Using the positive deviance (PD) approach as the conceptual basis for their study titled “Practical Limitations to a Positive Deviance Approach for Identifying Dietary Patterns Compatible with the Reduction of Cancer Risk”, M. Vossenaar and colleagues aimed to evaluate the viability of constructing healthy eating guides in four international settings, which included the Netherlands, Scotland, Mexico and Guatemala. This broad aim, it seems, was generated from the observation that the recommendations set by the World Cancer Research Fund and the American Institute for Cancer Research (WCRF/AICR) regarding the need for specific dietary behavior deemed as essential for cancer prevention are to a large extent considered impractical in modern settings (Vossenaar et al, 2010). Consequently, one of the secondary aims of the study, according to the authors, was “to provide ‘how to’ guidance for achieving healthful eating and life-style behaviors based on the positive practices found among respondents in the survey” (p. 383). The results demonstrated an overarching basis for PD guidance for developing dietary recommendations for cancer avoidance across all social hierarchies in Guatemala, insignificant for Mexico and Scotland, and effectively impractical for the Nordic country.
Due to the nature and complexity of this particular study, it is imperative to note here that the critique and synthesis will revolve around three thematic propositions, namely: suitability of the PD framework in generating the needed results about the study aims; appropriateness of the literature reviewed to support data, and; population samples and their ability to generate the expected results.
In selecting the framework of positive deviance (PD) approach to provide direction to preventive public health programming, the authors’ of this concordance study acknowledge the fact that several predetermined constructs or variables must be present if the aim of demonstrating ways and means in which dietary practices and life-style behaviors can be realized across four international settings. Indeed, they note that the PD approach is a “format that seeks to guide the development of educational interventions for positive behavioral changes based on the practical examples of those who successfully achieve the recommendations” (Vossenaar et al, 2010, p. 384). This is a good framework for any researcher interested in evaluating the viability of developing healthy eating habits and life-styles patterns aimed at cancer prevention among populations as it takes into account the best practices in dietary habits and life-style changes in a cross-section of the population and attempts to replicate the same to populations at risk. However, the authors are neither clear about the independent variable(s) and the dependent variable(s) that may guide such a construct, nor do they succinctly cover any extraneous variables that may arise, hence posing limitations to this study. These issues need to be put into perspective to not only provide the reader with the kind of flow needed in professional academic articles, but also in ensuring that the study aims and objectives are holistically met.
Issues of reviewing the relevant literature are of germane importance in any research if it is to draw any credibility. In this particular study, however, the authors make tepid attempts at generating relevant literature to support their propositions, particularly with regard to the preventive recommendations of the WCRF/AICR, and with regard to the PD approach. In one particular instance, the authors posit that “it is possible to use survey data to determine how closely population samples are in concordance with the tenets of the original 1997 WCRF/AICR recommendations” (Vossenaar et al, 2010, p. 383), but do not take any initiative to outline these recommendations in a way that will enable the reader understand the basis of their evaluation. The PD approach, which is a central tenet in this study, is again vaguely explained, with the authors preferring to cite too many sources but no detail. A more comprehensive review of the above two issues, in my view, could have assisted the authors make objective comparisons with the data received from the field, hence coming up with more authoritative findings.
Lastly, the authors must receive credit for not only engaging a large population in the study (n=1574) but also for ensuring that it cuts across four countries, thus ensuring that the issue of cultural diversity is comprehensively covered. Indeed, according to Vossenaar et al (2010), “the fundamental precepts, that behavioral modifications are overtly compatible with a culture and are feasible within economic constraints, remain sound and valid” (p. 390). Also, as postulated by the authors, culture is a major influencer of dietary habits, hence the need to include respondents from diverse backgrounds to evaluate how their eating habits and life-style patterns interrelate with the prevalence or prevention of cancer. The issue of including cultural diversity in considering and selecting the population and sample, in my view, was critical if the researchers were to succeed in evaluating the viability of constructing healthy eating models aimed at cancer prevention across the four countries. However, the authors erred in selecting samples from lower and higher socio-economic status in each of the countries. A homogenous sample (either rich or poor) could have resulted in more generalizable results bearing in mind that dietary habits are intrinsically tied to the size of the pocket. Indeed, some dietary habits of the rich, and which predisposes individuals to the risk of cancer, cannot be found among the poor as they are unable to purchase such diets. This oversight by the researchers should be rectified if the findings are to be generalizable and without bias.
Reference List
Vossenaar, M., Bermudez, O.I., Anderson, A.S., & Solomons, N.W. (2010). Practical limitations to a positive deviance approach for identifying dietary patterns compatible with the reduction of cancer risk. Journal of Human Nutrition & Dietetics, 23(4), 382-392.