The relationship between general mental health and number of health conditions in middle-aged females in the general U.S. population

Purpose: Research has shown that poor mental health is related to chronic health conditions; however, this relationship has not been assessed specifically in middle- aged women. Therefore, the purpose of this study was to determine whether general mental health differs by number of health conditions in middle-aged females in the general population. Methods: This cross-sectional analysis used 2017 data from the Behavioral Risk Factor Surveillance System (BRFSS) for middle-aged females ages 45 to 64 from Maine (N=2112), Massachusetts (N=1285), Minnesota (N=3131), Mississippi (N=1036), and New Hampshire (N=1262). Ordered logistic regression analyses were conducted by state to analyze the relationship between mental health (low, moderate, high) and number of health conditions while controlling for health-related, demographic, and socioeconomic factors. Results: Across states, the results indicated that about one-third of females ages 45 to 64 years reported low to moderate general mental health, and over two-thirds reported one or more health conditions. The results of adjusted analysis indicated that mental health was inversely related to health conditions with the relationship strengthening as the number of health conditions increased. Conclusion: Across states, mental health issues and multiple health conditions were prevalent and moderately to highly related in middle-aged women. Thus, practitioners should automatically screen for both in this target population and treat concurrently.


Introduction
Poor mental health, including depression, has become increasingly widespread as depression alone affects one in ten individuals in the general population and over 300 million people may currently have it [1][2][3]. Poor mental health is associated with an increased risk of suicide, morbidity, and mortality as well as with increased direct and indirect medical costs [4,5]. In addition, mental health issues can be exacerbated by substance use [2,3,6] and can differ by socioeconomic factors including income level, education, and occupational status and by demographic factors such as age, gender, country of origin, and marital status [7,8].
Moreover, mental health may be related to having chronic health conditions. For example, poorer mental health has been linked to specific health conditions such as diabetes, asthma, hypertension, arthritis, ulcers, heart disease, back/neck problems, chronic headaches, and obesity as well as to having multiple conditions and increased severity of pre-existing health conditions [4][5][6][7][8][9][10][11]. However, few studies focus on the relationship between mental health issues and health conditions specifically for middle-aged females in the general population who may be at higher risk for both, or on general mental health versus diagnosed mental illnesses [1,2]. Therefore, the purpose of this study is to assess the relationship between general mental health and number of health conditions in middle-aged females in the U.S. general population.

Design
This study was a cross-sectional analysis of data from the 2017 Behavioral Risk Factor Surveillance System (BRFSS), which is conducted by the Center of Disease for Control and Prevention (CDC) [12]. The purpose of BRFSS is to collect uniform data from all 50 states and several U.S. territories on preventative health practices and risk behaviors. The data was collected through telephone surveys with noninstitutionalized, civilian adults ages 18 and older with a landline or cell phone using Random Digit Dialing (RDD) techniques. The CDC compiles this data and makes deidentified data available to researchers to conduct secondary data analysis. This study was given exempt status by the Institutional Review Board of The University of North Texas Health Science Center.

Sample
The samples for this analysis included females ages 45 to 64 years old in Maine (N=2112), Massachusetts (N=1285), Minnesota (N=3131), Mississippi (N=1036) and New Hampshire (N=1262) with data for mental health and health conditions. These states were chosen from the BRFSS 2016 prevalence survey data maps because of their higher proportions of middle-aged females and varied mental health [13].

Data
The outcome, mental health, was originally measured as low, moderate, or high number of days in the past 30 days in which mental health was "not good," including "stress, depression and problems with emotions." We reversed this variable to represent number of "good days" as "low" (0-16 days), "moderate" (17-29 days), and "high" (30 days). The factor of interest was calculated as the number of "yes" responses to ever being diagnosed with any of the following: arthritis; asthma; cancer, other; cancer, skin; coronary heart disease (CHD); chronic obstructive pulmonary disorder (COPD); diabetes; heart attack; high blood pressure; high cholesterol; kidney disease; or stroke. We then categorized this value as "none," "one," "two," "three," or four or more" health conditions.
The control variables included alcohol use, tobacco use, age, ethnicity/race, marital status, education level, employment status, and income level, all of which are related to adult mental health in the literature [1-3,6-8]. Alcohol use categories were "use in last 30 days" vs. "no use in last 30 days." Tobacco use categories were "smoker" vs. "non-smoker." We used the BRFSS age categories of "45-54" and "55-64" years old. For ethnicity/race, we used "white, non-Hispanic" vs. "other." Marital status was categorized as "married" vs. "other." Education level was measured as "graduated college/technical school" vs. "did not graduate college/technical school." Employment status was measured as "employed" vs. "not employed." Income level was measured as "$50,000 or more" vs. "less than $50,000."

Analysis
We analyzed state data separately in order to assess patterns in variable relations across similar samples. As such, similar results in 3 or more of the 5 states were considered consistent evidence for relations. Ordered logistic regression by state was used to determine the relationship between mental health and number of health conditions after controlling for health-related, demographic, and socioeconomic factors. In ordered logistic regression, the proportional odds produced for each factor is interpreted as follows: for a one unit increase in the factor (i.e., comparing the designated group to the referent group), the resulting AOR applies to the odds of reporting the highest group of the outcome versus the lower groups of the outcome, as well as to the odds of reporting the high and middle groups of the outcome versus the low group of the outcome, after controlling for all other variables in the model. Any observations with missing data for any variables were excluded from adjusted analysis. All analyses were conducted in STATA 15 (©1985-2017 StataCorp LLC).

Adjusted statistics
As shown in table 2, the results of ordered logistic regression analysis for middle-aged females in Maine, Massachusetts, Minnesota, Mississippi, and New Hampshire indicated that after controlling for all other variables in the models, mental health differed by number of health conditions. Across states, compared to those with no health conditions, those who with one health condition were about 1.5 to 1.6 times less likely to report each successive level of mental health; those with two health conditions were about 1.5 to 2.2 times less likely to report each successive level of mental health; those with three health conditions were about 1.8 to 2.6 times less likely to report each successive level of mental health; and those with four or more health conditions were about 3 to 4 times less likely to report each successive level of mental health.

Discussion
The purpose of this study was to assess the relation between general mental health and number of health conditions in middle-aged women in the general population. The results indicated that about one-third of females ages 45 to 64 years reported low or moderate general mental health, and over two-thirds reported one or more health conditions, the most prevalent diagnoses being high blood pressure, high cholesterol, and arthritis, followed by diabetes, asthma, and cancer (other than skin). The results of this study showed an inverse relationship between reported mental health and health conditions and the relationship strengthened as the number of health conditions increased: middleaged females with one condition were about 1.5 times less likely, those with two conditions were about 1.5 to 2 times less likely, those with three conditions were about 2 to 2.5 times less likely, and those with four or more diagnoses were about 3 to 4 times less likely, to report each successive level of mental health. Our findings were consistent with previous studies that indicated depression and anxiety are related to having chronic health conditions in various target populations of adults [4][5][6][7][8][9][10][11]. To our knowledge, this is the first study to focus on the relation between general mental health and health conditions solely in this target population who may be at higher risk for both mental health issues and health conditions [1,2].
However, this study is not without limitations. Although we used a current general mental health measure rather than a diagnosis of the past and we had large number of health conditions to utilize, there was no information on the severity or management strategies for mental health issues or health conditions, including medication use, which could impact relations. Future research should include the assessment of management strategies in the relationship between mental health and chronic health conditions.

Conclusions
Because this was a population-based study, the results may be generalizable to middle-aged females in a primary care setting. Providers may expect about one-third of middle-aged women to report mental health issues and about two-thirds to report one or more health conditions. Because both may be prevalent and moderately-to highlyrelated in this target population, providers should automatically screen for mental health issues and health conditions in women ages 45 to 64 regardless of whether symptoms present. Treatments and side effects of treatments for comorbid conditions should be assessed concurrently.  Table 2. Adjusted results across states AOR=adjusted odds ratio; 95% CI=95% confidence intervals; ref=referent group. The model controlled for alcohol use, tobacco use, age, ethnicity/race, marital status, education level, employment status, and income level. For inverse relations, take the inverse of significant odds ratios to obtain effect size (1/AOR), which then reads as "participants in the designated group are 1/AOR times LESS likely than participants in the referent group to report each successive level of mental health status."