Food Availability and Link to Obesity

Purpose 

In the article by Chen, Jaenicke & Volpe (2016), the researchers sought to investigate the relationships between overweight and different elements of food environments particularly at home and neighboring areas. According to the study, food environments play a crucial role in influence one's diet. The research question for the study was to examine 'how several factors classified as individual-level, household-level and neighborhood-level factors influence obesity and overweight status.’ The overall hypothesis of this study was that obesity is directly related to the place from where food is obtained/consumed (food environments). In the second article, Morland & Evenson (2009) conducted a study that aimed at measuring the relationship between disparities in accessing food and health outcomes especially obesity in a sample size of 1295 adults based in the southern part of United States. The study was based on a research question intended to ’measure how food establishments such as supermarkets, grocery stores, and fast food outlets affected the development of obesity and overweight status.’ In this particular study, the researchers hypothesized that obesity has a direct relationship to proximity to food establishments such as fast food restaurants, supermarkets, and grocery stores. However, these local food environments vary with regard to the extent to which they influence the development of obesity. 

Research Methods 

In the first article by Chen, Jaenicke & Volpe (2016), they adopted analytical observational design to collect and examine data on a series of the case in what could be termed a cross-sectional study. For instance, in the first article, Chen, Jaenicke, and Volpe compiled data on different levels including individual, household and neighborhood recorded from 2008 to 2012 and observe risk factors to obesity and overweight status outcomes such as body mass index, age and gender. The second article was also based on an observational design in which a cross-sectional study was conducted from January to July 2003. The study compiled and examined data on a random digit dialed phone about non-institutionalized adult population. 

The two studies adopted two data collection methods including observations and documents and records. Observations were used to collect data on various dynamics of the problem, frequency of the target traits or other behaviors. Observations provided both quantitative and qualitative data, for example, food category, body mass index, age and ethnicity among others. Documents and records were used to examine existing data for example telephone logs, reports, and purchases. 

The data collected was mostly quantitative data. The data is categorized as quantitative because it can be measured, for example, in the first article, individual level data such as BMI, age, amount of fast food, income, and gender. 

Observation being the major data collection methods has various limitations. For example, the method is time-consuming as it requires a detailed analysis. For example, Chen et al. (2016) had to examine data from 2008 to 2012 on 38,650 individuals living in 18,381 households. Additionally, it is hard to study past problems using observations. The method also requires some special tools or instruments to work effectively. 

In both studies, descriptive statistics or analysis was used to analyze the collected data. Chen et al. (2016) computed descriptive statistics for individual variables for every level. Descriptive statistics used in the article involved the use of means and standard deviations for continuous variables while percentages were utilized in calculating observations that equaled 1. The Stata software version 1.3 was utilized in the statistical analysis of conditional intraclass correlation coefficient allocated to each model. In the second article Morland and Evenson (2009) used SAS to calculate 11 types of food stores/service places and results coded with 1 versus 0. Measures of central tendency such as mean and median were used to find the range and cut point. Descriptive analyses were then used to describe the sample population to obtain elements such as prevalence ratios and confidence intervals by use of SAS version 9.1. 

However, descriptive statistics analysis method has a broad range of limitations, for example, extreme values easily influence a method such as range. The use of data analysis software may also consume a lot of time and may become difficult to use various methods to calculate the range. Additionally, data applications also require expertise and knowledge to be able to analyze data. 

The articles used various key demographics in describing their study. For example, Chen et al. (2016) described their demographics based on age, gender, ethnicity, race education, income, poverty rates, food store types, and metro status. Similarly, Morland & Evenson (2009) described the population using several characteristics such as age, sex, race, ethnicity, employment, and education. Inclusion criteria included age (between 2 and 17 years for children and 18 years and above for adults), the amount of exercise per day, and central tracts that met low-income and low-access thresholds. Exclusion criteria were not clearly established in both articles. 

Key Findings 

In the first article, Chen et al. (2016) found that approximately a third of the total sample was overweight while another third of the sample was obese. The results in this article were consistent with statistics published in other national surveys including the National Health and Nutrition Examination Survey. Specifically, about 85% of the households were non-Hispanic white while more than average had college-level education and meant average income was found to be $69,000. Also, age and gender were positively related to obesity and overweight status. The USDA score had a negative relationship to the probability of obesity, however; other household level measures showed a significant socioeconomic disparity in overweight status or obesity. After individual- and household-level aspects were adjusted, most store count estimates of the neighborhood food environments were not significantly related to obesity. 

In the second article, the Morland and Evenson discovered that the prevalence of obesity or overweight lesser in regions with supermarkets but higher in those place that had small grocery stores or fast food outlets. These findings were consistent with what was published in other studies indicating that food choices are influenced by food establishments hence diet-related health outcomes. Specifically, the incidence of obesity was lowered by 0.73 in regions with at least one supermarket. The same was observed in areas that had at least one limited service restaurant or food store. Each mile near a supermarket was linked to 6% higher prevalence of obesity. 

Limitation of the Study 

The first limitation is that the study involved too large study sample, for example, more than 38,000 participants. Comparing results from such a large population is difficult. The sample is also not representative of the nationwide population, for example, the second study was only based on people living in the southern part of the United States. However, the studies adopted a statistical method which is efficient for a large amount of data, allows for easy comparison, and can be altered based on the characteristics. Some of the demerits include technicality to use, only efficient on experimental research and time-consuming. Lastly, the limitation is that it is not applicable to qualitative data. 

Major Conclusions of the Study 

The results of this study as discussed by Chen et al.(2016) is that concluded that neighborhood food environment features, for example, food desert status are related to obesity status even when home food environment factors are controlled. On the other hand, Morland & Evenson (2009) concluded that physical access to some specific types of food establishments such as restaurants influences food choices and diet-related health outcomes. These findings can be compared with what has been published in other studies for example, by Alter & Eny (2005); who also reported that the access to particular types of food stores as well as restaurants tend to influence food choices and diet-related diseases such as obesity. 

The study contributes to the scientific literature in various ways by helping in providing useful information on how proximity to food establishments or neighborhood food environments affects population health. For example, by comparing how living near a supermarket instead of a fast food restaurant contributes to obesity. Lastly, further investigations could be carried out because of this study, for example, by examining how social amenities such as health centers affect overweight status or obesity. 

References 

Chen, D., Jaenicke, E. C., & Volpe, R. (2016). Food Environments and Obesity: Household Diet Expenditure versus Food Deserts. AJPH Research 106(5), pp.881-888. 

Morland, K.B., & Evenson, K.R. (2009). Obesity Prevalence and the Local Food Environment. Health Place 15(2), p.491-495.