A comparative study of serum levels of glucose and aggression: A cross-sectional study

Abstract

Diabetes mellitus is a metabolic disorder associated not only with physical complications but also with psychological and behavioral disturbances. Emerging evidence suggests that blood glucose fluctuations may influence mood, impulse control, and aggressive behavior. This prospective, observational study included 278 adult patients (171 with type 2 diabetes and 107 without diabetes) admitted to the Internal Medicine Outpatient Clinic of Istanbul Training and Research Hospital between June and August 2025. Participants completed the Buss–Perry Aggression Questionnaire, which measures physical aggression, verbal aggression, anger, and hostility. Clinical data, including fasting blood glucose, insulin, lipid profile, and glycated hemoglobin (HbA1c) levels, were recorded. Individuals with diabetes were further stratified by HbA1c levels to examine the association between glycemic control and aggression. Aggression scores were significantly higher in patients with diabetes compared to those without (P < .05), and HbA1c levels positively correlated with Buss–Perry Aggression Questionnaire scores within the diabetes group (R = 0.549, P < .05). Fasting blood glucose, insulin, total cholesterol, and low-density lipoprotein levels were also elevated in individuals with diabetes. Poor glycemic control was associated with higher aggression levels, potentially reflecting alterations in mood regulation and impulse control. Study limitations include the single-center design, the observational nature, and the short follow-up period, which restrict causal inferences. These findings highlight the importance of comprehensive diabetes care that integrates behavioral assessments alongside metabolic monitoring and suggest the need for further longitudinal research to clarify the relationship between glycemic dysregulation and aggression.

1. Introduction

Diabetes mellitus (DM) is a metabolic disorder marked by hyperglycemia, which arises through different mechanisms depending on its subtype. In some types, it results from insulin resistance in peripheral tissues, while in others, it is caused by autoimmune destruction of pancreatic beta cells. It presents with symptoms such as polyuria, polydipsia, and weight loss.^[1] According to studies conducted in 2021, the number of people with diabetes worldwide was found to be over 540 million, and this number is projected to exceed 640 million by the year 2031.^[2] Recognizing and managing the complications of such a highly prevalent disease has become a global focus over the years. Microvascular and macrovascular complications related to diabetes have been well defined, and recommendations for their management have been presented to physicians worldwide through consensus reports and guidelines developed by expert panels and medical societies.^[3,4]

Studies have shown that the quality of life in patients with diabetes varies depending on their blood glucose regulation and the presence of diabetes-related complications. The levels of physical, social, psychological, and environmental functioning in these patients differ according to the type of complications associated with diabetes.^[5]

In recent years, studies on diabetes have primarily focused on blood glucose regulation, the management of diabetic complications, and the cardio–renal–metabolic disorders caused by the disease.^[6–10] However, there is a noticeable lack of research examining the effects of blood glucose regulation on psychological well-being and emotional state. Diabetes is a metabolic disease that can lead to various psychological disorders. One of these conditions is diabetic distress. It is a serious clinical condition accompanied by emotional disturbances, stress, feelings of guilt, and avoidance of treatment. Studies focusing on the psychological effects of diabetes are increasing, highlighting the growing importance of this issue.^[11]

Blood glucose fluctuations experienced by individuals with diabetes, along with mood changes and fears related to hypoglycemic episodes, suggest that these individuals may be more aggressive, hostile, and angry in their daily lives compared to those with normal blood glucose levels. In this research, we aimed to investigate the effects of blood glucose levels on patients’ aggression, hostility, and anger in people living with type 2 DM.

2. Materials and methods

This prospective research was implemented on 278 individuals who presented to the Internal Medicine Outpatient Clinic of Istanbul Training and Research Hospital between June 15, 2025, and July 15, 2025. Among the participants, 171 had type 2 DM, while 107 did not have a diagnosis of diabetes. The inclusion criteria were: being between 18 and 65 years of age, voluntarily agreeing to participate by reading and signing the informed consent form, and being able to independently complete the Buss–Perry Aggression Questionnaire (BPAQ). Patients were excluded if they were younger than 18 or older than 65 years; had a diagnosis of type 1 DM; had a history of major psychiatric disorders such as schizophrenia, bipolar disorder, or schizoaffective disorder; were currently using or had recently used (within the last 6 months) medications affecting the serotonergic or dopaminergic systems, including selective serotonin reuptake inhibitors, serotonin–norepinephrine reuptake inhibitors, tricyclic antidepressants, monoamine oxidase inhibitors, antipsychotics, or mood stabilizers; had diagnosed cognitive impairment, dementia, or intellectual disability; had a documented history of alcohol or narcotic drug abuse; had a history of incarceration, criminal records involving violence, or known incidents of physical aggression in public or domestic settings; had any acute infections, inflammatory conditions, or severe systemic illness at the time of data collection; were unable to complete the BPAQ due to language or comprehension difficulties; or had missing clinical data.

All individuals furnished written informed consent after being informed about the objectives and procedures of the study.

2.1. Assessment of aggression

Aggression levels in the included patients were assessed using the BPAQ. This self-report instrument consists of 29 items and is designed as a comprehensive measure of aggression in adults. The scale evaluates aggression across 4 subdimensions: physical aggression, verbal aggression, anger, and hostility.

Each item is rated on a 5-point Likert scale, ranging from 1 (“extremely uncharacteristic of me”) to 5 (“extremely characteristic of me”). The questionnaire is composed of the following item groupings:

• Physical aggression: items 1, 6, 9, 13, 14, 20, 22, 29.

• Verbal aggression: items 2, 5, 12, 21.

• Anger: items 4, 7, 15, 16, 19, 23, 26.

• Hostility: items 3, 8, 10, 11, 17, 18, 24, 25, 27, 28.

Some items in the scale are reverse-scored due to being phrased in a manner that reflects low aggression. These items are: 9, 10, 11, 12, 16, 22, 25, 26, and 28. These responses are reversed during scoring to maintain consistency in the direction of aggression assessment. The total score reflects the overall level of aggression, with higher scores indicating a greater tendency toward aggressive behavior.^[12] Patients completed the questionnaire independently by reading and responding to the items themselves, without assistance.

2.2. Assessment of laboratory measurements

Clinical and laboratory data were gathered from patient files and the hospital information management system. Following an overnight fast of 12 hours, venous blood samples were collected in the morning from all participants. Fasting blood glucose, fasting insulin, total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein cholesterol, triglyceride levels, and glycated hemoglobin (HbA1c) levels were documented. Individuals with diabetes were stratified into subgroups based on their HbA1c levels: 6.5% to 7.9%, 8.0% to 9.9%, and ≥10.0%.

2.3. Statistical analysis

Descriptive statistics of the data were presented as mean, standard deviation, median, minimum, maximum, frequency, and percentage values. The distribution of variables was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests. For the analysis of quantitative independent variables that were not normally distributed, the Mann–Whitney U test was used. The chi-square test was employed for the analysis of categorical independent variables. The relationship between variables was assessed through Spearman rank-order correlation analysis. All statistical analyses were conducted using the SPSS software, version 28.0 (IBM Corporation, Armonk).

2.4. Ethics approval and consent to participate

Ethics committee approval for this research was obtained from the Ethics Committee of Istanbul Training and Research Hospital (approval number: 143, dated June 13, 2025).. All procedures performed in the research were in accordance with the 1964 Helsinki Declaration. Written informed consent was obtained from all participants prior to their inclusion in the study.

3. Results

According to the inclusion and exclusion criteria of the study, only 278 out of 2866 screened patients were included (Fig. 1).

Figure 1.

Participant flowchart.

The mean age of the participants was 53 ± 8.5 years. One hundred twenty-seven (45.7%) of the patients were male and 151 (54.3%) were female. Of the participants, 39.6% were high school graduates, 37.1% had completed a university or doctoral degree, and 23.4% were primary school graduates. Regarding marital status, 24.1% of the participants were single and 75.9% were married. Fifty-five percent of the patients were smokers. Among the participants, 61.9% (n = 172) reported no alcohol consumption, 30.2% (n = 84) identified as social drinkers, and 7.9% (n = 22) reported regular alcohol use (Table 1).

Table 1.

Sociodemographic profile, lifestyle factors, and metabolic parameters of individuals included in the study.

		Min–Max			Median	Mean ± SD/n (%)
Age (yr)		23.0	–	67.0	53.0	53.0	±	8.5
Gender	Male					127		45.7%
	Female					151		54.3%
Education Level	Primary School					65		23.4%
	High school					110		39.6%
	University or Doctorate Degree					103		37.1%
Marital Status	Single					67		24.1%
	Married					211		75.9%
Smoking Status	(-)					125		45.0%
	(+)					153		55.0%
Alcohol Usage	(-)					172		61.9%
	Social drinker					84		30.2%
	Regularly					22		7.9%
Presence of diabetes mellitus	(-)					107		38.5%
	(+)					171		61.5%
FBG (mg/dL)		72.0	–	285.0	122.0	136.6	±	60.8
Fasting insulin (mU/L)		5.6	–	18.4	9.8	10.4	±	3.3
HbA1c (%)		5.2	–	14.3	6.8	7.4	±	2.1
Total cholesterol (mg/dL)		112.0	–	310.0	205.0	204.9	±	50.6
HDL (mg/dL)		32.0	–	87.0	51.0	51.4	±	12.2
LDL (mg/dL)		43.0	–	193.0	121.0	124.2	±	39.8
Triglycerides (mg/dL)		46.0	–	513.0	134.0	157.0	±	96.4
Patients’ HbA1c Levels	<6.5%					107		38.5%
	6.5–7.9%					99		35.6%
	8–9.9%					30		10.8%
	≥10%					42		15.1%
BPAQ score		36.0	–	124.0	56.0	60.4	±	16.7

Values are presented as minimum, maximum, median, mean ± standard deviation and number (%).

BPAQ = Buss–Perry Aggression Questionnaire, FBG = fasting blood glucose, HbA1c = glycated hemoglobin, HDL = high-density lipoprotein, LDL = low-density lipoprotein.

The mean fasting blood glucose level of the study participants was 136.6 ± 60.8 mg/dL. The mean fasting insulin level was 10.4 ± 3.3 mU/L. The mean HbA1c level was found to be 7.4 ± 2.1%. The mean total cholesterol level of the participants was 204.9 ± 50.6 mg/dL, while the mean high-density lipoprotein cholesterol level was 51.4 ± 12.2 mg/dL and the mean LDL cholesterol level was 124.2 ± 39.8 mg/dL. The mean triglyceride level was calculated as 157.0 ± 96.4 mg/dL (Table 1).

Among all participants included in the study (n = 278), 107 individuals (37.5%) had HbA1c levels below 6.5%. Additionally, 99 participants (35.6%) had HbA1c levels between 6.5% and 7.9%, 30 participants (10.8%) had levels between 8.0% and 9.9%, and 42 participants (15.1%) had HbA1c levels of ≥ 10%. All percentages were calculated based on the total study population (Table 1).

The mean BPAQ score of all participants was 60.4 ± 16.7 (Table 1).

The comparison between participants with and without diabetes in terms of demographic characteristics, laboratory parameters, and BPAQ scores is summarized in Table 2.

Table 2.

Comparative analysis of study variables in patients with and without diabetes mellitus.

		Patients without DM (n: 107)				Patients with DM (n: 171)				P
		Mean ± SD/n (%)			Median	Mean ± SD/n (%)			Median
Age (yr)		53.0	±	9.0	53.0	53.0	±	8.1	52.0	.697*
Gender	Male	60		56.1%		67		39.2%		.006 †
	Female	47		43.9%		104		60.8%
Education level	Primary school	8		7.5%		57		33.3%		.000 †
	High school	44		41.1%		66		38.6%
	University or Doctorate Degree	55		51.4%		48		28.1%
Marital status	Single	36		33.6%		31		18.1%		.003 †
	Married	71		66.4%		140		81.9%
Smoking status	(-)	34		31.8%		91		53.2%		.000 †
	(+)	73		68.2%		80		46.8%
Alcohol usage	(-)	56		52.3%		116		67.8%		.010 †
	Social drinker	51		47.7%		33		19.3%
	Regularly	0		0.0%		22		12.9%
FBG (mg/dL)		82.8	±	7.7	81.0	170.3	±	55.0	144.0	.000 *
Fasting insulin (mU/L)		9.0	±	1.9	9.2	11.3	±	3.7	11.8	.000 *
Total cholesterol (mg/dL)		195.2	±	48.1	185.0	211.0	±	51.4	207.0	.008 *
HDL (mg/dL)		49.8	±	11.4	44.0	52.4	±	12.6	52.0	.057*
LDL (mg/dL)		114.7	±	41.0	112.0	130.2	±	38.0	134.0	.018 *
Triglycerides (mg/dL)		153.4	±	83.3	138.0	159.3	±	104.0	111.0	.631*
BPAQ score		54.0	±	10.1	53.0	64.5	±	18.6	59.0	.000 *

Bold indicates a P-value of <.05 was considered statistically significant.

BPAQ = Buss–Perry Aggression Questionnaire, DM = diabetes mellitus, FBG = fasting blood glucose, HbA1c = glycated hemoglobin, HDL = high-density lipoprotein, LDL = low-density lipoprotein.

^*Mann–Whitney U test.

^†Chi-square test.

In the group with DM, the level of education and the rates of smoking and alcohol consumption were significantly lower than the group without DM (P < .05). Fasting blood glucose, fasting insulin, total cholesterol, and LDL cholesterol levels were significantly higher in the group with DM (P < .05). The BPAQ score was also significantly higher in the DM group than the group without diabetes (P < .05) (Table 2).

A statistically significant (P < .05) moderate positive correlation (R = 0.549) was found between HbA1c levels and BPAQ scores based on Spearman correlation analysis (Table 3, Fig. 2).

Table 3.

Spearman correlation between glycated hemoglobin levels and Buss–Perry Aggression Questionnaire scores.

		BPAQ score
HbA1c (%)	r	0.549
	P	.000
Spearman correlation

Bold indicates a P-value of <.05 was considered statistically significant.

BPAQ = Buss–Perry Aggression Questionnaire, HbA1c = glycated hemoglobin.

Figure 2.

Spearman correlation between glycated hemoglobin levels and Buss–Perry Aggression Questionnaire Scores.

4. Discussion

Patients with DM require psychological support throughout their lives starting from the time of diagnosis. For instance, the lifelong obligation to adhere to dietary restrictions, along with intermittent hypoglycemic episodes caused by medications or the disease itself, may lead to the development of an anxious, panicked, and aggression-prone personality. In cases where patients fail to comply with effective treatment regimens, the disease often remains suboptimally controlled, resulting in the progression of diabetes-related complications, a decline in quality of life, and organ failure due to end-organ damage. These outcomes contribute to increased utilization of healthcare services and place a substantial burden on healthcare systems. Despite this well-recognized reality, the provision of psychosocial support remains generally inadequate, primarily due to the complex nature of these needs and the additional strain they impose on healthcare infrastructures.^[13]

In our research, it was observed that individuals with type 2 DM had a lower educational level compared to those without the disease. Similarly, a study demonstrated that the prevalence of diabetes increased among women with lower educational attainment.^[14] This finding suggests that, over the years, the Turkish population has been relatively well-informed about diabetes through public awareness campaigns, the efforts of civil society organizations, diabetes associations, and health policy initiatives. As a result, diabetes has become a widely recognized and frequently discussed condition in Türkiye, with considerable emphasis placed on education and prevention. Recent studies have also highlighted the benefits of integrating modern technologies such as telemedicine, digital platforms, and social media into diabetes management. These tools have been shown to contribute to the control of body mass index, fasting plasma glucose, fat mass, and HbA1c levels in diabetic patients. Furthermore, they have been associated with improved diabetes-related knowledge and enhanced medication adherence.^[15]

In our research, smoking and alcohol consumption were found to be lower among individuals with diabetes as opposed to those without diabetes. The detrimental effects of smoking and alcohol on the cardiovascular system are well-established facts supported by numerous scientific studies over the years.^[16,17] Several studies have demonstrated that smoking cessation is associated with a reduction in mortality due to cardiovascular, respiratory, and neoplastic diseases.^[18] The lower rates of alcohol and tobacco use observed among individuals with diabetes in our study may reflect the effectiveness of physicians in explaining to their patients during routine 3-month diabetes follow-up visits that smoking and alcohol use have an additional negative impact on both the microvascular and macrovascular complications of diabetes.

As expected, in our study, fasting plasma glucose, fasting insulin, total cholesterol, and LDL cholesterol levels were significantly higher in individuals with diabetes compared to those without. The presence of elevated fasting plasma glucose and fasting insulin levels is consistent with the underlying pathophysiology of type 2 DM, which is characterized by insulin resistance.^[19] The coexistence of type 2 DM and hyperlipidemia is a frequently observed clinical condition, and both have been shown in previous studies to be independent risk factors for atherosclerotic cardiovascular diseases.^[20]

Although studies investigating the psychiatric effects of DM remain relatively scarce in the literature, both global and national research has increasingly focused on the relationship between diabetes and mental health conditions such as depression and anxiety. In a study conducted in Türkiye, Altunoğlu E.G. and colleagues demonstrated a significant association between insulin resistance and depression, suggesting that this relationship may be confounded by abdominal obesity.^[21] Similarly, Kim Y.C. et al analyzed data from the 2022 Korea National Health and Nutrition Examination Survey, evaluating 3 major psychiatric parameters: depression, anxiety, and suicidal ideation. Their findings indicated that depression was significantly more common in individuals with diabetes compared to controls. Moreover, among individuals with a diabetes duration of 15 years or more, both depression and suicidal ideation were found to be significantly higher. Based on these results, the authors proposed that routine psychiatric screening in patients with long-standing diabetes could serve as a public health strategy.^[22] In another study, Kassem S. and colleagues emphasized the bidirectional link between diabetes and depression, noting that each condition may exacerbate the other and contribute to worse clinical outcomes. The authors attributed this association to underlying pathophysiological mechanisms such as dysregulation of the hypothalamic–pituitary–adrenal axis and chronic inflammation.^[23] Furthermore, Boakye M.D.S. and colleagues undertook a comprehensive review of the psychosocial burden among individuals with diabetes in Africa, identifying 83 relevant studies published between 2000 and 2024. The pooled prevalence rates were reported as 43.3% for depressive symptoms, 38.8% for anxiety symptoms, 48.8% for moderate-to-severe diabetes-related distress, and 43.9% for poor mental quality of life.^[24] These findings underscore the substantial mental health burden associated with diabetes and the urgent need to integrate psychological care into diabetes management globally.

Studies examining diabetes in terms of aggression and hostility are quite limited in the literature. In a study conducted by Tilov B. and colleagues, it was demonstrated that individuals with diabetes, hypertension, and musculoskeletal disorders showed statistically significant differences in levels of anger and hostility at the 95% confidence level. Physical aggression levels were reported to be highest among patients with hypertension. The highest average level of verbal aggression was observed in patients with diabetes, while the highest levels of anger and hostility were again seen in individuals with hypertension.^[25] In another study conducted by R. Sadiq and colleagues involving 100 individuals with diabetes and 100 without, levels of physical aggression, verbal aggression, anger, and hostility were assessed. The results showed that verbal aggression was significantly higher among individuals with diabetes compared to those without. However, no significant difference was noted in terms of physical aggression.^[26] Consistent with similar findings in the literature, our study also revealed significantly higher aggression levels among patients with diabetes compared to those without. Moreover, it was determined that as HbA1c levels increased, the frequency of aggression also increased.

The brain relies entirely on glucose to meet its energy demands. Notably, the prefrontal cortex responsible for decision-making, impulse control, and regulation of social behavior is highly sensitive to fluctuations in blood glucose levels and hypoglycemia. It has been demonstrated that prefrontal cortical dysfunction during hypoglycemia leads to increased impulsivity and aggressive behavior.^[27] Sympathetic nervous system activation during hypoglycemic episodes often results in physiological symptoms such as sweating, palpitations, and irritability. This state of physiological stress lowers individuals’ tolerance thresholds, thereby predisposing them to increased anger and heightened verbal or physical aggression. Multiple studies have established a significant and positive association between DM and both anxiety disorders and elevated anxiety symptoms.^[28,29] Supporting this relationship at the neuropsychological level, a study conducted by Walters et al involving violence-prone male individuals proposed the “limited capacity model,” which suggests that the frontal lobe has a finite capacity to simultaneously regulate both metabolic and behavioral functions. When this capacity is exceeded, impairments in impulse control and aggression regulation may occur. Furthermore, the study reported that individuals with a predisposition to violent behavior exhibited both impaired frontal lobe functioning and poor glycemic control. These findings suggest that glucose levels are not only metabolically relevant but also play a key role in behavioral regulation, particularly aggression.^[30] The results of our study are consistent with the aforementioned literature. Aggression levels were significantly higher among individuals with diabetes compared to those without diabetes. Moreover, subgroup analysis revealed that, within the diabetic cohort, increasing HbA1c levels were associated with a greater tendency toward aggression.

The serotonin system plays an important role in the neurobiological basis of aggressive behavior. An inverse relationship has been demonstrated between central serotonin levels and impulsive aggression. It has been reported that as serotonin release decreases, impulsive anger and aggressive behaviors increase, with significantly elevated anger–hostility scores observed in affected individuals. Serotonin is a key neurotransmitter involved in higher-order cognitive and behavioral regulatory processes, particularly in the frontal cortex, and it also interacts with glucose regulation. Fluctuations in glucose levels may disrupt both serotonergic activity and frontal cortical control, thereby facilitating the emergence of impulsive behaviors.^[31] In this context, the positive correlation observed in our study between blood glucose levels and aggression scores may be attributed to a neurobiological mechanism mediated by serotonin. The increased tendency toward aggression among individuals with higher HbA1c levels could potentially be explained by reduced serotonergic tone and diminished frontal cortical inhibition. These findings highlight that glycemic dysregulation may have implications not only for metabolic health but also for neuropsychiatric functioning.

Study limitations include the single-center design, the observational nature of the study, and the short recruitment period, which restrict causal inferences. Additionally, our sample size, although sufficient for detecting moderate associations, may limit the generalizability of the findings to broader populations. Finally, the reliance on self-reported behavioral measures, such as the BPAQ, may introduce response biases despite the use of validated instruments. Future researches with multicenter designs, longer follow-up periods, and objective behavioral assessments are warranted to confirm and expand upon our findings.

5. Conclusion

This research observed an association between dysregulated blood glucose levels and increased aggression in adults, particularly among individuals with DM. Higher BPAQ scores in patients with diabetes, along with a positive correlation between HbA1c levels and aggression, indicate a potential link between glycemic control and aggressive tendencies. These findings are consistent with neurobiological models, including the limited capacity hypothesis and serotonergic dysregulation, which suggest that glucose fluctuations may influence impulse control and behavioral regulation. Although causality cannot be established, the results highlight the importance of considering behavioral and psychological assessments in the management of patients with diabetes. Future research using neuroimaging and longitudinal designs may help clarify the nature of the relationship between glucose regulation and aggression. A multidisciplinary approach addressing both metabolic and behavioral health may be beneficial for improving patient outcomes.

Author contributions

Conceptualization: Vahit Can Cavdar, Feray Akbas.

Data curation: Vahit Can Cavdar.

Formal analysis: Vahit Can Cavdar, Feray Akbas.

Investigation: Vahit Can Cavdar, Feray Akbas.

Methodology: Vahit Can Cavdar, Feray Akbas.

Project administration: Vahit Can Cavdar, Feray Akbas.

Resources: Vahit Can Cavdar, Feray Akbas.

Software: Vahit Can Cavdar, Feray Akbas

Supervision: Vahit Can Cavdar, Feray Akbas.

Validation: Vahit Can Cavdar, Feray Akbas.

Visualization: Vahit Can Cavdar, Feray Akbas.

Writing – original draft: Vahit Can Cavdar, Feray Akbas.

Writing – review & editing: Vahit Can Cavdar, Feray Akbas.