Emotional reactivity to daily stress,spousal emotional support,and fasting blood glucose among patients with type 2 diabetes

Abstract

Stress is associated with higher blood glucose in patients with diabetes, but the strength of this association varies considerably across patients. The current daily diary study of 129 patients with type 2 diabetes examined whether individual differences in emotional stress reactivity were associated with fasting blood glucose and whether emotional support provided by spouses moderated this association. Greater stress reactivity was related to greater variability in patients’ fasting glucose readings and, among patients with less support, to higher fasting glucose levels. Investigating individual differences in emotional stress reactivity may help to clarify the role of stress in blood glucose control.

Keywords

blood glucose emotional support stress reactivity type 2 diabetes

Managing type 2 diabetes is challenging, as optimal glucose control often requires patients to make substantial changes to their lifestyles and to adhere to a treatment regimen that usually involves careful monitoring of diet, exercise, medication, and blood glucose levels (American Diabetes Association, 2014). Adherence to the dietary component of the treatment regimen can be especially difficult (Glasgow et al., 2001) because strict adherence may involve giving up preferred foods or having to forgo sharing foods with others in social settings (and possibly having to provide explanations for not sharing; Gallant et al., 2007; Lin et al., 2008). Not surprisingly, patients often report that they find the management of their type 2 diabetes to be burdensome (Wagner, 2007). The challenges of adhering to the treatment regimen may be compounded, moreover, by daily stress that arises from other life demands (e.g. financial difficulties, household problems, interpersonal conflicts).

Stress reactivity and blood glucose

Stress can adversely affect blood glucose in patients with diabetes (see review by Morris et al., 2011), both by triggering the release of hormones, such as epinephrine and cortisol, that cause glucose to be secreted into the bloodstream (Cox and Gonder-Frederick, 1992; Surwit and Schneider, 1993) and by depleting self-regulatory resources needed for effective treatment adherence (Hagger et al., 2009). For example, stress may arouse negative affect that leads people to consume high-carbohydrate “comfort foods” in an effort to reduce their distress (Tomiyama et al., 2011), even though such foods contribute to elevated blood glucose and yield short-lived mood benefits (Gibson, 2012). Stress also may disrupt the self-monitoring and inhibitory control needed to maintain adherent behavior (Heatherton and Wagner, 2011).

The role of stress in blood glucose control, however, may depend on how individuals respond emotionally to stress. Although most people experience psychological distress, or negative affect, in response to stress (Mroczek and Almeida, 2004), they vary in the degree of distress that is aroused (Bolger and Zuckerman, 1995; Nesselroade, 1988). Similarly, not all individuals with diabetes manifest a relationship between stress and blood glucose levels (Kramer et al., 2000). Riazi et al. (2004) found considerable individual variability in the strength and direction of the association between blood glucose and daily stress among patients with type 1 diabetes. Individual differences in emotional stress reactivity, or the extent to which people experience negative affect in response to stress, may partly account for this variability (Semmer et al., 2004), but little research has examined this possibility. The current study accordingly investigated the association between individual differences in stress reactivity and fasting blood glucose among individuals with type 2 diabetes. We examined fasting blood glucose because it is a key therapeutic target in the management of diabetes (Woerle et al., 2007).

Potential moderators of the association between stress reactivity and blood glucose

It is also important to identify factors that may moderate the association between stress reactivity and blood glucose control (Skaff et al., 2009). Of particular interest are factors that might help to mitigate the potential adverse effects of stress reactivity. A key factor that has received considerable attention in the literature is social support. Support from family members, particularly spouses, has been linked to better chronic disease management and diabetes-related outcomes (Beverly et al., 2008; Miller and Brown, 2005). Evidence suggests, moreover, that social support may help to reduce the adverse effects of stress on blood glucose control in adults with diabetes. For example, in one study of individuals with diabetes (Griffith et al., 1990), a high level of life stress was associated with poor blood glucose control only among those with low perceived social support. Social support can help to reduce feelings of distress aroused by life stress and can also serve as an external resource that bolsters patients’ self-regulatory resources (Beverly et al., 2008) when they are depleted by stress.

We accordingly investigated whether emotional support provided by spouses reduced the association between emotional stress reactivity and fasting blood glucose among individuals with type 2 diabetes. We used a daily diary design to be able to compute within-person estimates of emotional stress reactivity. We predicted that patients with type 2 diabetes who were more emotionally reactive to daily stress (reflected in a stronger association between daily stress and daily negative affect) would exhibit worse fasting blood glucose control (reflected in higher mean fasting blood glucose and greater fasting blood glucose variability) over a 24-day period. We further predicted that the association between stress reactivity and fasting blood glucose would be reduced among patients who received greater emotional support from their spouse.

Methods

Participants

Data for this study were derived from a sample of 129 older couples in which one spouse was diagnosed with type 2 diabetes. Eligibility criteria included the following: (1) reporting a primary medical diagnosis of type 2 diabetes, (2) being ⩾55 years of age, (3) being married or in a marital-like relationship, and (4) having recently received a recommendation from a health care provider to improve their adherence to dietary guidelines for type 2 diabetes management. Additionally, patients had to be cognitively functional and had to be free of any major hearing, speech, or language problems that would have prevented the comprehension and completion of interviews in English. Of 235 couples screened for eligibility, 58 couples were found to be ineligible and 48 declined to participate, yielding a final sample size of 129 couples (72.9% response rate among eligible couples).

Procedure

Study procedures were reviewed and approved by a university’s Institutional Review Board. Patients were recruited from medical offices, diabetes education centers, and senior citizen organizations, as well as through newspaper advertisements and radio announcements. Recruitment brochures described the study, eligibility criteria, and financial incentives. Patients who expressed interest in participation provided their contact information to project staff, who later telephoned them to provide additional details about the study and to determine eligibility. Following verbal assent, consent forms were sent to the patient, and assessments were scheduled.

Data for this study came from the baseline assessment, in which each patient completed an in-person interview, self-administered questionnaire, and 24 end-of-day electronic diaries on a laptop computer provided for his or her use in their home. Each daily diary could be accessed during a 4-hour period during the evening (8:00–11:59 pm), using a participant-specific password; each diary was time- and date-stamped on completion (see Stephens et al., 2013 for additional information about the study design and procedures).

Measures

All key study variables were derived from daily diary data, with each measure assessed daily for 24 days. All covariates, except for mean daily dietary adherence, were assessed in the in-person interview or self-administered questionnaire.

Daily stress

To assess perceived daily stress, patients were asked each day, “How trying or stressful was your day?” Patients rated on a 3-point scale (1 = not at all, 2 = somewhat, 3 = very much) how stressful their day had been (cf. Armeli et al., 2005).

Negative affect

Daily negative affect was assessed each day with an abbreviated version of the negative affect subscale of the Positive and Negative Affect Schedule (Watson et al., 1988). Patients rated on a 5-point scale (1 = very slightly or not at all, 5 = extremely) the extent to which they experienced six negative emotions (distressed, upset, scared, nervous, afraid, and guilty) that day. Items were averaged to create a measure of negative affect experienced during the day (Cronbach’s α = 0.94).

Emotional reactivity to daily stress

Emotional reactivity to daily stress was operationalized as the extent to which each patient’s daily stress was associated with daily negative affect over the 24 days of diary assessments. The measures of daily stress and negative affect described above were used to create the individual difference estimate of emotional stress reactivity. Specifically, we used hierarchical linear modeling (HLM 6.08; Raudenbush and Byrk, 1992) to compute a stress reactivity variable for each participant. Day-to-day (level 1) variability in negative affect was first examined as a function of daily stress. Person-specific (level 2) slope coefficients were then derived from the analysis and were used to represent each participant’s average emotional reactivity to daily stress across the daily diary assessment (Sliwinski et al., 2009). This approach is used when emotional reactivity is treated as an independent variable (i.e. an individual difference that might predict blood glucose control) rather than as a dependent variable (e.g. Bolger and Zuckerman, 1995).

Spousal emotional support

Four items asked patients to report the extent to which their spouse had helped them with their concerns about their ability to adhere to their physician-recommended diet. We emphasized support for patients’ diet-related concerns because dietary adherence is a key aspect of the treatment regimen for type 2 diabetes (American Diabetes Association, 2014) and one that patients find to be especially challenging (Beverly et al., 2008; Glasgow et al., 2001). For each item, patients rated on a 3-point scale (1 = not at all, 2 = somewhat, 3 = very much) how much support they received from their spouse. Sample items included “Today your husband [wife] … listened to your concerns about managing your diabetic diet” and “Today your husband [wife] … tried to comfort you when you were worried about making poor food choices.” Items were averaged for each day, and the daily values were then averaged to create a composite measure of spousal emotional support provided over the 24-day daily diary assessment (Cronbach’s α = 0.95).

Fasting blood glucose

To assess daily fasting blood glucose, patients were asked to record each of their blood glucose readings every day, in accordance with their physician’s instructions, up to a maximum of seven possible readings. We examined the first blood glucose reading each day because patients with type 2 diabetes are advised to take their initial blood glucose reading before eating. Values were averaged to create a composite measure of fasting blood glucose level across the 24-day daily diary assessment. In addition, the standard deviation (SD) of participants’ daily fasting blood glucose levels was computed to create a composite measure of fasting blood glucose variability across the 24-day daily diary assessment. Fasting blood glucose variability has gained interest among researchers as a potential predictor of poor clinical outcomes among individuals with type 2 diabetes (e.g. an increased risk of vascular events; Hirakawa et al., 2014), and the SD is an accepted index of such variability (DeVries, 2013; Siegelaar et al., 2010). Five patients who did not record any blood glucose readings were excluded from the analyses.

Covariates

The data analyses included controls for patient’s sex (male = 0, female = 1), duration of time diagnosed with type 2 diabetes (in years), the number of diagnosed co-morbid chronic health conditions (from a list of 19 conditions; for example, heart disease, arthritis), depressive symptoms, body mass index (BMI), and mean daily dietary adherence. Depressive symptoms were assessed with the 20-item Center for Epidemiologic Studies Depression Scale (Radloff, 1977). Items were rated on 4-point scale that reflected how often in the past week the participant had felt the way described (0 = rarely or none of the time (less than 1 day); 3 = most of the time (5–7 days)). Sample items included “I felt depressed” and “I felt that everything I did was an effort.” The items were summed to create a composite measure of depressive symptoms (Cronbach’s α = 0.91). Daily dietary adherence was assessed with five items adapted from the diet subscale of the Summary of Diabetes Self-Care Activities Measure (Toobert et al., 2000). For example, patients were asked to indicate the extent to which, during the course of the day, they had eaten five or more servings of fruits, had avoided high-fat foods such as red meat or full-fat dairy products during the day and had spaced carbohydrates evenly throughout the day (1 = not at all, 2 = somewhat, 3 = very much). Items were averaged to create a composite measure of daily dietary adherence across the 24-day diary assessment (Cronbach’s α = 0.80).

Data analyses

We conducted multiple linear regression analyses to examine the main and interactive effects of emotional stress reactivity and spousal emotional support in predicting both fasting blood glucose level and fasting blood glucose variability, controlling for patient sex, number of years diagnosed with diabetes, number of co-morbid health conditions, BMI, and mean daily dietary adherence. In addition, the analysis of blood glucose variability also controlled for patients’ average blood glucose level because these two facets of blood glucose control are often correlated (r = 0.56, DeVries, 2013), as was the case in this study (r = 0.53), leading some researchers to recommend adjusting for average blood glucose level when blood glucose variability is examined (e.g. DeVries, 2013). The stress reactivity and spousal emotional support variables were centered before computing the interaction term (Aiken and West, 1991). Variables were entered in the following stepwise order: covariates (step 1), stress reactivity and spousal emotional support (step 2), and the interaction between stress reactivity and spousal emotional support (step 3). The nature of the hypothesized interaction effect, if significant, was examined by plotting the predicted values of the dependent variables (fasting blood glucose level, fasting blood glucose variability) for low and high values of stress reactivity (−1/+1 SD of the mean) at low and high values of spousal support (−1/+1 SD of the mean). In addition, the significance of simple slopes at each level of spousal support was examined.

Results

Sample characteristics

On average, patients were 66.05 years old (SD = 7.71). Approximately half of the patients were female (n = 65, 53.5%). Most patients were non-Hispanic White (74.4%), 24 percent were African American, and the remaining 1.6 percent were Asian American or American Indian/Alaskan Native. All but two participants (1.5% of the sample) had been married at least 5 years, with an average marital duration of 37.90 years (SD = 13.81).

Patients reported that they had been diagnosed with type 2 diabetes for an average of 11.80 years (SD = 9.41), with 86.76 percent having been diagnosed for 2 years or longer. Patients reported an average of 1.21 (SD = 1.35) additional chronic health conditions. Their average BMI was 31.16 (SD = 7.49), which is considered obese (Ogden et al., 2012). Self-reported glycated hemoglobin (HbA1c) scores were available for 91 participants, and physician-reported scores were available for 66 patients. Mean values for these scores were 7.10 (SD = 1.25) and 7.06 (SD = 1.11), respectively.

Most patients were taking medication to control their diabetes, with 79.1 percent taking oral medication, 36.4 percent taking insulin, and 25.6 percent taking both oral medication and insulin. Additionally, two-thirds (66.7%) of the participants reported that they had experienced some medical complications, with heart disease, nerve damage, and foot problems reported as the most common complications. Substance use (specifically, alcohol and cigarette consumption) was low in this sample. In the baseline interview, 40.6 percent of the patients reported no alcohol intake in the past 7 days and, among those did report alcohol intake, 84.3 percent reported having consumed one drink per day and only 3.9 percent of reported having consumed more than two drinks per day. Only 11.6 percent of patients reported having smoked in the past 7 days.

Descriptive analyses

Table 1 presents the means, SDs, and intercorrelations for the key study variables. As shown in the table, mean fasting blood glucose levels were not related to mean daily stress (r = 0.15, p = 0.096) and mean negative affect (r = 0.15, p = 0.096), but they were significantly related to greater emotional reactivity to stress (r = 0.22, p = 0.014). Blood glucose variability was related to mean daily stress (r = 0.21, p = 0.020), mean negative affect (r = 0.25, p = 0.006), and greater emotional reactivity to stress (r = 0.39, p = 0.001).

Table 1.

Means, SDs, and intercorrelations for key study variables (N = 129).

Variable	Range	M	SD	1	2	3	4	5	6
1. Daily stress	1–3	1.58	0.41	–	0.69***	0.18*	0.53***	0.15	0.21*
2. Negative affect	1–5	1.23	0.31		–	0.18*	0.75***	0.15	0.25**
3. Spousal emotional support	1–3	1.47	0.51			–	0.08	0.16	0.10
4. Emotional stress reactivity^a	0.18–0.86	0.22	0.16				–	0.22**	0.39**
5. Fasting blood glucose level	82–230	130.63	24.41					–	0.53**
6. Fasting blood glucose variability	4.86–88.32	22.82	16.12						–

Mean value reflects average over a 24-day assessment. Fasting blood glucose variability = standard deviation (SD) of daily fasting blood glucose readings over a 24-day assessment.

Within-person estimate of the extent to which daily stress is associated with negative affect over a 24-day assessment.

p < 0.05; **p < 0.01; ***p < 0.001.

Emotional stress reactivity, spousal emotional support, and fasting blood glucose level

The analysis of fasting blood glucose levels revealed no main effects of stress reactivity or spousal support (cf. Griffith et al., 1990) but did reveal a significant interaction between stress reactivity and spousal support (β = −0.24, p = 0.003; see Table 2, left panel). Specifically, among patients who reported less spousal support, greater stress reactivity was associated with significantly higher fasting blood glucose (simple slope = 46.21(18.81), t = 2.46, p = 0.015; see Figure 1). Among patients who reported more spousal support, in contrast, greater stress reactivity was unrelated to fasting blood glucose (simple slope = −21.31(19.40), t = −1.10, n.s.). The pattern in Figure 1 also suggested, unexpectedly, that patients who received greater spousal support generally had higher fasting glucose levels than did patients who received less spousal support.

Table 2.

Emotional stress reactivity, spousal emotional support, and fasting blood glucose.

	Fasting blood glucose level				Fasting blood glucose variability
	B	SE B	β	T	B	SE B	β	T
Covariates
Sex (female = 1)	−2.92	3.98	−0.06	<−1	1.92	2.45	0.06	<1
Diabetes duration (years)	0.35	0.23	0.13	1.54	0.46	0.15	0.26	3.19**
Co-morbid health conditions	1.56	1.58	0.09	< 1	0.58	0.99	0.05	<1
Depressive symptoms	0.11	0.24	0.05	< 1	−0.16	0.14	−0.11	−1.13
BMI	0.71	0.28	0.22	2.51**	−0.25	0.18	−0.11	−1.40
Mean dietary adherence	−11.86	6.11	−0.16	−1.94⁺	−5.79	3.75	−0.12	−1.54
Fasting blood glucose level	–	–	–	–	0.30	0.06	0.45	5.33***
Independent variables
Emotional stress reactivity	12.73	15.46	0.08	< 1	31.13	9.33	0.30	3.34^**
Spousal emotional support	8.40	4.16	0.17	2.02*	1.09	2.57	0.03	<1
Interaction
Emotional stress reactivity × spousal emotional support	−69.47	23.09	−0.24	−3.01**	3.35	14.37	0.02	<1
Adjusted R²	0.21				0.39

SE: standard error; BMI: body mass index.

Fasting blood glucose level = average (M) of daily fasting blood glucose readings over a 24-day assessment. Fasting blood glucose variability = SD of daily fasting blood glucose readings over a 24-day assessment. Mean dietary adherence = mean daily dietary adherence over a 24-day assessment. Emotional stress reactivity = within-person estimate of the extent to which daily stress is associated with negative affect over a 24-day assessment.

p < 0.06; *p < 0.05; **p < 0.01; ***p <0 .001.

Figure 1.

Fasting blood glucose level as a function of emotional stress reactivity and spousal emotional support.

Emotional stress reactivity, spousal emotional support, and fasting blood glucose variability

The analysis of fasting blood glucose variability revealed a significant main effect of stress reactivity (β = 0.30, p = 0.001; see Table 2, right panel) but no main effect of spousal support or interaction between stress reactivity and spousal support. Patients who exhibited greater emotional reactivity to daily stress also exhibited greater variability in their fasting blood glucose over the 24-day period of the daily diary assessment, and this association was not affected by the level of support provided by the patients’ spouses.

Supplemental analyses

We conducted two sets of supplemental analyses to aid in the interpretation of the associations between stress reactivity and fasting blood glucose. Results of these analyses are available from the first author (K.S.R) on request.

Does average daily stress over a 24-day period predict average blood glucose levels or variability?

First, we examined whether participants’ average daily stress over the 24-day period of the daily diary assessment predicted their average blood glucose levels or variability during the same period. We conducted two multiple regression analyses that had the same structure as the analyses described above except that we replaced the measure of stress reactivity with the measure of mean daily stress and modified the interaction term to test an interaction between mean daily stress and spousal emotional support. In both analyses, neither the main effect of daily stress nor the interaction between daily stress and spousal support were significant. Comparing these results with those described above suggests that emotional stress reactivity was better able to explain participants’ fasting blood glucose than was their average daily stress per se.

Does the inclusion of controls for mean daily stress, mean negative affect, and mean diabetes distress alter the findings?

We also repeated our primary regression analyses with controls included for the patient’s mean daily stress and mean negative affect (cf. Piazza et al., 2013) in order to evaluate the effects of stress reactivity independent of the patient’s overall level of stress or negative affect over the 24-day daily diary assessment. The results of these analyses mirrored those shown in Table 2 (left and right panels). These results, along with the inclusion of depressive symptoms as a covariate in our primary analyses, help to rule out the possibility that the effects of greater stress reactivity are due to a general tendency to experience more stress, more negative affect, or more depressive symptoms. We also conducted supplemental analyses that included the patient’s diabetes-related distress (assessed with an abbreviated version of the Problem Areas in Diabetes Scale; Polonsky et al., 1995) as a covariate. The results were unchanged, which indicates that the effects of patients’ stress reactivity are unlikely to be attributable to their diabetes-related distress.

Does the inclusion of controls for marital duration and marital quality alter the findings?

We also repeated our primary regression analyses with controls included for the patient’s marital duration (in years) and marital quality (assessed with a 5-item version of the Quality of Marriage Index; Norton, 1983). The results were unchanged when these covariates were included, which suggests that the counterintuitive effects of the spouse’s emotional support were unlikely to be attributable either to newer marriages (in which spouses may still have been adjusting to the patient’s illness and their role as a support provider) or to less satisfying marriages (in which the spouse’s emotional support might have poorly attuned to the patient’s needs).

Discussion

Evidence suggests that life stress can adversely affect blood glucose control in patients with type 2 diabetes (Morris et al., 2011), perhaps by depleting the psychological resources needed for effective diabetes self-management (Hagger et al., 2009) or by triggering the release of hormones that contribute to elevated blood glucose (Surwit and Schneider, 1993). Yet the association between stress and blood glucose control exhibits considerable individual variability (Riazi et al., 2004), and this variability may be due, in part, to individual differences in the degree to which stress arouses emotional distress (Soo and Lam, 2009). This study investigated this possibility by computing within-person estimates of emotional reactivity to daily stress and examining their association with fasting blood glucose in a sample of older adults with type 2 diabetes. In view of evidence linking social support to better blood glucose control in diabetic adults (Griffith et al., 1990), this study also examined the extent to which emotional support provided by spouses moderated the association between stress reactivity and participants’ blood glucose.

The results of our multivariate analyses revealed no evidence of a main effect of participants’ emotional stress reactivity on their blood glucose levels but did reveal an interaction between stress reactivity and emotional support provided by the spouse. Greater stress reactivity was associated with higher fasting blood glucose among participants with less spousal support, consistent with our hypothesis. Contrary to expectation, however, participants with greater spousal support exhibited relatively elevated blood glucose regardless of their level of stress reactivity. Particularly surprising was the elevated blood glucose among participants who were low in stress reactivity and high in spousal support.

One question raised by the results, therefore, is how to interpret this unexpected finding. One possibility is that the patients who were low in stress reactivity and who received a high level of spousal support may have been experiencing stressors that adversely affected their blood glucose control but for which the spouse’s provision of diet-related emotional support was mismatched and, therefore, not helpful (cf. Reinhardt et al., 2006). Another possibility is that the spouses’ support may have inadvertently exacerbated their partners’ illness-related concerns (Gallant et al., 2007). For example, the spouses themselves may have been worried about their partners’ high glucose levels, and the emotional support they provided to their partner may have compounded their partners’ worries (Coyne et al., 1988). Receiving social support can also entail costs, including costs to the recipient’s self-efficacy and feelings of indebtedness (e.g. McClure et al., 2014; Martire et al., 2002). Subtle or implicit forms of support that avoid these costs may be the most effective (Bolger and Amarel, 2007), although little is known about the feasibility or effectiveness of implicit emotional support in the context of chronic illness. Further research that examines spouses’ reasons for offering support and both patients’ and spouses’ perceptions of, and reactions to, their illness-related support transactions will be needed to tease apart these possible explanations.

In contrast to the unexpected finding that emerged among patients with high spousal support, we did find, as expected, that patients with low spousal support who were high in emotional stress reactivity were likely to exhibit elevated fasting blood glucose. Moreover, greater emotional stress reactivity was related to greater blood glucose variability, irrespective of the level of support provided by the spouse. These findings are consistent with previous research that has linked greater stress to worse blood glucose control among individuals with diabetes (Morris et al., 2011; Soo and Lam, 2009), but the results of this study highlight the role of individual differences in stress reactivity as a factor that may influence blood glucose control. Our data do not allow us to determine whether stress reactivity was associated with neuroendocrine processes or behavioral patterns that may have contributed to higher or more variable blood glucose. For example, exercise, diet, or medication management may be disrupted by the coping demands associated with greater stress reactivity; such demands may drain finite motivational and self-regulatory resources that could otherwise be directed toward maintaining greater adherence to the treatment regimen (Hagger et al., 2009). Investigating the physiological and behavioral processes that link greater stress reactivity to worse blood glucose control is an important goal for future research (cf. Skaff et al., 2009).

Limitations and directions for future research

In assessing the results of this study, several limitations need to be considered. First, the self-report nature of the data is a limitation, although we have no reason to believe that participants systematically distorted their reports of daily stress or negative affect over the 24-day assessment period or were inaccurate in recording their daily blood glucose values. The estimates of participants’ stress reactivity were derived empirically, moreover, rather than reported subjectively, which increases confidence that these estimates may not have been particularly vulnerable to self-report biases. Nonetheless, the use of methods to supplement self-reports is a worthwhile goal for future research. In a related vein, patients’ dietary adherence was measured using self-reports. Future research would benefit from the use of more objective measures (e.g. blood glucose meter downloads, medication refills, fitness assessment, accelerometer readings) to assess patients’ self-care behaviors. In addition, more complete HbA1c data and assessment of within-day fluctuations in blood glucose (among patients who have multiple daily readings) would provide a better understanding of patients’ glycemic control (cf. Torimoto et al., 2013). This study also did not examine the specific types of daily stress that participants were encountering in the course of their daily lives. A useful direction for future research, therefore, would be to identify particular classes of everyday stressors that are especially likely to be upsetting (Bolger et al., 1989) and to interfere with blood glucose control among individuals with type 2 diabetes. In a related vein, this study did not investigate the specific mechanisms that may mediate the association between stress reactivity and blood glucose levels, and delving into these mechanisms would be valuable in future research.

The sample in this study was predominantly non-Hispanic White, which limits the generalizability of our findings to minority populations, in which both rates of type 2 diabetes (Centers for Disease Control and Prevention, 2014) and levels of everyday stress tend to be high (Thoits, 2010). In addition, participants in this study were older adults, with a mean age of 66.05 years, and negative affect generally declines with advancing age (Charles and Carstensen, 2010). Stress reactivity, and its associations with blood glucose control, might be greater in younger individuals with type 2 diabetes.

Conclusion

Despite these limitations, this study adds to the literature on the association between stress and blood glucose among individuals with type 2 diabetes by highlighting the potential effects of individual differences in emotional stress reactivity or the extent to which stress arouses negative affect. Continued investigation of such individual differences may help to forge a more complete understanding of how the stress experienced by people with diabetes in the course of their daily lives compromises their blood glucose control.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Institute on Aging (grant no. R01 AG24833) and the National Center for Advancing Translational Sciences (grant no. KL2 TR000147).

References

Aiken

West

(1991) Multiple Regression: Testing and Interpreting Interactions. Thousand Oaks, CA: SAGE.

American Diabetes Association (2014) Standards of medical care in diabetes—2014. Diabetes Care 37: S14–S80.

Armeli

Todd

Mohr

(2005) A daily process approach to individual differences in stress-related alcohol use. Journal of Personality 73: 1657–1686.

Beverly

Miller

Wray

(2008) Spousal support and food-related behavior change in middle-aged and older adults living with type 2 diabetes. Health Education & Behavior 35: 707–720.

Bolger

Amarel

(2007) Effects of social support visibility on adjustment to stress: Experimental evidence. Journal of Personality and Social Psychology 92: 458–475.

Bolger

Zuckerman

(1995) A framework for studying personality in the stress process. Journal of Personality and Social Psychology 69: 890–902.

Bolger

Delongis

Kessler

. (1989) Effects of daily stress on negative mood. Journal of Personality and Social Psychology 57: 808–818.

Centers for Disease Control and Prevention (2014) National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, 2014. Atlanta, GA: US Department of Health and Human Services.

Charles

Carstensen

(2010) Social and emotional aging. Annual Review of Psychology 61: 383–409.

10.

Cox

Gonder-Frederick

(1992) Major developments in behavioral diabetes research. Journal of Consulting and Clinical Psychology 60: 628–638.

11.

Coyne

Wortman

Lehman

(1988) The other side of support: Emotional overinvolvement and miscarried helping. In: Gottlieb

(ed.) Marshaling Social Support: Formats, Processes, and Effects. Thousand Oaks, CA: SAGE, pp. 305–330.

12.

DeVries

(2013) Glucose variability: Where it is important and how to measure it. Diabetes 62: 1405–1408.

13.

Gallant

Spitze

Prohaska

(2007) Help or hindrance? How family and friends influence chronic illness self-management among older adults. Research on Aging 29: 375–409.

14.

Gibson

(2012) The psychobiology of comfort eating: Implications for neuropharmacological interventions. Behavioural Pharmacology 23: 442–460.

15.

Glasgow

Toobert

Gillette

(2001) Psychosocial barriers to diabetes self-management and quality of life. Diabetes Spectrum 14: 33–41.

16.

Griffith

Field

Lustman

(1990) Life stress and social support in diabetes: Association with glycemic control. International Journal of Psychiatry in Medicine 20: 365–372.

17.

Hagger

Wood

Stiff

. (2009) The strength model of self-regulation failure and health-related behaviour. Health Psychology Review 3: 208–238.

18.

Heatherton

Wagner

(2011) Cognitive neuroscience of self-regulation failure. Trends in Cognitive Sciences 15: 132–139.

19.

Hirakawa

Arima

Zoungas

. (2014) Impact of visit-to-visit glycemic variability on the risks of macrovascular and microvascular events and all-cause mortality in type 2 diabetes: The ADVANCE trial. Diabetes Care 37: 2359–2365.

20.

Kramer

Ledolter

Manos

. (2000) Stress and metabolic control in diabetes mellitus: Methodological issues and an illustrative analysis. Annals of Behavioral Medicine 22: 17–28.

21.

Lin

Anderson

Hagerty

. (2008) Diabetes self-management experience: A focus group study of Taiwanese patients with type 2 diabetes. Journal of Clinical Nursing 17: 34–42.

22.

McClure

Craw

. (2014) Understanding the costs of support transactions in daily life. Journal of Personality 82: 563–574.

23.

Martire

Stephens

MAP

Druley

. (2002) Negative reactions to received spousal care: Predictors and consequences of miscarried support. Health Psychology 21: 167–176.

24.

Miller

Brown

(2005) Marital interactions in the process of dietary change for type 2 diabetes. Journal of Nutrition Education and Behavior 37: 226–234.

25.

Morris

Moore

Morris

(2011) Stress and chronic illness: The case of diabetes. Journal of Adult Development 18: 1–11.

26.

Mroczek

Almeida

(2004) The effect of daily stress, personality, and age on daily negative affect. Journal of Personality 72: 355–378.

27.

Nesselroade

(1988) Sampling and generalizability: Adult development and aging research issues examined within the general methodological framework of selection. In: Schaie

Campbell

Meredith

. (eds) Methodological Issues in Aging Research. New York: Springer, pp. 13–42.

28.

Norton

(1983) Measuring marital quality: A critical look at the dependent variable. Journal of Marriage and the Family 45: 141–151.

29.

Ogden

Carroll

Kit

. (2012) Prevalence of obesity in the United States, 2009–2010. JAMA 307: 491–497.

30.

Piazza

Charles

Sliwinski

. (2013) Affective reactivity to daily stressors and long-term risk of reporting a chronic physical health condition. Annals of Behavioral Medicine 45: 110–120.

31.

Polonsky

Anderson

Lohrer

. (1995) Assessment of diabetes-related distress. Diabetes Care 18: 754–760.

32.

Radloff

(1977) The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1: 385–401.

33.

Raudenbush

Byrk

(1992) Hierarchical Linear Models: Applications and Data Analysis Methods. Newbury Park, CA: SAGE.

34.

Reinhardt

Boerner

Horowitz

(2006) Good to have but not to use: Differential impact of perceived and received support on well-being. Journal of Social and Personal Relationships 23: 117–129.

35.

Riazi

Pickup

Bradley

(2004) Daily stress and glycaemic control in type 1 diabetes: Individual differences in magnitude, direction, and timing of stress-reactivity. Diabetes Research and Clinical Practice 66: 237–244.

36.

Semmer

McGrath

Beehr

(2004) Conceptual issues in research on stress and health. In: Cooper

(ed.) Handbook of Stress, Medicine and Health (2nd edn). New York: CRC Press, pp. 1–45.

37.

Siegelaar

Holleman

Hoekstra

. (2010) Glucose variability; Does it matter? Endocrine Reviews 31: 171–182.

38.

Skaff

Mullan

Almeida

. (2009) Daily negative mood affects fasting glucose in type 2 diabetes. Health Psychology 28: 265–272.

39.

Sliwinski

Almeida

Smyth

. (2009) Intraindividual change and variability in daily stress processes: Findings from two measurement-burst diary studies. Psychology and Aging 24: 828–840.

40.

Soo

Lam

(2009) Stress management training in diabetes mellitus. Journal of Health Psychology 14: 933–943.

41.

Stephens

MAP

Franks

Rook

. (2013) Spouses’ attempts to regulate day-to-day dietary adherence among patients with type 2 diabetes. Health Psychology 32: 1029–1037.

42.

Surwit

Schneider

(1993) Role of stress in the etiology and treatment of diabetes mellitus. Psychosomatic Medicine 55: 380–393.

43.

Thoits

(2010) Stress and health: Major findings and policy implications. Journal of Health and Social Behavior 51: S41–S53.

44.

Tomiyama

Dallman

Epel

(2011) Comfort food is comforting to those most stressed: Evidence of the chronic stress response network in high stress women. Psychoneuroendocrinology 36: 1513–1519.

45.

Toobert

Hampson

Glasgow

(2000) The summary of diabetes self-care activities measure: Results from 7 studies and a revised scale. Diabetes Care 23: 943–950.

46.

Torimoto

Okada

Mori

. (2013) Relationship between fluctuations in glucose levels measured by continuous glucose monitoring and vascular endothelial dysfunction in type 2 diabetes mellitus. Cardiovascular Diabetology 12: 1.

47.

Wagner

(2007) Coping with diabetes: Psychological determinants of diabetes outcomes. In: Martz

Livneh

(eds) Coping with Chronic Illness and Disability: Theoretical, Empirical, and Clinical Aspects. New York: Springer, pp. 215–239.

48.

Watson

Clark

Tellegen

(1988) Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology 54: 1063–1070.

49.

Woerle

Neumann

Zschau

. (2007) Impact of fasting and postprandial glycemia on overall glycemic control in type 2 diabetes: Importance of postprandial glycemia to achieve target HbA1c levels. Diabetes Research and Clinical Practice 77: 280–285.