The Symptom Burden of Autonomic Neuropathy Is Associated with Decreased Quality of Life in 6961 People with Diabetes

Abstract

Background: Diabetes often causes microvascular complications such as neuropathy. Autonomic neuropathy remains under-recognized, and its impact on quality of life (QoL) is unclear. This study investigated associations between symptoms of autonomic dysfunction, including organ-specific subdomains, and QoL in individuals with type 1 (T1D) and type 2 diabetes (T2D). Methods: A cross-sectional population-based survey was conducted in the North Denmark Region among individuals with T1D and T2D, assessing autonomic symptom burden with the Composite Autonomic Symptom Score-31 (COMPASS-31), general well-being with the Short Form Health Survey (SF-36), and psychological well-being with the Hospital Anxiety and Depression Scale. Multivariate linear regression assessed associations between autonomic symptom scores and QoL outcomes. Results: The COMPASS-31 scores were 8.9 (2.9; 22.8) in T1D and 12.4 (5.3; 26.1) in T2D. SF-36 physical composite scores were 52.1 (43.2; 56.4) in T1D and 49.3 (40.3; 54.8) in T2D, with similar mental composite scores (50.7 (40.3; 56.9) vs. 51.4 (41.2; 57.2)). Signs of moderate to severe anxiety were observed in 9.9% (95% confidence interval (CI): 8.1–11.9) of T1D and 8.9% (95% CI: 8.1–9.6) of T2D, while depression was present in 5.9% (95% CI: 4.5–7.6) and 5.1% (95% CI: 4.5–5.7). Higher autonomic symptom burden, especially pupillary, vasomotor, and bladder domains, was associated with lower SF-36 score and higher anxiety and depression scores. Conclusions: the Autonomic symptom burden is associated with reduced QoL and increased psychological distress in individuals with diabetes. These findings emphasize the importance of assessing and managing autonomic symptoms in diabetes care to support overall well-being.

1. Introduction

Autonomic neuropathy is a serious complication of diabetes, and one of the key challenges of autonomic neuropathy is that its symptoms can be subtle and slowly progressing and are therefore easily overlooked. Autonomic neuropathy is characterized by dysfunction of the vagus nerve, a key component of the parasympathetic nervous system connecting the brain to the organs, which can indeed be mirrored in its target organs. This can lead to a diverse range of symptoms and health issues from multiple organ systems such as the eyes, cardiovascular system, urinary and digestive system, but also sweat abnormalities. A recent Danish study found that 37% of individuals with type 1 diabetes (T1D) and 44% of individuals with type 2 diabetes (T2D) are burdened with symptoms of autonomic dysfunction despite access to a universal free public healthcare system. The most reported issues were pupillary abnormalities and orthostatic intolerance [1]. The pathogenesis of autonomic neuropathy differs from T1D to T2D. T1D has earlier onset, longer disease duration, and autoimmune comorbidities, whereas T2D often has a prolonged undiagnosed period and is associated with metabolic syndrome, both increasing neuropathy risk.

Diabetes and pre-diabetes have been linked to diminished health-related quality of life (QoL) [2,3], likely due to the combined effects of high disease burden and unresolved or poorly managed complications. Chronic conditions can negatively impact QoL [4], and symptoms such as pupillary discomfort (e.g., photophobia, dry eyes, blurred vision), dizziness, and fear of falling due to orthostatic hypotension, may contribute to emotional distress and social isolation, ultimately diminishing both psychological and physical well-being [5,6,7]. In adolescents with T1D, abnormal autonomic function tests and high symptom burden had the lowest WHO-5 well-being index [8]. Moreover, in both healthy individuals and patients with orthostatic intolerance, greater severity of autonomic symptoms has been associated with symptoms of depression, anxiety and/or stress [9]. Even among the elderly without known neurodegenerative diseases, evidence suggests that autonomic symptoms are associated with a lower QoL [10].

The daily management of diabetes requires a high level of personal commitment, often demanding substantial lifestyle adjustments and adherence to complex treatment regimens to prevent long-term complications. However, the extensive list of potential complications can be overwhelming, and patients are often more focused on glucose levels and the immediate demands of treatment [11], which may mask the subclinical complications that may gradually diminish QoL without overt clinical signs. The impact of autonomic symptom domains on QoL is not well characterized in diabetes. Thus, this study aims to investigate the association between self-reported symptoms of autonomic dysfunction and QoL indicators in the North Denmark Region, including psychological and general well-being.

2. Materials and Methods

2.1. Study Population and Data Source

The methodology is described in detail elsewhere [1,12,13]. In short, this cross-sectional study was conducted as an online survey targeting adults in the North Denmark Region, which accounts for approximately 10% of the Danish population [14]. The Business Intelligence Unit at Aalborg University Hospital identified all adults with T1D or T2D in the region using the Danish National Health Insurance Service Register. Individuals were identified based on International Classification of Diseases 10th Revision codes DE101–109 and DE110–119, or through records of chronic care services linked to T2D (0131 or 0132), provided that there was no termination code (0133). The identified individuals with access to Digital Post (the National secure electronic communication system between public authorities and citizens) were invited to a survey hosted on REDCap (Research Electronic Data Capture, Vanderbilt University, USA) [15,16]. The survey was open from November 2022 to February 2023. This study was registered with the Regional Data Security Authority (ref.no. F2022-089). Approval from the ethical committee was not required per Danish legislation.

2.2. Survey Information

The survey contained multiple questionnaires. In the present study, the results from the validated Danish version of the Composite Autonomic Symptom Score-31 (COMPASS-31) [17], the Hospital Anxiety and Depression Scale (HADS) [18] and Short Form Health Survey (SF-36) [19] are analyzed.

The COMPASS-31 questionnaire comprises 31 questions and aims to quantify symptoms of autonomic dysfunction in 6 domains: orthostatic, vasomotor, secretomotor, gastrointestinal, bladder, and pupillomotor function. Items are scored 0–100, with higher scores indicating more impaired function. The COMPASS-31 score was determined using Sletten’s method, assigning weights to every subdomain [20]. Based on the total COMPASS-31 score, the study population was classified into three groups according to the severity of autonomic symptoms: no self-reported autonomic dysfunction (SAD−; COMPASS-31 score < 16), early self-reported autonomic dysfunction (SAD+; score 16–31), and established self-reported autonomic dysfunction (SAD++; score ≥ 32).

The SF-36 questionnaire contains 36 questions measuring the QoL concerning eight different roles: physical functioning, limitations due to physical problems, bodily pain, general health, vitality/energy, social functioning, limitations due to emotional problems, and mental health. Notably, the higher the score, the better. Mental and physical composite scores were calculated according to the SF-36 manual [21]. The SF-36 composite scores reflect general well-being and serve as indicators of QoL.

The HADS questionnaire is a 14-point survey used for identifying and quantifying anxiety and depression, with seven questions regarding each. Higher scores indicate higher degrees of anxiety or depression. Scores were calculated as described by Zigmond [18]. A symptom score of ≥11 indicates moderate to severe anxiety or depression. HADS anxiety and depression scores are used as measures of psychological well-being.

In addition, demographic data, including age, sex, height, weight, body mass index (BMI), disease duration, and socioeconomic parameters were collected. Income levels were converted to the categories: low, middle, and high, using the 33rd and 67th percentiles as thresholds.

2.3. Statistical Analysis

All statistical analysis was performed in Stata (2023. Stata Statistical Software: Release 18). Data normality was tested using the Shapiro–Wilk test and QQ-plots and expressed accordingly, i.e., either mean ± standard deviation or median (interquartile range (IQR)). Data were analyzed with the appropriate statistical test, e.g., t-test or the Mann–Whitney U test, ANOVA or the Kruskal–Wallis test, corrected for multiple pairwise comparisons using Benjamini–Hochberg correction. The proportion of participants with anxiety and depression was estimated as percentages, with 95% confidence intervals (CIs) calculated using the exact method for binomial distributions.

Moreover, the association between COMPASS-31 scores and both mental and physical health parameters was examined using multivariate linear regression, adjusted for diabetes duration, age, BMI, unemployment, and retirement/disability pension status. Given the large sample size, a linear modeling approach was applied despite not fully meeting all model assumptions. Figures were created using R version 4.5.0 (2023. R: A Language and Environment for Statistical Computing) with the forest plot package to visualize the slope coefficients and 95% CI from the multivariate analysis.

3. Results

3.1. Response Rate

This In total, 29,155 people with diabetes were identified in the North Denmark Region. Of these, 5949 (20.4%) did not have access to Digital Post and were excluded. The remaining 23,206 were invited to the online survey, and 9913 responses were received. Of these, 2192 (22%) were excluded for reasons including declining participation, duplication, non-regional residency, or incomplete demographics. Lastly, 6961 unique complete responses to the three included questionnaires were available (Figure 1).

3.2. Clinical Characteristics

The clinical characteristics are summarized in

Table 1. Demographics of the study participants. IQR: interquartile range, BMI: body mass index, COMPASS-31: Composite Autonomic Symptom Score-31, SF-36: Short Form Health Survey. PCS: physical composite score, MCS: mental composite score. HADS: Hospital Anxiety and Depression Scale (A: anxiety, D: depression).

	Type 1 Diabetes	Type 2 Diabetes
n (%)	981 (14.1)	5980 (85.9)
Age, median (IQR)	59 (48; 68)	67 (60; 74)
BMI, median (IQR)	25.7 (23.2; 29.1)	29.4 (26.2; 33.5)
Sex, n (%)
Female	432 (44.0)	2328 (38.9)
Male	545 (55.6)	3641 (60.9)
Other	<5 (.)	11 (0.2)
Diabetes duration, median (IQR)	26 (13; 40)	10 (4; 15)
Education, n (%)
Municipal primary/lower secondary school (10 years)	230 (23.4)	1978 (33.2)
Upper secondary school/Academy profession graduate (11–15 years)	385 (39.2)	2048 (34.3)
University graduate, BSc/MSc (16 ≥ years)	366 (37.3)	1939 (32.5)
Employment status, n (%)
Employed/self-employed/student	554 (56.4)	3063 (34.6)
Unemployed	41 (4.2)	192 (3.3)
Retirement/disability pension	387 (39.4)	3704 (61.2)
Income, n (%)
Low	244 (24.9)	1756 (29.4)
Middle	318 (32.4)	2044 (34.2)
High	266 (27.1)	1105 (18.5)
Unknown	153 (15.6)	1075 (18.0)
Questionnaire scores, median (IQR)
COMPASS-31	8.9 (2.9; 22.8)	12.4 (5.3; 26.1)
SF-36 PCS	52.1 (43.2; 56.4)	49.3 (40.3; 54.8)
SF-36 MCS	50.7 (40.3; 56.9)	51.4 (41.2; 57.2)
HADS-A	4 (1; 7)	3.0 (1; 6)
HADS-D	2 (1; 5)	2 (0; 4)
Moderate or severe anxiety, % (95 CI)	9.9 (8.1–11.9)	8.9 (8.1–9.6)
Moderate or severe depression, % (95 CI)	5.9 (4.5–7.6)	5.1 (4.5–5.7)

. As expected, individuals with T1D were slightly younger, had a lower BMI, and a longer diabetes duration compared to the individuals with T2D. Moreover, the T2D group comprised a higher proportion of retired individuals (incl. those receiving disability pension), resulting in a relatively lower percentage of participants categorized as having high income. In the questionnaire results, participants with T1D had slightly lower COMPASS-31 scores, while SF-36 and HADS scores were otherwise similar between groups. Signs of moderate to severe anxiety were observed in approximately 10% of individuals with T1D and T2D, and around 5% reported depressive symptoms.

3.3. Association Between Levels of Autonomic Dysfunction and General and Psychological Well-Being

Figure 2 presents the median and interquartile range (IQR) for SF-36 physical and mental composite scores, as well as anxiety and depression scores, stratified by levels of self-reported autonomic dysfunction (SAD severity). Both SF-36 physical and mental composite scores decreased progressively with increasing SAD severity, while anxiety and depression scores increased, reaching their highest values in the SAD++ group. Comparisons across SAD severity levels were statistically significant for all outcomes (all p  <  0.001).

3.4. Impact of Autonomic Symptom Burden on General and Psychological Well-Being

Multivariate analysis examined the correlation between autonomic symptom burden and measures of psychological and general well-being, as shown in Figure 3. The total COMPASS-31 score, and all subdomains were negatively associated with the SF-36 physical and mental composite scores and positively associated with the anxiety and depression scores. In terms of general physical well-being (SF-36 PCS), the strongest negative impact was seen with vasomotor (β = −2.37, 95% CI [−2.61, −2.13], R² = 0.10) and pupillary symptoms (β = −2.40, 95% CI [−2.59, −2.20], R² = 0.19). Similarly, for mental health (SF-36 MCS), pupillary symptoms showed the largest impact (β = −3.61, 95% CI [−3.85, −3.35], R² = 0.18), followed by vasomotor and bladder symptoms. Regarding psychological well-being, pupillary symptoms showed the greatest impact on both anxiety score (β = 1.22, 95% CI [1.14, 1.30], R² = 0.20) and depression score (β = 1.00, 95% CI [0.92, 1.07], R² = 0.19). Vasomotor and bladder symptoms followed. Overall, all autonomic symptom domains appear to influence both general and psychological well-being.

4. Discussion

In this study, we demonstrate that higher autonomic symptom burden is associated with reduced general and psychological well-being. Both physical and mental composite scores of the SF-36 decreased progressively with increasing severity of autonomic dysfunction. Concurrently, HADS scores for anxiety and depression increased, underscoring the psychological impact of autonomic dysfunction. Among the symptom domains, pupillary disturbances, such as delayed or reduced constriction of the pupil, causing dry eyes, photophobia, or blurry vision, had the greatest impact on both psychological well-being and general well-being. Moreover, bladder symptoms, e.g., urinary urgency, increased frequency, or difficulties in bladder emptying, along with vasomotor symptoms, e.g., flushing, also showed high impact. Thus, autonomic symptom burden is associated with reduced QoL.

A Swedish population-based study investigating health-related QoL in individuals with diabetes also utilized the SF-36, reporting a physical composite score of 50 in T1D and 47 in T2D, which is comparable to our findings of 52 and 49, respectively. Similarly, the mental composite score in their cohort was 49 for T1D and 51 for T2D, aligning closely with our results of 51 in both groups [22]. These similarities are likely influenced by the comparable healthcare systems that offer universal access and structured diabetes care. Moreover, this study found an average anxiety score of 4 for T1D and 3 for T2D, which is considered normal, but around 10% of the individuals with T1D and T2D showed signs of moderate to severe anxiety, where meta-analyses show prevalence rates from 14 to 40% in diabetes populations [23,24]. This is likely due to the population-based nature of this study, which includes more individuals with better glycemic control or more proactive healthcare engagement, reducing the overall burden of psychological distress compared to hospital-based samples. Furthermore, the use of a conservative cut-off score (≥11) on the HADS questionnaire, which omits mild cases, likely contributed to the lower estimate. Given the absence of clinical diagnostic confirmation, this strict threshold was considered appropriate to minimize false positives. Depending on the definition, the T1D Exchange Clinic Registry estimated a depression prevalence of 4–10%, which is comparable to our finding of 5% [25].

The results of this study support previous findings that diabetes is associated with reduced QoL [2,3]. Consistent with Rasmussen (2025), who reported that a high COMPASS-31 score was linked to the lowest QoL in adolescents [8], this study found that individuals with SAD++ had the lowest general well-being scores with increased emotional distress as well. Good physical health is the foundation for the overall well-being, and here, a higher symptom burden, particularly in pupillary and vasomotor domains, was associated with lower SF-36 physical composite scores, reflecting the physical limitations imposed by autonomic dysfunction affecting daily life. Pupillary dysfunction, leading to visual disturbances such as impaired light adaptation, can heighten fall risk, indirectly causing anxiety and loss of confidence in daily mobility, especially when exacerbated by orthostatic hypotension in diabetes [26]. Vasomotor dysfunction with impaired blood vessel regulation, orthostasis, regulation of skin blood flow, temperature sensitivity, and flushing can be quite uncomfortable, particularly in social situations. Autonomic symptoms negatively affect QoL in otherwise healthy individuals [10]. The impact is likely even greater in people with diabetes, who must manage autonomic symptom burden alongside the daily challenges of a complex, chronic condition.

Mental health, as assessed by the SF-36 mental composite score, was similarly affected by autonomic symptoms. Again, pupillary symptoms exhibited the most pronounced negative impact, followed by vasomotor and bladder domains. The unpredictable and often invisible nature of vasomotor symptoms, e.g., sudden flushing or temperature dysregulation, may contribute to feelings of embarrassment or social withdrawal. Symptoms of bladder dysfunction also performed consistently in our models. This is unsurprising, as urinary urgency, frequency, and incomplete emptying are well-documented features of diabetic autonomic neuropathy [27]. Gastrointestinal symptoms, such as fecal incontinence and diarrhea, also contributed to declines in both physical and mental scores (−0.9 and −1.23, respectively). These symptoms can interfere with daily activities, potentially developing coping behaviors that foster social isolation and psychological distress, which would explain the observed correlation with both anxiety and depression scores as well.

Overall, the models explained between 10% and 31% of the variance in health outcomes, underscoring the substantial contribution of autonomic symptom burden to both psychological distress and quality of life measures. Importantly, these findings reinforce that autonomic dysfunction is not merely a peripheral complication but plays a central role in the lived experience of diabetes, affecting both physical capacity and emotional well-being. These findings reinforce the multifaceted burden of autonomic dysfunction, where not only physical discomfort but also the chronic, unpredictable nature of symptoms can erode mental well-being. The findings of this study support the Danish national guidelines [28], which recommend that individuals are directly asked about the presence of autonomic symptoms as including sexual dysfunction, gastrointestinal complaints, angina equivalents, and dizziness upon postural change.

Sample size is a major strength of this study; however, the use of patient-reported outcomes is both a strength and a limitation. While it offers valuable insight into the subjective symptom burden experienced by individuals, these measures cannot be clinically validated within the current design. Moreover, the survey did not collect detailed information on comorbidities or concomitant medications. This precluded adjustment for conditions with symptomatic overlap with autonomic dysfunction, such as pupillomotor symptoms from allergies or hot flashes from menopause. Thus, residual confounding cannot be excluded, and may have contributed to the observed associations. Consequently, there is a risk of symptom misclassifications, and diabetes cannot be confirmed as the underlying cause of the reported symptoms as autonomic neuropathy may also arise from other conditions such as neurodegenerative disorders (e.g., Parkinson’s disease, multiple system atrophy), autoimmune diseases, chronic alcohol use, amyloidosis, vitamin deficiencies, infections, or certain medications. Other peripheral nerve disorders can coexist and contribute to the symptom burden such as Guillain-Barre syndrome, chronic inflammatory demyelinating polyradiculoneuropathy and carpal tunnel syndrome as well. Moreover, the study design preferentially included those with higher health literacy particularly by excluding those exempts from Digital Post, a vulnerable group who may have higher disease severity. Thus, some selection bias exists. The cross-sectional nature of the study also limits causal inference, and a longitudinal design would be necessary to determine whether autonomic symptoms predict changes in quality of life over time. Finally, the study cohort represents a relatively well-managed diabetes population, which may limit generalizability to individuals with more severe disease, suboptimal glycemic control or other countries with different healthcare systems or ethnic composition. In Denmark, diabetes care is tax-funded and universally accessible, with regular checkups and a strong emphasis on preventive care, which may contribute to better disease management compared with settings with more limited healthcare access. However, the findings are likely generalizable to countries with similarly organized healthcare systems.

5. Conclusions

In conclusion, this study highlights the association between autonomic symptom burden and reduced QoL in individuals with diabetes. Autonomic symptom burden was linked to higher levels of anxiety and depression, as well as reduced physical and mental health status, with pupillary, vasomotor, and bladder symptoms emerging as key contributors. These findings emphasize the need to consider autonomic symptoms in diabetes care, as they may play a role in the psychological well-being and overall QoL of the affected individuals.