In our study, U.S. adults reported an average of about 3 days during the past 30 days when they felt "sad, blue or depressed." Our results are consistent with previous studies documenting the increased prevalence of depressive symptoms among the following groups: women [27–31], in certain minority racial and ethnic groups , people with lower levels of education and income [32, 33], people of lower employment status [27, 34, 35], people formerly married or living together but not married [27, 36], and in those with limited or no access to health care [32, 37, 38]. The gap in the number of SBDD between men and women was less pronounced as socioeconomic status improved. Respondents who reported a higher number of SBDD also reported engaging in unhealthy behaviors such as cigarette smoking, binge drinking, and physical inactivity. Underweight and obese adults also reported higher numbers of SBDD than did normal or overweight adults. These findings extend previous public health studies that have documented an association between self-reported mental distress and behaviors risky to health [39–42].
Given the cross-sectional design of the BRFSS, we were unable to determine whether risky behaviors preceded or followed SBDD. Nonetheless, our findings provide additional evidence for the association of considerable public health importance between negative mood and unhealthful behaviors . For example, in the prospective Stirling County Study (1952–1992), subjects who became depressed were more likely to initiate smoking, continue smoking, and refrain from quitting smoking than those who had never become depressed .
Negative mood adversely influences self-efficacy to adopt and maintain healthful behaviors and may thwart other self-motivating processes (e.g., attitudes, outcome expectations, and goals) associated with engaging in healthful behaviors . Perceived inefficacy can foster additional despondency. This finding has implications for public health interventions. For example, psychosocial interventions that elicit positive emotions, instill confidence in adopting health-promoting behavior, and improve people's coping skills might be more effective for individuals with despondent mood than interventions designed to arouse fear regarding the consequences of engaging in risky behaviors–which can foster inefficacy and increased despondency .
Our findings support the construct validity of the SBDD measure in this study because SBDD were associated with other physical and mental HRQOL domains in expected ways. Groups with progressively higher numbers of physically unhealthy days, activity limitation days, and pain days reported a higher number of SBDD. Moreover, these associations were more pronounced with mentally unhealthy days and anxiety days, than with physically unhealthy days, activity limitation days, and pain days. We found an exception to the linear relationship between SBDD and HRQOL measures with our measure for sleeplessness. Adults who reported 1–2 days of sleeplessness reported fewer SBDD than those who reported no days of sleeplessness. Sleep disturbance, both insomnia and hypersomnia are symptoms of depression. Those reporting no days of sleeplessness, but more SBDD, might be those with hypersomnia. Additional studies are warranted to examine this hypothesis.
Besides the cross-sectional design, this study has other limitations. Only 38 states and the District of Columbia included the HRQOL supplemental module that assessed SBDD. All states and the District of Columbia, however examined mentally unhealthy days–the number of days respondents experienced poor mental health due to stress, depression or problems with emotions. Mean mentally unhealthy days in the states that assessed SBDD with the HRQOL supplemental module did not differ significantly from that in states that did not. Given the positive correlation between mentally unhealthy days and SBDD (r = 0.6), states that did not assess SBDD would most likely report similar SBDD as states that did include this measure, suggesting similar study results had all states assessed SBDD. Second, BRFSS excludes people who do not have telephones, live in institutions, and persons younger than 18 years. Third, BRFSS may under represent the severely impaired because functional capacity is required to participate in BRFSS. Including this group however, would probably only strengthen the associations we found because the variability of SBDD would increase because the severely impaired would be more likely to report more SBDD. Finally, because our findings on SBDD are based on respondents' self-reports rather than on professionally administered psychiatric evaluations, people who experience SBDD may differ from people with clinical depression.