Clinical characteristics and associated complications in diabetic patients with Grade 1 stress injury based on a TriNetX Retrospective Study

Abstract

Diabetes mellitus is a global health burden often complicated by Stage 1 Pressure Injury (PI), Peripheral Arterial Occlusive Disease (PAOD), amputation, and mental illness. These complications impair quality of life and increase healthcare demands. This study examined the associations among PI, PAOD, amputation, and mental illness in diabetic patients to enhance clinical strategies. A retrospective cohort study was conducted using TriNetX multicenter data from 2014 to 2023. Diabetic patients with or without Stage 1 PI were analyzed. Kaplan-Meier survival analysis estimated hazard ratios (HRs) and 6-month all-cause mortality for PAOD, amputation, and mental illness. Among 7697 diabetic patients, the HR for PAOD rose from 2.74 to 2.90, while mortality decreased slightly. The HR for amputation increased from 2.74 to 4.24, with minimal change in mortality. Mental illness risk slightly decreased (HR from 1.91 to 1.88), with a moderate reduction in mortality. First-degree PI in diabetic patients is associated with higher risks of PAOD and amputation, and slightly lower risk for mental illness. Early screening and prevention for at-risk individuals are essential to reduce complications and improve outcomes.

Introduction

Background

Diabetes mellitus is an important global health issue, and with changing lifestyles and the gradual aging of the global population, the prevalence of diabetes mellitus continues to climb, posing a major challenge in the health and socio-economic spheres^1,2. The increased risk of diabetes is strongly associated with age, ethnicity, family history, smoking, obesity, and physical inactivity and complications³. Since diabetes accounts for approximately 2–3% of the total healthcare budget in all countries, any increase in its incidence and prevalence would have a significant economic impact⁴.

PAOD is a disease of obstructive atherosclerosis of the lower extremities; patients may experience abnormalities in other vascular suggestions⁵. Diabetes mellitus is one of the risk factors in patients with PAD and is present in 70% of lower limb amputation cases⁶. This disease is commonly seen in diabetic patients, especially those with poor glycemic control leading to an increased risk of PAD⁷. The serious consequences of PAOD include ulceration, and tissue necrosis; it may ultimately lead to amputation of the limb⁸. Pressure injuries are caused by prolonged and sustained pressure on the skin and underlying tissues and are common in elderly patients⁹. Pressure injuries are categorized according to the degree of injury, with grade I pressure injuries being the mildest form, presenting as localized erythema of the skin without skin breakdown¹⁰.

Objectives

Since the number of samples that can be obtained by a single hospital in most of these studies is limited, and the results of the studies may also be limited in patient populations depending on the region, this study aims to investigate the correlation between PAOD and first-degree stress injury and its influencing factors in diabetic patients by using the large-scale, multicenter data from the TriNetX global health research network platform. The aim is to provide a scientific basis for clinical prevention and management of diabetes. This study aims to investigate the correlation between PAOD and first-degree stress injury in diabetic patients and its influencing factors using the TriNetX global health research network platform, in order to provide a scientific basis for clinical prevention and management.

Methods

Study design

Figure 1 shows the steps of data collection in this study using TriNetX and the final results after PSM. This is a retrospective study analyzed using the TriNetX Analytics Network, a global health research and analytics platform that provides researchers with secure and compliant access to a large and diverse set of real-world clinical data. TriNetX provides access to data from a wide range of healthcare providers, including electronic health records, claims, and clinical trial data, providing a comprehensive and highly detailed resource for analyzing the health of TriNetX with data provided from a wide range of healthcare providers, including electronic health records, claims and clinical trial data, providing a comprehensive and highly detailed resource for analyzing patient populations, treatment outcomes and disease patterns.

Fig. 1.

Research flow chart.

The TriNetX user interface can be used to generate groups with specific inclusion and exclusion criteria and balance the potential confounding variables of these groups by using an integrated propensity score matching (PSM) algorithm. The results of the matching are then compared and illustrated by examining the results of interest for a given time period. For more detailed information about TriNetX, its data, sources, and capabilities, please visit https://trinetx.com.

Participants

Inclusion

This study included patients aged 18 years and older with a confirmed diagnosis of diabetes mellitus (DM), as defined by ICD-10 codes E08–E13, between January 1, 2014, and December 31, 2023. Two distinct cohorts were identified: (1) patients who developed a first-documented Stage 1 pressure injury following their initial diabetes diagnosis, and (2) patients with diabetes who did not experience any pressure injury during the observation period. These results were derived from 1,043,838 patients, representing 20.96% of the total cohort of 4,978,463 individuals, with a missing data rate of approximately 79.04%.

Exclusion

To focus on incident outcomes and minimize confounding from pre-existing conditions, patients were excluded if they had a documented diagnosis of peripheral arterial occlusive disease (PAOD; ICD-10: I73), lower limb amputation (CPT code: 1005524), or any mental illness (ICD-10: F01–F99) prior to their diabetes diagnosis. Only cases where these outcomes occurred within one year after the diagnosis of diabetes were included in the analysis.

Statistical analysis

This study employed the built-in propensity score matching (PSM) function within the TriNetX platform to minimize baseline differences between the pressure injury and non-pressure injury cohorts. Baseline covariates used for matching included demographic variables, relevant clinical characteristics, Clinical laboratory values and ICD-10, which were selected based on their potential influence on the outcomes of interest and their prevalence in literature. Following propensity score matching, Cox proportional hazards models were employed to estimate hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) for one-year outcomes. Kaplan–Meier survival analyses were subsequently performed, and group differences were assessed using the log-rank test. As the TriNetX platform does not support imputation of missing data, all analyses were restricted to patients with complete data for the variables included in the matching algorithm. Although formal sensitivity analyses—such as testing alternative caliper widths or employing different matching algorithms—were not feasible due to platform limitations, the quality of matching was assessed through standardized mean differences, and the consistency of outcome patterns across subgroups was examined to evaluate the robustness and stability of the findings.

Results

Demographic and clinical characteristics

This study included 4,978,463 patients with diabetes mellitus, of whom 7697 (mean age 70.3 ± 12.3) had a first-degree stress injury and 4,970,766 (mean age 60 ± 14.6) did not. Table 1 summarizes the key demographic characteristics and comorbidities for each cohort.

Table 1.

Comparison of clinical characteristics of the study cohort before and after propensity score matching.

	Initial populations				P-value	SMD	Propensity score matched populations				P-value	SMD
	Have pressure ulcer N = 7697		Non pressure ulcer N = 4,970,766				Have pressure ulcer N = 7697		Non pressure ulcer N = 7697
Age; mean ± SD, years	70.3	12.3	60.0	14.6	< 0.001	0.762	70.3	12.3	70.3	12.2	0.865	0.003
Sex
Female	3343	45.62%	2,313,517	48.00%	< 0.001	0.063	3343	45.62%	3359	45.86%	0.795	0.004
Male	3985	54.38%	2,506,428	52.00%	0.018	0.027	3985	54.38%	3966	54.14%	0.759	0.005
Race; n, %
White	4809	66.86%	2,555,279	54.74%	< 0.001	0.225	4809	66.86%	4800	65.17%	0.881	0.002
Black or African American	1019	14.17%	736,130	15.77%	< 0.001	0.045	1019	14.17%	1014	13.77%	0.905	0.002
Unknown Race	989	13.75%	1,079,495	23.12%	< 0.001	0.236	989	13.75%	1166	15.83%	< 0.001	0.066
Asian	376	5.23%	297,441	6.37%	< 0.001	0.048	376	5.23%	385	5.23%	0.738	0.005
Laboratory tests, mean ± SD
Glucose	150.6	75.6	162.4	87.1	< 0.001	0.146	150.5	75.6	151.8	76.2	0.378	0.017
Body mass index	30.0	8.7	32.1	7.8	< 0.001	0.261	30.0	8.7	30.8	7.0	< 0.001	0.101
Hemoglobin A1c/Hemoglobin. Total in Blood	7.2	2.0	7.6	2.1	< 0.001	0.205	7.2	2.0	7.3	1.8	0.185	0.039
Cholesterol	139.4	50.0	176.9	51.7	< 0.001	0.738	139.4	50.0	167.5	44.2	< 0.001	0.595
Braden scale total score	18.8	3.5	19.0	2.8	< 0.001	1.016	15.8	3.5	18.3	2.8	< 0.001	0.789

Impact of pressure injury on six-month health outcomes

A higher proportion of patients with pressure injuries developed peripheral arterial occlusive disease (PAOD) at six months (355/7697) compared to those without pressure injuries (137/7697). Kaplan-Meier survival analysis indicated a significantly increased risk of PAOD in the pressure injury group (HR: 2.74, 95% CI: 2.25–3.34; Fig. 2(a) and Table 2). The incidence of amputation was also higher among patients with pressure injuries (29/7697) than those without (11/7697), with a corresponding hazard ratio of 2.74 (95% CI: 1.37–5.48; Fig. 2(b) and Table 2). Additionally, 921 patients with pressure injuries developed mental illness within six months, compared to 514 patients without pressure injuries. Kaplan-Meier analysis showed a significantly elevated risk of mental illness in the pressure injury group (HR: 1.91, 95% CI: 1.71–2.13; Fig. 2(c) and Table 2). A forest plot summarizing these three outcomes is presented in Fig. 3.

Fig. 2.

Kaplan-Meier curves for three diseases at 6 months.

Table 2.

Risk of disease outcome six months after pressure injury versus no pressure injury.

Outcomes	Hazard ratio	P-value
PAOD	2.74 (2.45, 3.34)	0.687
Amputation	2.74 (1.37, 5.48)	0.136
Mental	1.91 (1.71, 2.13)	0.012

Fig. 3.

Forest plot of risk of outcomes.

To highlight the key findings observed at the 6-month follow-up, the following summary outlines the differences in disease incidence and associated risks between patients with and without pressure injuries:

Peripheral Arterial Occlusive Disease (PAOD):

• 355 patients with pressure injuries developed PAOD compared to 137 without;

• HR: 2.74 (95% CI: 2.25–3.34), indicating a significantly higher risk.

Amputation:

• 29 patients with pressure injuries underwent amputation versus 11 without;

• HR: 2.74 (95% CI: 1.37–5.48), suggesting a more than twofold increased risk.

Mental Illness:

• 921 patients with pressure injuries developed mental illness compared to 514 without;

• HR: 1.91 (95% CI: 1.71–2.13), reflecting elevated psychological burden.

Impact of pressure injury on one-year health outcomes

Patients with pressure injuries exhibited a higher one-year incidence of peripheral arterial occlusive disease (PAOD), with 475 cases out of 7697, compared to 181 cases among patients without pressure injuries. Kaplan-Meier survival analyses confirmed a significantly elevated risk of PAOD in the pressure injury group (HR: 2.90, 95% CI: 2.45–3.44; Figs. 2(a), 4(a), and Table 3). Similarly, the risk of amputation at one year was markedly higher in patients with pressure injuries (50/7697) compared to those without (13/7697), with a corresponding hazard ratio of 4.20 (95% CI: 2.30–7.81; Fig. 4(b) and Table 3). In terms of mental health outcomes, 1225 patients with pressure injuries developed mental illness within one year, versus 732 in the non-pressure injury group, indicating a significantly increased risk (HR: 1.80, 95% CI: 1.71–2.06; Fig. 4(c) and Table 3). A forest plot summarizing these three outcome measures is presented in Fig. 5.

Table 3.

Risk of disease outcome one year after pressure injury versus no pressure injury.

Outcomes	Hazard ratio	Log-rank p-value
PAOD	2.90 (2.45, 3.44)	0.602
Amputation	4.24 (2.30, 7.81)	0.104
Mental	1.88 (1.71, 2.06)	0.702

Fig. 4.

Kaplan-Meier Curves for three diseases at 12 months.

Fig. 5.

Forest plot of risk of outcomes.

To highlight the key findings observed at the one-year follow-up, the following summary outlines the differences in disease incidence and associated risks between patients with and without pressure injuries:

Peripheral Arterial Occlusive Disease (PAOD):

• 475 patients with pressure injuries developed PAOD compared to 181 without;

• HR: 2.90 (95% CI: 2.45–3.44), indicating a substantially higher vascular risk.

Amputation:

• 50 patients with pressure injuries underwent amputation versus 13 without;

• HR: 4.20 (95% CI: 2.30–7.81), suggesting a more than fourfold increased risk.

Mental Illness:

• 1225 patients with pressure injuries developed mental illness compared to 732 without;

• HR: 1.80 (95% CI: 1.71–2.06), reflecting a significant psychological burden.

Discussion

Interpretation of one-year disease risk following pressure injury

The findings of this study reveal a significant association between early-stage pressure injuries and an elevated one-year risk of adverse clinical outcomes among patients with newly diagnosed diabetes. Specifically, patients with pressure injuries demonstrated a 2.90-fold increased risk of developing peripheral arterial occlusive disease (PAOD), a 4.20-fold increased risk of amputation, and a 1.80-fold increased risk of psychiatric disorders compared to those without pressure injuries. These results suggest that pressure injuries may serve as an early surrogate marker of systemic vulnerability, reflecting underlying pathophysiological processes such as impaired microvascular circulation, chronic inflammation, and suboptimal glycemic control. This study highlights the need for a paradigm shift in the clinical and administrative management of pressure injuries among diabetic patients—from reactive treatment of localized skin breakdown to proactive prevention and systemic risk mitigation. Addressing pressure injuries in this context may offer a valuable opportunity to interrupt the progression toward severe vascular, surgical, and neuropsychiatric outcomes, ultimately improving patient quality of life and reducing healthcare resource utilization. Smart healthcare technologies have gained increasing popularity in recent years and are now being widely applied across various medical domains, including cardiology¹⁸, cancer screening¹⁹, and clinical care^20–23. Among these applications, the integration of health information systems remains a critical priority. Based on the findings of this study, we propose that incorporating automated alerts into electronic health records—specifically for the early detection of pressure injuries—may enable timely identification and facilitate more proactive intervention strategies. Furthermore, the strong association between pressure injuries and psychiatric comorbidities underscores the importance of embedding mental health screening and support services into comprehensive diabetes management frameworks. While most existing studies have focused on pressure injury prevention and detection in intensive care unit (ICU) settings^12–25, the present study suggests the value of expanding such approaches to broader diabetic populations in routine care environments.

Implications of increased disease risk after pressure injury

Although many studies have pointed out the association between diabetes mellitus and pressure injuries, a number of studies have investigated the role of diabetes mellitus as a risk factor in the development of pressure injuries¹³, and diabetes mellitus and surgical factors have not been adequately considered in traditional models of pressure injuries¹⁴. Diabetes mellitus is also the most common predictor in applied stress injury risk prediction models¹⁵. The association between diabetes and various mental health disorders is becoming clearer¹⁶ and there is a correlation between psychiatric disorders and increased mortality¹⁷. This study found that diabetic patients with first-degree stress injuries occurring with symptoms of psychiatric disorders had a higher mortality rate than those with PAOD and amputation, and that the longer the duration of the concomitant episode, the lower the chances of survival. In patients with amputation, pressure injuries appear among the factors¹⁸, and patients with amputation due to diabetic foot infections have a higher Laboratory Risk Indicator for necrotizing fasciitis (LRINEC) score compared to patients without amputation¹⁹. This study found that the risk of amputation increased with time leading to an increase in risk from 2.74 to 4.24. The risk of PAOD did not affect the risk due to the length of time, whereas the Mental risk was found to decrease with the length of time.

This study was based on data obtained from a large-scale electronic health record (EHR) database (TriNetX), which, although comprehensive, is inherently limited by issues such as incomplete or inconsistent data entry, potential coding errors, and diagnostic inaccuracies. As patient data are recorded by a variety of healthcare providers across multiple institutions, variability in documentation practices may introduce selection bias and compromise the accuracy of clinical classifications. Moreover, this study was unable to investigate in depth the etiological mechanisms underlying the development of pressure injuries in patients with diabetes. Nor could it assess the influence of care-related factors such as nurse-to-patient ratios, mobility protocols, or preventive interventions implemented during hospitalization. Future research should integrate both retrospective and prospective approaches and utilize data from acute care settings to capture more detailed clinical, behavioral, and environmental information to better elucidate these associations. In addition, the study cohort was restricted to individuals with a confirmed diagnosis of diabetes mellitus, which may limit the generalizability of the findings to non-diabetic populations. Future investigations should explore whether similar associations exist in other high-risk groups. By addressing these limitations through more granular data collection, standardized clinical assessments, and prospective study designs, future research can build upon the current findings to inform more targeted preventive strategies and improve patient outcomes.

Conclusions

This study is strengthened by its large sample size, extended follow-up duration, inclusion of real-world data from an ethnically diverse population, and incorporation of laboratory test values. By evaluating outcomes at different time intervals, the study minimized concerns related to reverse causality and demonstrated that diabetic patients with stress injuries—particularly those who later develop peripheral arterial occlusive disease (PAOD) or undergo amputation—face a significantly elevated risk of poor outcomes. Although patients with mental illness showed a slightly lower incidence of stress injuries, their overall survival rate remained lower, suggesting the presence of more complex underlying health burdens. Given these findings, early identification and targeted prevention strategies for diabetic patients at high risk of developing pressure injuries are strongly recommended. Clinicians should consider implementing routine skin integrity assessments and pressure injury risk evaluations as part of standard diabetes care, particularly within the first year following diagnosis. Multidisciplinary management involving endocrinologists, wound care specialists, mental health professionals, and vascular teams may be crucial in mitigating the downstream complications associated with stress injuries in this population. Integrating alert systems within electronic health records to flag at-risk individuals could further support proactive intervention.

Future research should focus on prospective longitudinal studies and randomized controlled trials (RCTs) to evaluate the effectiveness of targeted interventions—such as early mobilization programs, advanced wound care protocols, or mental health integration—in reducing the incidence and severity of stress injuries among diabetic patients. Additionally, studies incorporating patient-reported outcomes and quality-of-life measures would provide valuable insight into the broader impact of these complications and the effectiveness of holistic care models.

Author contributions

Y-HC, P-HC, M-YC, A-LK, M-RW, I-CC: Data acquisition. Y-HC and J-LX: Supervision or Mentorship. S-YH helped revise the manuscript. Y-HC had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Funding

Funding support was also provided by the Nursing Department of TCVGH under the research grant TCVGH-1143701 F.

Data availability

The datasets used and/or analysed during the current study available from the first author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of Taichung Veterans General Hospital (IRB No. CE22308B). This study was conducted as a retrospective analysis of de-identified data. The requirement for informed consent was waived by the Institutional Review Board (IRB) due to the use of anonymized data and the minimal risk to participants. All methods were performed in accordance with the relevant guidelines and regulations or declaration of Helsinki.