The Impact of Health Locus of Control and Anxiety on Self-Monitored Blood Glucose Concentration in Women with Gestational Diabetes Mellitus

Abstract

Background:

The purpose of this study was to evaluate the associations between health locus of control (HLC), anxiety, and glycemic control from the time of diagnosis of gestational diabetes (GDM) to the end of pregnancy.

Methods:

The study group comprised 165 women with GDM. Baseline HLC (∼27 weeks of gestation) was assessed by the Multidimensional HLC Scale. The level of anxiety was measured at baseline and follow-up (37 weeks of gestation) by the State-Trait Anxiety Inventory. Using questionnaires, we collected information about the level of fear related to measuring blood glucose several times per day, dietary regimen, and insulin therapy, as well as fear for the baby and its health, patient's own health, and having diabetes in the next pregnancy. Glycemic control was evaluated by self-monitored fasting and postprandial blood glucose levels.

Results:

Baseline state anxiety was significantly higher than trait anxiety. From baseline to follow-up, the state anxiety and percentage of women with increased fear for their infant's health, diabetic diet, self-monitoring of blood glucose, and insulin injection significantly decreased. In comparison to women with blood glucose in the low tertile, those with blood glucose in the high tertile had significantly higher scores in the chance HLC subscale and a similar level of state anxiety. Blood glucose was positively correlated with the chance HLC score.

Conclusions:

Chance HLC beliefs seem to be associated with poorer glycemic outcomes in women with GDM. Our results suggest the need for further efforts to reduce the GDM-associated state anxiety.

Introduction

Gestational diabetes mellitus (GDM) is defined as glucose intolerance first diagnosed during pregnancy. Although the exact causes of GDM are unknown and being still investigated, this condition tends to develop in the second half of pregnancy and represents the imbalance between insulin release and insulin sensitivity.¹ GDM may increase the risk of serious fetal, newborn, and maternal complications, such as infant macrosomia, injury during birth, stillbirth, neonatal hypoglycemia, and preeclampsia.¹ Moreover, GDM-related long-term complications include a higher risk of type 2 diabetes (T2DM) in both mothers and their offspring. All of these health risks may induce feelings of anxiety, fear and despair, loss of control, failure, and guilt, as well as worries about one's own health and the health of the baby.^2
–4

An additional source of negative emotions and distress in mothers with GDM may arise from the need for blood glucose self-monitoring, dietary restrictions, and the necessity of insulin treatment. These factors may overlap with and aggravate the psychological changes associated with pregnancy. In a normal pregnancy, maternal distress and anxiety may increase the risk of premature delivery,^2,3 low child birth weight,⁴ postpartum depression,^3,5 or sleep problems during infancy and toddlerhood.^6,7 In patients with diabetes, the psychosocial attributes, which are mainly driven by diabetes-related emotional stress, may contribute to difficulty with diabetes self-management, poor glycemic control, and worsening diabetes status over time.⁸ It is believed that these psychological attributes may be related, at least partially, to perceived health locus of control (HLC). HLC is commonly used in health psychology to evaluate self-management behavior in chronic medical conditions.⁹ It is conceptualized as either internal or external. People with internal HLC believe that they can control their life, whereas those with external HLC believe that their decisions and life are controlled by extraneous factors that they cannot influence, or by chance or fate.¹⁰ In patients with diabetes, it has been suggested that changes in HLC can precede changes in self-care behavior, thus leading to changes in blood glucose concentrations.¹¹ Indeed, some earlier studies demonstrated that patients with T2DM who exhibited internal HLC demonstrated better overall adjustment to diabetes regimens, better adherence to self-control, and better metabolic control of diabetes.^12
–14 Similarly, in patients with type 1 diabetes (T1DM) receiving continuous subcutaneous insulin infusion, high internal HLC represented the most important HLC pattern for achieving good glycemic control.¹⁵ On the other hand, other studies¹⁶ and a recent meta-analysis¹⁰ failed to confirm the association between HLC and glycemic control. To date, there have been no reports on the associations of HLC with glycemic outcomes in GDM.

The purpose of this study was to evaluate the associations between HLC and changes in anxiety level and glycemic control from the time of diagnosis of GDM to the end of pregnancy. We were also interested in identifying modifiable factors that may affect anxiety in GDM.

Materials and Methods

Study participants

The study group comprised women with GDM treated in a tertiary care center (Outpatient Clinic for Diabetic Pregnant Women in Szczecin, Poland). GDM was diagnosed by a 75-g oral glucose tolerance test by using the cut-off values given by current guidelines.^17,18 We excluded cases with overt diabetes, prior GDM, and multiple pregnancies. Of the 225 women included, 165 fulfilled all of the study procedures. All patients participated in a diabetes education program provided by a trained diabetes nurse educator, and patient education skills were regularly reviewed and reinforced. Education was focused on achieving effective self-monitoring of blood glucose concentration by using a glucometer with at least four measurements per day (fasting, 1 hour after main meals, and additionally as required) as well as therapeutic goals, which were defined as fasting glucose level 70–90 mg/dL, below 120 mg/dL 1 hour after meals, and above 60 mg/dL between 2:00 A.M. and 4:00 A.M.¹⁷ All subjects used the same type of glucometer as well as test strips (Accu-Chek Go; Roche Diabetes Care, Poland). All blood glucose values were collected by using the Accu-Chek Smart Pix software version 2.0. We recorded blood glucose measurements taken within the entire study period from all 165 women, and the mean value and standard deviation (SD) were calculated for the whole group. The median number of measurements per woman was 392; of those, 98 were fasting and 294 were postprandial. Additional measurements were considered postprandial. Routine treatment included an individually tailored diabetic diet and physical activity as well as long-acting and/or short-acting insulin. The dose of insulin was self-adjusted in case of inadequate glycemic control.

In all study participants, baseline information on demographic variables, pregravid weight, height, weight at evaluation, gestational age, co-morbidities, current and past treatments received, as well as current symptoms were obtained from medical charts. The baseline visit took place at the time of diagnosis (27.94 ± 1.65 weeks of gestation), and the follow-up visit occurred at gestational age 37.40 ± 1.7 weeks. The study protocol was approved by the Pomeranian Medical University Ethics Committee.

Assessment of patients' fear related to GDM

Using a self-administered, paper-and-pencil, 8-item Likert-type questionnaire, we collected information about the level of fear related to measuring blood glucose several times per day, dietary regimen, frequent finger puncturing, and insulin therapy, as well as fear for the baby and its health, patient's own health, and having diabetes in the next pregnancy. The fear of blood glucose measuring was defined as the fear of obtaining blood glucose levels outside the targets. We also asked women to rate the level of support received from their family. Participants rated their answers on a scale ranging from 1 (“not at all”) to 5 (“very much”). This questionnaire was completed twice, once at baseline and once at follow-up.

State-Trait Anxiety Inventory

The State-Trait Anxiety Inventory (STAI) is a 40-item self-reported questionnaire with four possible responses to each item. It is widely used to estimate the severity of state (20 items) and trait anxiety (20 items).¹⁹ State anxiety reflects a transitory emotional state or condition that is characterized by subjective, consciously perceived feelings of tension and apprehension, and heightened autonomic nervous system activity. It may fluctuate over time and can vary in severity. In contrast, trait anxiety refers to an individual's general level of stress, is defined as a feeling of stress, nervousness, and discomfort that is experienced on a day-to-day basis, and indicates a predisposition to be anxious in normal situations. The raw STAI scores were expressed on the sten scale, by using the age- and sex-specific norms for Polish adults determined by the Psychological Test Laboratory of the Polish Psychological Association.²⁰ The level of anxiety was categorized as low (1–4 sten scores), average (5–6), or high (7–10). The Cronbach alpha of the State-Trait Inventory ranged between 0.89 and 0.96, suggesting good reliability.²⁰ Trait and state anxiety were assessed at baseline. State anxiety was additionally assessed at the follow-up visit.

Multidimensional HLC scale

Form B of the Multidimensional HLC scale was administered to all study participants only at baseline. The form was constructed to measure general health-related control beliefs without being specific to any health behavior or condition. The HLC scale consists of the three 6-item locus of control subscales (Internality, Powerful Others, and Chance), and it measures the extent to which respondents believe that a given factor affects their health status or the progress of their disease.²¹ Participants rate the degree of agreement with each item on a Likert rating scale ranging from 1 (strongly disagree) to 6 (strongly agree). The scores on the subscales are independent of each other, and high scores on the Internality subscale indicate high internal control. The Cronbach alpha of the Multidimensional HLC Scale was 0.74 (Internal), 0.69 (Chance), and 0.54 (Powerful Others).²²

We used culturally adapted and validated Polish translations of STAI^19,20 and HLC.²²

Statistical analysis

Descriptive statistics included frequency distributions (number with condition and percentage) for categorical variables and means, SD, and range for continuous variables. To assess changes from baseline to follow-up, we used a paired t-test or nonparametric Mann-Whitney U-test for continuous variables and Chi-square test with Yates' correction for categorical variables. Spearman's rank correlation coefficients were used to test for associations between blood glucose and STAI and HLC scores. The STAI and HLC scores were adjusted for age and pregravid body mass index by using multiple regression. All analyses were carried out in Statistica (StatSoft, Poland).

Results

Description of the study population

Baseline characteristics of the study population are shown in Table 1. The mean age was 30.4 years; 29 women (14.5%) were 35 years of age or older. The frequencies of pregravid overweight (body mass index between 25.0 and 29.9 kg/m²) and obesity (body mass index above 30.0 kg/m²) were 27.3% and 14.5%, respectively. A high proportion of women had a history of prior childbirth (44%) and T2DM in first-degree relatives (49%). The mean Internality and Powerful Others HLC scores were similar and significantly higher than the Chance score (p < 0.05). Results from the STAI questionnaire revealed a significantly higher state anxiety score in comparison with trait anxiety (p = 0.007).

Table 1.

Baseline Characteristics of Study Participants

Variables	Women with GDM (n = 165)
Age (years)	30.42 ± 4.84
Height (cm)	164.86 ± 5.31
Weight before pregnancy (kg)	66.98 ± 13.72
Prepregnancy body mass index (kg/m²)	24.60 ± 4.67
Gestational age at the first visit (weeks)	27.94 ± 2.65
Diabetes in first-degree relatives, n (%)	81 (49.1)
Prior childbirth, n (%)	73 (44.2)
Multidimensional health locus of control
Internal (score)	27.05 ± 3.96
Powerful others (score)	25.50 ± 5.33
Chance (score)	19.73 ± 5.35
State-trait anxiety inventory
State anxiety
Raw value	42.21 ± 6.12
Sten score	6.30 ± 2.33
Trait anxiety
Raw value	36.68 ± 5.86
Sten score	3.97 ± 1.92

Data are mean ± standard deviation or number (percentage).

GDM, gestational diabetes.

Outcome changes between baseline and follow-up

During the study period, 86 women (52%) required insulin therapy, which was initiated on average at 31.1 ± 2.3 weeks of gestation. At the follow-up visit, the mean level of state anxiety was 5.83 ± 1.82 sten scores, which was significantly lower than the baseline score (6.30 ± 2.33 sten scores; p = 0.011). Frequency distribution of changes in self-reported level of fear and family support in women with GDM are shown in Table 2. At the follow-up visit, the percentage of women with increased fear of treatment and diagnostic procedures, such as diabetes diet, self-monitoring of blood glucose, puncturing the fingers, and insulin injection, as well as fear for the baby and its health significantly decreased. On the other hand, the percentage of those with a high fear for their own health, having diabetes in the next pregnancy, and low family support remained unchanged. Interestingly, the percentage of women with a high fear of insulin injection was similar, regardless of whether they had received insulin therapy or not after the baseline visit. Among the factors that were directly related to GDM, higher fear of treatment or diagnostic procedures and lower family support were associated with higher state anxiety scores; this trend was observed at both baseline and follow-up (Table 3). At the follow-up visit, women reported a significantly lower anxiety associated with diabetic diet and blood glucose self-monitoring, as well as fear for the baby and its health. However, they reported a similar level of fear of finger puncturing and insulin therapy, and similar support from family. We did not find significant differences in state anxiety associated with fear for one's own health and fear of having diabetes in the next pregnancy (data not shown).

Table 2.

Changes from Baseline to Follow-Up in Self-Reported Level of Fear and Family Support

Variable	Baseline, n (%)	Follow-up, n (%)	p
Fear of diet
Not at all	45 (27.3)	78 (47.3)	0.0002
A little	55 (33.3)	58 (35.2)	0.9893
Moderately	42 (25.5)	20 (12.1)	0.0025
Quite a bit	21 (12.7)	8 (4.9)	0.0001
Very much	2 (1.22)	1 (0.61)	0.5619
Fear of blood glucose measurement several times per day
Not at all	64 (38.8)	119 (72.1)	0.0001
A little	57 (34.6)	38 (23.0)	0.0276
Moderately	30 (18.2)	5 (3.03)	0.0001
Quite a bit	13 (7.88)	2 (1.22)	0.0092
Very much	1 (0.61)	1 (0.61)	1.00
Fear of puncturing the fingers to measure blood glucose
Not at all	92 (55.8)	140 (84.9)	0.0001
A little	31 (18.8)	17 (10.3)	0.0209
Moderately	23 (13.9)	6 (3.64)	0.0016
Quite a bit	15 (9.09)	2 (1.21)	0.0496
Very much	4 (2.42)	1 (0.6)	0.1310
Fear of insulin injections^a
Not at all	3 (1.82)	71 (43.0)	0.0001
A little	9 (5.45)	22 (13.3)	0.0116
Moderately	10 (6.06)	19 (11.5)	0.0903
Quite a bit	50 (30.3)	22 (13.3)	0.0002
Very much	93 (56.4)	31 (18.8)	0.0001
Fear of baby and its health
Not at all	9 (5.45)	62 (37.6)	0.0001
A little	21 (12.7)	36 (21.8)	0.0322
Moderately	43 (26.1)	30 (18.2)	0.0803
Quite a bit	60 (36.4)	20 (12.1)	0.0001
Very much	32 (19.4)	17 (10.3)	0.0758
Fear of own health
Not at all	20 (12.1)	16 (9.70)	0.5619
A little	26 (15.8)	38 (23.0)	0.1095
Moderately	32 (19.4)	42 (25.5)	0.1892
Quite a bit	48 (29.1)	50 (30.3)	0.8423
Very much	39 (23.6)	19 (11.5)	0.0040
Fear of having diabetes in the next pregnancy
Not at all	22 (13.3)	16 (9.70)	0.3007
A little	17 (20.3)	20 (12.1)	0.6006
Moderately	38 (23.0)	34 (20.6)	0.5940
Quite a bit	39 (23.6)	42 (25.5)	0.6123
Very much	49 (29.7)	53 (32.1)	0.2996
Support received from family
Not at all/a little	10 (6.1)	11 (6.7)	0.6982
Moderate	55 (33.3)	42 (25.5)	0.1162
Quite a bit	68 (41.2)	73 (44.9)	0.5779
Very much	32 (19.4)	39 (23.6)	0.4872

Including 86 women who were treated with insulin.

Table 3.

Relationship Between Self-Rated Changes in Gestational Diabetes-Related Factors, Family Support, and Changes in State Anxiety Level

Factors related to GDM	State anxiety (sten score)
Factors related to GDM	Baseline	p	Follow-up	p
Fear of diabetic diet
Not at all (referent)	4.515 ± 1.08		4.012 ± 1.54^a
A little	6.076 ± 1.29	0.0048	5.787 ± 2.09^a	0.0039
Somewhat	6.509 ± 2.02	0.0020	6.424 ± 1.99	0.0011
Quite a bit	7.980 ± 1.77	0.0001	7.123 ± 1.84^a	0.0008
Very much	7.650 ± 1.34	0.0001	6.841 ± 1.42^a	0.0009
Fear of blood glucose self-monitoring
Not at all (referent)	4.595 ± 2.08		3.994 ± 2.32^a
A little	6.744 ± 2.91	0.0003	5.767 ± 2.23^a	0.0002
Somewhat	7.503 ± 2.82	0.0001	7.431 ± 1.99	0.0001
Quite a bit	7.995 ± 2.03	0.0003	7.209 ± 2.13^a	0.0001
Very much	7.990 ± 1.14	0.0001	7.213 ± 1.64^a	0.0001
Fear of finger puncturing
Not at all (referent)	5.107 ± 3.24		4.890 ± 2.64
A little	6.980 ± 2.62	0.0055	6.888 ± 2.05	0.0089
Somewhat	7.069 ± 2.39	0.0074	6.989 ± 2.99	0.0067
Quite a bit	6.927 ± 2.26	0.0361	6.724 ± 2.09	0.0170
Very much	7.625 ± 1.54	0.0278	7.349 ± 1.69	0.0348
Fear of insulin injections
Not at all (referent)	4.552 ± 2.91		4.686 ± 2.82
A little	4.861 ± 2.56	0.7351	4.752 ± 2.96	0.2571
Somewhat	5.170 ± 2.35	0.4490	5.201 ± 2.15	0.0624
Quite a bit	6.070 ± 2.99	0.0392	6.110 ± 2.86	0.0415
Very much	7.263 ± 2.09	0.0010	7.174 ± 2.99	0.0019
Fear of the baby and its health
Not at all/a little (referent)	4.582 ± 2.02		3.844 ± 1.68
Somewhat	6.503 ± 1.91	0.0097	5.293 ± 2.36^a	0.0260
Quite a bit/very much	6.430 ± 2.33	0.0119	5.502 ± 2.14^a	0.0021
Support received from family
Very much (referent)	6.224 ± 2.30		4.824 ± 2.27
Quite a bit	5.991 ± 2.35	0.2342	4.951 ± 2.05	0.0140
Somewhat	6.560 ± 2.36	0.1672	5.642 ± 2.02	0.0039
A little/not at all	6.998 ± 1.89	0.0675	7.352 ± 2.36	0.0021

Data are mean ± standard deviation. p-Values refer to comparisons with the reference level of each variable.

p < 0.05 versus baseline.

Blood glucose control

Throughout the entire study period, self-monitored blood glucose values were symmetrically distributed with a mean of 106.9 mg/dL (5.94 mmol/L) and an SD of 6.3 mg/dL (0.35 mmol/L) (Table 4). Overall, 63% of fasting and 59% postprandial blood glucose values collected from all measurements were within the target range. Patients with combined fasting and postprandial blood glucose levels in the low tertile (below 107.0 mg/dL) had a similar state anxiety score as those with blood glucose in the high tertile (above 112.0 mg/dL) (Table 5). Similarly, we did not find any differences in Internal and Powerful Others HLC scores between patients with blood glucose in the low tertile and those with blood glucose in the high tertile; however, patients with blood glucose in the high tertile had significantly higher scores in the chance subscale of HLC.

Table 4.

Self-Measured Blood Glucose Concentrations During the Entire Study Period

Blood glucose (mg/dL)	Mean	Standard deviation	Range
Fasting	87.51	5.58	68–112
1 hour after breakfast	117.91	10.02	95–154
1 hour after lunch	117.21	7.99	93–148
1 hour after dinner	117.88	8.43	95–155
All measurements	109.96	6.27	89–129

Table 5.

Anxiety and Health Locus of Control Scores According to Blood Glucose Tertiles

	Glucose <107 mg/dL (low tertile)	Glucose >112 mg/dL (high tertile)	p
State-trait anxiety (sten score)
State anxiety	6.17 ± 2.01	5.85 ± 2.37	0.306
Trait anxiety	3.28 ± 2.78	2.69 ± 2.59	0.426
Health locus of control (score)
Internal	27.87 ± 4.03	26.79 ± 3.48	0.080
Powerful others	25.05 ± 5.10	25.75 ± 5.07	0.449
Chance	18.58 ± 2.25	20.47 ± 4.68	0.026

Anxiety and health locus of control were measured at baseline. Glucose levels are based on all pre- and post-prandial measurements performed within the entire study period. Data are adjusted for age and body mass index.

No significant correlations were found between blood glucose levels and state anxiety (R = 0.107), trait anxiety (R = −0.091), and perceived internal (R = −0.182) and powerful others HLC (R = 0.132). There was a weak but significant correlation of blood glucose with the chance HLC score (R = 0.212; p = 0.048).

Discussion

In this study, we report for the first time on a potential impact of HLC beliefs on blood glucose levels in women with GDM. Earlier reports addressed this issue for other types of diabetes, mainly T2DM, but yielded conflicting results. Some studies suggested that patients with a high internal¹⁵ or a high internal and low chance HLC²³ achieved better metabolic control, as assessed by the glycated hemoglobin (A1C) level, than those with high external beliefs. This can be explained by better overall adjustments to diabetes regimens and greater self-motivation among patients scoring in the internal control direction.^8,12,24 However, other studies demonstrated a positive association between internal HLC and A1C (worsening of metabolic control),^13,14 a negative association of powerful others (external) beliefs with A1C (better metabolic control),²⁵ or lack of any association between HLC and A1C.^10,16,26

In GDM, a recent study suggests the highest percentage of overall satisfactory adherence to diabetes regimen among women with a high Powerful Others HLC.²⁷ Further, the Chance and Powerful Others subscales of the Multidimensional HLC Scale were related to psychological distress in GDM.²⁸ However, these studies did not assess the relationships between all three dimensions of HLC and metabolic control. Our results suggest the association of high chance beliefs and higher blood glucose levels. In addition, we found a positive correlation between the chance HLC and state anxiety, suggesting that high chance beliefs may negatively affect the GDM-related factors that increase state anxiety, such as fear of diabetic diet, self-monitoring of blood glucose, finger puncturing, and insulin therapy.

In this study, women with GDM had significantly higher baseline state anxiety than trait anxiety. State anxiety refers to a temporary condition in response to some perceived threat such as GDM and can potentially be modifiable. In contrast, trait anxiety refers to the tendency to experience state anxiety in response to a threat, and it describes a personality characteristic rather than a temporary feeling. It has been suggested that subjects with higher trait anxiety tend to experience increased state anxiety under stressful or pressure-laden conditions.²⁹ This may be important in a pregnancy complicated by GDM, as women with GDM frequently have a higher level of state anxiety than healthy pregnant women, presumably due to a greater level of stress.³ To date, data on anxiety in GDM are scarce. Crowther et al. did not find a difference in state anxiety between women who underwent routine versus intensive treatment for GDM.³⁰ On the other hand, Daniells et al.³¹ demonstrated that at ∼30 weeks of gestation, women with GDM had higher levels of state (but not trait) anxiety, in comparison with healthy pregnant women. However, these differences were not seen at the end of pregnancy and in the postpartum period. In this study, we found that state anxiety in GDM was at a moderate to high level and was strongly associated with fear of diabetes-related treatments and procedures, such as diabetes diet, self-monitoring of blood glucose, insulin injections (regardless of whether or not they received insulin therapy), and finger puncturing, as well as fear for the baby and its health. From baseline to follow-up, the mean state anxiety score significantly decreased. In contrast, the level of fear of insulin injection and finger puncturing was relatively stable. Interestingly, although the number of women in high-fear groups (respondents who answered “quite a bit” or “very much”) significantly decreased between baseline and follow-up visits, the percentage of those with high fear of having diabetes in the future remained unchanged.

Fear of self-injecting insulin, similar to the fear of finger puncturing, is well known among patients starting insulin therapy and is likely to compromise glycemic control and induce psychological stress.³² It was shown that a fear of blood and pain may lead to less frequent self-testing, hamper the intensive insulin therapy, and affect adherence and glycemic control.^32,33 In addition, patient barriers to accepting insulin therapy also include the risk of hypoglycemia, reluctance to adapt the timing and self-adjusting of scheduled insulin doses, as well as concerns of a negative impact of insulin on the child.³⁴

There were certain limitations to our study. First, due to its observational design, associations presented between the independent and outcome variables do not necessarily represent causal relationships. In particular, regardless of perceived HLC, the poorer metabolic control of GDM may be associated with genetic predisposition, low physical activity, hypertriglyceridemia, low muscle mass and excess body fat, or concomitant diseases known to affect blood glucose level. These factors were not assessed in this study. Second, we did not assess metabolic control of GDM by A1C. Instead, we used a large number of fasting and postprandial glucose measurements, which were performed at least four times per day throughout the entire study period. Because the observational period was ∼9 weeks, blood glucose testing seems to be a better indicator of short- and long-term glucose control than A1C, as the latter reflects the mean glucose level in the proceeding 14 weeks. Finally, we did not assess the level of education, marital and socioeconomic status, knowledge about GDM, support from healthcare providers, and weight gain during pregnancy. These factors are known to affect the level of stress in GDM³⁵ and might influence the level of state anxiety. Although we found a statistical association between glycemic control and chance HLC beliefs, our results should be interpreted cautiously because HLC is greatly influenced by several external factors that are not evaluated in this study, including (1) type and level of patient education received; (2) socioeconomic status and self-care tools available; (3) education level; and, importantly, (4) degree of empathic communication between patients and healthcare providers.^36,37

In conclusion, our results suggest that in women with GDM, high chance HLC beliefs may be associated with poorer glycemic outcomes that are evaluated by blood glucose levels. Our findings may have potential clinical implications. First, despite intensive baseline education provided by trained professionals, at the follow-up visit, there was still a high percentage of women who had increased fear of insulin injections, fear for their own health and the infant's health, and fear of having diabetes in the next pregnancy. These psychological aspects of living with GDM may be considered potentially modifiable and, therefore, need more attention, motivation, support, regular reinforcement, and counseling, provided by trained staff, preferably diabetes educators. Second, according to the HLC theory, high chance HLC tends to increase the likelihood of negative forms of health behavior⁸ and, hence, may be considered an unfavorable feature in GDM. However, we recognize that diabetes control may be influenced by many other factors. Therefore, despite our findings, it remains uncertain as to whether universal testing for high chance HLC would improve the identification of subjects with GDM who are at risk of poor glycemic control. This issue needs further studies. Finally, our results suggest the need for further efforts to improve the management of GDM to reduce the associated state anxiety.

Footnotes

Acknowledgment

This research did not receive any grants from funding agencies in the public, commercial, or not-for-profit sectors.

Author Disclosure Statement

No competing financial interests exist.

References

Alfadhli

. Gestational diabetes mellitus. Saudi Med J, 2015; 36:399–406.

Orr

, Reiter

, Blazer

, James

. Maternal prenatal pregnancy-related anxiety and spontaneous preterm birth in Baltimore, Maryland. Psychosom Med, 2007; 69:566–570.

Alder

, Fink

, Bitzer

, Hösli

, Holzgreve

. Depression and anxiety during pregnancy: A risk factor for obstetric, fetal and neonatal outcome? A critical review of the literature. J Matern Fetal Neonatal Med, 2007; 20:189–209.

Mulder

, Robles de Medina

, Huizink

, van den Bergh

, Buitelaar

, Visser

. Prenatal maternal stress: Effects on pregnancy and (unborn) child. Early Hum Dev, 2002; 70:3–14.

Mallikarjun

. Prevention of postnatal depression. J R Soc Promot Health, 2005; 125:221–226.

O'Connor

, Caprariello

, Blackmore

, et al. Prenatal mood disturbance predicts sleep problems in infancy and toddlerhood. Early Hum Dev, 2007; 83:451–458.

Austin

, Hadzi-Pavlovic

, Leader

, Saint

, Parker

. Maternal trait anxiety, depression and life event stress in pregnancy: Relationships with infant temperament. Early Hum Dev, 2005; 81:183–190.

Zulman

, Rosland

, Choi

, Langa

, Heisler

. The influence of diabetes psychosocial attributes and self-management practices on change in diabetes status. Patient Educ Couns, 2012; 87:74–80.

Wallston

, Wallston

, DeVellis

. Development of the multidimensional health locus of control (MHLC) scales. Health Educ Monogr, 1978; 6:160–170.

10.

Hummer

, Vannatta

, Thompson

. Locus of control and metabolic control of diabetes: A meta-analysis. Diabetes Educ, 2011; 37:104–110.

11.

Chen

, Lin

. Relationships among health locus of control, psychosocial status and glycemic control in type 2 diabetes adults. J Nurs Care, 2014; 3:162.

12.

Weng

, Hung

, Chi

. Psychosocial and biological factors associated with glycemic control for patients with type 2 diabetes: An application of structural equation modeling analysis. Pan-Pacific Manag Rev, 2010; 13:33–47.

13.

O'Hea

, Moon

, Grothe

, et al. The interaction of locus of control, self-efficacy, and outcome expectancy in relation to HbA1c in medically underserved individuals with type 2 diabetes. J Behav Med, 2009; 32:106–117.

14.

Williams

, Lynch

, Voronca

, Egede

. Health locus of control and cardiovascular risk factors in veterans with type 2 diabetes. Endocrine, 2016; 51:83–90.

15.

Indelicato

, Mariano

, Galasso

, et al. Influence of health locus of control and fear of hypoglycaemia on glycemic control and treatment satisfaction in people with type 1 diabetes on insulin pump therapy. Diabetic Med, 2017; 34:691–697.

16.

Graco

, Hutchinson

, Barker

, Lawlor

, Wong

, Fourlanos

. Glycemic outcome not predicted by baseline psychological measures in a diabetes management program. Popul Health Manag, 2012; 15:163–167.

17.

Polish Diabetes Association. Clinical recommendations in diabetes care 2014. Clin Diabetol, 2014; 3(Suppl. A):1–71.

18.

American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care, 2014; 37(Suppl. 1):81–90.

19.

Bąk

. Measuring the state, trait, expression and control of anger. The Polish adaptation of STAXI-2. Pol Forum Psychol, 2016; 21:93–122.

20.

Wrześniewski

Inwentarz stanu i cechy lęku STAI: Polska adaptacja STAI. Warsaw, Poland: Pracownia Testów Psychologicznych Polskiego Towarzystwa Psychologicznego, 2011 [Polish].

21.

Wallston

. Multidimensional health locus of control scales. In: Gellman

, Turner

, eds. Encyclopedia of behavioral medicine. New York: Springer, 2013:1266–1269.

22.

Juczyński

Narzędzia pomiaru w promocji i psychologii zdrowia. Warsaw, Poland: Pracownia Testów Psychologicznych Polskiego Towarzystwa Psychologicznego, 2009 [Polish].

23.

Stenstrom

, Wikby

, Andersson

, Rydén

. Relationship between locus of control beliefs and metabolic control in insulin-dependent diabetes mellitus. Br J Psychol, 1998; 3:15–21.

24.

van der Heijden

, Pouwer

, Romeijnders

, Pop

. Testing the effectiveness of a self-efficacy based exercise intervention for inactive people with type 2 diabetes mellitus: Design of a controlled clinical trial. BMC Public Health, 2012; 12:331.

25.

Fuscaldi

, Balsanelli

, Grossi

. Locus of control in health and self-esteem in type-2 diabetic patients. Rev Esc Enferm USP, 2011; 45:853–859.

26.

Büyükkaya Besen

, Günüşen

, Arda Sürücü

, Koşar

Predictor effect of Locus Of Control (LOC) on self-care activities and metabolic control in individuals with type 2 diabetes. PeerJ, 2016; 4:e2722.

27.

Rizk

, Ossman

. The relationship between health locus of control and adherence to diabetes regimen among gestational diabetic women. Life Sci J, 2013; 10:3428–3438.

28.

Spirito

, Ruggiero

, McGarvey

, Coustan

, Low

. Maternal and fetal health locus of control during pregnancy: A comparison of women with diabetes and nondiabetic women. J Reprod Infant Psychol, 1990; 8:195–206.

29.

Horikawa

, Yagi

. The relationships among trait anxiety, state anxiety and the goal performance of penalty shoot-out by university soccer players. PLoS One, 2012; 7:e35727.

30.

Crowther

, Hiller

, Moss

, McPhee

, Jeffries

, Robinson

. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med, 2005; 352:2477–2486.

31.

Daniells

, Grenyer

, Davis

, Coleman

, Burgess

, Moses

. Gestational diabetes mellitus: Is a diagnosis associated with an increase in maternal anxiety and stress in the short and intermediate term?. Diabetes Care, 2003; 26:385–389.

32.

Feitosa

, Sampaio

, Batista

, Pinheiro

. Frequency of fear of needles and impact of a multidisciplinary educational approach towards pregnant women with diabetes. Rev Bras Ginecol Obstet, 2013; 35:111–116.

33.

Berlin

, Bisserbe

, Eiber

, et al. Phobic symptoms, particularly the fear of blood and injury, are associated with poor glycemic control in type I diabetic adults. Diabetes Care, 1997; 20:176–178.

34.

Lapolla

, Di Cianni

, Di Benedetto

, et al. Quality of life, wishes, and needs in women with gestational diabetes: Italian DAWN Pregnancy Study. Int J Endocrinol, 2012; 2012:784726.

35.

Kopec

, Ogonowski

, Rahman

, Miazgowski

. Patient-reported outcomes in women with gestational diabetes: A longitudinal study. Int J Behav Med, 2015; 22:206–213.

36.

Abredari

, Bolourchifard

, Rassouli

, Nasiri

, Taher

, Abedi

. Health locus of control and self-care behaviors in diabetic foot patients. Med J Islam Rep Iran, 2015; 29:283.

37.

Decety

, Fotopoulou

. Why empathy has a beneficial impact on others in medicine: Unifying theories. Front Behav Neurosci, 2014; 8:457.