Food Safety Network

Show me, don't tell me

29.jun.07, Brae V. Surgeoner, Food Protection Trends 27(6): 508

29.jun.07, Brae V. Surgeoner, Food Protection Trends 27(6): 508
foodsafety.ksu.edu
Brae Surgeoner is a research assistant with the International Food Safety network at Kansas State University
The study of consumer food-handling practices has relied almost exclusively on data obtained in self-report surveys. In Redmond and Griffith’s review of food safety studies on consumer food-handling in the home (2003) they found that self-reported practice questions were used in 92 per cent of survey studies. Ad nauseam, researchers ask questions like how often primary food preparers wash their hands after touching raw meat, wash utensils used to handle raw food before they are used for cooked foods and, of course, use a thermometer to check the internal temperature of meat, poultry, and egg dishes. The advantage of the self-reported approach is that data can be collected from representative samples to estimate the prevalence of improper food-handling practices in the population. Likewise, in-depth profiles of consumer food-handling behavior can be generated to document risk factors and offer insight into the relative contribution of both intrinsic and extrinsic factors that affect engagement in safe food-handling behaviors.
Consumer reports about what they say they do in their kitchens are essential to food safety professionals and inspire many research endeavors. However, the validity of self-reported data cannot be taken for granted.
The problem is that people often lie.
Errors in self-reporting may result from an inability to comprehend the question being asked, errors in recalling information from memory and, as is the case in food preparation, focus on recurring events that may blend together in memory and be difficult to report in isolation. Previous research has also found that social desirability bias is pervasive in self-report responses. In other words, the tendency to over-report the frequency with which one engages in socially desirable behaviors (“I always wash my hands after using the bathroom”), and under-report the frequency with which one engages in socially undesirable behaviors (“I never wash my hands after using the bathroom”). Still, self-report data positions researchers to identify strategies that enhance the information that consumers report.
In 1999, a team of Australian researchers, in their article, “A Video Study of Australian Domestic Food-Handling Practices,” impressed upon readers of the Journal of Food Protection the discrepancy that exists between what consumers say they do, and what they actually do. Comparing responses to a food-safety questionnaire administered prior to video surveillance of participants in their home kitchens, the researchers found significant deviations between stated and actual behavior.  For example, there was a highly significant difference between self-reported and observed hand-washing practices.
Observational studies are not without limitations. Often they are too expensive, obtrusive, and/or time consuming to conduct. Moreover, researchers require ethical approval from their Institutional Review Board (IRB) to protect the rights, welfare, and privacy of human subjects who agree to participate. In some sense, this is a daunting task. But, as various government bodies and institutions work to assess the risk of the public health impact from foodborne pathogens associated with a particular food, increasingly, this type of reality-research is required to inform mathematical models. As it stands, current models must, for the most part, assign wide bands of uncertainty for many inputs that are dependent on human behavior (i.e. cooking method).
In the study presented in this issue, Walter and colleagues survey government employees in Osceola County, Florida to determine their in-home food handling practices and to assist in the development of educational materials. The latter objective raises the question of how to evaluate the effectiveness of the material.
There is considerable debate in the health promotion literature over the definition and measurement of relevant outcomes to health promotion, including the use of evaluation methodologies that assess both the outcome achieved and the process by which it was achieved. Nevertheless, the evaluation component is paramount to encourage funding decision-makers to continue to support food safety efforts.
It’s widely accepted in the field of nutrition that education programs possess considerable potential for increasing knowledge and producing change, but documenting these changes have proven difficult for researchers, who tend to favour program takeoffs, not their endings. Consequently, the focus has been on design and implementation of training and development programs rather than evaluation.
The food safety literature increasingly contains calls to action to there is a call to action to for researchers and health promotion folks to develop more compelling safe food-handling messages, yet similar to the field of nutrition, there has been limited evaluation of the effectiveness of various food safety interventions. Without observing actual behavior, food safety educators may be developing interventions that are successful in changing what individuals report they do, but may do little in changing what they actually do.