Identifying barriers to nursing management of postoperative stress hyperglycaemia in non-diabetic patients: a TDF-guided qualitative study in a Chinese tertiary hospital

Abstract

Background

Stress-induced hyperglycaemia (SIH) in non-diabetic surgical patients is under-recognised and is associated with infection, prolonged hospital stay, and adverse outcomes. Nurses play a central role in perioperative glucose monitoring, yet barriers to effective SIH management in routine surgical wards remain insufficiently understood.

Objective

To explore surgical nurses’ perceived barriers to postoperative SIH management in non-diabetic patients using the Theoretical Domains Framework (TDF).

Methods

We conducted a TDF-guided descriptive qualitative study in a tertiary hospital in Jiangsu, China. Semi-structured interviews were undertaken with 25 surgical nurses across six specialties. Data were analysed using a hybrid inductive–deductive approach: inductive coding procedures informed by Colaizzi’s analytic steps, followed by deductive mapping to the 14 TDF domains. Trustworthiness was supported through member checking, an audit trail, and reflexive journaling.

Results

Barriers clustered within nine TDF domains, including limited SIH-specific knowledge and skills, unclear professional roles and decision boundaries, low self-efficacy, fragmented interprofessional coordination, environmental and resource constraints, cognitive overload, lack of standardised protocols, and limited patient/family cooperation. Nurses described uncertainty about thresholds and targets for non-diabetic patients, challenges in interpreting glucose trends and supporting insulin-related care, and difficulty prioritising glycaemic concerns within busy ward workflows.

Conclusion

Multiple interrelated barriers hinder the effective nursing management of SIH in non-diabetic patients. Interventions should target modifiable domains through training programs, standardised protocols, decision-support systems, and collaborative care models. Addressing these barriers may improve timely recognition and escalation of postoperative dysglycaemia, supporting patient safety and postoperative recovery.

Stress-induced hyperglycaemia (SIH) refers to a transient elevation in blood glucose (≥ 7.8 mmol/L) triggered by surgical stress in patients without pre-existing diabetes and is a common yet under-recognised perioperative complication [1]. Evidence indicates that perioperative hyperglycaemia in patients without diabetes is associated with increased risks of postoperative complications and worse outcomes; in some studies, the relative risk is comparable to or even greater than that in patients with established diabetes [2–5]. In addition, a proportion of patients with SIH may develop persistent dysglycaemia after discharge, underscoring the clinical relevance of timely recognition and management [6]. In contrast to patients with known diabetes—who typically have clearer glycaemic targets, monitoring routines, and established inpatient pathways—hyperglycaemia in patients without diabetes is more likely to be interpreted as transient or clinically secondary, creating uncertainty about thresholds, targets, and escalation in routine surgical-ward practice [7–9].

Despite these risks, real-world studies have documented quality gaps in perioperative glycaemic screening, monitoring, and escalation, particularly among patients without a prior diabetes diagnosis [7, 10]. Several factors may contribute to under-recognition in routine wards, including diagnosis-label–driven prioritisation, competing postoperative care demands, and limited operationalisation of SIH-specific procedures for non-diabetic patients in everyday workflows [8, 9]. In routine surgical wards, nurses are central to the practical delivery of glycaemic care, including initiating or coordinating monitoring, recognising abnormal values, escalating concerns, implementing glucose-related orders while managing competing perioperative demands, and engaging patients and families to support adherence [7, 9, 11–13]. However, existing barrier-focused evidence has largely been generated in critical care settings or examined postoperative glycaemic management at a broader team level, leaving nurse-specific barriers to SIH management in non-diabetic postoperative patients in routine surgical wards insufficiently characterised [9, 14, 15].

To address this gap, we adopted the Theoretical Domains Framework (TDF), which synthesises constructs from multiple behaviour change theories into 14 domains and is widely used in implementation research to diagnose why evidence-based practices are not consistently enacted [16–18]. Importantly, the TDF supports translation from behavioural determinants (barriers and facilitators) to implementation-relevant strategies (e.g., training, standard operating procedures, audit and feedback, and decision support) to improve clinical practice [17, 18]. Therefore, this TDF-guided qualitative study aimed to explore barriers perceived by surgical nurses when managing postoperative SIH in non-diabetic patients in a Chinese tertiary hospital, providing a theory-informed foundation for designing targeted and feasible implementation interventions.

Methods

Study design

This study adopted a TDF-guided descriptive qualitative design using semi-structured interviews. A descriptive qualitative approach was selected to generate a pragmatic, practice-oriented account of barriers encountered in routine surgical-ward workflows. The TDF was used as an analytic lens to systematically identify behavioural determinants (i.e., barriers and facilitators) underpinning nurses’ SIH-related practices and to inform implementation-relevant intervention development. Reporting followed the Consolidated Criteria for Reporting Qualitative Research (COREQ).

Participants and setting

Purposive sampling was used to recruit participants from surgical wards of a tertiary hospital in Jiangsu Province, China, between January and March 2025. Nurses were eligible if they:

1. Held a valid registered nurse license;

2. Had at least 1 year of continuous work experience in a surgical ward;

3. Had direct responsibility for at least one of the following in routine postoperative care: bedside blood glucose testing, documentation/reporting of glucose values, implementation of glucose-related medical orders (e.g., insulin administration/infusion adjustment as per orders), or escalation/communication regarding abnormal glucose values;

4. Had encountered and managed at least one non-diabetic postoperative patient with suspected or documented SIH during routine work (e.g., performed glucose monitoring and/or implemented related orders and/or escalated abnormal glucose values), enabling them to reflect on SIH management barriers in real practice.

5. Were willing and able to provide informed consent and participate in a one-on-one interview.

Eligibility was confirmed during recruitment using brief screening questions about nurses’ recent postoperative glucose-management experience with non-diabetic patients.

Nurses were purposively recruited from six surgical specialties (general surgery, gastrointestinal oncology surgery, breast surgery, urology, orthopedics, and cardiothoracic surgery). Sampling and preliminary analysis proceeded iteratively until thematic saturation was achieved. Specifically, after approximately 20 interviews, no substantively new codes or themes emerged and the coding framework became stable; an additional five interviews were conducted to confirm stability. No eligible nurses who were approached declined participation, and no participants withdrew after enrolment.

Data collection

An interview guide was developed using the 14 TDF domains [19], informed by relevant literature [9, 14], expert consultation, and pilot testing (Table 1). The draft guide was reviewed by three purposively selected experts (n = 3) with expertise in perioperative glycaemic management, surgical/endocrinology nursing management, and qualitative research. Feedback focused on clinical relevance, clarity, and completeness of domain coverage, and revisions were made accordingly. The revised guide was pilot-tested with two eligible surgical nurses (n = 2) who were not included in the final sample; minor refinements were made to wording, the sequence of prompts, and probing questions to better elicit barriers and contextual influences within each domain.

Table 1.

Sample interview questions mapped to TDF domains

TDF Domain	Representative Questions
Knowledge	“Describe your understanding of stress-induced hyperglycaemia in non-diabetic surgical patients. How did you acquire this knowledge?”
Skills	“What specific skills are needed to effectively manage SIH? How confident do you feel in your ability to perform these skills?”
Social/Professional Role & Identity	“How do you view your role and responsibilities in managing blood glucose for non-diabetic patients who develop hyperglycaemia?”
Beliefs about Capabilities	“Describe situations where you feel most and least capable of managing postoperative hyperglycaemia in non-diabetic patients.”
Optimism	“What is your outlook regarding improvements in SIH management? What gives you hope or concern?”
Beliefs about Consequences	“What are the potential outcomes of inadequate SIH management? How serious do you consider these consequences?”
Reinforcement	“Describe any feedback or recognition you receive related to glucose management. How does this influence your practice?”
Intentions	“What factors encourage or discourage your active involvement in SIH management?”
Goals	“What are your personal or professional goals related to glucose management for non-diabetic patients?”
Memory, Attention & Decision Processes	“Describe challenges you face in remembering glucose monitoring protocols or making decisions about interventions.”
Environmental Context & Resources	“What environmental factors or resources impact your ability to manage SIH effectively?”
Social Influences	“How do colleagues, physicians, or institutional policies influence your glucose management practices?”
Emotion	“Describe your emotional responses when caring for non-diabetic patients with hyperglycaemia. How do these emotions affect your actions?”
Behavioral Regulation	“What strategies—personal or organizational—help you maintain consistent glucose management practices?”

Note: Probes were used for each domain to elicit barriers/enablers (e.g., “What makes this difficult in practice?” “What factors hinder or facilitate your actions?”)

Interviews were conducted face-to-face in a quiet, private room. Written informed consent was obtained prior to participation. Each semi-structured interview lasted 30–60 min and was conducted by two trained nursing graduate student interviewers and TDF-informed interviewing, under supervision of the research team. Interviews were audio-recorded with permission and accompanied by field notes. All interviews were conducted in Mandarin Chinese and transcribed verbatim within 24 h. Transcripts were not returned to participants for line-by-line correction; instead, member checking was conducted on the interpreted themes/findings (see Sect. “Data Analysis and Trustworthiness”). Questions were designed to elicit behavioural determinants, recognising that these may operate as barriers or facilitators depending on conditions (e.g., “What makes this difficult or easier in practice?”).

Data analysis and trustworthiness

Interviewers were two nursing graduate students with formal training in qualitative interviewing and TDF-informed interviewing, and they conducted interviews under supervision of the research team. Data were analysed using a hybrid inductive–deductive approach. First, transcripts were coded inductively using procedures informed by Colaizzi’s analytic steps (e.g., familiarisation, extracting significant statements, formulating meanings, and clustering themes) [20]. Second, the emergent themes were mapped deductively to the 14 TDF domains to organise barriers across individual, social, and environmental levels. Two researchers independently coded transcripts, compared codes and themes, and resolved discrepancies through consensus; NVivo supported data management.

Saturation was assessed concurrently with analysis by comparing newly coded transcripts with the existing codebook. Recruitment ceased when thematic redundancy was observed and subsequent interviews (n = 5) confirmed stability of the coding framework.

TDF integration and analytic decisions. First, themes were generated inductively through Colaizzi’s steps. Second, emergent themes were mapped to the 14 TDF domains to organise barriers across individual, social, and environmental levels. Domains were reported as stand-alone findings only when they contained distinct, barrier-relevant themes with sufficient depth across participants and specialties; overlapping or sparsely represented content was merged with adjacent domains or not reported separately. Reinforcement was largely absent. Optimism and Beliefs about Consequences were briefly mentioned and embedded within other domains rather than forming distinct barrier themes. Emotion was infrequently expressed as an independent determinant and was captured within workload- and uncertainty-related accounts (Environmental Context and Resources; Memory, Attention and Decision Processes). Goals did not emerge as a distinct domain and was therefore incorporated within Intentions and Behavioural Regulation. To minimise the risk of forcing data into predefined categories, we maintained an audit trail of mapping decisions and resolved ambiguous coding through consensus discussions; an illustrative example is provided in Box 1.

Box 1.

Illustrative example of the analytic pathway from raw data to TDF domain (audit trail)

Raw quotation (significant statement): “Glucose management involves more pharmacological interventions, which should be more of a doctor’s responsibility. Where our responsibility ends remains unclear.” (N8)

Formulated meaning: The nurse perceives unclear role boundaries and hesitates to take initiative in SIH management.

Theme cluster: Role ambiguity and unclear professional boundaries in perioperative glycaemic management. Mapped TDF domain: Social/Professional Role and Identity.

Trustworthiness was strengthened through member checking (themes/findings) and peer debriefing (credibility), purposive sampling across specialties and professional titles (transferability), an audit trail and transparent coding procedures (dependability), and reflexive journaling (confirmability). Interviews were conducted by two trained nursing graduate students who were not affiliated with participants’ departments, which reduced hierarchical pressure. Reflexive journals and team debriefings were used to identify and challenge preconceptions and support balanced interpretation.

Ethical considerations

This study was approved by the Ethics Committee of the participating hospital (Approval No. 2025NL-127-01). All participants were informed of their right to withdraw at any time without consequence. Data confidentiality was strictly maintained, and anonymized codes were used in place of names during analysis and reporting.

Results

Sample demographics

Twenty-five nurses from six surgical specialties participated (general surgery n = 6; gastrointestinal oncology surgery n = 4; breast surgery n = 4; urology n = 3; orthopedics n = 4; cardiothoracic surgery n = 4). Most participants were female (23/25) and held a bachelor’s degree (15/25). The median age was 31 years (range 23–48) and the median clinical experience was 7 years (range 1–25). Participant characteristics are presented in Table 2.

Table 2.

Demographic characteristics of participants (n = 25)

ID	Gender	Age (years)	Education	Title	Experience (years)	Department
N1	Female	24	Bachelor	Nurse	2	General Surgery
N2	Female	26	Associate	Nurse	3	General Surgery
N3	Female	25	Bachelor	Nurse	3	GI Oncology Surgery
N4	Female	28	Bachelor	Senior Nurse	4	GI Oncology Surgery
N5	Male	30	Bachelor	Senior Nurse	5	GI Oncology Surgery
N6	Female	29	Associate	Senior Nurse	4	Breast Surgery
N7	Female	32	Bachelor	Senior Nurse	7	Breast Surgery
N8	Female	35	Bachelor	Charge Nurse	10	Breast Surgery
N9	Female	40	Bachelor	Charge Nurse	15	General Surgery
N10	Female	33	Associate	Senior Nurse	8	Urology
N11	Male	31	Bachelor	Senior Nurse	6	Urology
N12	Female	45	Master	Associate Chief Nurse	20	General Surgery
N13	Female	27	Associate	Senior Nurse	5	Orthopedics
N14	Female	34	Bachelor	Senior Nurse	9	Orthopedics
N15	Female	38	Bachelor	Charge Nurse	12	Orthopedics
N16	Female	42	Bachelor	Charge Nurse	16	Cardiothoracic Surgery
N17	Female	29	Associate	Senior Nurse	6	Cardiothoracic Surgery
N18	Female	36	Bachelor	Charge Nurse	11	Cardiothoracic Surgery
N19	Female	23	Associate	Nurse	1	GI Oncology Surgery
N20	Female	26	Associate	Nurse	3	General Surgery
N21	Female	31	Bachelor	Senior Nurse	7	Urology
N22	Female	39	Bachelor	Charge Nurse	14	Breast Surgery
N23	Female	48	Master	Associate Chief Nurse	25	Orthopedics
N24	Female	30	Associate	Nurse	8	General Surgery
N25	Female	37	Bachelor	Nurse	12	GI Oncology Surgery

Findings

Guided by the TDF, we identified barriers to nursing management of postoperative SIH in non-diabetic patients across nine domains. Across domains, nurses described a “diabetes-first” heuristic, whereby glucose care was triggered mainly by a diabetes label and SIH in non-diabetic patients received less attention. Participants reported inconsistent monitoring, delayed escalation, and hesitation around insulin-related care, alongside patient/family scepticism about repeated finger-stick testing. Table 3 provides a summary mapping of the retained TDF domains to the corresponding barrier themes and their SIH-specific manifestations in non-diabetic postoperative care.

Table 3.

Mapping of TDF domains to barrier themes identified in this study

TDF domain	Barrier theme(s) in this study
Knowledge	Limited SIH-specific knowledge and diabetes-first assumptions in non-diabetic postoperative care
Skills	Challenges in insulin-related care and glucose-trend interpretation for non-diabetic SIH
Social/Professional Role and Identity	Role ambiguity in escalation and coordination for non-diabetic postoperative hyperglycaemia
Beliefs about Capabilities	Low confidence in SIH-related actions, particularly in non-diabetic patients
Intentions and Goals	Unclear targets and competing priorities for non-diabetic SIH care
Memory, Attention and Decision Processes	Information overload and interruptions affecting timely SIH recognition and escalation
Environmental Context and Resources	Resource and workflow constraints that deprioritise non-diabetic SIH care
Social Influences	Patient and family scepticism about glucose monitoring in non-diabetic patients
Behavioural Regulation	Lack of ward-level SIH procedures and limited audit/feedback for non-diabetic patients

Knowledge domain: limited SIH-specific knowledge, particularly in non-diabetic postoperative care

Nurses described gaps in SIH-related knowledge, which were particularly evident when postoperative hyperglycaemia occurred in patients without a prior diabetes diagnosis.

Limited awareness of SIH-related guidance. Several participants reported that they had not encountered SIH-specific guidance in practice: “I haven’t seen any guidelines for stress-induced hyperglycaemia management” (N1). Uncertainty about when to intervene was also common: “Patients’ blood glucose rises after surgery, but why it increases and at what level it requires intervention is unclear. I don’t really understand the specific mechanisms” (N2).

Diabetes-focused perceptions of glucose care. Many nurses described a tendency to prioritise glucose monitoring and management for patients with known diabetes, while paying less attention to non-diabetic patients unless values were markedly elevated: “Postoperatively, we mainly focus on diabetic patients’ blood glucose. For regular patients, unless glucose is particularly elevated, we generally don’t pay special attention” (N24).

Skills domain: challenges in insulin-related care and glucose-trend interpretation in non-diabetic SIH

Participants reported difficulties in implementing SIH-related skills, especially interpreting glucose trends and supporting insulin-related care in patients without diabetes, and attributed these challenges to limited training opportunities.

Insulin administration and titration challenges. Nurses described uncertainty when adjusting insulin for non-diabetic patients with postoperative hyperglycaemia: “For postoperative hyperglycaemia, doctors provide a range for us to adjust insulin infusion rates, but for non-diabetic patients, determining timing and methods for glucose trend adjustment remains challenging” (N5).

Limited SIH-specific training. A lack of targeted training was frequently noted: “There’s no training specifically for postoperative glucose management in non-diabetic patients. I don’t feel adequately professional in this area” (N12).

Social/professional role and identity: role ambiguity in escalation and coordination for non-diabetic postoperative hyperglycaemia

Nurses described uncertainty about professional boundaries and escalation responsibilities, particularly when hyperglycaemia occurred in non-diabetic postoperative patients.

Unclear responsibility boundaries. Some nurses perceived glycaemic management—particularly medication-related decisions—as primarily a physician responsibility and were uncertain about where nursing responsibility begins and ends: “Glucose management involves more medication interventions, which should be more of a doctor’s responsibility. Where our responsibility ends remains unclear” (N8).

Limited team collaboration. Participants also reported a lack of structured collaboration mechanisms for SIH management, including the absence of a dedicated glucose management team or unified coordination across specialties: “Our department lacks a dedicated glucose management team. Each specialty works independently without unified coordination” (N14).

Beliefs about capabilities: low confidence in SIH-related actions, particularly in non-diabetic patients

Nurses reported low confidence when responding to abnormal glucose values, especially in non-diabetic patients where thresholds, targets, and insulin-related actions were less clear.

Diminished self-efficacy. Some participants described uncertainty even when physician orders were available: “When encountering particularly high or low glucose levels, despite having physician orders, I still feel uncertain during implementation” (N13). Others emphasised workload pressure and perceived limits to what could be prioritised during busy shifts: “We work very hard daily with many patients. Providing additional attention to glucose management feels overwhelming. Sometimes I think successfully completing postoperative treatment orders is already sufficient” (N17).

Intentions and goals: unclear targets and competing priorities for non-diabetic SIH care

Participants reported that glycaemic targets and monitoring plans were rarely explicit for non-diabetic postoperative patients, and SIH-related tasks were often deprioritised amid competing perioperative demands.

Absence of explicit glucose targets. Nurses reported uncertainty about acceptable glucose ranges, target values, and monitoring frequency for non-diabetic patients with postoperative hyperglycaemia: “We don’t know what glucose range is acceptable for non-diabetic patients. Are there specific target values? Doctors haven’t clearly specified” (N5). This was also reflected in routine goal-setting, which prioritised other postoperative milestones: “In our daily plans, we have goals like ‘ambulation on day X’ or ‘drain removal on day Y.’ Unless the patient is diabetic, we really don’t set glucose targets” (N9).

Competing priorities and limited proactive initiative. Participants reported that attention to moderate hyperglycaemia could be displaced by other urgent tasks during busy shifts: “We handle many tasks daily; glucose monitoring is just one item. Often while measuring glucose, other emergencies arise and attention shifts” (N19). Others described delaying escalation when multiple problems co-occurred: “If I observe hyperglycaemia, I prepare to inform the doctor or address it later, because hyperglycaemic patients often have infection issues, so I focus on those first” (N4).

Memory, attention and decision processes: information overload and interruptions affecting timely SIH recognition and escalation

Nurses described cognitive overload and frequent interruptions that undermined timely recognition, documentation, and escalation of postoperative hyperglycaemia, especially when monitoring and escalation pathways were not standardised.

Cognitive overload in information processing. Participants reported challenges in recalling and integrating information required for glucose management, particularly across shifts: “Each patient’s glucose variations, medication status, dietary conditions—there’s too much information to remember. Sometimes after a night shift, many details are forgotten the next day” (N7). They also described uncertainty when multiple steps and rules needed to be considered: “Glucose management involves too many knowledge points—when to test, how to test, how to analyze results, how to intervene. It’s easy to get confused, especially when busy” (N6).

Interrupted attention in busy ward environments. Participants noted that frequent interruptions and a noisy environment affected concentration during glucose-related work: “The ward is very noisy with various alarms, call bells, and family conversations. Glucose testing is frequently interrupted, affecting concentration” (N15). Similarly, competing demands during a shift could lead to task switching: “One shift involves handling many tasks; glucose monitoring is just one. Often while measuring glucose, other emergencies arise and attention transfers” (N16).

Environmental context and resources: resource and workflow constraints that deprioritise non-diabetic SIH care

Participants described equipment, documentation, and staffing constraints that delayed monitoring and follow-up, with some noting that limited glucometers were often prioritised for patients with known diabetes.

Limited equipment and information system support. Nurses reported that insufficient glucometers and the absence of electronic integration increased the burden of monitoring and documentation: “Our unit has only one functional glucometer, which prioritizes diabetic patients” (N10). Manual transcription of readings into electronic records was also described as time-consuming, particularly during busy periods: “Glucometer readings require manual computer entry; there’s no information system upload to electronic medical records for direct physician review. When we’re busy, timely reporting sometimes isn’t possible” (N22).

Staffing shortages and workload pressure. Participants highlighted that heavy workloads limited the time available for glucose monitoring, recording, and follow-up: “Our department has few nurses; each person manages many patients with heavy workloads. Glucose monitoring, recording, and analysis all require time, but we lack sufficient time” (N23). Night shifts were described as particularly challenging due to low staffing levels: “Night shift has only one person managing over 40 patients in the entire unit. Handling glucose abnormalities is very difficult; we often can’t manage everything” (N11).

Social influences: patient and family scepticism about glucose monitoring in non-diabetic patients

Nurses reported that some patients and families questioned the need for repeated glucose monitoring in the absence of a diabetes diagnosis, which could delay testing and reduce adherence to monitoring plans.

Questioning the need for monitoring among non-diabetic patients. Nurses described resistance when patients or families perceived glucose checks as unnecessary: “When checking patient glucose, they think since they’re not diabetic, why do we need daily finger sticks for glucose testing?” (N21). Some also reported refusal or interruption of monitoring when families prioritised rest in the immediate postoperative period: “Our postoperative patients returning to the ward are basically in a sleepy state. When we try to test glucose, families wave us away, wanting the patient to rest well, refusing our monitoring because they consider it unimportant” (N3).

Behavioural regulation: lack of ward-level SIH procedures and limited audit/feedback for non-diabetic patients

Participants described limited behavioural regulation supports for SIH management, including the absence of standardised ward-level procedures and minimal audit/feedback for postoperative hyperglycaemia in non-diabetic patients.

Lack of standardised protocols. Nurses reported variation in physicians’ instructions and an absence of unified criteria guiding monitoring frequency and escalation: “Each doctor handles postoperative hyperglycaemia differently—some require testing every 4 hours, others every 2 hours. There’s no unified standard for what glucose ranges require specific monitoring protocols” (N25).

Limited quality monitoring and feedback. Participants reported that SIH-related care for non-diabetic patients was rarely audited or evaluated, with organisational oversight perceived as diabetes-focused: “Whether these patients’ glucose management is done well isn’t evaluated by anyone. Nursing department inspections only focus on diabetic patients” (N20). Some noted the absence of routine indicator tracking and feedback for improvement: “Our department has never analyzed glucose management quality indicators like glucose target achievement rates or complication incidence rates. We don’t know how we’re performing” (N18).

Discussion

Knowledge and skill gaps: diabetes-centric mental models and uncertainty in non-diabetic SIH care

This study suggests that suboptimal SIH management in non-diabetic postoperative patients is partly driven by gaps in nurses’ foundational knowledge and practical skills, alongside a diabetes-centric mental model in routine ward practice. Similar deficits in glycaemic management knowledge and confidence have been reported in ICU contexts, indicating that these challenges are not confined to a single setting. However, in surgical wards, the problem may be amplified by diagnostic and decision uncertainty for non-diabetic patients, where clear thresholds, targets, and escalation pathways are less consistently operationalised in daily workflows [8, 15].

In particular, nurses described uncertainty in interpreting glucose trends and implementing insulin-related orders for non-diabetic SIH, reflecting limited opportunities for structured training beyond diabetes-focused routines. This finding aligns with evidence that targeted education and protocol-based training can improve nurses’ competencies in glucose monitoring and insulin titration, and that implementing evidence-based pathways can strengthen knowledge, attitudes, and practice consistency in inpatient hyperglycaemia care [21, 22]. Importantly, the training needs emerging from our data were skill-specific rather than generic, centring on (i) recognising SIH in patients without diabetes and distinguishing it from “expected transient fluctuations”, (ii) interpreting glucose trajectories and identifying when escalation is warranted, and (iii) safely supporting insulin-related care within orders (e.g., implementing infusion-rate adjustment ranges as prescribed and communicating abnormal values promptly).

Importantly, the “transient/low-priority” framing of hyperglycaemia in patients without diabetes may be clinically consequential. A systematic review and meta-analysis in ICU survivors found that stress hyperglycaemia was associated with increased odds of incident diabetes and prediabetes diagnosed ≥ 3 months after discharge, suggesting that stress hyperglycaemia can unmask latent dysglycaemia rather than representing a purely benign, self-limited phenomenon [23]. Although the underlying populations differ from postoperative surgical wards, these data strengthen the rationale for timely recognition, documentation, and appropriate follow-up planning for non-diabetic patients who develop SIH, and they reinforce the need for SIH-specific capacity building for nurses across the perioperative pathway.

Taken together, these results support the need for SIH-specific capacity building that is tailored to routine surgical-ward workflows—emphasising recognition of SIH in non-diabetic patients, practical interpretation of glucose trajectories, and standardised escalation and documentation practices—to improve timely identification and safe responses to postoperative hyperglycaemia.

Role ambiguity, interprofessional coordination, and cognitive workload as behavioural barriers

Our findings indicate that nurses’ SIH-related actions in routine surgical wards were largely framed as task execution under medical orders, with limited clarity regarding decision boundaries (e.g., when to escalate, how to interpret trends, and how to prioritise glycaemic concerns alongside competing perioperative demands). Such role ambiguity may reduce nurses’ perceived autonomy and willingness to initiate proactive escalation, particularly for non-diabetic patients whose hyperglycaemia is often viewed as transient or secondary.

Participants also described fragmented interprofessional communication, suggesting that the absence of structured coordination mechanisms (e.g., agreed escalation pathways, shared targets, or regular case review) can further weaken consistency of SIH management. While some international models allow expanded nursing roles (including nurse prescribing or protocol-driven insulin adjustment), these practices are strongly shaped by local scope-of-practice regulations and may not be directly transferable to all Chinese hospital settings. Although some international systems support expanded nursing roles (e.g., nurse prescribing or protocol-driven insulin adjustment), these models depend on jurisdiction-specific scope-of-practice regulations, credentialing, and governance structures and should not be assumed applicable in Chinese surgical wards. Therefore, the immediate implication is not role substitution, but clearer definition of nursing responsibilities within multidisciplinary workflows and protocol-enabled decision support [24].

In addition, nurses reported substantial cognitive and workload pressures during peak care periods, which may compromise timely recognition, documentation, and escalation of abnormal glucose values. Consistent with cognitive workload and information overload literature, excessive concurrent demands can impair attention and decision quality. These findings support pragmatic strategies such as workload-aware task allocation, streamlined documentation, and workflow-integrated prompts or checklists to reduce cognitive burden and strengthen safe escalation in SIH care [25].

System-level resource constraints and gaps in standardisation

Our findings highlight that resource constraints and limited standardisation jointly hinder consistent SIH management in routine surgical wards. Participants described practical barriers such as limited access to glucometers, heavy workloads, and time pressure, which may delay monitoring, documentation, and escalation. These constraints can interact in a reinforcing cycle—insufficient equipment and staffing increase workflow disruption, which may in turn compromise the timeliness and reliability of glycaemic care [26].

Given the time-sensitive nature of postoperative glucose monitoring, participants’ accounts suggest a need for workflow-integrated supports that reduce reliance on individual vigilance alone. As potential future avenues (rather than immediate recommendations), digital solutions such as automated documentation or electronic prompts may help streamline monitoring and escalation. Technologies including continuous glucose monitoring (CGM) have shown promise in hospital settings for providing more granular data and supporting timely detection; however, implementation in general surgical wards would require local feasibility assessment (e.g., cost, staff training, data integration, and governance) [27]. Similarly, predictive approaches may support early risk stratification and decision-making, but their utility should be considered as a future direction rather than an immediate practice recommendation [28].

At the organisational level, nurses frequently reported reliance on variable physician instructions in the absence of shared targets and standard operating procedures. Establishing SIH-specific ward protocols—such as agreed monitoring frequency, escalation thresholds, and documentation requirements—together with audit–feedback mechanisms, may provide a pragmatic foundation for improving consistency while remaining feasible within current scope-of-practice boundaries.

Limited patient and family support as an implementation barrier

Effective SIH management in non-diabetic postoperative patients often depends on patient and family understanding and cooperation, particularly when repeated bedside glucose testing is required. In our study, nurses reported that some families questioned the necessity of monitoring in the absence of a diabetes diagnosis, which could lead to resistance, delayed testing, or reduced adherence to recommended monitoring plans.

This finding aligns with evidence that health literacy, perceived necessity, and self-efficacy can influence engagement with care and adherence behaviours [29].

Therefore, implementation efforts may benefit from structured, low-burden education strategies embedded into routine perioperative care (e.g., brief bedside scripts, visual aids explaining stress hyperglycaemia and its risks, and targeted counselling at key transition points such as immediate postoperative return to the ward and discharge planning). Such strategies may help reduce misunderstandings and support shared goals between nurses, patients, and families.

In addition, repeated finger-stick testing can be perceived as uncomfortable or burdensome, and finger-prick pain/discomfort has been described as a barrier that can undermine adherence to glucose monitoring in other contexts [30]. In routine surgical wards, this burden may be amplified when patients and families do not perceive an immediate benefit because “the patient is not diabetic,” highlighting the need for brief, standardised explanations that link SIH monitoring to tangible postoperative risks and safety.

Comparison with prior theory-informed studies and contribution of a TDF lens

This study extends the existing literature by applying the TDF to examine nurse-specific barriers to postoperative SIH management in non-diabetic surgical patients—a context that is often managed using diabetes-centric assumptions. Consistent with prior theory-informed work on postoperative glycaemic management at the surgical team level, we similarly identified barriers related to limited knowledge/skills, variable practices, and system constraints [9]. Our findings also align with evidence from critical care settings showing that knowledge gaps and resource/workflow pressures can undermine glycaemic care [15]. Notably, prior work has highlighted that stress hyperglycaemia is frequently under-recognised and deprioritised in patients without diabetes, with clinical attention often driven by diagnostic labels rather than glycaemic trajectories [9].

However, many prior studies either addressed postoperative glycaemic management at a broader team level or did not explicitly characterise nurse-specific behavioural determinants in routine wards; our data add granularity by showing how a “diabetes-first” heuristic operates in everyday surgical workflows and shapes monitoring, escalation, and insulin-support behaviours for non-diabetic SIH. Importantly, the non-diabetic surgical context highlighted several distinct implementation challenges. Nurses described diagnostic and priority uncertainty (i.e., SIH perceived as secondary to other perioperative risks), role ambiguity at nurse–physician boundaries when escalation and titration decisions are required, and patient/family scepticism toward repeated glucose testing in the absence of a diabetes diagnosis. These barriers help explain why SIH may remain under-recognised and inconsistently managed in routine wards, despite its clinical relevance.

In addition, several TDF domains (e.g., Reinforcement, Emotion, Optimism, Beliefs about Consequences) were not retained as stand-alone findings, which may reflect that nurses framed SIH management challenges primarily as capability- and context-driven (knowledge/skills, workload, resources, and standardisation) rather than as motivation- or emotion-dominated barriers. This pattern is broadly consistent with theory-informed implementation work in inpatient settings where capability and environmental determinants dominate, whereas reinforcement- or affect-related domains become more salient when formal audit–feedback and accountability mechanisms are prominent. This pattern is informative for intervention design, suggesting that feasible improvements may be achieved by prioritising workflow-integrated supports and standardised processes over strategies that rely mainly on individual motivation.

Implications for practice and intervention development

Our findings suggest that intervention development should target key modifiable TDF domains (e.g., Knowledge, Skills, Environmental Context and Resources, Social Influences, and Behavioural Regulation). The TDF can support implementation-oriented design by linking behavioural determinants to intervention functions and behaviour change techniques (BCTs) [18, 31].

In practical terms, future interventions should be multi-level: (1) nurse-focused components to strengthen SIH-specific knowledge and skills (e.g., trend interpretation, escalation communication, and safe insulin-support practices per orders); (2) system-focused components to reduce cognitive and workflow burden (e.g., ward SOPs, documentation simplification, equipment/IT support, and audit–feedback); and (3) interdisciplinary components to align shared targets and escalation pathways (e.g., agreed thresholds, clear consult triggers, and case review mechanisms when needed). In the short term, feasible ward-level actions may include SIH-specific education and skills training (e.g., case-based learning and bedside coaching), locally agreed escalation thresholds and pathways, and standard operating procedures supported by audit–feedback [32].

Given evidence that stress hyperglycaemia can be associated with increased risk of subsequent dysglycaemia after discharge, implementation packages may also consider adding a simple documentation-and-follow-up prompt (e.g., flagging SIH episodes in discharge summaries or recommending outpatient glucose reassessment when appropriate), while remaining aligned with local scope-of-practice and institutional policies [23].In the longer term, system-level strategies could focus on workflow-integrated digital supports (e.g., EMR-enabled prompts/alerts [33] and streamlined documentation) and broader information-system integration to reduce reliance on individual vigilance. Overall, a multi-component approach combining training, workflow supports, and organisational reinforcement may be warranted. Future studies should pilot such domain-informed interventions and evaluate feasibility and effectiveness in routine surgical-ward workflows.

Study limitations

Several limitations should be considered. First, this was a single-centre study in one tertiary hospital, which may limit transferability to other organisational contexts. Second, we interviewed nurses only; perspectives from surgeons, endocrinology teams, and patients/families were not included and could further triangulate barriers across the care pathway. Third, as interviews relied on self-report, responses may have been influenced by recall and social desirability; future studies could incorporate observational data or document review to strengthen contextual interpretation. Finally, although we used an audit trail to support transparent mapping, domain mapping may still risk under-representing less frequently articulated determinants; multi-site studies with broader sampling may help further validate domain coverage.

Conclusion

This TDF-guided qualitative study adds to the SIH literature by clarifying nurse-specific, implementation-relevant barriers to postoperative SIH management in non-diabetic surgical patients in routine wards. The findings highlight a “diabetes-first” heuristic in which glycaemic care is triggered mainly by a diabetes label, while hyperglycaemia in non-diabetic patients is often viewed as transient or lower priority, leading to inconsistent monitoring, delayed escalation, and hesitation around insulin-related care. The most urgent actionable implications are to provide SIH-specific training (especially on threshold awareness, trend interpretation, and safe support of insulin-related care under orders), establish clear ward-level escalation pathways and role boundaries, and implement standardised procedures with practical workflow supports (e.g., prompts/checklists and audit–feedback) to improve timely recognition and management of postoperative SIH in non-diabetic patients.

Author contributions

Nan Wang and Jie Zhang contributed equally to this work. Conceptualization: Li Yang and Peibei Duan. Methodology: Li Yang, Nan Wang, and Jie Zhang. Investigation and Data Collection: Nan Wang, Jie Zhang, and Chaonan Fei. Formal Analysis: Nan Wang, Jie Zhang, and Zhenzhen Sun. Validation: Peibei Duan and Li Yang. Writing – Original Draft: Nan Wang, Jie Zhang, and Chaonan Fei. Writing – Review & Editing: Li Yang and Peibei Duan. Supervision: Li Yang. All authors read and approved the final manuscript.

Funding

The authors declare that no financial support was received for the research, authorship, and/or publication of this article.

Data availability

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

Declarations

Ethical approval

This study was approved by the Ethics Committee of Jiangsu Province Hospital of Chinese Medicine (Approval No. 2025NL-127-01). Written informed consent was obtained from all participants before the interviews. All participants were informed about the study’s purpose, confidentiality of information, voluntary participation, and the right to withdraw at any time. Participant anonymity was preserved by assigning identification codes instead of personal information. The study was conducted in accordance with the Declaration of Helsinki and complied with all relevant ethical guidelines and regulations.

Competing interests

The authors declare no competing interests.