Framework to Classify Physical Activity Intervention Studies for Older Adults

There is an urgent need to increase physical activity levels among older adults (1). Although robust evidence supports many physical activity interventions, implementation remains a challenge (2). Physical activity programs are often considered complex interventions, comprising multiple interacting components (3). Poor reporting of complex interventions can hinder effective implementation because it can be difficult for end users to replicate the intervention. There is no agreed framework to classify physical activity intervention studies for older adults. This lack of consensus affects how results from intervention studies are reported and interpreted, precluding the ability to make meaningful comparisons and recommendations across different populations and contexts. Existing checklists, such as the Template for Intervention Description and Replication (TIDieR) and the Consensus on Exercise Reporting Template (CERT), are valuable in providing a comprehensive structure to classify interventions at a level sufficient for replication (4,5). However, these checklists do not classify characteristics of the study population, which are important for policymakers and program providers to consider when implementing interventions in real-world settings. When conducting our previous scoping review of physical activity interventions for older adults, we recognized the need to develop a standardized method to classify studies to assist with program implementation as well as study reporting, data extraction, and consolidation of evidence (6). The early iteration of our framework has since undergone extensive testing, modification, and evaluation during a rapid review on the effectiveness of physical activity programs and services for older adults (6). In this commentary, we outline the two-phase development process for and present the final version of a framework for classifying physical activity intervention studies for older adults. PHASE 1: INITIAL DEVELOPMENT OF THE FRAMEWORK An expert panel of seven clinicians and researchers in physical activity, aging, and implementation science participated in regular meetings to develop the framework. International physical activity experts, including from the World Health Organization (WHO), were also consulted throughout the development process. The structure and content of the framework was informed by (i) the WHO Global Action Plan on Physical Activity (GAPPA) (7), (ii) the TIDieR checklist (4), (iii) the CERT (5), (iv) the Prevention of Falls Network Europe (ProFaNE) taxonomy (8), (v) the WHO International Classification of Functioning, Disability and Health (ICF) (9), (vi) the World Report on Ageing and Health (10), (vii) the International Consortium for Health Outcomes Measurement (ICHOM) Standard set of health outcome measures for older persons (11), (viii) the recommended minimum data set of implementation indicators (12), and (ix) the Consolidated Standards of Reporting Trials (CONSORT) checklist (13). The GAPPA calls for physical activity programs and services to improve opportunities for physical activity for older adults (GAPPA Action 3.4), outlining a range of program settings and approaches to physical activity program provision and identifying important subpopulations (7). We considered all of these elements and included additional items to provide a greater level of detail about the type of physical activity provided, the outcomes achieved, and the broader context of the intervention. The TIDieR checklist, CERT, and ProFaNE taxonomy informed the development of the intervention section of the framework (4,5,8). Where appropriate, terminology from these resources was modified to optimize relevance to a physical activity audience. The item descriptors provide further information about how TIDieR and CERT items relate to the items within the current framework. The ICF, the World Report on Ageing and Health, the ICHOM Standard set of health outcome measures for older persons, and the recommended minimum data set of implementation indicators were used to guide the development of the outcomes section of the framework (9–12). The ICF concepts of “Body Structure and Function,” “Activities,” and “Participation” informed the “Intrinsic Capacity,” “Functional Ability,” and “Social Domain” items of our framework (9). Additional items were included to focus on outcomes targeted specifically by physical activity interventions. Outcomes relevant to physical activity interventions were included from the ICHOM Standard set of outcome domains for older people (11). The CONSORT checklist was used throughout the development process to ensure all relevant aspects of an intervention study—participants, interventions, and outcomes—were captured (13). The CONSORT checklist contains additional items critical to the reporting of clinical trials, but they are beyond the intended scope of our framework (e.g., participant recruitment, randomization, and statistical methods) (13). Our intention was for our proposed framework be used in parallel with the relevant CONSORT checklist when reporting intervention studies. PHASE 2: PILOT TESTING AND REFINEMENT OF THE FRAMEWORK The framework was extensively pilot tested during our previous scoping review of systematic reviews of physical activity interventions and a recent rapid review of intervention studies (6,14). We further tested the framework using primary studies that evaluated different types of physical activity interventions in various settings, locations, and populations. Modifications were made to the framework based on group discussion to achieve consensus, with final decisions made by the senior author (C.S.). We also received feedback from experts at WHO through a series of meetings presenting the framework development progress. This iterative process continued throughout the scoping review and rapid review, when the framework was continuously tested, evaluated, and refined during data extraction and synthesis of primary studies. OVERVIEW OF THE FRAMEWORK The framework uses the Population Intervention Comparator Outcome (PICO) structure to classify intervention studies (15) (Table 1). The PICO structure allows for the classification of all aspects that are relevant to study design and implementation (15). For each PICO item, there are further PICO aspects of interest, under which there are up to four levels of classification (Table 1). For example, a strengthening intervention would be classified as intervention (PICO item) – type of activity (PICO aspect of interest) – structured exercise (classification level 1) – strength, resistance, power (classification level 2) – yes (classification level 3) – specify (classification level 4). Each framework item has an accompanying descriptor, which outlines the item definition and any relevant information (table, Supplemental Content 1, https://links.lww.com/TJACSM/A225). The accompanying descriptors provide detailed information for each framework item to enable use in research and program implementation. TABLE 1 - Framework for Classifying Physical Activity Programs and Services for Older Adults. PICO Item PICO Aspect of Interest Framework Classification Level 1 Framework Classification Level 2 Framework Classification Level 3 Framework Classification Level 4 Population—Older adults Context Country High income Specify Upper-middle income Specify Lower-middle income Specify Low income Specify Mixed Specify Not specified Remoteness Rural/remote Urban Both Not specified Other Specify Individual factors Age ≥50 years ≥60 years ≥65 years ≥70 years ≥75 years ≥80 years ≥85 years ≥90 years Other Specify Gender Any Female Male Diverse or underserved cultural/socioeconomic background Yes a Cultural Linguistic Migrant Indigenous Socioeconomic No a Living arrangements Yes a Own home Public housing complex Retirement village Residential aged care facility No a Physical activity level Yes a Low physical activity No a Physical impairments/limitations Yes a Mobility Difficulty walking/slow walking/limited distance Walking aid use Wheelchair use Frailty Prefrail Frail Prefrail/frail Osteopenia Yes Sarcopenia Yes Falls risk Past falls Falls risk Nutrition Undernutrition Overweight/obesity Vision Mild vision impairment Hearing Mild hearing impairment Pain Chronic pain No a Cognitive impairment or low mood Yes a Cognitive impairment Mild cognitive impairment Low mood Symptoms of depression Symptoms of anxiety Other—specify No a Health conditions Yes a Mixed chronic conditions Neurological Stroke Parkinson’s disease Multiple sclerosis Traumatic brain injury Spinal cord injury Mixed neurological Other—specify Cognition or mood Dementia Depression Anxiety Mixed cognition or mood Other—specify Musculoskeletal Osteoarthritis Inflammatory arthritis Osteoporosis Hip fracture Mixed musculoskeletal Other—specify Respiratory Chronic obstructive pulmonary disease Coronavirus disease 2019 Mixed respiratory Other—specify Cardiovascular Heart failure Acute myocardial infarction Hypertensive diseases Mixed cardiovascular Other—specify Metabolic Type 1 diabetes Type 2 diabetes Mixed metabolic Other—specify Cancer Specify Mixed type of cancer Blindness Deafness Multimorbidity Other Specify No a Intervention Type of program Physical activity delivered Physical activity promoted Coaching Referral Brief/very brief intervention eHealth/mHealth Other Specify Type of physical activity Overall activity Structured exercise b Balance, functional, neuromotor Yes Specify type, intensity, prescription, and progression No Strength, resistance, power Yes Specify type, intensity, prescription, and progression No Walking/wheeling Yes Specify type, intensity, prescription, and progression No Endurance Yes Specify type, intensity, prescription, and progression No High-intensity interval training Yes Specify type, intensity, prescription, and progression No Multi-component Yes Specify type, intensity, prescription, and progression No Recreation Tai chi Specify Yoga/Pilates Specify Dance Specify Noncompetitive sport Specify Other Specify Competitive sport Standard Specify Modified Specify Exercise-based videogame Structured exercise Specify Recreation/sporting activity Specify Who prescribed Professional Health professional Specify Physical activity leader Specify Volunteer No age restriction volunteer Peer volunteer Carer/caregiver Professional carer/caregiver Family/peer-support Who delivered Professional Health professional Specify Physical activity leader Specify Volunteer No age restriction volunteer Peer volunteer Carer/caregiver Professional carer/caregiver Family/peer-support Other Specify Not applicable Who with Individual One other person Group Combination Delivery mode Synchronous/live In person Video Specify Phone App/Web based/online Specify Asynchronous/prerecorded Video/DVD Specify App/Web based/online Specify Exercise-based videogame Specify Mixed Specify Location b Community facility Yes No Own home Yes No Health facility Yes Inpatient facility Specify Outpatient facility Specify Other Specify No Residential aged care facility Yes No Retirement village Yes No Workplace Yes No Public housing complex Yes No Faith-based facility Yes No Park/sports field Yes No Research center Yes No Not applicable Prescribed dose Interventions with supervised/unsupervised component Supervised component Yes No Unsupervised component Yes No Session duration Supervised Unsupervised Session frequency Supervised Unsupervised Program duration Supervised Unsupervised Type of supervision In person Online Both Not applicable Additional strategies to support physical activity Yes Activity monitor Yes No Mentoring Yes No Social activity Yes No Incentives Yes No Booklet Yes No Other Specify No Tailoring Yes Specify No Personnel time Staff time Volunteer time Technology-related time Additional staff time available Yes No Number of staff contacts (physical activity promoted) Equipment Yes Specify No Comparison No intervention Specify Physical activity Same physical activity Different dose Specify Different delivery mode Specify Additional strategies to support physical activity Specify Different physical activity Specify Education Specify Nutrition Specify Combination of interventions Specify Other intervention Specify Outcome Physical activity Overall physical activity Steps Device used Light physical activity Device used Questionnaire used Moderate-vigorous physical activity Device used Questionnaire used Upright time Device used Questionnaire used Total physical activity Device used Questionnaire used Domains of physical activity Leisure Specify Transport Specify Household Specify Occupation/education Specify Other Specify Falls Falls Rate of falls Self-reported, proxy Routinely collected Proportion of fallers Self-reported, proxy Routinely collected Fall-related injuries Rate of fall-related injuries Self-reported, proxy Routinely collected Proportion of people with fall-related injuries Self-reported, proxy Proportion of fallers Routinely collected Intrinsic capacity Physical domain Bone mineral density Body composition Specify Cardiometabolic indicators Specify Strength Specify Fitness Specify Pain Specify Other Specify Functional ability Physical domain Mobility and balance Self-report Specify Performance—combined Specify Performance—single task Specify Self-care Self-report Specify Performance Specify Overall function Self-report Specify Other Specify Social domain Self-report Questionnaire used Observation Tool used Cognitive and emotional domain Cognitive Self-report Specify Performance Specify Emotional Self-report Specify Well-being and quality of life Well-being Self-report Specify Quality of life Self-report Specify Frailty Specify Sarcopenia Specify Mortality Specify Polypharmacy Specify Sleep Specify Diet Specify Implementation measures Implementation outcomes Adoption Delivery of the intervention Delivery of the implementation strategies Dose delivered Delivery of the intervention Delivery of the implementation strategies Reach Delivery of the intervention Delivery of the implementation strategies Fidelity (adherence) Delivery of the intervention Delivery of the implementation strategies Sustainability Delivery of the intervention Delivery of the implementation strategies Implementation determinants Context Delivery of the intervention Delivery of the implementation strategies Acceptability Delivery of the intervention Delivery of the implementation strategies Adaptability Delivery of the intervention Delivery of the implementation strategies Feasibility Delivery of the intervention Delivery of the implementation strategies Compatibility (appropriateness) Delivery of the intervention Delivery of the implementation strategies Cost Delivery of the intervention Delivery of the implementation strategies Culture Delivery of the intervention Delivery of the implementation strategies Dose (satisfaction) Delivery of the intervention Delivery of the implementation strategies Complexity Delivery of the intervention Delivery of the implementation strategies Self-efficacy Delivery of the intervention Delivery of the implementation strategies Resource use Individual Specify Healthcare system Specify Adverse events Specify Study type Efficacy/effectiveness Explanatory trial Pragmatic trial Hybrid type 1 trial Implementation/scale-up Implementation Hybrid type 2 trial Hybrid type 3 trial Scale-up a “Yes” at classification level 2 for individual factors indicates that studies specified that particular factor as an inclusion criterion. “No” indicates studies did not specify the factor as an inclusion criterion.b Items under this level are not mutually exclusive (i.e., more than one item can be selected). SUGGESTED USES This framework can be used for three purposes: 1) to classify and extract data for existing studies, 2) to assist with designing and reporting new studies and study replication testing, and 3) to support end users in implementing physical activity interventions for older adults. First, this framework will streamline data classification and extraction, particularly when conducting systematic reviews of physical activity interventions for older adults. Interventions are often specialized to address the needs of this age group, and we have taken this into account when developing the framework to ensure that reviews of this type of intervention can be conducted. The framework can be used as a comprehensive guide for data extraction for all PICO study aspects, which we can verify through conducting two previous reviews using the framework (6,14). This will enable the identification of gaps in the literature. By identifying specific details regarding the intervention components and delivery, the framework can also be used to support meta-analyses and cost-effectiveness analyses. Second, we suggest that physical activity researchers consult this framework when designing and reporting studies to ensure that all relevant aspects of implementation are considered and reported (e.g., considering socioeconomic or cultural/linguistic diversity or the geographical remoteness of the study population). While conducting our previous review (14), we noticed that such characteristics are often overlooked in the reporting of trials, despite being important considerations for study replication and implementation. Third, we developed this framework with the goal of supporting implementation at the forefront. Despite strong evidence for many physical activity programs for older adults, implementation remains challenging (2,16). The resolution adopting the WHO GAPPA observed the request by member states for assistance with implementation, in particular, for tools and resources for the policy action areas (17). By comprehensively outlining intervention components and including the minimum data set of implementation determinants and outcomes (12), our framework will support policymakers and program providers in implementing physical activity programs in different contexts and populations. LIMITATIONS AND STRENGTHS The framework was developed through a rapid but rigorous process which occurred in parallel with data extraction for two reviews of physical activity literature (6,14). By necessity, this framework was developed in a limited timeframe; hence, we did not use an established approach (e.g., Delphi process) to achieve consensus. Although we have extensively pilot tested the framework, we suggest independent evaluation to verify its applicability. The PICO structure is an important strength of this framework. This means that all key aspects of a program or service that are needed for implementation or scale-up are captured. Although the TIDieR checklist and the CERT both provide a comprehensive structure to classify interventions, they do not classify the study population. Knowledge of the characteristics of the population in which an intervention has been delivered is important for implementation, so policymakers and program providers can understand which interventions are most effective in different populations. Ultimately, this framework can be used to classify and extract data, assist with study design and reporting, and support physical activity program implementation.