The association of depression and anxiety with glycemic control among Mexican Americans with diabetes living near the U.S.-Mexico border

Diabetes is a leading public health concern of particular relevance to Mexican Americans. Between 2003–2006, the estimated age-adjusted prevalence of diabetes among Mexican Americans adults ≥ 20 years of age was 16.3%, compared with 9% among non-Hispanic Whites [1]. Moreover, the prevalence of diabetes was recently estimated to be over 30% among Mexican American adults residing near the U.S.-Mexico border [2]. Overweight/obesity prevalence is also alarmingly high among Mexican Americans at 81.2%, compared with 66.7% among non-Hispanic Whites [3]. Numerous studies have shown that obesity and weight gain are linked with the onset of diabetes [4‐6]. Unfortunately, Mexican Americans with diabetes, particularly those born in the U.S., are at greater risk of all-cause mortality than non-Hispanic Whites [7]. In order to address diabetes-related health disparities, focused research will be needed to better understand the physiological, psychosocial, and behavioral factors that contribute to the onset and course of diabetes among Mexican Americans.

Research indicates that the relationship between depression and diabetes is bi-directional [8‐13]. Depression is associated with the development of diabetes [8, 12]. Conversely, diabetes is associated with the development of depression [8‐10]. Given that diabetes increases risk for depression, it is not surprising that higher rates of depression are found among diabetic patients in cross-sectional studies. Further, Mexican Americans with diabetes are more likely than other racial/ethnic groups to report symptoms of depression [14]. In fact, rates of clinically significant depressive symptoms have been reported to be as high as 40% among individuals of Mexican origin with diabetes who live on either side of the South Texas border [15]. The presence of depression in diabetes is noteworthy because depression may influence glycemic control among individuals with diabetes through physiological and behavioral pathways (for reviews, see [16‐19]).

Less is known about how anxiety might influence glycemic control. Initial research suggests that adults with diabetes may have elevated levels of anxiety and greater anxiety disorder prevalence relative to the general population [20‐23]. Higher levels of anxiety have also been reported among Mexicans with diabetes [24]. However, initial findings related to the impact of anxiety on glycemic control among individuals with diabetes are equivocal [20, 25, 26]. In a meta-analysis, Anderson et al. [25] reported that the relationship between anxiety and glycemic control only approached significance, though anxiety measured via diagnostic interviews was significantly associated with hyperglycemia. Since then, Gois et al. [26] reported that anxiety (measured via self-report questionnaire) was not associated with glycemic control (i.e., HbA1c < 8 or ≥ 8). Conversely, Collins et al. [20] reported that having a high level of anxiety (measured via self-report questionnaire) was associated with poor perceived glycemic control and a greater number of diabetes complications. As such, research is needed to clarify the relations between anxiety and glycemic control in general and specifically among Latinos/Hispanics with diabetes.

Depression is associated with hyperglycemia and elevated HbA1c levels among individuals with diabetes [27‐31]. Behavioral factors that contribute to poor glycemic control among individuals with diabetes may include poor adherence to recommended health behaviors including diet, weight control, physical activity, glucose monitoring, and medication regimens [17, 29, 32‐34]. McKellar et al. [35] reported that depression had a negative impact on glucose regulation through self-care behaviors including poor adherence to diabetes dietary guidelines and poor medication adherence. Similarly, Chiu et al. [31] reported that depressive symptoms negatively influenced glycated hemoglobin (HbA1c) through diabetes-related behavioral factors including inadequate physical activity and greater body mass index (BMI). Recently, the link between depression and elevated HbA1c has been demonstrated specifically among Latinos/Hispanics with diabetes [36]. Plausibly, the same behavioral factors may be influenced by anxiety among individuals with diabetes. Additional research is needed to characterize the relations of depression and anxiety with the behavioral management of diabetes and glycemic control among Latinos/Hispanics.

The primary purpose of the current study was to evaluate the cross-sectional associations of depression and anxiety with modifiable factors known to influence glycemic control (i.e., BMI, waist circumference, and physical activity) and more direct measures of glycemic control (i.e., HbAlc and fasting glucose) among Mexican Americans with diabetes residing near the Texas-Mexico border. Potential moderators of the associations of depression and anxiety with diabetes-related factors were also evaluated including age, gender, education, preferred language, and birth country.

Overall, higher scores on measures of depression and anxiety among Mexican Americans with diabetes were associated with greater BMI and waist circumference, and engaging in less physical activity. In addition, greater depression was associated with higher fasting glucose, while greater anxiety was associated with higher HbA1c. Thus, the findings suggest that depression and anxiety have a negative influence on the behavioral management of diabetes and glycemic control among Mexican Americans residing near the U.S.- Mexico border.

Depressed/anxious participants differed from those who were not depressed or anxious in a variety of ways. Similar to nationally representative samples [50, 51], depressed/anxious Mexican Americans with diabetes were of greater age, more likely to be female, and less educated than their non-depressed/anxious counterparts; they also had greater BMI and waist circumference, and engaged in less physical activity. In contrast with other research focused on anxiety disorder prevalence (e.g., [52]), anxious participants were more likely than non-anxious participants to have been born in Mexico than the U.S., and preferred to complete study assessments in Spanish. Perhaps this finding may be understood in the context of research indicating that acculturative stress is associated with greater anxiety and depression among Mexican Americans [53]. In addition, Breslau et al. [54] found that immigration from Mexico predicted the subsequent onset of anxiety. Plausibly, the higher levels of anxiety among Mexican-born participants may have been related to acculturative stress, though more research will be needed to understand potential links between nativity, acculturative stress, and anxiety.

Notably, age, years of education, assessment language, and birth country each functioned as moderators of the relationship between depression/anxiety with modifiable factors and glycemic control. Specifically, anxiety was inversely associated with hours of physical activity among younger participants, but not among participants ≥ 52 years of age. Depression was positively associated with fasting glucose among younger participants, while there was no relation among older participants. Among younger participants with diabetes, only 43% (vs. 71.2% among older participants) had been told by a doctor that they had diabetes and only 36.6% (vs. 63.8% among older participants) were taking diabetes medication. Plausibly, depression may have had a greater impact on fasting glucose among those who were undiagnosed and therefore not taking any medications to control diabetes. Conversely, the fasting glucose of those taking medications to control blood sugar may be less affected by depression.

Interestingly, greater depression and anxiety were associated with higher HbA1c among participants born in the U.S., those who were more educated, and those who completed the study assessments in English. Conversely, depression/anxiety was not associated with HbA1c among those born in Mexico, those who were less educated, and those who completed the study assessments in Spanish. Taken together, one interpretation of these findings might be that glycemic control was more influenced by depression among acculturated participants than those who were less acculturated. In the current sample, those born in the U.S. had greater education than those born in Mexico (11.84 vs. 7.71 years) and similarly, those who completed study assessments in English had greater education than those who completed assessments in Spanish (13.87 vs. 7.76 years on average). Perhaps more acculturated and educated participants had a better understanding of how to influence HbA1c through behavior, as well as greater access to diabetes-related health care (through greater education) than their less acculturated and less educated counterparts. Vega et al. [55] reported that acculturation was positively associated with cardiovascular disease-related health knowledge among Mexican Americans, and that acculturation was strongly associated with education. Thus, depression may have a greater negative impact on glycemic control among individuals who are knowledgeable about diabetes care and have been actively managing their diabetes through behavior. Alternatively, it is possible that depression may be associated with dietary responses that differ by acculturation or education. More research will be needed to understand the links between nativity, language preference, education, and glycemic control.

Given that diabetes is associated with depression and anxiety, it seems plausible that the assessment and treatment of depression and anxiety might improve self-management of diabetes and glycemic control. However, in a recent review it was concluded that while psychosocial interventions may be effective for the treatment of depression among individuals with diabetes, it is not clear that such interventions have a beneficial impact on diabetes-related outcomes [56]. Similarly, the findings of initial research suggest that pharmacological treatments for depression improve mood, but have little impact on glycemic control (for a review, see [34]). Additional studies are needed to evaluate the impact of more comprehensive interventions designed to target both depression and diabetes management, and to identify other factors that are related to both mood and glycemic control. For example, recent research suggests that depression is associated with increased risk for the onset of dementia in individuals with diabetes [57]. Finally, treatments designed to increase physical activity may be particularly beneficial for the dual purpose of reducing depression and improving glycemic control [58].

The current study has several strengths and limitations. A notable strength of the study was the sample comprised entirely of Mexican Americans with diabetes who were recruited from within the community of Brownsville, TX (on the U.S.-Mexico border), thereby avoiding the bias inherent in clinic-based studies. Latinos/Hispanics are a vulnerable and understudied group as it relates to diabetes, and even less attention has been paid to the health of those living along the border. A limitation of the study was the cross-sectional design. As such, we were not able to determine whether there might be a causal relation between depression, modifiable behavioral factors, and glycemic control among Mexican Americans with diabetes. Similarly, we could not determine whether depression precedes or develops subsequent to poor glycemic control. Plausibly, modifiable factors, such as BMI and waist circumference, may mediate the relations between depression and glycemic control among Mexican Americans as in other samples (e.g., [59]). However, given the cross-sectional nature of the data these mediational relationships were not evaluated. Finally, it is notable that the physical activity measure (from which METS were calculated) changed during the course of the study (i.e., IPAQ to Godin) and some participants completed the physical activity assessment after a delay (i.e., assessment not completed concurrently with other measures) which may have impacted the findings. However, please note that depression and anxiety remained significantly related to physical activity (METs) even after controlling for the specific physical activity measure (Godin vs. IPAQ vs. delayed assessment).

Greater depression and anxiety were associated with poorer behavioral management of diabetes management and poorer glycemic control among Mexican Americans living near the U.S.-Mexico border. Findings highlight the need for comprehensive interventions designed to target depression and anxiety in the context of diabetes management among border residents. Understanding the psychosocial and behavioral factors associated with diabetes management among Latinos/Hispanics is a first step towards reducing and eliminating diabetes-related health disparities.