Effectiveness of interventions to promote physical activity and/or decrease sedentary behaviour among rural adults: a systematic review and meta-analysis

Physical Activity Among Rural Men: Barriers and Preferences

INTRODUCTION

Physical activity positively affects health. Although 94% of Americans know the health benefits of regular physical activity, more than 75% do not achieve recommended levels. The objective of our study was to identify and define the key components of a physical activity intervention tailored to rural American men.

METHODS

We recruited rural men (N = 447) via Amazon's Mechanical Turk online platform to complete a needs assessment survey focused on their interest in a physical activity intervention, preferred intervention features, and potential intervention objectives. Data were summarized by using descriptive statistics. A cumulative logistic regression model examined associations between the men's perceived importance of physical activity to health and their interest in a physical activity intervention.

RESULTS

Almost all participants (97.7%) rated physical activity as "at least somewhat important" to their health, and 83.9% indicated they would be "at least somewhat interested" in participating in a physical activity intervention. On a scale of 1 (not at all a barrier) to 5 (very much a barrier), motivation (mean 3.4; 95% CI, 3.3-3.5), cold weather (mean, 3.4; 95% CI, 3.3-3.5), and tiredness (mean, 3.3; 95% CI, 3.2-3.4) were rated the biggest barriers to physical activity. Becoming fitter (54.1%) was the top reason for joining a physical activity program. Preferred delivery channels for receiving an intervention were mobile application (ranked from 1 being the most preferred and 9 being the least preferred: mean, 2.8; 95% CI, 2.70-3.09) and e-mail (mean, 4.2; 95% CI, 3.92-4.36). Rural men preferred interventions that taught them how to exercise and that could be done from home.

CONCLUSION

Our findings suggest US men in rural areas are receptive to physical activity programs. A systematic approach and a clear model of development are needed to tailor future physical activity interventions to the special needs of rural men.

Physical activity and cognitive function: A comparison of rural and urban breast cancer survivors

PURPOSE

Increasing evidence suggests rural breast cancer survivors (BCS) may experience greater burden in symptoms known to be associated with cancer-associated cognitive decline (CACD). Yet, little is known about CACD in rural BCS. This study (1) examined differences in cognitive function, moderate-to-vigorous physical activity (MVPA), and other CACD correlates and (2) tested the effects of MVPA on cognitive function in rural versus urban BCS.

METHODS

Rural and urban BCS (N = 80), matched on age, education, and time since diagnosis from a larger study, completed cognitive tasks assessing processing speed (Trails-B, Mazes, Task-Switch) and working memory (spatial working memory) and questionnaires assessing subjective memory impairment (SMI), MVPA, and CACD correlates (i.e., sleep quality, fatigue, anxiety/depression). Some participants (n = 62) wore an accelerometer to objectively estimate MVPA. Multiple linear regression and multivariate analysis of covariance were used to test study aims.

RESULTS

Rural BCS (n = 40, M = 61.1±8.4 years-old) performed significantly slower on Trails-B (p<0.01) compared with urban BCS (n = 40, M = 61.0±8.2 years-old) and engaged in less objectively-estimated daily MVPA (mean difference = 13.83±4.73 minutes; p = 0.01). No significant differences in SMI, self-reported MVPA, or CACD correlates were observed (all p>0.28). Regression models did not reveal a significant interaction between MVPA and cognitive performance (all p>0.1); however, estimated marginal means models indicated that the effect of MVPA on processing speed was evident only among rural BCS (Trails-B, p = 0.04; Mazes, p = 0.03).

CONCLUSIONS

Findings suggest rural BCS may suffer greater CACD and engage in less MVPA. Additional research is warranted to further examine CACD and more effectively promote MVPA in rural BCS.

Machine learning and ontology in eCoaching for personalized activity level monitoring and recommendation generation

Leading a sedentary lifestyle may cause numerous health problems. Therefore, passive lifestyle changes should be given priority to avoid severe long-term damage. Automatic health coaching system may help people manage a healthy lifestyle with continuous health state monitoring and personalized recommendation generation with machine learning (ML). This study proposes a semantic ontology model to annotate the ML-prediction outcomes and personal preferences to conceptualize personalized recommendation generation with a hybrid approach. We use a transfer learning approach to improve ML model training and its performance, and an incremental learning approach to handle daily growing data and fit them into the ML models. Furthermore, we propose a personalized activity recommendation algorithm for a healthy lifestyle by combining transfer learning, incremental learning, the proposed semantic ontology model, and personal preference data. For the overall experiment, we use public and private activity datasets collected from healthy adults (n = 33 for public datasets; n = 16 for private datasets). The standard ML algorithms have been used to investigate the possibility of classifying daily physical activity levels into the following activity classes: sedentary (0), low active (1), active (2), highly active (3), and rigorous active (4). The daily step count, low physical activity, medium physical activity, and vigorous physical activity serve as input for the classification models. We first use publicly available Fitbit datasets to build the initial classification models. Subsequently, we re-use the pre-trained ML classifiers on the real-time MOX2-5 dataset using transfer learning. We test several standard algorithms and select the best-performing model with optimized configuration for our use case by empirical testing. We find that DecisionTreeClassifier with a criterion "entropy" outperforms other ML classifiers with a mean accuracy score of 97.50% (F1 = 97.00, precision = 97.00, recall = 98.00, MCC = 96.78) and 96.10% (F1 = 96.00, precision = 96.00, recall = 96.00, MCC = 96.10) in Fitbit and MOX2-5 datasets, respectively. Using transfer learning, the DecisionTreeClassifier with a criterion "entropy" outperforms other classifiers with a mean accuracy score of 97.99% (F1 = 98.00, precision = 98.00, recall = 98.00, MCC = 96.79). Therefore, the transfer learning approach improves the machine learning model performance by ≈ 1.98% for defined datasets and settings on MOX2-5 datasets. The Hermit reasoner outperforms other reasoners with an average reasoning time of 1.1-2.1 s, under defined settings in our proposed ontology model. Our proposed algorithm for personalized recommendations conceptualizes a direction to combine the classification results and personal preferences in an ontology for activity eCoaching. The proposed method of combining machine learning technology with semantic rules is an invaluable asset in personalized recommendation generation. Moreover, the semantic rules in the knowledge base and SPARQL (SPARQL Protocol and RDF Query Language) query processing in the query engine helps to understand the logic behind the personalized recommendation generation.

Effectiveness of Physical Activity Interventions among Rural Cancer Survivors: A Systematic Review and Meta-Analysis

This review estimated the effectiveness of behavior change interventions to increase physical activity (PA) among rural adult cancer survivors. PubMed Medline, CINAHL, and PsychINFO were systematically searched through July 2020. Two independent investigators screened citations to identify studies to increase PA in adults residing in rural areas who had received any cancer diagnosis. Meta-analyses were conducted to assess proportion of participants achieving PA goal, paired mean difference (MD) in aerobic PA and strength training, and retention from baseline to post-intervention. Seven studies met inclusion criteria encompassing a total of 722 participants (591 in intervention and 131 controls). Overall quality of evidence was low to medium. The pooled proportion of participants achieving PA goals (150-225 min/wk) was 39% [95% confidence interval (CI), 18%-62%]. The mean time spent engaging in aerobic PA increased from baseline to post-intervention (range, 6-52 weeks) by 97.7 min/wk (95% CI, 75.0-120.4), and the MD in time spent on strength training was 12.2 min/wk (95% CI, -8.3-32.8). The pooled retention rate was 82% (95% CI, 69%-92%) at 6 to 78 weeks. Because of the modest intervention effects, low quality of evidence, and small number of studies, further rigorously designed behavior change interventions, including randomized controlled trials with long-term follow up, are needed to confirm efficacy for increasing PA in rural cancer survivors and to test innovative implementation strategies to enhance reach and effectiveness.

Effectiveness of a Lifestyle Modification Program Delivered under Real-World Conditions in a Rural Setting

Whilst there is considerable evidence to support the efficacy of physical activity and dietary interventions in disease and death prevention, translation of knowledge into practice remains inadequate. We aimed to examine the uptake, retention, acceptability and effectiveness on physical activity, physical function, sitting time, diet and health outcomes of a Healthy Eating Activity and Lifestyle program (HEAL^TM) delivered under real-world conditions. The program was delivered to 430 adults living across rural South Australia. Participants of the program attended weekly 2 h healthy lifestyle education and exercise group-based sessions for 8 weeks. A total of 47 programs were delivered in over 15 communities. In total, 548 referrals were received, resulting in 430 participants receiving the intervention (78% uptake). At baseline, 74.6% of participants were female, the mean age of participants was 53.7 years and 11.1% of participants identified as Aboriginal and/or Torres Strait Islander. Follow-up assessments were obtained for 265 participants. Significant improvements were observed for walking, planned physical activity, incidental physical activity, total physical activity, 30 s chair stand, 30 s arm curl, 6 min walk, fruit consumption and vegetable consumption, sitting time and diastolic blood pressure. Positive satisfaction and favourable feedback were reported. The healthy lifestyle program achieved excellent real-world uptake and effectiveness, reasonable intervention attendance and strong program acceptability amongst rural and vulnerable communities.

Identifying key barriers to effective breast cancer control in rural settings

Breast cancer is the most common cancer and the second most common cause of cancer mortality among women in the United States. Efforts to promote breast cancer control in rural settings face specific challenges. Access to breast cancer screening, diagnosis, and treatment services is impaired by shortages of primary care and specialist providers, and geographic distance from medical facilities. Women in rural areas have comparable breast cancer mortality rates compared to women in urban settings, but this is due in large part to lower incidence rates and masks a substantial rural/urban disparity in breast cancer survival among women diagnosed with breast cancer. Mammography screening utilization rates are slightly lower among rural women than their urban counterparts, with a corresponding increase in late stage breast cancer. Differences in breast cancer survival persist after controlling for stage at diagnosis, largely due to disparities in access to treatment. Travel distance to treatment centers is the most substantial barrier to improved breast cancer outcomes in rural areas. While numerous interventions have been demonstrated in controlled studies to be effective in promoting treatment access and adherence, widespread dissemination in public health and clinical practice remains lacking. Efforts to improve breast cancer control in rural areas should focus on implementation strategies for improving access to breast cancer treatments.

Trail Conditions and Community Use: Utilizing Geospatial Video to Guide the Adoption of a Spatial-Temporal Trail Audit Tool (STAT)

Physical activity (PA), associated with all-cause mortality, morbidity, and healthcare costs, improves vitamin D absorption, immune response, and stress when completed outdoors. Rural communities, which experience PA inequities, rely on trails to meet PA guidelines. However, current trail audit methods could be more efficient and accurate, which geospatial video may support. Therefore, the study purpose was (1) to identify and adopt validated instruments for trail audit evaluations using geospatial video and a composite score and (2) to determine if geospatial video and a composite score motivate (influence the decision to use) specific trail selection among current trail users. Phase 1 used a mixed-method exploratory sequential core design using qualitative data, then quantitative data for the development of the Spatial-temporal Trail Audit Tool (STAT). Geospatial videos of two Northeast Ohio trails were collected using a bicycle-mounted spatial video camera and video analysis software. The creation of STAT was integrated from Neighborhood Environment Walkability Scale (NEWS), Walk Score, and Path Environment Audit Tool (PEAT) audit tools based on four constructs: trail accessibility, conditions, amenities, and safety. Scoring was determined by three independent reviewers. Phase 2 included a mixed-method convergent core design to test the applicability of STAT for trail participant motivation. STAT has 20 items in 4 content areas computing a composite score and was found to increase trail quality and motivation for use. STAT can evaluate trails for PA using geospatial video and a composite score which may spur PA through increased motivation to select and use trails.

Student-Led Motivational Interviewing for Physical Activity Promotion among Rural Adults: A Feasibility and Acceptability Trial

In many countries, rural residents have lower life expectancies and poorer health outcomes than urban residents. Adults living in rural Australia have lower physical activity levels than major city counterparts, contributing to this observed health disparity. As physical activity interventions in rural populations have shown minimal success, there is an urgent need for innovative and affordable interventions that facilitate active lifestyles in this vulnerable population. This study assessed the feasibility of physical activity-focused motivational interviewing, delivered by university health sciences undergraduates in a rural Australian region. “Health age” was assessed at baseline (n = 62) from physiological and behavioral measures, immediately followed by the motivational interview, with health age again assessed at 8 weeks follow-up. Mixed methods using a questionnaire (n = 41 at both time points) and one-on-one interview (n = 8) identified aspects of intervention acceptability and feasibility. A large majority rated the motivational interview as meaningful (98%), empathetic (96%), autonomy-focused (88%), and likely to lead to sustained behavior change (98%). Interviews highlighted several potential attitudinal and structural factors that might influence long-term behavior change. Further development of this strategy in rural regions will depend on a deeper understanding of individuals’ and communities’ awareness, attitudes, and beliefs in relation to active lifestyles.

Reasoned Action Approach Correlates of Physical Activity Among African American Men Living With HIV: A Cross-Sectional Study

ABSTRACT

Physical activity (PA) is associated with a reduced incidence of chronic diseases. However, little is known about the predictors of PA among African American men living with HIV. We report secondary analyses examining the strength of the relationships between the reasoned action approach constructs, attitude, subjective norm, descriptive norm, and self-efficacy and intention to exercise and self-reported adherence to Department of Health and Human Services PA guidelines using baseline data from a trial of a PA intervention with 302 African American men, between 40 and 88 years of age (M = 53.9; SD = 7.2) living with HIV. Multiple regression revealed that attitude, subjective norm, and self-efficacy were positively associated with intention. Logistic regression revealed higher odds of meeting PA guidelines as self-efficacy increased. Self-efficacy and descriptive norm predicted meeting the aerobic guideline. Self-efficacy predicted meeting the muscle-strengthening guideline. Interventions targeting reasoned action approach variables may contribute to efforts to improve PA among African American men living with HIV.

A physical activity coaching intervention can improve and maintain physical activity and health-related outcomes in adult ambulatory hospital patients: the Healthy4U-2 randomised controlled trial

Background

The Healthy 4 U-2 study sought to evaluate the effect of a twelve-week, physical activity (PA) coaching intervention for changes and maintenance in PA, anthropometrics and health-related outcomes in adults presenting to an ambulatory hospital clinic.

Methods

One hundred and twenty insufficiently active adults were recruited from an ambulatory hospital clinic and randomised to an intervention group that received an education session and five 20-min telephone sessions of PA coaching, or to a control group that received the education session only. ActiGraph GT3X accelerometers were used to measure moderate-to-vigorous physical activity (MVPA) at baseline, post-intervention (3-months) and follow-up (9-months). Secondary outcome measures (anthropometrics, PA self-efficacy, and health-related quality of life) were also assessed at the three time points.

Results

At baseline, the mean age and body mass index of participants were 53 ± 8 years and 31 ± 4 kg/m², respectively. Relative to control, the intervention group increased objectively measured MVPA at post-intervention (p < 0.001) and 9 months follow-up (p < 0.001). At the 9-month follow-up the intervention group completed 22 min/day of MVPA (95% CI: 20 to 25 min/day), which is sufficient to meet the recommended PA guidelines. The intervention group exhibited beneficial changes in body mass (p < 0.001), waist circumference (p < 0.001), body mass index (p < 0.001), PA self-efficacy (p < 0.001), and health-related quality of life (p < 0.001) at the 9-month follow-up.

Conclusions

This study demonstrates that a low contact PA coaching intervention results in beneficial changes in PA, anthropometrics and health-related outcomes in insufficiently active adults presenting to an ambulatory care clinic. The significant beneficial changes were measured at post-intervention and the 9-month follow-up, demonstrating a maintenance effect of the intervention.

Trial registration

Prospectively registered on the Australian and New Zealand Clinical Trials Registry (ANZCTR, Trial registration number: ACTRN12619000036112.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12966-020-01063-x.

Implementation of Physical Activity Interventions in Rural, Remote, and Northern Communities: A Scoping Review

Compared with urban centers, rural, remote, and northern communities face substantial health inequities and increased rates of noncommunicable disease fuelled, in part, by decreased participation in physical activity. Understanding how the unique sociocultural and environmental factors in rural, remote, and northern communities contribute to implementation of physical activity interventions can help guide health promotion policy and practice. A scoping review was conducted to map literature describing the implementation of physical activity interventions in rural, remote, and/or northern communities. Databases MEDLINE, PsycINFO, EMBASE, CINAHL, and SPORTDiscus were searched using a predetermined search strategy. Outcomes of interest included community demographics, program characteristics, intervention results, measures of implementation, and facilitators or barriers to implementation. A total of 1672 articles were identified from a search of databases, and 8 from a targeted hand search. After screening based on inclusion and exclusion criteria, 12 articles were summarized in a narrative review. Prominent barriers to physical activity program implementation included transportation, lack of infrastructure, sociocultural factors, and weather. Facilitators of program success included flexibility and creativity on the part of the implementation team, leveraging community relationships, and shared resources. Few papers reported on traditional implementation outcomes such as fidelity, dose, and quality. There is a lack of rigorous implementation evaluations of physical activity interventions delivered in rural, remote, or northern communities. Positive aspects of rural life, such as social cohesion and willingness to share resources, appear to contribute to successful program implementation.

Rural physical activity interventions in the United States: a systematic review and RE-AIM evaluation

Background

Previous reviews of rural physical activity interventions were focused on intervention effectiveness and had reported overall mixed findings. The purpose of this systematic review was to apply the Reach, Efficacy, Adoption, Implementation and Maintenance (RE-AIM) framework to evaluate the extent to which rural physical activity interventions in the U.S. have reported on dimensions of internal and external validity and to offer suggestions for future physical activity interventions for rural U.S. populations.

Methods

Pubmed, PsychINFO, CINAHL, PAIS, and Web of Science were searched through February 2019 to identify physical activity intervention studies conducted in rural regions in the U.S. with adult populations. Titles, abstracts, and full texts of articles were reviewed against inclusion and exclusion criteria. Data extraction from included articles included a summary of study details, rural classification system used, and the presence or absence of a total 61 RE-AIM indicators, including reach (n = 13), efficacy/effectiveness (n = 10), adoption (n = 21), implementation (n = 9), and maintenance (n = 8).

Results

A total of 40 full-text articles representing 29 unique studies were included. Classifications of rurality included self-statements by authors (n = 19, 65.5%), population/census-based definitions (n = 3, 10.3%), Rural Urban Continuum Codes (n = 3, 10.3%), Rural Urban Commuting Area codes (n = 2, 6.9%), the 2014 Alabama Rural Health Association classification system (n = 1, 3.4%) and the U.S. Office of Management and Budget classification system (n = 1, 3.4%). Individual studies reported between 14.8 to 52.5% of total RE-AIM indicators. Studies reported 15.4 to 84.6% indicators for reach; 20.0 to 70.0% indicators for efficacy/effectiveness; 4.8 to 47.6% indicators for adoption; 11.1 to 88.9% indicators for implementation; and 0 to 25.0% indicators for maintenance.

Conclusions

We found an overall poor reporting of components related to external validity, which hinders the generalizability of intervention findings, and a lack of consistency in the definition of rurality. Future research should focus on balancing factors of internal and external validity, and should aim to develop a greater understanding of how rurality influences health and behavior to provide contextual knowledge needed to advance the translation of physical activity interventions into practice in rural communities and reduce rural health disparities.

Trial registration

The review protocol was registered with PROSPERO: CRD42019116308.

Participant Evaluation of a Prediabetes Intervention Program Designed for Rural Adults

The type 2 diabetes epidemic is a global crisis threatening the health and economies of many nations. This study aimed to evaluate a prediabetes intervention program designed for rural adults in southwestern Ontario based on the feedback of participants. Rural adults with prediabetes were referred by physicians to an intervention program developed to assist with unique barriers rural adults face related to the built environment and socioeconomic status when adopting a healthy lifestyle. After 6 monthly education sessions offered by a dietitian and a nurse, participants completed a questionnaire to share their program experience. In addition, 6 focus groups consisting of 5-9 participants were conducted to assess program acceptability, feasibility, and practicality. Of 49 enrolled, 35 participants aged 60.8 ± 7.1 (mean ± SD) evaluated the program. Participants reported finding the program to be acceptable, feasible, and practical due to the interactive nature of the sessions, the group setting and the availability of health professionals. This prediabetes lifestyle intervention program was perceived as successfully addressing rural adults' needs in terms of adopting a healthy lifestyle. Feedback received through program evaluation, which included a participant experience survey and focus groups has helped improve this program and may benefit other prediabetes education intervention programs.

Exploring Australian residents cycling engagement - Differences in self-reported cycling behaviour between urban and rural dwelling Queenslanders

ISSUE ADDRESSED

Cycling participation rates in Australia and Queensland remain low. Rural residents typically do not have distinct cycling infrastructure available for use. Investigating current cycling participation and barriers to cycling engagement will inform future work to promote cycling engagement.

METHODS

Data were collected through the inclusion of cycling-related questions in the 2012 Queensland Social Survey.

RESULTS

The majority of the sample had not cycled in the previous 12 months (66%; n = 831). Significantly more rural residents reported not cycling due to environmental concerns (P < 0.05) and preferring other modes of transport or exercise (P < 0.01). Rural cyclists (31.4%; n = 113) had higher levels of cycling engagement in the previous week (41.6% to 32.6% urban; χ²  = 7.83; n = 420; P < 0.05) but lower cycling durations than urban cyclists (41.6% rural residents cycling for 30+ minutes vs 57.4% urban residents; χ²  = 8.25; n = 418; P < 0.01). Rural cycling engagement was independently associated with being male (OR = 2.34; 95% CI = 1.40-3.91); sufficient physical activity engagement (OR = 1.86; 95% CI = 1.10-3.15); and having children at place of residence (1-2 children: OR = 3.21; 95% CI = 1.88-5.49; 3+ children: OR = 3.01; 95% CI = 1.17-7.75).

CONCLUSIONS

One-third of all adults cycled in the previous 12 months irrespective of location; however reasons for not cycling varied by urban/rural location. Results indicate that environmental factors appear to be inhibiting cycling participation in rural areas. Advocating for government commitment for infrastructure development to support safe cycling across Queensland including outside of metropolitan areas is required. SO WHAT?: This research explores self-reported cycling engagement amongst a sample of urban and rural dwelling Queenslanders. Differences in cycling exposure and reasons for lack of engagement offer insights into what may encourage greater cycling participation.

Effectiveness of interventions to improve cardiovascular healthcare in rural areas: a systematic literature review of clinical trials

The objective of this systematic literature review is to examine the impact of interventions to improve cardiovascular disease healthcare provided to people living in rural areas. Systematic electronic searches were conducted in Medline, CINAHL, Embase, Scopus, and Web of Knowledge in July 2018. We included clinical trials assessing the effectiveness of interventions to improve cardiovascular disease healthcare in rural areas. Study eligibility assessment, data extraction, and critical appraisal were undertaken by two reviewers independently. We identified 18 trials (18 interventions). They targeted myocardial infarction (five interventions), stroke (eight), and heart failure (five). All the interventions for myocardial infarction were based on organizational changes (e.g. implementation of mobile coronary units). They consistently reduced time to treatment and decreased mortality. All the interventions for heart failure were based on the provision of patient education. They consistently improved patient knowledge and self-care behaviour, but mortality reductions were reported in only some of the trials. Among the interventions for stroke, those based on the implementation of telemedicine (tele-stroke systems or tele-consultations) improved monitoring of stroke survivors; those based on new or enhanced rehabilitation services did not consistently improve mortality or physical function; whereas educational interventions effectively improved patient knowledge and behavioural outcomes. In conclusion, a number of different strategies (based on enhancing structures and providing patient education) have been proposed to improve cardiovascular disease healthcare in rural areas. Although available evidence show that these interventions can improve healthcare processes, their impact on mortality and other important health outcomes still remains to be established.

Revisiting the evolutionary origins of obesity: lazy versus peppy-thrifty genotype hypothesis

The recent global obesity epidemic is attributed to major societal and environmental changes, such as excessive energy intake and sedentary lifestyle. However, exposure to 'obesogenic' environments does not necessarily result in obesity at the individual level, as 40-75% of body mass index variation in population is attributed to genetic differences. The thrifty genotype theory posits that genetic variants promoting efficient food sequestering and optimal deposition of fat during periods of food abundance were evolutionarily advantageous for the early hunter-gatherer and were positively selected. However, the thrifty genotype is likely too simplistic and fails to provide a justification for the complex distribution of obesity predisposing gene variants and for the broad range of body mass index observed in diverse ethnic groups. This review proposes that gene pleiotropy may better account for the variability in the distribution of obesity susceptibility alleles across modern populations. We outline the lazy-thrifty versus peppy-thrifty genotype hypothesis and detail the body of evidence in the literature in support of this novel concept. Future population genetics and mathematical modelling studies that account for pleiotropy may further improve our understanding of the evolutionary origins of the current obesity epidemic.

Cognitive function and the agreement between self-reported and accelerometer-accessed physical activity

Background

Numerous studies have reported weak or moderate correlations between self-reported and accelerometer-assessed physical activity. One explanation is that self-reported physical activity might be biased by demographic, cognitive or other factors. Cognitive function is one factor that could be associated with either overreporting or underreporting of daily physical activity. Difficulties in remembering past physical activities might result in recall bias. Thus, the current study examines whether the cognitive function is associated with differences between self-reported and accelerometer-assessed physical activity.

Methods

Cross-sectional data from the population-based Activity and Function in the Elderly in Ulm study (ActiFE) were used. A total of 1172 community-dwelling older adults (aged 65–90 years) wore a uniaxial accelerometer (activPAL unit) for a week. Additionally, self-reported physical activity was assessed using the LASA Physical Activity Questionnaire (LAPAQ). Cognitive function was measured with four items (immediate memory, delayed memory, recognition memory, and semantic fluency) from the Consortium to Establish a Registry for Alzheimer’s Disease Total Score (CERAD-TS).

Results

Mean differences of self-reported and accelerometer-assessed physical activity (MPA) were associated with cognitive function in men (r_s = −.12, p = .002) but not in women. Sex-stratified multiple linear regression analyses showed that MPA declined with high cognitive function in men (β = −.13; p = .015).

Conclusion

Results suggest that self-reported physical activity should be interpreted with caution in older populations, as cognitive function was one factor that explained the differences between objective and subjective physical activity measurements.

Electronic supplementary material

The online version of this article (10.1186/s12877-018-0747-x) contains supplementary material, which is available to authorized users.

Personalized Physical Activity Coaching: A Machine Learning Approach

Living a sedentary lifestyle is one of the major causes of numerous health problems. To encourage employees to lead a less sedentary life, the Hanze University started a health promotion program. One of the interventions in the program was the use of an activity tracker to record participants' daily step count. The daily step count served as input for a fortnightly coaching session. In this paper, we investigate the possibility of automating part of the coaching procedure on physical activity by providing personalized feedback throughout the day on a participant's progress in achieving a personal step goal. The gathered step count data was used to train eight different machine learning algorithms to make hourly estimations of the probability of achieving a personalized, daily steps threshold. In 80% of the individual cases, the Random Forest algorithm was the best performing algorithm (mean accuracy = 0.93, range = 0.88-0.99, and mean F1-score = 0.90, range = 0.87-0.94). To demonstrate the practical usefulness of these models, we developed a proof-of-concept Web application that provides personalized feedback about whether a participant is expected to reach his or her daily threshold. We argue that the use of machine learning could become an invaluable asset in the process of automated personalized coaching. The individualized algorithms allow for predicting physical activity during the day and provides the possibility to intervene in time.

Relationship between objectively measured intensity of physical activity and self-reported enjoyment of physical activity

Physical activity is an important factor for the maintenance of health. Enjoyment of physical activity is essential to motivate persons to engage in sufficient physical activity. We examined whether self-reported enjoyment of PA is associated with objective measurement of the intensity of PA. A cardiovascular examination program was provided for individuals aged 40-75 years without a history of cardiovascular events in Greifswald, Germany between 2012 and 2013. Participants (n = 255) were asked to wear a three-axial accelerometer device (ActiGraph, GT3X +, Pensacola, Florida, USA) for 7 consecutive days. After wearing the device, the participants were asked to complete the 18-item self-administered physical activity enjoyment scale (PACES). Participants' (n = 200) daily minutes of moderate-to-vigorous physical activity (MVPA) and their enjoyment of PA were analysed in a linear regression approach. The mean age of the participants was 56.3 ± 9.7 years, 41.0% were male. The average MVPA duration was 44.4 ± 27.3 min per day. In the regression analysis, enjoyment of PA was positively associated with MVPA (β = 0.18, 95% CI (0.05; 0.31), p = 0.009), participants with higher enjoyment of PA showed higher MVPA. We found a positive association between MVPA and enjoyment of PA, although for male participants only. Between bouted MVPA and enjoyment of PA there was no significant relationship.