Abstract
As occupational therapy research advances and the body of evidence supporting effective interventions continues to grow, there is a need to expedite the transfer of research findings into practice, and the use of intervention development frameworks becomes increasingly essential. In this column, we introduce the Multiphase Optimization Strategy (MOST), an engineering-inspired framework for the development, optimization, and evaluation of multicomponent interventions. MOST allows for the systematic development and evaluation of optimized interventions that prioritize effectiveness within constraints like affordability, scalability, and efficiency. Using MOST while developing an intervention may reduce the delay between intervention development and real-world implementation. Moreover, adopting MOST will bolster the use of rigorous research designs in occupational therapy studies and foster shared terminology with other disciplines that have successfully applied this framework across a range of health priorities and conditions. Thus, we advocate integrating MOST into occupational therapy intervention development research.
This The Issue Is column introduces the Multiphase Optimization Strategy (MOST), an engineering-inspired framework for the development, optimization, and evaluation of multicomponent occupational therapy interventions.
Ensuring the use of evidence-based practice (EBP) is a priority within the field of occupational therapy across clinical settings and client populations (American Occupational Therapy Association, 2007; Juckett et al., 2019). As occupational therapy research advances and the body of evidence supporting its effective interventions continues to grow, there is a growing imperative, both within and outside the profession, to enhance translational research efforts and expedite the transfer of research findings into practice (also referred to as the research-to-practice pipeline; Juckett et al., 2019). Given these developments, the integration of intervention development frameworks becomes increasingly necessary.
Translational frameworks for intervention development provide a comprehensive, systematic, and rigorous approach to translating empirical evidence from basic science into practice. Central to these frameworks is the concept of optimization, which involves a purposeful, iterative, and data-driven process aimed at enhancing both the intervention and its implementation within resource constraints. An optimization process is best achieved through the early adoption of a guiding framework. Optimization differs from the post hoc adaptations that are often required when interventions are tailored to new contexts, as occurs in intervention adaptation processes (Guastaferro & Pfammatter, 2023).
Guastaferro and Pfammatter (2023) contrasted the core characteristics and laid out key considerations regarding the selection and use of the following translational intervention development frameworks: the Multiphase Optimization Strategy (MOST; Collins, 2018), the Medical Research Council Guidelines for Complex Interventions (Craig et al., 2008; Skivington et al., 2021), and the National Institutes of Health’s Obesity-Related Behavioral Intervention Trials model (Czajkowski et al., 2015). Of the frameworks reviewed by Gustaferro and Pfammatter as presented in their 2023 article, we believe MOST appears to hold value for and relevance to the profession of occupational therapy. In this column, we contemplate why MOST aligns with the current objectives of occupational therapy research and advocate its integration into the field.
Overview of MOST
MOST, an engineering-inspired framework for the development, optimization, and evaluation of multicomponent interventions, has been applied to behavioral, biobehavioral, and biomedical interventions across public health priorities, including, but not limited to, substance use (Windsor et al., 2018), cancer treatment (Green et al., 2022), palliative care (Wells et al., 2021), diabetes (MacPherson et al., 2022), HIV prevention (Mistler et al., 2023), smoking cessation (Piper et al., 2018), cardiac health (Huffman et al., 2020), and oral health (Ihab et al., 2023). MOST offers a systematic approach to guide the process of developing and evaluating optimized (rather than ultimate or best) interventions—those that are most effective in achieving outcomes under implementation constraints imposed by the need for affordability, scalability, and efficiency. Implementation constraints include human-related aspects (e.g., staff time and expertise, participants' time) and logistical considerations (e.g., cost, space, equipment; Collins, 2018). MOST ensures the development of interventions that balance effectiveness against affordability, scalability, and efficiency by including implementation considerations in the early stages of development. Table 1 defines the key terms used in MOST.
MOST Key Terms and Definitions
Note. MOST = Multiphase Optimization Strategy.
MOST comprises three phases: (1) preparation, (2) optimization, and (3) evaluation. The preparation phase lays the foundation for the subsequent phases and has two main objectives. The first is to develop an empirically and theoretically derived conceptual model that depicts the mechanisms through which the outcome will be effected. The conceptual model also facilitates the identification of candidate intervention components. Components are referred to as candidates because whether they are included in the optimized intervention will be empirically decided in the subsequent optimization phase. The second objective of the preparation phase is to identify the optimization objective for the intervention, that is, the manner in which effectiveness will be strategically balanced with affordability, scalability, and efficiency (Guastaferro & Collins, 2019). At the end of the preparation phase there is a clear depiction of the intervention to be tested in the subsequent phase.
In the optimization phase the objective is to empirically identify the optimized intervention. To accomplish this objective, rigorous and efficient experimental designs—referred to as optimization randomized controlled trials (ORCTs)—are used that allow for the examination of the effect of each component on the outcome of interest, independently and in combination. Suitable experimental designs are from the factorial experimental design family, including factorial or fractional factorial trials; sequential, multiple assignment, randomized trials; and microrandomized trials. Space precludes a full discussion of these experimental designs; readers are referred to Collins (2018) for a comprehensive discussion of MOST. The optimized intervention is identified through an empirical decision-making process that accounts for the optimization objective (Strayhorn et al., 2024). For example, empirical data from a factorial experiment may identify only two of three components producing a desired effect; thus, the optimized intervention would comprise these components, which also align with the optimization objective. Last, the evaluation phase is designed to determine the effectiveness of the optimized intervention compared with a suitable control, frequently by means of a two-arm randomized controlled trial (RCT).
Although the progression of the three phases appears to follow a linear path, the process is iterative and seldom linear (Guastaferro & Pfammatter, 2023). It may be necessary to return to the preparation phase after an ORCT should there not be a clear optimized intervention (i.e., the resource management principle). MOST also allows for a continuous process of improvement once new insights and advances occur (i.e., the continual optimization principle), thereby further streamlining the effectiveness, affordability, scalability, or efficiency of multicomponent interventions (Collins, 2018).
The Value of Applying MOST in Occupational Therapy
MOST is not specific to any one discipline or population, and thus it can be applied across the scope of the profession’s practice. It focuses on the development of complex interventions (i.e., those that comprise multiple interacting components; Collins, 2018; Craig et al., 2008); therefore, it is compatible with occupational therapy interventions, which typically integrate various components and address the dynamic interplay among an individual, their environment, and the occupations in which they engage. Incorporating MOST into the development of occupational therapy interventions offers benefits that may potentially promote the profession’s intervention science.
MOST ensures that the intervention’s ingredients are clearly defined as measurable components ready for testing. This is in contrast to the classical treatment package approach, in which the intervention is assembled and then evaluated by means of an RCT as a treatment package (Collins, 2018; Guastaferro & Collins, 2019). The RCT, therefore, limits the investigator’s ability to examine the performance of the components of the intervention independently and in combination (Collins et al., 2024). Even in the case of proven efficacy, the contribution of each intervention component, and the interaction between, them remains unknown. For example, assume an intervention that includes the following three components: (1) strategy acquisition and practice, (2) caregiver training, and (3) home visits. Suppose, then, following the classical treatment package approach, an RCT found the intervention to produce clinically and significantly meaningful effects on the outcome of interest (e.g., engagement in everyday occupations). Although effectiveness may have been determined, the two-arm RCT design neither provides empirical information about the individual effects of the three intervention components nor clarifies how one component may influence the effect (positively or negatively) of another component. Without this information, components that do not contribute meaningfully to the outcome of interest, and that may be costly or burdensome to implement, may be included in the intervention unnecessarily.
Another example that illustrates the limitations of the classical treatment approach and the need to define the intervention’s ingredients emerges from an RCT conducted by Rovner et al. (2020). They examined the efficacy of an occupational therapy behavioral intervention versus a community health worker–delivered diabetes self-management education to improve glycemic levels among African Americans with mild cognitive impairment, low medication adherence, and challenges in maintaining glycemic targets. Although the occupational therapy intervention had a strong conceptual foundation and successful implementation, the researchers recognized a notable limitation: It was challenging to break down the intervention into distinct components and to precisely understand the impact of each treatment activity on the outcome (Rovner et al., 2020). MOST offers solutions to these constraints while ensuring the use of rigorous experimental designs. Such designs are crucial for addressing a wide range of clinical questions; enhancing EBP; and ensuring replicability, reproducibility, and treatment fidelity. An example of a hypothetical occupational therapy intervention that follows the MOST phases is depicted in Figure 1.
An example of a hypothetical occupational therapy intervention that follows the MOST phases.
Compared with the alternative frameworks, MOST provides a prescriptive and detailed approach, from the initial development stage to the implementation and dissemination stages, directing the researcher to follow systematic steps to make sure decisions are data driven while balancing effectiveness against real-life considerations. This is important in the current nature of occupational therapy services, where resource considerations (e.g., productivity, budget, and time constraints) play a critical role and may change over time. A scoping review of occupational therapyinterventions for community-dwelling older adults identified key implementation barriers: practitioners’ time constraints, cost concerns, organizational barriers, and caregiver perspectives (Juckett & Robinson, 2018). Therefore, to ensure that EBP is successfully adopted and implemented in practice settings, there is a need to understand practitioners’ perspectives on real-life challenges and opportunities to implement EBP. Thus, if factors affecting implementation and dissemination are considered during intervention development, then there is an increased likelihood of developing an intervention that is effective and feasible to implement without adaptations, thereby potentially accelerating the intervention implementation.
MOST is valuable in evolving areas of practice in which new evidence is needed to justify services to certain populations. In addition, MOST is suitable for improving already-established interventions because it incorporates data from earlier work as a foundation to improve the intervention on the basis of current needs, advancements, and circumstances (following the continual optimization principle).
Conclusion
In this column, we have introduced the MOST framework to the occupational therapy profession. MOST may be a promising avenue to help advance the science of occupational therapy interventions across domains and settings. Moreover, it can guide a systematic process of developing new interventions and improving existing ones. Using MOST enables the delivery of effective optimized interventions while considering implementation factors and fostering rigorous intervention research. We believe that the integration of the MOST framework into occupational therapy ensures a systematic intervention development process. It may also enhance visibility, recognition, and the adoption of shared terminology with other disciplines already using this innovative approach to develop interventions that mitigate diverse health and behavioral problems.