Abstract
The growing prevalence of overweight and obesity is a major public health concern. Among the U.S. adult population, the prevalence of obesity (defined as a body mass index ranges from 30.0 kg/m2 to 99.8 kg/m2) increased from approximately 20% in 2000 to 27% in 2008. Previous researches indicated that obesity might be significantly associated with depression. Several researches conducted in the United States and Canada have indicated associations between obesity and depressive symptoms, measures of psychological distress, and history of depression. However, none of the studies to date has been conducted based on the 2006 Behavioral Risk Factor Surveillance System Survey (BRFSS). The propounded study participants will be 2006 Behavioral Risk Factor Surveillance System (BRFSS) responders. The 2006 BRFSS was chosen due to a significant set of “healthy days” questions along with some optional modules and state- added questions. Findings from this research can provide instructions to government officers on making social policy decisions to help people in need of mental health services.

Key words: obesity, depression, PHQ, BRFSS

Specific Aims

The growing prevalence of overweight and obesity is a major public health concern. Among the U.S. adult population, the prevalence of obesity (defined as a body mass index ranges from 30.0 kg/m2 to 99.8 kg/m2) increased from approximately 20% in 2000 to 27% in 2008 [1]. Obesity can lead to many kinds of disease problems, such as diabetes, high blood pressure, high cholesterol, asthma, arthritis and poor health status. Usually, the overweight or obesity prevalence is perceived due to the consequence of an energy imbalance, with energy intake exceeding that of energy expenditure. It is estimated that each year, among the U.S. death persons, 280,000 of whom are attributed to obesity or overweight [2]. Obesity-related morbidity is estimated to account for 9.1% of total annual U.S. medical expenditures each year [3].
Previous researches indicated that obesity might be significantly associated with depression [4,5]. Depression is one of the most prevalent mental disorders [6]. The National Institute of Mental Health (NIMH) in 2000 estimated that 9.5% of the U.S. populations suffer from a depressive illness in any given year [7]. The National Survey on Drug Use and Health (NSDUH) estimated that, during the years 2005-2006, 11.29% of total U.S. adults had experienced serious psychological distress in past year [8]. Several researches conducted in the United States and Canada have indicated associations between obesity and depressive symptoms [9], measures of psychological distress [10], and history of depression [11]. It is reported that the relationship between depression and obesity is dependent upon different gender, age, and race/ ethnics. People younger than 65 years old are much more prone to get depressed than their counterparts [9]. Besides, significant positive associations between depression and obesity are observed among women but not men [12,13,14]. However, when it comes to the relationship between depression and obesity dependent upon different races, there are some discrepancies. One report indicated that the Non-Hispanic Whites had a higher rate of depression compared to Non-Hispanic Blacks, Hispanics, and Asians [15]. Another report observed that Hispanics are much more prone to get depressed than Whites and Blacks [9]. #p#分頁標題#e#
The association between obesity and anxiety or substance use disorders is poorly understood. In epidemiology study, anxiety symptoms have indicated moderate positive associations with obesity in community and clinic samples [16]. Alcohol abuse has been associated with a lower risk of obesity [17].
We conduct this research to seek for the answer to the paradoxical results regarding the ethnicity differences and to investigate the relationship between obesity and substance use disorders. Meanwhile, to our knowledge, none of the studies to date has been conducted based on the 2006 Behavioral Risk Factor Surveillance System Survey (BRFSS). In the 2006 BRFSS, depression was measured using the Patient Health Questionnaire (PHQ-8) instead of using just a single question in the 2001 BRFSS. The PHQ can evaluate the severity of depressive symptoms as well as establish provisional diagnoses of major and minor depression. Findings from this research can give instructions to government officers on making social policy decisions to help people in need of mental health services.

Background and Significance

Understanding the causes of obesity is pivotal for improving health balance and reducing obesity-related diseases. Most researchers have investigated that the combination of a redundant nutritional food intake and a sedentary lifestyle are the main reasons for the rapid acceleration of overweight and obesity in Western countries in the last quarter of the 20th century. In spite of the widespread availability of nutritional information in schools, doctors’ offices, on the Internet and on groceries, it is evident that overeating remains a substantial problem. For instance, reliance on energy-dense fast-food meals tripled between 1977 and 1995, and calorie intake quadrupled over the same period. However, dietary intake in itself is inadequate to explain the phenomenal increase in levels of obesity in many of the industrialized world in recent years [18].

Weight regulation is very complex due to series of individual-level influences. In the United States, the prevalence of obesity is higher among middle-aged and older adults than that among younger adults [19]. Besides, obesity is more common among women than men [20]. In developed countries, obesity is inversely related to income and education level [21,22].
Previous researches indicated that obesity might be significantly associated with depression [4,5]. Depression is one of the most prevalent mental disorders [6]. The National Institute of Mental Health (NIMH) in 2000 estimated that 9.5% of the U.S. populations suffer from a depressive illness in any given year [7]. The National Survey on Drug Use and Health (NSDUH) estimated that, during the years 2005-2006, 11.29% of total U.S. adults had experienced serious psychological distress in past year [8]. Several researches conducted in the United States and Canada have indicated associations between obesity and depressive symptoms [9], measures of psychological distress [10], and history of depression [11]. It is reported that the relationship between depression and obesity is dependent upon different gender, age, and race/ ethnics [9]. One previous study examined the association between obesity and depression dependent on sex differences, reporting a significant positive association among women but not men [12,13,14]. Another national study which examined the association between the depressive mood and obesity based on the results of 2001 BRFSS reported a stronger relationship between obesity and depression among those younger than 65 years [9]. However, when it comes to this relationship dependent upon ethnicity differences, there are some discrepancies. One study used data from the National Co-morbidity Survey Replication (NCS-R) to examine the relationship between obesity and a range of mood, anxiety, and substance use disorders in the U.S. adult population. The results indicated that the Non-Hispanic whites had a higher rate of depression compared to Non-Hispanic blacks, Hispanics, and Asians [15]. Another study used data from the 2001 BRFSS observed that Hispanics are much more prone to get depressed than Whites and Blacks [9].#p#分頁標題#e#

The association between obesity and anxiety or substance use disorders is poorly understood. In epidemiology study, anxiety symptoms have indicated moderate positive associations with obesity in community and clinic samples [16]. Alcohol abuse has been associated with a lower risk of obesity [17].

Hypotheses of the Research
1. Young obese women are much more prone to be depressed than non-obese women.
2. Hispanics are much more prone to be depressed than Whites and Blacks, OR
3. Non-Hispanic Whites have a higher rate of depression compared to Non-Hispanic Blacks, Hispanics, and Asians.
4. Smokers are more likely to get depressed than non-smokers.
5. Heavy drinkers have a higher rate of depression than those who are not.

Research Design and Methods
Data from the 2006 Behavioral Risk Factor Surveillance System (BRFSS) will be analyzed for the study. The BRFSS is established by the Centers for Disease Control and Prevention (CDC) in 1984. It is the largest, continuously conducted telephone health survey system in the world. The BRFSS is designed to identify and monitor risk factors for diseases such as diabetes, cancer, obesity, asthma, nutritional related maladies and more. Mental health is chosen as one of these risk factors. This system can provide abundant data information based on individual- level healthy behaviors of U.S. adults each year. During the survey, the participants would be asked to answer several questions that related to healthy behaviors, such as dietary intake (fruits and vegetables), nutrition and physical activity, tobacco and alcohol use, health care access, hypertension, as well as some biological factors including height and weight [23].
Study Subject

The propounded study participants will be 2006 Behavioral Risk Factor Surveillance System (BRFSS) responders. These individuals responded to the 10 questions regarding anxiety and depression. The 2006 BRFSS was chosen due to a significant set of “healthy days” questions along with some optional modules and state- added questions.

The Patient Health Questionnaire (PHQ-8)

The Patient Health Questionnaire (PHQ-8) was conducted in 41 states and territories; it consisted of 8 questions based on the Statistical Manual of Psychiatric Disorders, Fourth Edition (DSM-IV) [24,25]. It was modified from a self-report paper format to a format compatible to telephone interview in order to be used in the 2006 BRFSS. The responders were asked to answer the questions on what they had experienced about emotions or behaviors related to depression during the past two weeks (Appendix 1). The total score for the PHQ-8 ranged from 0 to 14, which also meant the total number of days for a responder that might have experienced emotions or behaviors.
In order to separate varying levels of depression, the CDC developed three algorithms (Appendix 2). The total number of days was converted to a point scale ranging from 0 to 3 (0 = “0 to 1 day”, 1 = “2 to 6 days”, 2 = “7 to 11 days”, and 3 = “12 to 14 days”) [26]. Depending on the algorithm used, the total points were used to determine whether the responders were depressed at the time they were interviewed. #p#分頁標題#e#

In this study, we will use the CDC developed Algorithm 3 (Depression severity score is divided into two groups: less than 10 and greater than 10) due to its validity in detecting depression in the general population and its simplicity in use.

In addition to the eight questions regarding emotions and behaviors, there are two more questions on whether the responder had ever been told by a healthcare provider that he/she had been diagnosed of depression/anxiety.

Demographic Characteristics

Demographic characteristics, including age, sex, race/ ethnicity, height and weight, were obtained from the participants’ self-reports. Obesity was defined by BMI [weight (kg) / height2 (m2)] and was classified into three groups: non-overweight/ obese (BMI<25); overweight (25? BMI<30); obese (BMI ? 30). Previous studies indicated that self-reported height and weight were highly correlated with physical measurements [27,28], but self-reports tended to underestimate weight and overestimate height [28,29], resulting in lower estimates of overweight and obesity.
Age, sex, race/ ethnicity, and health status are regarded as moderators in this study. Age groups are defined as young (age 18-64 years) and old (age 65 + years). Race is defined as White/ Non-Hispanic, African American, Hispanic, and Other, by self-report. Health status is defined as Smoking Status (Not At Risk and At Risk) and Heavy Drinking Status (Not At Risk and At Risk). Besides, we will use the following socio-economic variables: education level is defined as less than High School (H.S.), H.S. or General Educational Development (G.E.D.), Post-H.S., and College Graduates; income level is defined as $15,000 to $24,999, $25,000 to $49,999, and $50,000 +; marital status is defined as Married, Divorced/ Separated, Widowed, Never Married and Unmarried.

Statistical Analysis

We will apply a normalized weight to each participant based on a sample weight variable provided in the BRFSS dataset in all analyses. This can ensure unbiased estimates for the general populations. The descriptive data on depression are treated as prevalence, which are used as a function of age, sex, race, smoking status, and heavy drinking status.
Logistic regression models will be used for comparative and moderator analyses. Meanwhile, we will control for the socio-economic variables. The results of the comparative analyses are presented as OR with the confidence interval parameter set at 95%. Confidence intervals are used to estimate statistical significance in the comparative analyses to determine whether there are any difference between the variables and the prevalence of depression. 95% CI means if repeated same sample size from the same population were collected or the prevalence rates of depression were recalculated, approximately 95% of the newly estimated intervals would contain the true rate, or we can be 95% confidence that the samples would include the true rate.
For the moderator analysis, we will test appropriate intersection terms (i.e., obesity ? age, obesity ? sex, obesity ? race, obesity ? smoking status, and obesity ? heavy drinking status) based on their main effects and socio-economic variables [30].#p#分頁標題#e#

Appendix 1. PHQ-8 Questions on the BRFSS
Now I am going to ask you some questions about your mood. When answering these questions, please think about how many days each of the following has occurred in the past 2 weeks.
1. Over the last 2 weeks, how many days have you had little interest or pleasure in doing things?
2. Over the last 2 weeks, how many days have you felt down, depressed or hopeless?
3. Over the last 2 weeks, how many days have you had trouble falling asleep or staying asleep or sleeping too much?
4. Over the last 2 weeks, how many days have you felt tired or had little energy?
5. Over the last 2 weeks, how many days have you had a poor appetite or ate too much?
6. Over the last 2 weeks, how many days have you felt bad about yourself or that you were a failure or had let yourself or your family down?
7. Over the last 2 weeks, how many days have you had trouble concentrating on things, such as reading the newspaper or watching?
8. Over the last 2 weeks, how many days have you moved or spoken so slowly that other people could have noticed? Or the opposite- being so fidgety or restless that you were moving around a lot more than usual?
9. Has a doctor or other healthcare provider EVER told you that you had an anxiety disorder (including acute stress disorder, anxiety, generalized anxiety disorder, obsessive-compulsive disorder, panic disorder, phobia, posttraumatic stress disorder, or social anxiety disorder)?
10. Has a doctor or other healthcare provider EVER told you that you have a depressive disorder (including depression, major depression, dysthymia, or minor depression)?

Appendix 2. CDC Developed Algorithms for the PHQ-8
Algorithm 1*:
Score: Depression Severity:
0 to 4.9 No depression
5 to 9.9 Mild depression
10 to 14.9 Moderate depression
15 to 19.9 Moderately severe depression
20 Severe depression
*Scores that are greater than ten are classified as “current depression”.

Algorithm 2:
Score Depression Severity:
0 to 1 No depression
• Negative response (< 7 days) to Questions 1 & 2 OR
• Positive response (7+ days) to < 2 questions
2 to 4 Minor Depression
• Positive response (7+ days) to Questions 1 or 2 AND
• Positive response (7+ days) to between 2-4 questions
5 to 8 Major Depression
• Positive response (7+ days) to Questions 1 or 2 AND
• Positive response (7+ days) to a total of 5+ questions

Algorithm 3:
Score: Depression Severity:
0 to 9 Depression Severity Score Less than 10
10 or greater Depression Severity Score Greater than 10

References
1. BRFSS Prevalence Data Comparison: 2000 v.s. 2008 http://apps.nccd.cdc.gov/brfss/display_c.asp?yr_c=2008&yr=2000&cat=OB&state=US&bkey=20000020&qkey=4409&qtype=C&grp=0&