizations, and other entities jointly contributing to the production of community health. This response has involved stakeholders from numerous sectors who have been faced with new decisions, objectives, and constraints. We examined the cross-sector organizational decision landscape that formed in response to the COVID-19 pandemic in North Carolina. Methods: We conducted virtual semi-structured interviews with 44 organizational decision-makers representing nine sectors in North Carolina between October 2020 and January 2021 to understand the decision-making landscape within the first year of the COVID-19 pandemic. In line with a complexity/systems thinking lens, we defined the decision landscape as including decision-maker roles, key decisions, and interrelationships involved in producing community health. We used network mapping and conventional content analysis to analyze transcribed interviews, identifying relationships between stakeholders and synthesizing key themes. Results: Decision-maker roles were characterized by underlying tensions between balancing organizational mission with employee/community health and navigating organizational vs. individual responsibility for reducing transmission. Decision-makers' roles informed their perspectives and goals, which influenced decision outcomes. Key decisions fell into several broad categories, including how to translate public health guidance into practice; when to institute, and subsequently loosen, public health restrictions; and how to address downstream social and economic impacts of public health restrictions. Lastly, given limited and changing information, as well as limited resources and expertise, the COVID-19 response required cross-sector collaboration, which was commonly coordinated by local health departments who had the most connections of all organization types in the resulting network map. Conclusions: By documenting the local, cross-sector decision landscape that formed in response to COVID-19, we illuminate the impacts different organizations may have on information/misinformation, prevention behaviors, and, ultimately, health. Public health researchers and practitioners must understand, and work within, this complex decision landscape when responding to COVID-19 and future community health challenges.

Use of Modeling to Inform Decision Making in North Carolina during the COVID-19 Pandemic: A Qualitative Study.

Background. The COVID-19 pandemic has popularized computer-based decision-support models, which are commonly used to inform decision making amidst complexity. Understanding what organizational decision makers prefer from these models is needed to inform model development during this and future crises. Methods. We recruited and interviewed decision makers from North Carolina across 9 sectors to understand organizational decision-making processes during the first year of the COVID-19 pandemic (N = 44). For this study, we identified and analyzed a subset of responses from interviewees (n = 19) who reported using modeling to inform decision making. We used conventional content analysis to analyze themes from this convenience sample with respect to the source of models and their applications, the value of modeling and recommended applications, and hesitancies toward the use of models. Results. Models were used to compare trends in disease spread across localities, estimate the effects of social distancing policies, and allocate scarce resources, with some interviewees depending on multiple models. Decision makers desired more granular models, capable of projecting disease spread within subpopulations and estimating where local outbreaks could occur, and incorporating a broad set of outcomes, such as social well-being. Hesitancies to the use of modeling included doubts that models could reflect nuances of human behavior, concerns about the quality of data used in models, and the limited amount of modeling specific to the local context. Conclusions. Decision makers perceived modeling as valuable for informing organizational decisions yet described varied ability and willingness to use models for this purpose. These data present an opportunity to educate organizational decision makers on the merits of decision-support modeling and to inform modeling teams on how to build more responsive models that address the needs of organizational decision makers. Highlights: Organizations from a diversity of sectors across North Carolina (including public health, education, business, government, religion, and public safety) have used decision-support modeling to inform decision making during COVID-19.Decision makers wish for models to project the spread of disease, especially at the local level (e.g., individual cities and counties), and to help estimate the outcomes of policies.Some organizational decision makers are hesitant to use modeling to inform their decisions, stemming from doubts that models could reflect nuances of human behavior, concerns about the accuracy and precision of data used in models, and the limited amount of modeling available at the local level.

Using decision analysis to support implementation planning in research and practice.

BACKGROUND: The process of implementing evidence-based interventions, programs, and policies is difficult and complex. Planning for implementation is critical and likely plays a key role in the long-term impact and sustainability of interventions in practice. However, implementation planning is also difficult. Implementors must choose what to implement and how best to implement it, and each choice has costs and consequences to consider. As a step towards supporting structured and organized implementation planning, we advocate for increased use of decision analysis. MAIN TEXT: When applied to implementation planning, decision analysis guides users to explicitly define the problem of interest, outline different plans (e.g., interventions/actions, implementation strategies, timelines), and assess the potential outcomes under each alternative in their context. We ground our discussion of decision analysis in the PROACTIVE framework, which guides teams through key steps in decision analyses. This framework includes three phases: (1) definition of the decision problems and overall objectives with purposeful stakeholder engagement, (2) identification and comparison of different alternatives, and (3) synthesis of information on each alternative, incorporating uncertainty. We present three examples to illustrate the breadth of relevant decision analysis approaches to implementation planning. CONCLUSION: To further the use of decision analysis for implementation planning, we suggest areas for future research and practice: embrace model thinking; build the business case for decision analysis; identify when, how, and for whom decision analysis is more or less useful; improve reporting and transparency of cost data; and increase collaborative opportunities and training.