Obesity and depression are prominent health conditions that have warranted public health attention. In this project, the specific research question asked was: “Is there a statistically significant difference in the mean value of depression scores for those who are obese compared to those who are not?” To offer informational insight between the sample population and the variables of obesity and depression scores, the SAS software was used for statistical analysis. The research hypothesis was H1: ?1 > ?2, where the reported mean values in depression scores would be higher in those who are obese compared to those who are not.
The connections between obesity(OBESE) and depression scores (DESPRESYR) were investigated by a statistical analysis of a sample population from a dataset named “COH602” was derived from a subset of variables from the California Health Interview Survey (CHIS) 2011-2012 dataset. Prior to analyzing the data, we defined the research variables using the Data Dictionary retrieved from SASHELP.
First, obesity is labeled as OBESE and coded as 1(OBESE=1) and coded as 0 (NOT OBESE=0) for not-obese. An individual with Obese status is classified as a body mass index ? 30 kg/m2. Next, depression scores are labeled as DESPRESYR and defined as serious psychological depression in past year using the Kessler 6 Score (K6) nonspecific distress scale; where scores range from 6-30 on the CHIS 2011-2012 Survey (CHIS, 2014.). The K6 asks: During the worst emotional month in the past 12 months, about how often did you feel ·1) nervous; 2) hopeless; 3) restless or fidgety; 4) so depressed that nothing could cheer you up; 5) that everything was an effort; 6) worthless? (Prochaska et al.,2012.) K6 response categories and item scores are: all the time (1), most of the time (2), some of the time (3), a little of the time (4), or none of the time (5). (Kessler ey al.,2003.) Serious mental distress and moderate mental stress were identified using reserve coding by the criteria of scores: K6 ? 13, and 5 ? K6 <13, respectively (Prochaska et. al., 2012). For the purpose of this analysis “Depression score” and “K6 score” will be used synonymously and interchangeably to identify the variable DEPRESSYR. See Table 1. Variables and Characteristics.
Table 1. Variables and Characteristics
- Variable Label Code Notation
- Obese Obese 0=Not obese
- 1= Obese Not-obese
- SRSEX Gender 1=male
- 2=female male
- DEPRESSYR Depression Score 6-30 K6 Score= depression sc
Descriptive and Inferential Statistics were analyzed using SAS data tools and procedures. Data was summarized by frequencies (%) using PROC FREQ in variables of obese status, depression scores and gender; basic statistical measures of mean and standard deviation(SD) using PROC UNIVARIATE in variables of obese status, depression scores and gender. Chi-square tests were performed using PROC FREQ in order to examine the percentages between obese status and individual K6 scores and to summarize the point estimates and 95% confidence limits between obese status and gender. Also, a Two-independent Samples T-Test was performed, using PROC TTEST, in order to determine if there was a significant difference between the two mean values of K6 Scores in those who were obese and those who were not. The Null Hypothesis was H0: ?1 = ?2, meaning no change between the population parameter and the hypothesized value. The Decision Rule to REJECT the null hypothesis will result if p-value is less than the alpha value(?=0.05).
Please see separate SAS Codes and Outputs_ Final Analysis Project word document, for all statistical procedures performed using SAS.
In a descriptive statistics analysis, the individual data of the chosen variables was summarized, See Table 2. For instance, of the total amount of participants surveyed (n=42935), 74.68% (n=32063) were considered not-obese (OBESE=0); the other 25.32% (n=10872) were Obese (OBESE=1). The Obese sample averaged at 0.253220 (SD=0.43486). Next, in measuring the depression variable in the surveyed participants (n=42714) the average score for the K6 scale was 26.70 (SD= 3.79247). The median score was 28. Lastly, in the gender variable, 41.57% (n=17848) were males and 58.43% (n=25087) were female; the sample averaged at 1.58430185 (SD=0.49284)
Table 2. Descriptive Statistics Summary
- Variable Frequency (n=) Percent (%) Mean Standard Deviation
- Sample Population 42935 100% Not Obese 32063 74.68% 0.25 0.44
- Obese 10872 25.32% Male 17848 41.57% 1.58 0.49
- Female 25087 58.43% K6 Score 42714 26.70 3.72
In inferential statistics, the two-way table of OBESE by DEPRESSYR illustrates all the possible combinations of the row and column variables, obese and K6 scores, respectively. The highest K6 scores were 30 at 24.91% (n=10638) followed by K6 score 29 at 15.17% (n=6479) and K6 score 28 at 14.87% (n=6353). The two-sample t-test resulted in the average depression scores of those who were obese (? K6 OBESE=0) was 26.87 (SD=3.67) and average depression scores of those not-obese (? K6 NOT OBESE=1) was 26.21(SD=4.23) The T-test further revealed that the Equality of Variances Test of Pr>F was <.0001 which is less than <0.05; therefore, the Satterthwaite Method results were reported. This method assumes the variances between the two samples as unequal. The data further reports the test statistic value (t-value) as 14.58. and the p-value significance (Pr > |t|) as <.0001. The odds ratio of Obese status was 0.98 with a 95% confidence interval of (0.93, 1.01).
Almost a quarter of survey takers, including both obese and not-obese, scored a K6 score of 30- indicating psychological distress. According to the statistical data produced, since the p-value is less than the alpha level of 0.05, we enact the decision rule to Reject the null hypothesis, H0: ?1 = ?2. which indicates a change between the population parameter and the hypothesized value. Thus, the statistical evidence conveyed a significant difference between the mean values of depression scores in obese individuals. Additionally, our research hypothesis is likely to be true based on the aforementioned results and conclude that the mean value of K6 scores among obese individuals is significantly higher than the mean value of K6 scores among non-obese individuals. Lastly, in male to female comparison the risk of being obese is higher for women than in men.
- California Health Interview Survey. CHIS 2011-2012 Adult Survey. UCLA Center for Health Policy Research. Los Angeles, CA: June 2014.
- Kessler RC, Barker PR, et al. (2003) Screening for Serious Mental Illness in the General Population. Archives of General Psychiatry, 60(2): 184-189.
- Prochaska, J. J., Sung, H. Y., Max, W., Shi, Y., & Ong, M. (2012). Validity study of the K6 scale as a measure of moderate mental distress based on mental health treatment need and utilization. International journal of methods in psychiatric research, 21(2), 88-97.
Cite this essay
Final Analysis Project – Obesity and Depression. (2019, Dec 02). Retrieved from https://studymoose.com/final-analysis-project-obesity-and-depression-essay