Practical Opportunities for Healthy Diet and Physical Activity: Relationship to Intentions, Behaviors, and Body Mass Index

The purpose of this study was twofold: first, the authors sought to revalidate a previously developed method of measuring practical opportunities for healthy diet and exercise in a larger, more diverse patient population. Second, the study sought to better understand whether these opportunities influenced patients' behavioral intentions, dietary choices, physical activity levels, and body mass indices (BMI).

The authors write that clinical assessment and interventions regarding dietary and other lifestyle modifications currently focus primarily on patients' behavioral intentions and motivation with limited success. This, they argue, calls for new, more efficacious methods to promote lifestyle changes that decrease the morbidity and mortality associated with obesity. The authors offer the capability approach as an alternative framework for understanding individuals' health behaviors. This biopsychosocial model theorizes that merely instructing patients to improve their diets and activity levels is insufficient, and that achieving the desired outcome requires providers to be more mindful of the factors influencing both patients' intentions and behaviors. This approach emphasizes the complexity of behavioral change, and suggests that evaluating practical opportunities or capabilities for change, including resources and conversion factors such as autonomy and health literacy, is crucial in helping patients achieve these goals.

Patients were recruited from eight family practice groups in six cities in Texas. Those over the age of 18 who spoke either English or Spanish were included in the study; exclusion criteria were limited to an unwillingness to participate and cognitive impairments significant enough to limit participation. The population was predominantly Hispanic and white, with 54.3% of 746 participants identifying as Hispanic, 30.6% as Non-Hispanic white, 11.5% as Non-Hispanic black, and 3.6% as other. Nearly 60% of patients had some college education, and two-thirds of participants were female. The majority of participants reported a monthly income range less than $2000, with 35.4% reporting a range less than $1000. The mean age of included patients was 44.4. Although 78.1% of participants had a BMI >25, nearly 55% reported achieving 150 minutes of physical activity each week, per the CDC recommendations. The student group thought it unlikely that the percentage of overweight or obese individuals according to BMI would be so high if the level of physical activity was as high as patients reported. The group also found the monthly incomes interesting given the level of educational attainment, and questioned the accuracy of these figures. The group also discussed the fact that despite population demographics, 92.2% of individuals were surveyed in English and only 7.8% in Spanish. While the authors mention that surveys were offered in both Spanish and English, they fail to address whether both versions were offered to all participants or whether the Spanish form was distributed only upon request. Moreover, the authors do not address the level of language fluency of those surveyed in English; lower levels of familiarity and comfort with the language would invariably affect the reliability of patients' responses. The group found this to be a potential weakness of the study, and would have liked to know how it was decided which version of the survey individuals were given, and who was responsible for making that decision.

The student discussion group also raised questions about the process through which patients were recruited. The authors provided limited information regarding how patients were identified, by whom, and whether the process was standardized across all eight sites. The group would have liked to know how surveys were administered; distributing the surveys in person and having someone available not only to answer questions but also to review the forms for completion would provide more reliable results as compared to sending the surveys to participants via mail. Moreover, the authors admit that due to an error in distribution of survey materials, the STC measure was only included in packets at two of the eight sites; consequently, only 200 of the 746 participants received this particular document and results were reported using imputed data. The discussion group viewed this as a significant limitation of this study and was hesitant to give much credence to the author's conclusions given the lack of information about standardization of methods across the various recruitment sites, missing survey materials, as well as degree of completion of the surveys included in study packets.

The authors utilized a twenty-five item survey called the Capability Assessment for Diet and Activity (CADA), the Starting the Conversation (STC) instrument, as well as the International Physical Activity Questionnaire (IPAQ) in order to measure patients' practical opportunities for diet and exercise, and current behaviors. The group acknowledges that the IPAQ, despite being self-reported, has been shown to be highly reliable but would have liked to see the other instruments in order to get a better idea of the kind of questions asked, and to evaluate whether each would serve its intended purpose. The authors write that the STQ tool is a 7 item scale that assesses servings or occasions per week of consumption of various foods. The group questioned, however, whether the three possible responses (target behavior, need for improvement, and significant need for improvement) were adequate and reliable indicators of patient behaviors. The discussion group argued that ideas about target behavior and the degree of improvement needed is likely to vary from one individual to the next, and that the study could have been strengthened by correlating the self-reported values with quantified measurements.

This study used a factor structure previously evaluated in an earlier study. The four factors measured were 1) diet resources; 2) conversion factors; 3) physical activity resources; and 4) physical activity conversion factors. Confirmatory factor analysis was based on the 14 dietary items and 11 physical activity items shown in table 1, and measured the degree to which various factor loadings impacted the factors' associations with intentions for healthy diet and exercise as well as BMI. Results were similar to those in the previous study, supporting the use of this algorithm on a larger scale. Again, the student group would have liked to know the response options and questioned the impact of language proficiency on the results. For instance, the item "know how to eat healthy foods" was interpreted in three different ways by our group alone; the group felt strongly that similar discrepancies in interpretations were likely to have occurred among study participants, particularly when these items were translated into Spanish. However, the standard deviations for these measures were low, reflecting a low level of variation in responses. The authors found that conversion factors are important for both diet and exercise, but that resources are less predictive of change. This suggests that patients require not only access to resources but education regarding how to best utilize the resources available to them in order to achieve their goals. Healthcare providers should, therefore, emphasize achieving a better understanding of individuals' barriers to lifestyle changes, in order to most effectively counsel them. This is supported by figures 1a and 1b, which demonstrate that conversion factors have the highest value in impacting physical activity levels.

Table 3 predicted the impact of resources and conversion factors of these factors on BMI using two models. The dietary model predicted BMI on the basis of age, race, income, diet conversion, diet resource, and diet intention. The physical activity model predicted BMI using age, race, income, activity conversion, activity resource, and activity intention. Dietary and activity factors were kept isolated from one another. The authors found that the dietary factor accounted for 14% of the variance in BMI while the activity factor accounted for 11% of the variance in BMI. Conversion factors, intention, age, and African American and non-Hispanic race were significant. However, the discussion group was surprised to find that income was not. The group postulated that combining diet and physical activity may demonstrate which has a greater impact on BMI.

The group discussed how it might approach a similar study given the chance to repeat it. Members decided that it would be crucial to ensure distribution of all study materials to all participants, and to check items for completion. They also agreed that a larger sample size would be beneficial in allowing for stratification on the basis of age, educational level, or other factors; it was felt that participants in different age groups or those with varying educational attainment would face diverse barriers to achieving dietary and activity level changes. The group also agreed that it would like to gain a better understanding of educational methods, whether handouts, educational seminars, or others, patients prefer in order to maximize providers' impact in clinical practice.

Competing interests: None declared