Secondary Logo

Journal Logo

Original Investigation

How Credible Is Online Physical Activity Advice? The Accuracy of Free Adult Educational Materials

Thomas, Jafrā D.1; Cardinal, Bradley J.2

Author Information
Translational Journal of the ACSM: May 1, 2020 - Volume 5 - Issue 9 - p 82-91
doi: 10.1249/TJX.0000000000000122

Abstract

Physical activity provides salutary effects across all domains of preventive medicine, is just as effective or more effective than many pharmacological treatments of disease or infirmity, and is a cost-effective strategy for health promotion (1). Around the world, nations have recognized this by issuing guidelines aimed at helping people realize the diverse health benefits that physical activity provides. There is great consistency across these recommendations, as well as the underlying science supporting the recommendations. The similarity across national guidelines go beyond the quantity and types of physical activity being recommended; they also uniformly stress the need for inclusive physical activity promotion as a means of advancing health equity (i.e., physical activity for all, opportunities for all). For example, in its national action plan to improve health equity, the American College of Sports Medicine posited that equity in physical activity is a prerequisite to equity in health (1). Keeping with an equity perspective, strategies to promote physical activity can be evaluated in terms of how well they increase the odds or probability that diverse population segments will meet physical activity guidelines. Physical activity equity can also be assessed by examining the quality of educational materials that are made available to the general public (2).

On that topic, adults across diverse demographic groups voluntarily search for media-based health information. Mediated health communications can increase health literacy, which relates to the skill and motivation to obtain, understand, and apply credible health information (3). Low health literacy contributes to behavioral health disparities and hospitalizations that are preventable, whereas high health literacy positively predicts regular engagement in prevention-oriented health behaviors, including exercise (4). Educational materials can reinforce physical activity inequity if their content is inconsistent with health literacy principles (5). US adults have reported being confused about the physical activity guidelines, with limited knowledge about the guidelines being common (6,7). Educational materials are of limited to no value if they confuse adults or fail to educate them appropriately, prompting reliance on resources that not all people have access to in order to supplement or understand the material’s content (6,7). As unsatisfying as such scenarios may be, they may be commonplace, which should prompt concern for all interested in equitable physical activity promotion. The Internet is a popular source among adults for physical activity educational materials, and it may even be a first source (8). Because physical activity equity is partially dependent on educational material quality, the aim of this study was to determine the extent that online materials located using lay search strategies presented valid advice with respect to national physical activity guidelines.

BRIEF LITERATURE REVIEW

Before the widespread accessibility of Internet content being what it is today, leaders within the medical and allied health professions urged scholars and professional bodies to ensure that only quality information (i.e., accurate, credible, readable, and actionable) was being disseminated via the Internet (9,10). Nearing the close of the 20th century, for example, Silberg et al. (11) published an article in the Journal of the American Medical Association with the subtitle, “Caveant lector et viewor—let the reader and viewer beware.” Just over a decade later, Bonnar-Kidd and colleagues (12) published an empirical study in Health Communications expressing little confidence lay users will be able to easily find quality online physical activity information. Two decades after Silberg et al.’s (11) study was published, Devine and colleagues (13) urged academic and professional leaders to make the quality of online health-related websites a national public health priority. The inspirations for these calls to action are based in part on a larger movement toward patient-centered care, where a key objective is to empower patients to take responsibility and exercise authority over their health management (14). The calls to actions are also inspired by cultural shifts that encourage individuals to obtain health information and use it to manage their health risk via the adoption of healthy lifestyles (14). Health literacy is critical to patient empowerment, and a fundamental approach to health literacy promotion is educational media that is primarily text based (15).

Lay websites with physical activity–related content are also wrought with the sort of common problems to quality that have been identified in systematic reviews of research literature, which include the presentation of inaccurate and incomplete information with respect to scientific health guidelines (16). For example, Bonnar-Kidd et al. (12) reported that only 22% of their sample of 41 websites intended for the general public had communicated physical activity–related information with high quality, and none of the websites they assessed were highly accurate based on national guidelines from two sources: the American College of Sports Medicine (published in 1998) and the US Department of Health and Human Services (published in 2000). Gorczynski et al. (17) reported that 59% of their sample of 17 websites provided information not supported by the Canadian Physical Activity Guidelines for Adults 18–64 yr (published in 2011). Results from studies on websites for specific population segments or practitioners raise further concerns about the credibility of online physical activity information. For example, Jetha et al. (18) evaluated 30 websites with physical activity–related content specific to individuals with spinal cord injuries. Although the websites in their study were found to be of high technical quality, only 20% presented specific physical activity advice or suggestions. Of those 20%, all were found to fall short on satisfactory accuracy. This phenomenon has been replicated by others. For example, Ahmed et al. (19) combined a systematic search process with a purposive sampling method to obtain 41 websites presenting prevention/management information on sports concussions to the general public. The research team assessed the websites for accuracy using a checklist that was developed using the most recent scientific consensus statement at the time of their study. Just over half of the statements observed in their sample of websites were consistent with their 22-item list. In other words, the information presented was at best incomplete and at worst highly inaccurate.

The quality and accuracy of health-related websites seem to be suboptimal regardless of domain type or intended audience segment. Bonnar-Kidd et al. (12) compared information quality and accuracy by website domain type and found preliminary evidence that domain type is associated with quality and accuracy. In their sample, dot-gov and dot-net/edu were associated with quality and accuracy, whereas only dot-net/edu was associated with accuracy (12). Notably, the accuracy of website content for these significant associations between production source and quality/accuracy was less than desirable (i.e., moderate), especially given the sources of production (12). In the physical activity domain, questionable validity of website quality certification has also been documented. Ahmed et al. (19) reported no significant difference in information accuracy between HONcode-certified (i.e., “Health on the Net Code”) lay educational sport concussion websites and those that were not certified. They postulated that quality certified content may not be regularly updated by site webmasters or operators as a reason for there being no difference between certified and noncertified websites. Irrespective of source or topic areas, it seems that lay adults will be hard-pressed to locate credible content having to do with physical activity.

Study Justification

The literature previously reviewed suggests that the presence of physical activity guidelines may often be absent or incomplete in online educational material, and that the advice or suggestions given tend to be inconsistent with the guidelines. However, the ability to generalize the findings from previous studies to web articles freely available through the Internet is limited in at least two ways. First, most previous studies delimited their focus to resources for a specific subpopulation, who may have physical activity guidelines distinct from those designed for the majority of people residing in the United States. Second, the effect of production source on content consistency with national physical activity guidelines is rarely evaluated, although the source of production may meaningfully influence how consistent the content of educational media is with national physical activity guidelines.

Study Purpose

The purpose of this study was to address the limitations of the extant literature. Specifically, lay search strategies using the Google search engine were used to locate a diverse sample of web articles focused on physical activity promotion, which were used to test two hypotheses on the basis of previous research reports. Hypothesis 1 was that most web articles will present behavioral advice or recommendations inconsistent with the United States’ 2008 Physical Activity Guidelines for Americans (PAGs) 18 to 64 yr of age. Hypothesis 2 was that when messages consistent with the PAGs were observed, most would correspond to guidelines for aerobic (endurance) physical activity rather than muscle strengthening.

METHODS

The present study was conducted as part of a series of independent investigations administered concurrently and using the same sample of materials. As the first to be submitted for peer-reviewed publication, this write-up presents in detail the methodology that served as the basis for the full set of studies.

Inclusion Criteria

The inclusion criteria used by Thomas and Cardinal (5) were generally followed with the following adaptations. To be included, the educational material must have been a web article developed to promote physical activity, which we defined as any effort or action to increase personal value toward physical activity and to motivate physical activity behaviors (20). Web article was defined as containing the following characteristics: a document primarily formatted for viewing as a webpage through an Internet web browser, the primary purpose of which was to provide information, discuss ideas, or provide suggestions pertaining to a specific topic indicated by the document title; the document content was complete in that it had an appreciable beginning, middle, and end; and the document was not part of a series of articles that required reference to a previous or subsequent publication. Second, the communication objective of the web article had to have been educational where the explicit goal was to teach, guide, or persuade the reader with regard to planning or completing some sort of behavior related to physical activity. The primary medium of communication must have been text, which is a preferred medium among adults if their goal is to learn. Specific to this study, the web articles obtained must have presented messages related to the 2008 Physical Activity Guidelines for Americans 18 to 64 yr of age. Of note, this study occurred before the release of the 2018 guidelines.

Web articles from four sources of production were sought: commercial, government, professional associations, and voluntary health agencies. Cardinal and Sachs (21) first proposed this classification scheme, and the definitions advanced by Thomas and Cardinal (5) were used to sort web articles into one of the four categories. Web articles whose primary function was to list ideas (e.g., “top 20 cardio exercises”) were ineligible for inclusion. To ensure that the results could be generalized to organizations for which the content was produced, entities whose primary function was to disseminate information produced by others were ineligible for inclusion (e.g., news articles, personal blogs, and press releases). Periodically, the term webpage is used in this write-up. Webpage is a generic term that represents a document containing content and hosted on a website (e.g., audio file, link repository, video file, and web article; [22]).

Sampling Procedures

A list of plain language search terms was generated between the first and second authors, then arranged into a logic grid format (23). The search terms were partitioned into two columns: query subject (e.g., “cardio AND exercise*”) and query goal/aim (e.g., ideas OR recommendations OR routines OR suggestions OR tips OR “workout plan*”). In addition to the search term logic grid, a list of sites to search directly was generated because they were known to produce physical activity promotion content (5). The search strategy document underwent four independent and critical reviews by three professionals with expertise on the information-seeking behaviors of lay adults. The experts were a web strategist and digital communication specialist (Oregon State University), an associate professor and science librarian (Oregon State University), and a public librarian responsible for selecting materials related to physical activity and exercise (Corvallis-Benton County Public Library, Oregon). Supplemental Content 1, http://links.lww.com/TJACSM/A96, presents the list of terms and organizations used to locate web articles for potential inclusion in the present study.

The Google search engine was used to search the Internet, which is the preferred Internet search engine of 83% of US adults (24). Internet searches were carried out with the web browser set to private to avoid browser history and other customizations from potentially biasing search engine results (25). Only the first four pages generated from initial search results were assessed, which is consistent with previous research and the general searching behavior of adults (26). In addition to a general Internet search, a direct search of websites was undertaken using the Google search engine and the established list of search terms (27). Directly searching websites using the Google search engine entailed entering the website top domain URL into the Google search bar followed by a colon (e.g., “cdc.gov:”). Then, the search terms generated for this study were entered after the top domain, followed by the command to search. At no time during the search processes were webpage hyperlinks activated to locate more potential samples. The search engine hyperlinked webpage results were themselves the webpages to be considered for inclusion in the present study.

A total of 436 web articles were located using the general Internet search. Because the predominant result of the general Internet search was commercial sources, the direct site search focused only on the government, professional, and voluntary health agency categories. There were 197 additional web articles obtained from the direct search of websites. After excluding web articles that did not meet the study inclusion criteria, there were 285 web articles eligible for inclusion, but group sizes based on production source were grossly uneven. To preserve the internal validity of the study, counterbalanced subgroups on the basis of production source were generated. First, 95% representative samples were randomly drawn stratified by production source from the 285 web articles eligible for inclusion. Second, the sample size of each subgroup was systematically reduced by bootstrapping until counterbalanced subgroups were created (28–30). This method generated 139 web articles distributed into the following production source categories: commercial (n = 36), government (n = 35), professional association (n = 32), and voluntary health agency (n = 36). For the present study, 123 of the web articles were evaluated for PAG-related messages after exclusion for one of two reasons: the intended article audience was older adults or women who were (expecting to be) pregnant. More than half (n = 72; 58.5%) of the reviewed articles contained PAG-related messages, which were analyzed in full for the present study.

Measures

Text-based messages embedded in the sample of web articles (N = 72) were compared with the 2008 National Physical Activity Guidelines for adults 18 to 64 yr of age. This delimitation produced 17 guidelines arranged into the following general set of categories: aerobic physical activity (n = 7), additional benefits through aerobic physical activity (n = 3), inactive adults (n = 5), and muscle-strengthening physical activity (n = 2). The unit of analysis was the web article, and the objective was to determine if at least one fully consistent message was observed for each guideline considered. Web articles were also labeled as inconsistent for a guideline if messages pertaining to the guideline were absent from the web article content. Rater reliability was measured using Cohen’s (31) κ calculated using the free webtool developed by Freelon (32). The coding form was pilot tested with web articles not part of the present study. On the basis of the interpretive cut points by Landis and Koch (33), the form demonstrated modest interrater reliability (κ = 0.46) and, after a 12-day grace period, substantial intrarater reliability (κ = 0.70). Supplemental Content 2, http://links.lww.com/TJACSM/A97, presents the study coding form. Raters easily reached full agreement on all discrepancies, and then the first author coded the entire sample of web articles.

Analysis Plan

All statistical analyses were performed in the Statistical Program for the Social Sciences (Version 25). Statistical significance was set at the probability value of P ≤ 0.05. Descriptive statistics were used to summarize attributes of the sample of web articles as a whole and on the basis of production source. Beyond the attributes assessed to address the specific research aims of this study, the following resource characteristics were documented: the publication or update (revision) year, topic areas covered, and intended adult audience (e.g., men, older adults, and women). For the main analysis, a χ2 goodness-of-fit test was used to determine if the consistent–inconsistent categorizations were unequal for the sample of web articles for each of the 17 PAGs considered. Likelihood χ2 tests (i.e., 2) were performed to determine if production source was associated with the degree that web articles were observed to present consistent PAG messages. Cramer’s V was used to measure the magnitude of association estimated by the 2 statistic using the following interpretive guidelines: 0.10, “minimal”; 0.30, “typical”; and 0.50, “substantial” (34).

RESULTS

Descriptive Results

The 72 web articles were obtained from 34 unique organizations that were primarily located in the United States (n = 67; 93.0%). The publication or update year for the sample ranged from 2008 to 2018 (M = 2016.34, SD = 2.02); however, 14 (19.4%) of the web articles did not provide publication date information (i.e., commercial, 7; government, 2; professional association, 0; and voluntary health agency, 5). Only one of the web articles (1.7%) was published/updated in 2008 and another in 2009 (1.7%), and the remainder were published/updated between 2013 and 2018 (96.6%). The sample was distributed into the following production source categories: commercial (n = 19), government (n = 21), professional association (n = 11), and voluntary health agency (n = 21).

Most of the web articles contained at least one graphic (n = 56; 77.8%) and seemed to target a gender-neutral audience or both men and women (n = 67; 93.1%). The most popular article topic was to provide physical activity or exercise ideas or suggestions (n = 42; 32.6%), followed by to encourage a physically active lifestyle (n = 16; 22.2%), the management of a specific health condition (n = 15; 11.6%), strategies to overcome exercise or physical activity barriers (n = 13; 10.1%), technical instructions (n = 12; 9.3%), and an overview of exercise training principles or benefits (n = 1; 1.4%). It was possible for the content of a given web article to have multiple foci areas. The largest number of web articles had as a general subject something other than aerobic (endurance) or muscular fitness promotion (n = 34; 47.2%; e.g., tips to get active, stay active, or get “in shape”), followed by aerobic (endurance) fitness promotion (n = 22; 30.6%), muscular fitness promotion (n = 14; 19.4%), and both aerobic (endurance) and muscular fitness promotion (n = 2; 2.8%).

Main Analysis

Table 1 presents the distribution of a web articles across the two categories—no consistent messages were observed, and yes, at least one consistent message was observed—for the 17 PAGs considered in this study; Table 1 also presents the results of the χ2 goodness-of-fit test testing if any of the no–yes web article distributions were different in magnitude. Across the 17 guidelines considered in this study, the vast majority of web articles (i.e., n = 44–72; 61.1%–100%) did not present a single message that was consistent with any guideline. This trend was significant (P < 0.001) for all but one of the 2008 PAGs assessed (i.e., PAG 4: engage in aerobic activity for at least 10 min, P = 0.059). Not a single message was consistent with 14 of the 17 guidelines for more than 75% of the web articles.

TABLE 1
TABLE 1:
χ 2 Goodness-of-Fit Test: Percent of Web Articles that Presented at Least One Message Consistent with Physical Activity Guidelines for Adults Aged 18–64 yr.

There were two guidelines where the majority of web articles presented at least one consistent message (i.e., the guidelines to accumulate 150 min of moderate intensity physical activity a week (PAG 1) for government sources only and the guidelines to perform aerobic (endurance) activity in at least 10-min bouts (PAG 4) for government and professional association sources only). The remaining 15 guidelines had more than 50% of web articles presenting messages inconsistent with the 2008 PAGs irrespective of production source.

For 12 of the 17 guidelines considered, the source of production was not associated with the degree of message consistency (all, P > 0.05). Variations between sources of production were minimal for these 12 guidelines (V = 0.11–0.26). Where significant association between production source and message consistency was observed, it had to do with aerobic (endurance) physical activity (all, P ≤ 0.05). Four of these guidelines were weekly recommendations for obtaining the most health benefits through aerobic (endurance) physical activity. The fifth was the advice to accumulate daily physical activity through bouts of at least 10 min in duration. Across these five guidelines, the government source subgroup contained the greatest portion of web articles with at least one consistent message (range, 19.0%–57.1%), followed by voluntary health agency (range, 4.8%–38.1%), commercial (range, 0%–15.8%), and last professional association (range, 0%–54.5%). The strength of association between production source and web article consistency was typical across the five guidelines (i.e., V = 0.32–0.39).

The full results to the χ2 tests to determine if consistency varied on the basis of production source are partitioned into the following tables: Table 2 (PAGs for health benefits via aerobic (endurance) activity), Table 3 (PAGs for additional health benefits via aerobic (endurance) activity), Table 4 (PAGs for health benefits via muscle-strengthening activity), and Table 5 (PAGs for adults who are inactive/sedentary).

TABLE 2
TABLE 2:
Comparison by Production Source: Percent of Web Articles that Presented at Least One Message Consistent with Aerobic Physical Activity Guidelines for Adults Aged 18–64 yr.
TABLE 3
TABLE 3:
Comparison by Production Source: Percent of Web Articles that Presented at Least One Message Consistent with Guidelines for Additional Health Benefits Via Aerobic Physical Activity for Adults Aged 18–64 yr.
TABLE 4
TABLE 4:
Comparison by Production Source: Percent of Web Articles that Presented at Least One Message Consistent with the Physical Activity Guidelines for Muscle-Strengthening for Adults Aged 18–64 yr.
TABLE 5
TABLE 5:
Comparison by Production Source: Percent of Web Articles that Presented at Least One Message Consistent with the Physical Activity Guidelines Adults Aged 18–64 yr with Inactive (Sedentary) Lifestyles.

DISCUSSION

Web articles are a popular source of health-related information among adults, with the Internet being a preferred source of physical activity information for many. Unfortunately, Internet-based educational materials may undermine physical activity efforts by promulgating advice that is inconsistent with the established physical activity guidelines. Of concern is that content contradictory to national physical activity guidelines fosters confusion or misinformation about how to develop or maintain activity routines conducive to health and well-being across the life-span. The efforts of lay readers are unlikely to lead to improved health literacy; at times, their efforts may even result in a reduction of their health literacy due to confusion caused by contradictory or inconsistent physical activity advice presented across sources (6,7). To advance understanding of these issues, two hypotheses were tested in this study: first, most web articles developed to promote physical activity would contain messages inconsistent with the 2008 Physical Activity Guidelines for Americans 18 to 64 yr of age (PAGs), and second, when consistent messages were observed, most message would have to do with aerobic (endurance) PAGs rather than muscle-strengthening PAGs.

The first hypothesis was supported. Most of the web articles presented advice that was inconsistent with the PAGs. Depending on the guideline, the portion of web articles lacking at least one consistent message ranged from approximately two-thirds of the sample to the entire sample of web articles. The sample portion that lacked at least one consistent message was significant for all but one guideline, the recommendations to perform aerobic (endurance) activity in at least 10-min intervals (which has been retracted in the 2018 edition of the guidelines). Notably, no web article was observed to present messages consistent with the guideline, “start with light- to moderate-intensity muscle-strengthening activity once per week.” Thus, no P value could be computed for this muscle-strengthening guideline. Regardless, this is a glaring omission.

The second hypothesis was partially supported. Although aerobic (endurance) activity guidelines contained a greater portion of web articles with at least one consistent message in most cases, this was not the case for all aerobic (endurance) activity guidelines. The portion of web articles with messages consistent to muscle-strengthening guidelines was larger than 6 of the 11 aerobic (endurance) activity guidelines. The aerobic (endurance) activity guidelines that typically contained a larger portion of web articles were those focused on accumulating a weekly amount of physical activity. By comparison with the two muscle-strengthening guidelines for substantial health benefits, 1.5 to 3.4 times as many web articles presented at least one consistent message to an aerobic (endurance) activity guideline.

We also tested if the rate of web article consistency with the PAGs varied on the basis of production source, the second apparent study to perform this type of comparative analysis. The first study to do so seems to have been conducted by Bonnar-Kidd et al. (12) 10 years ago. Our results provide evidence that the type of web article organization may influence the extent that the advice presented adheres to established guidelines, but this may only be true for certain aerobic (endurance) activity recommendations. Significant associations of typical magnitude were observed only for aerobic (endurance) activity recommendations having to do with weekly amounts of physical activity, and the advice to accumulate activity in 10-min intervals at a minimum (now retracted in the 2018 guidelines). Where significant association was observed, government sources of production often produced a larger portion of the web articles observed with messages consistent to the PAGs, which was followed by voluntary health agencies in most cases; professional associations and commercial sources had the fewest number of web articles with consistent messages.

Our findings are similar to those reported by Bonnar-Kidd et al. (12) 10 yr ago. On the basis of their reported values of explained variance derived from their analysis of variance, Bonnar-Kidd et al. observed a magnitude of association approaching that of typical between source producer and message consistency, which is also similar to our own findings (34). Their methods only allowed for a summative analysis of association, whereas our approach also demonstrates which type(s) of guidelines is associated with production source. The study-level patterns of consistency that we observed for our aggregate sample held true across the four production source categories. Our results add further evidence that web articles often may contain advice that is inconsistent with the established physical activity guidelines, regardless of production source.

There are several implications stemming from the observations described in this study. First, in addition to the evidence that web articles in circulation may often contain inconsistent messages to established physical activity guidelines, our results suggest that adults who are sedentary or modestly active are least likely to locate web articles with valid advice for achieving health and fitness benefits through physical activity. This produces an invisible disadvantage to adults who may experience social stigma due to their inactive lifestyle and who become motivated to locate advice for making a personal change. Short-term and maintenance of physical activity behavior changes among members of these two groups may offer the greatest health and economic benefits to individuals, organizations, and communities (35). Physical activity guidelines to support adoption and eventual maintenance of higher activity levels among adults with predominantly sedentary lifestyles had the fewest portion of web articles with consistent messages for most guidelines, less than 3% of web articles for most of these guidelines. For advice to work toward the modest weekly goal of 60 min of moderate-intensity aerobic (endurance) activity, less than 2% of web articles accurately conveyed this guideline.

Second, web articles may unintentionally prompt individuals to set unrealistic physical activity goals. Approximately 25% of the US adult population is physically inactive during leisure periods of their day, and the percentage of US adults who meet the recommendation of accumulating 150 min of moderate-intensity aerobic (endurance) activity or its equivalent is near 50%; both of these statistics have persisted for more than two decades to date and with modest variation (36) In our sample, there were two guidelines with which a large number of web articles presented consistent messages: engage in at least 10-min bouts of activity (now retracted in the 2018 PAGs), and avoid inactivity/any amount of activity is beneficial. Although the popularity of these messages suggests that content producers include appropriate advice for the least active, contextually these messages could also be delivered in messages to motivate moderately/highly active readers to stay active or increase their activity level. The gestalt of our findings—both at the study and production source level—suggests that this latter scenario may often be the case concerning Internet-disseminated physical activity information.

A third implication of our results is that information seekers may be more likely to encounter messages inconsistent with public health guidelines for muscle-strengthening activity than for aerobic (endurance) activity. Approximately 10% of web articles with PAG-related messages presented at least one consistent message with basic muscle-strengthening PAGs. In contrast, there were between ~15% and 38% of web articles that presented at least one consistent message with basic aerobic (endurance) activity guidelines—a 1½- to a 4-fold difference in magnitude for the aggregate sample. Although for a few cases more web articles presented messages consistent with muscle-strengthening PAGS than with aerobic (endurance) guidelines at the production source level, the trend was that most web articles had greater consistency with aerobic (endurance) activity PAGs.

Toward Addressing Issues of Quality in Physical Activity Resources

Individuals who produce or select materials developed for physical activity promotion have a model for how they can prevent dissemination of low-quality or inaccurate resources, irrespective of medium (i.e., print or electronic). Many studies have developed quality appraisal tools that take the form of checklists. These checklists allow for a fast review of information accuracy, readability, and how supportive content is to helping readers personally apply information to their own lives.

A consequential outcome that results from material producers or selectors appraising material before their dissemination using a standard form is consciousness raising. Individuals serving in these roles become sensitized to issues embedded in the materials that detract from the potential of materials to be effective, and individuals can develop skills at correcting the issues identified. Experience with performing quality checks using standard forms should support individuals’ ability to teach others to spot and address issues in behavioral resources as well. The Supplemental Content 2 file, http://links.lww.com/TJACSM/A97, provides the standard form used in this study to determine if web article messages were consistent with the 2008 Physical Activity Guidelines for Americans 18–64 yr of age. Given the large overlap between these guidelines and the 2018 second edition guidelines, updating the form to be used to check for advice consistent with the 2018 guidelines would be relatively straightforward.

Readers may reference the Suitability Assessment of Materials as one standard form to appraise resource quality on the basis of literacy demand, among other factors that influence the usefulness of a print or electronic resource (37). Modest use of the Suitability Assessment of Materials to evaluate physical activity behavioral resources has been observed (38), both for print (39) and web-based materials (40).

Study Limitations

There are several limitations to the present study that should be kept in mind. First, it is possible that edits to web article content could result in a decrease or an increase in the level of consistency with a given guideline. These potential changes are not captured using a cross-sectional study design. Second, although the Google search engine is the predominant choice among US adults, web article results obtained through it could modestly vary with results obtained through other search engines. The search for web articles also occurred during a relatively small window of time. Finally, although dichotomous classification systems similar to the one that we used are a popular method by researchers (12,17,18), such approaches do limit the conclusions that could be drawn of how consistent messages tend to be. It has been reported that websites may produce PAG consistent messages on one web article and an inconsistent message to the same PAG on another (12), but the magnitude of these occurrences has yet to be quantified. It is possible that at-odd messages could occur on the same web article, too. Quantitatively documenting the degree that messages are consistent with PAGs and how often inconsistencies occur (if at all) for a given guideline would help to fill a knowledge gap. We took a modest step toward this more nuanced methodology by first documenting if PAG-related messages were present and then quantifying the rate to which web articles presented at least one consistent message for a set of 17 possible PAGs.

Future Research Recommendations

There are several specific ways that future research can increase knowledge about the quality of physical activity advice intended for lay adult audiences presented via the Internet.

First, individuals are most likely to find advice appropriate for moderately to highly active adults. Future research could test this hypothesis by comparing websites based on their apparent population activity group and the extent that appropriate guidelines are communicated with the intended audience, which could potentially be determined from information extracted from titles and/or opening paragraphs. A precursor study could be to limit the web articles to those intended for adults with predominantly sedentary lifestyles and to assess the extent that the advice is consistent with guidelines for this demographic group.

Second, although the guidelines used in this study encompassed a broad range of individuals, future research could advance knowledge of the quality of advice generally communicated by assessing web articles intended for subpopulations whom adapted physical activity guidelines/consensus statements exist (e.g., women who are expecting to become pregnant/pregnant/postpartum; individuals with spinal cord injuries; and older adults, individuals with certain chronic conditions such as osteoarthritis). Research in this area may be especially pertinent if an ample period of time has occurred between two sets of physical activity guidelines for a given subpopulation (e.g., 5 yr).

Third, established physical activity guidelines are communicated in multiple languages. It would be informative to know the extent which non-English web articles may present advice consistent with physical activity guidelines, especially if the intended audience belongs to a group for which adapted, modified, or variable guidelines exist.

Fourth, research that contributes to theoretical knowledge about the process and context in which physical activity promotional messages are produced is needed. Such lines of inquiry would help to identify viable ways to improve the quality of advice presented via print and Internet-based media. The process and context through which physical activity behavioral resources are produced has been studied within kinesiology (41), but the amount of kinesiology research that has focused on process or context is meager (5).

Finally, the communication of inaccurate advice for achieving health and fitness through media is a persistent trend because the creators of that information seem to lack adequate knowledge. The results of content analytical studies support this hypothesis. So, too, do studies that assess knowledge of exercise prescription guidelines (the basis of many physical activity guidelines) among kinesiology college students and certified exercise professionals (42,43). However, professionals formally trained in exercise science are not the only individuals tasked with communicating health-related physical activity information (44). Consider the wide availability of media communications that are on health and fitness topics and produced by commercial, government, professional, and voluntary health agency organizations. Future research that tests educational strategies or modules designed for a broad audience and intended to promote adequate skill in communicating basic, valid, health-related physical activity advice would fill an overlooked knowledge translation gap within kinesiology (5).

CONCLUSION

Beliefs that physical activity is good for a person’s body, mind, and spirit stretch back millennia in human history. Scientific evidence accumulated in recent decades now substantiates what has long been believed. Physical activity is integral to health, independently and directly affecting life satisfaction, feelings of vitality, physical health, psychological health, and biologic risk for morbidity and premature mortality (1). Scientific consensus is strong, and national physical activity guidelines have been issued. This includes the second edition of the 2008 Physical Activity Guidelines for Americans (published in 2018 with substantial overlap with the 2008 edition). Wide and consistent communication of national physical activity guidelines is imperative to health equity promotion. Though often uncertain about what constitutes valid physical activity advice for health and vitality, many adults are receptive to working toward leading a more active lifestyle, and they regularly seek out information on physical activity using the Internet. The results of this study substantiate that most educational materials on physical activity are not credible; most of the web articles in our sample disseminated inconsistent and/or inaccurate information related to the 2008 Physical Activity Guidelines for Americans. Specifically, the majority of free web articles lacked at least one consistent message with 17 guidelines from the 2008 Physical Activity Guidelines for Americans. Strategies to make gains in health equity through gains in physical activity equity must also include attention to the quality and accuracy of widely disseminated behavioral resources, including those disseminated through online media.

We thank the following content experts for their review and feedback of our plan to locate potential study samples: Hannah Rempel (Associate Professor, College of Agricultural Sciences Librarian, and Graduate Student Service Coordinator, Oregon State University), John McQueen (Information Technology Web Strategist and Digital Communication Specialist, Graduate School, Oregon State University), and Lisa Stout (Selector for Collection Materials about Physical Activity and Exercise, Corvallis-Benton County Public Library). We also thank the following faculty at Oregon State University for their guidance and constructive feedback on an earlier iteration of this manuscript: Jessica Beck, Ph.D. (Assistant Dean, Graduate Student Development, Graduate School); M. Margaret Dolcini, Ph.D. (Professor, School of Social and Behavioral Health Sciences); Vicki Ebbeck, Ph.D. (Associate Dean for Student Success, College of Public Health and Human Sciences); and Aurora M. Sherman, Ph.D. (Associate Professor, School of Psychological Science).

The completion of the present study was supported by a 2018–2019 Thayer Raymond Fellowship that was awarded to the first author (funding source: College of Public Health and Human Sciences, Oregon State University).

The authors have no conflicts of interests to report. The views expressed in this article are not endorsed by the American College of Sports Medicine.

REFERENCES

1. Hasson RE, Brown DR, Dorn J, et al. Achieving equity in physical activity participation: ACSM experience and next steps. Med Sci Sports Exerc. 2017;49(4):848–58.
2. Chesser A, Burke A. Navigating the digital divide: a systematic review of ehealth literacy in underserved populations in the United States. Inform Health Soc Care. 2016;41(1):1–19.
3. Osborn CY, Paasche-Orlow MK, Bailey SC, Wolf MS. The mechanisms linking health literacy to behavior and health status. Am J Health Behav. 2011;35(1):118–28.
4. Taggart J, Williams A, Dennis S, et al. A systematic review of interventions in primary care to improve health literacy for chronic disease behavioral risk factors. BMC Fam Pract. 2012;13:49.
5. Thomas JD, Cardinal BJ. Gibberish in communicating written physical activity information: making strides at derailing a perpetual problem. Sociol Sport J. 2018;35(2):108–18.
6. U.S. National Cancer Institute. Physical activity: cancer risk perceptions and group differences in behavior. HINTS Briefs. 2006;4:1–2.
7. Knox ECL, Musson H, Adams EJ. Knowledge of physical activity recommendations in adults employed in England: associations with individual and workplace-related predictors. Int J Behav Nutr Phys Act. 2015;12(1):69.
8. Berry TR, Spence JC, Plotnikoff RC, Bauman A. Physical activity information seeking and advertising recall. Health Commun. 2011;26(3):246–54.
9. Eysenbach G, Powell J, Kuss O, Sa ER. Empirical studies assessing the quality of health information for consumers on the world wide web: a systematic review. JAMA. 2002;287(20):2691–700.
10. Shoemaker SJ, Wolf MS, Brach C. Development of the patient education materials assessment tool (PEMAT): a new measure of understandability and actionability for print and audiovisual patient information. Patient Educ Couns. 2014;96(3):395–403.
11. Silberg WM, Lundberg GD, Musacchio RA. Assessing, controlling, and assuring the quality of medical information on the Internet: Caveant lector et viewor—let the reader and viewer beware. JAMA. 1997;277(15):1244–5.
12. Bonnar-Kidd KK, Black DR, Mattson M, Coster D. Online physical activity information: will typical users find quality information? Health Commun. 2009;24(2):165–75.
13. Devine T, Broderick J, Harris LM, Wu H, Hilfiker SW. Making quality health websites a national public health priority: toward quality standards. J Med Internet Res. 2016;18(8):e211.
14. Ayo N. Understanding health promotion in a neoliberal climate and the making of health conscious citizens. Crit Public Health. 2012;22(1):99–105.
15. Samoocha D, Bruinvels DJ, Elbers NA, Anema JR, van der Beek AJ. Effectiveness of web-based interventions on patient empowerment: a systematic review and meta-analysis. J Med Internet Res. 2010;12(2):e23.
16. Zhang Y, Sun Y, Xie B. Quality of health information for consumers on the web: a systematic review of indicators, criteria, tools, and evaluation results. J Assoc Inf Sci Technol. 2015;66(10):2071–84.
17. Gorczynski P, Patel H, Ganguli R. Evaluating the accuracy, quality, and readability of online physical activity, exercise, and sport information for people with schizophrenia. Ment Health Phys Act. 2013;6(2):95–9.
18. Jetha A, Faulkner G, Gorczynski P, Arbour-Nicitopoulos K, Martin Ginis KA. Physical activity and individuals with spinal cord injury: accuracy and quality of information on the Internet. Disabil Health J. 2011;4(2):112–20.
19. Ahmed OH, Sullivan SJ, Schneiders AG, McCrory PR. Concussion information online: evaluation of information quality, content and readability of concussion-related websites. Br J Sports Med. 2012;46:675–83.
20. DiClemente RJ, Salazar LF, Crosby RA. Health Behavior Theory for Public Health: Principles, Foundations, and Applications. Burlington (MA): Jones & Bartlett Learning; 2013.
21. Cardinal BJ, Sachs M. The readability of exercise promoting literature. Res Q Exerc Sport. 1992;63(1):A32–2.
22. Mozilla.org. What is the difference between webpage, website, web server, and search engine? [Internet]. 2019. Available at: https://developer.mozilla.org/en-US/docs/Learn/Common_questions/Pages_sites_servers_and_search_engines.
23. Aromataris E, Riitano D. Constructing a search strategy and searching for evidence: a guide to the literature search for a systematic review. Am J Nurs. 2014;114(5):49–56.
24. Purcell K, Brenner J, Rainie L. Search Engine Use 2012 [Internet]. Washington (DC): Pew Internet & American Life Project; 2012. Available from: http://www.pewinternet.org/files/old-media/Files/Reports/2012/PIP_Search_Engine_Use_2012.pdf.
25. McInnes N, Haglund BJA. Readability of online health information: implications for health literacy. Inform Health Soc Care. 2011;36(4):173–89.
26. Pan B, Hembrooke H, Joachims T, Lorigo L, Gay G, Granka L. In Google we trust: users’ decisions on rank, position, and relevance. J Comput Mediat Commun. 2007;12(3):801–23.
27. Karch M. Use Google to search within a single website: narrow your search to a single website with this tip. Lifewire.com [Internet]. [cited 9 Oct 2018]; Available from: https://www.lifewire.com/Google-search-one-website-1616499.
28. Cardinal BJ. Development and evaluation of stage-matched written resources about lifestyle and structured physical activity. Percept Mot Skills. 1995;80:543–6.
29. Krejcie RV, Morgan DW. Determining sample size for research activities. Educ Psychol Meas. 1970;30:607–10.
30. Zhu W. Making bootstrap statistical inferences: a tutorial. Res Q Exerc Sport. 1997;68:44–55.
31. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20:37–46.
32. Freelon D. ReCal: intercoder reliability calculation as a web service. Int J Internet Sci. 2010;5:20–33.
33. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159.
34. Vaske JJ, Gliner JA, Morgan GA. Communicating judgments about practical significance: effect size, confidence intervals and odds ratios. Human Dimens Wildl. 2002;7(4):287–300.
35. Carlson SA, Adams EK, Yang Z, Fulton JE. Percentage of deaths associated with inadequate physical activity in the United States. Prev Chronic Dis. 2018;15:E38.
36. An R, Xiang X, Yang Y, Yan H. Mapping the prevalence of physical inactivity in US states, 1984–2015. PLoS One. 2016;11(12):e0168175.
37. Doak CC, Doak LG, Root JH. Teaching Patients With Low Literacy Skills [Internet]. 2nd rev. ed. Philadelphia: Lippincott; 1996, p. 224. Available from: https://www.hsph.harvard.edu/healthliteracy/resources/teaching-patients-with-low-literacy-skills/.
38. Thomas JD, Flay BR, Cardinal BJ. Are physical activity resources understandable as disseminated? A meta-analysis of readability studies. Quest. 2018;70(4):492–518.
39. Vallance JK, Taylor LM, Lavallee C. Suitability and readability assessment of educational print resources related to physical activity: implications and recommendations for practice. Patient Educ Couns. 2008;72(2):342–9.
40. Thomas JD, Cardinal BJ. Analyzing suitability: Are adult web resources on physical activity clear and useful? San Luis Obispo (CA): California Polytechnic State University; 2019. In review.
41. Lattimore D, Griffin SF, Wilcox S, et al. Understanding the challenges encountered and adaptations made by community organizations in translation of evidence-based behavior change physical activity interventions: a qualitative study. Am J Health Promot. 2010;24(6):427–34.
42. Ekkekakis P, Albee MJ, Zenko Z. Knowledge of exercise prescription guidelines across one 4-year kinesiology curriculum. Res Q Exerc Sport. 2016;87:124–30.
43. Zenko Z, Ekkekakis P. Knowledge of exercise prescription guidelines among certified exercise professionals. J Strength Cond Res. 2015;29(5):1422–32.
44. Das BM, DuBose KD, Peyton A. Active health care providers’ practices and views on counseling patients to be active. Transl J Am Coll Sports Med. 2018;3(24):190–5.

Supplemental Digital Content

Copyright © 2020 by the American College of Sports Medicine