Perioperative Nursing Workload, Missed Care, and Near-Miss Events: A Prospective Observational Study

Overview

This prospective, single-center observational study examines the relationship between perioperative nursing workload and safety-critical perioperative nursing care in the operating rooms of Agri Training and Research Hospital. Guided by the Donabedian structure-process-outcome framework, the study will directly observe eligible surgical procedures to record objective, procedure-level workload indicators (interruptions, task-switching, concurrent duties, procedure duration, shift timing, and equipment/supply shortages) together with the observed perioperative nurse's self-reported workload using the Turkish NASA Task Load Index (NASA-TLX), and a structured observation of nine safety-critical perioperative nursing care tasks (identification/procedure/site verification, allergy and critical preoperative information verification, positioning and pressure-injury prevention, normothermia, surgical counts, specimen/label verification, equipment and implant readiness, handoff, and documentation). Any delayed or omitted safety-critical care items, and any near-miss or process-deviation events identified and corrected before patient harm, will also be recorded using standardized forms. The study does not involve patients as direct research participants; only the perioperative work environment and voluntarily participating nurses are observed, and routine clinical care, staffing assignments, and surgical decisions will not be modified. The primary aim is to determine whether higher perioperative nursing workload is associated with a higher rate of missed/delayed safety-critical nursing care at the procedure level; secondary aims examine the association between missed safety-critical care and near-miss/process-deviation events, and, if sufficient events occur, whether missed safety-critical care mediates part of the relationship between workload and near-miss/process-deviation events.

Background and Rationale: Perioperative care is a safety-critical setting characterized by high cognitive demand, time pressure, concurrent tasks, and frequent workflow interruptions. Operating room nurses perform multiple safety-critical care duties within the same clinical workflow, including patient and procedure verification, safe positioning, surgical counts, specimen/label verification, equipment preparation, normothermia maintenance, handoff, and documentation. Missed nursing care has been defined in the nursing literature as any required aspect of patient care that is omitted, either in part or in total, or substantially delayed (Kalisch, Landstrom, \& Hinshaw, 2009). Recent literature calls for mechanism-focused research that moves beyond prevalence estimates of missed care toward understanding why care is missed and what outcomes result (Palese, 2026). Studies examining the perioperative setting through direct, procedure-level observation remain limited. While workflow interruptions and perceived workload have been linked to missed nursing care (Liu et al., 2026), it remains unclear which specific safety-critical perioperative nursing duties are most likely to be delayed or omitted under high workload, and whether this translates into near-miss or process-deviation events. This study addresses workload using both a subjective self-report tool (NASA-TLX) and directly observed objective indicators (interruptions, task-switching, concurrent duties, procedure duration, shift timing, and equipment/supply shortages). Conceptual Model: The study's conceptual model integrates the Donabedian structure-process-outcome framework with the missed nursing care literature: perioperative workload and workflow interruptions represent exposures related to the care environment (structure); missed/delayed safety-critical perioperative nursing care represents the proximal care process; and near-miss/process-deviation events represent the safety outcome. The pre-specified pathway is: workload and interruptions to missed/delayed safety-critical care to near-miss/process-deviation events. A distinguishing feature of this design is that missed nursing care is measured through direct, procedure-level observation of applicable safety-critical tasks rather than retrospective, general self-report, allowing nurse-level perceived workload, procedure-level objective workload, and directly observed care process to be linked within the same multilevel data structure. Aims and Research Questions: The primary aim is to examine the relationship between perioperative nursing workload and procedure-level missed/delayed safety-critical nursing care. Secondary aims are to evaluate the relationship between missed/delayed safety-critical care and near-miss/process-deviation events, to determine the independent contribution of objective workload indicators, and, if a sufficient number of near-miss events is reached, to exploratorily examine whether missed safety-critical care mediates part of the relationship between workload and near-miss/process-deviation events. Research questions: (1) What are perioperative nurses' procedure-level perceived and objective workload levels? (2) How frequently are safety-critical perioperative nursing duties delayed or omitted, and which duties are most affected? (3) Is higher perceived and objective perioperative workload independently associated with a higher Missed Safety-Critical Care Index? (4) Is missed/delayed safety-critical care associated with the occurrence of near-miss/process-deviation events? (5) If a sufficient number of near-miss events is reached, does missed/delayed safety-critical care explain part of the relationship between workload and near-miss/process-deviation events? Hypotheses: H1 (primary): As perioperative nursing workload increases, the rate of missed/delayed safety-critical care at the procedure level will increase. H1a: A higher Missed Safety-Critical Care Index will be associated with an increased likelihood of near-miss/process-deviation events. H1b (exploratory): If a sufficient number of events is reached, missed/delayed safety-critical care will explain part of the relationship between perioperative workload and near-miss/process-deviation events. Setting and Design: This is a single-center, prospective, analytic observational study conducted in the operating rooms of Agri Training and Research Hospital. Routine clinical practice will not be altered, and the researchers will not intervene in clinical decisions, team task allocation, surgical technique, or patient care. Data collection is planned to take place over approximately 8-12 weeks between October 2026 and December 2026, following institutional and ethics committee approval. The unit of analysis is the surgical procedure; because the same nurse may be observed across multiple procedures, the data have a clustered structure with procedures nested within nurses, which will be accounted for using appropriate multilevel models. Participants and Eligibility: The study population consists of perioperative nurses working in the operating rooms of Agri Training and Research Hospital and the eligible surgical procedures in which participating nurses are actively on duty during the data collection period. Sampling is non-probability, consecutive sampling of eligible procedures occurring during pre-defined observation shifts. Nurse inclusion criteria: actively on duty as a perioperative nurse at Agri Training and Research Hospital operating rooms; at least 3 months of operating room experience; voluntary agreement to participate and to complete the post-procedure NASA-TLX form. Nurse exclusion criteria: extended leave or medical report resulting in inactive duty during the data collection period; withdrawal from the study; failure to complete the post-procedure workload form. Procedure inclusion criteria: performed in the Agri Training and Research Hospital operating rooms; involves a participating perioperative nurse in active duty; is an elective or planned surgical procedure; and at least a portion of the safety-critical care items are directly observable. Procedure exclusion criteria: emergency/resuscitative circumstances that would affect the clinical flow being observed; procedures in which observation could not be completed; and observation records missing core variables. A minimum of 300 evaluable procedures is targeted, with data collection continuing for up to 330 procedures to allow for an anticipated data loss of approximately 10%. For inter-rater reliability, approximately 15% of procedures in the main dataset will be independently and simultaneously coded by two researchers, with percent agreement and Cohen's kappa calculated for categorical items and an appropriate intraclass correlation coefficient calculated for count/duration variables; approximately 10 procedures will be used as a pilot/standardization period before data collection begins and will not be included in the study sample. No patient identifiers, national identification numbers, or protocol numbers that could identify individual clinical outcomes will be collected; each procedure and nurse will be assigned a de-identified research code. Data Collection Tools: Data will be collected using a four-part standardized toolkit. (1) Procedure and Objective Workload Observation Form: a researcher-developed form (not a validated psychometric scale) recording procedure and nurse research codes, date/time, surgical branch and procedure group, surgical approach (open, laparoscopic/endoscopic, or other), procedure duration, shift timing (day, evening/night, or overtime), the observed nurse's operating room experience, the number of cases in that shift, the total number of interruptions, the number of interruption-related task-switching events, the number of concurrent-duty events, equipment or supply shortage events, workflow interruptions due to door traffic or external communication, and an indicator of time pressure, along with free-text observation notes. (2) NASA Task Load Index (NASA-TLX), Turkish adaptation: a widely used self-report instrument assessing perceived workload across six subscales (mental demand, physical demand, temporal demand, performance, effort, and frustration), each scored on a 0-100 scale, with the raw NASA-TLX score calculated as the arithmetic mean of the six subscale scores; the Turkish adaptation and its validity/reliability were established by Sonmez, Oguz, Kutlu, and Yildirim (2017); the original instrument was developed by Hart and Staveland (1988). The form will be completed by the observed perioperative nurse as soon as possible after the procedure is completed. (3) Safety-Critical Perioperative Nursing Care Observation Form: a standardized, procedure-level process observation tool developed by the researchers based on current perioperative safety practices and guidelines; it is not a psychometric self-report scale. Only items applicable to the given procedure are scored, using the categories timely/fully completed, delayed/partial, not done, or not applicable, across nine safety-critical domains: active participation in patient identification, procedure, and surgical-site verification; verification of allergy and critical preoperative information; positioning and pressure-injury prevention measures; normothermia/heat-loss prevention measures or verification; completion of the surgical count according to institutional protocol; specimen/label verbal verification; verification of equipment/implant/material readiness; completion of safe postoperative handoff information; and completion of required perioperative documentation. The primary outcome, the Missed Safety-Critical Care Index, is calculated per procedure as the number of delayed/partial and not-done applicable safety-critical items divided by the total number of applicable safety-critical items. (4) Near-Miss and Process Deviation Log Form: records safety deviations that were identified and corrected before patient harm and that required re-verification of a safety-critical process (for example, a count discrepancy requiring a recount, late identification of missing equipment, information corrected during a time-out, specimen/label correction, or late-completed handoff information); events are coded as indicators of process safety rather than individual fault, noting when the event was identified, whether it was corrected without patient harm, whether re-verification or repeat of a step was required, and the researchers' classification of the event as a near-miss, a process deviation, both, or unclassifiable. Data Collection Procedures: Data will be collected only after ethics committee and institutional approvals have been obtained. Written informed consent will be obtained from participating perioperative nurses; participation is entirely voluntary, and non-participation or withdrawal will not affect employment conditions or institutional rights. During an eligible procedure, the researcher will record objective workload indicators and safety-critical nursing care directly onto the standardized forms from a position that does not interfere with operating room workflow, without providing real-time performance feedback to the team and without intervening in routine care. The observed nurse will complete the Turkish NASA-TLX form as soon as possible after the procedure. Near-miss/process-deviation events will be recorded directly during observation, with the routine procedural record checked only when necessary to confirm whether an event occurred. Event classification will be reviewed against pre-defined categories by the researchers before analysis. No individually identifying data (patient identity, staff registry number, or individual performance ratings) will be entered into the analytic dataset, and individual observation results will not be reported to supervisors, colleagues, or third parties. Data Analysis: Data will be analyzed using IBM SPSS Statistics and R. Prior to analysis, a data dictionary, the primary exposure, the primary outcome, and key confounders will be finalized, and missing-data patterns, distributional characteristics, and outliers will be examined. Descriptive analyses will summarize categorical variables as counts and percentages and continuous variables as mean +/- standard deviation or median (interquartile range); NASA-TLX total/subscale scores, objective workload indicators, the number of applicable care duties, the missed/delayed care rate, and near-miss event frequency will be reported with 95% confidence intervals. The primary outcome, the Missed Safety-Critical Care Index, will be modeled at the primary analysis stage using an appropriate generalized linear mixed-effects binomial model accounting for random intercepts at the nurse level; the primary exposure is the raw NASA-TLX total score, with objective workload indicators as secondary exposures. The primary model will additionally adjust for pre-specified confounders including procedure duration, surgical approach/procedure group, shift timing, the nurse's operating room experience, and the number of staff on duty concurrently; continuous variables will be checked for linearity of association, and model results will be reported as adjusted effect estimates with 95% confidence intervals. The secondary analysis will evaluate the relationship between missed/delayed safety-critical care and the occurrence of at least one near-miss/process-deviation event using mixed-effects logistic regression; if a sufficient number of events is reached, the exploratory pathway workload to missed/delayed safety-critical care to near-miss/process-deviation events will be examined using multilevel mediation analysis; if the event count is insufficient, mediation modeling will not be attempted and associations will be reported as separate models. Inter-rater agreement in the double-coded subsample will be assessed using Cohen's kappa and percent agreement for categorical variables and an appropriate intraclass correlation coefficient for count/duration variables. Statistical significance will be defined as a two-sided p\<0.05; interpretation will emphasize effect sizes and 95% confidence intervals over p-values alone. Anticipated Difficulties: Direct observation in the operating room setting requires substantial researcher time and attention; the same nurse being observed across multiple procedures produces clustered data; awareness of being observed may introduce a Hawthorne effect; some safety-critical care items are not applicable to every procedure; near-miss events are expected to occur at a relatively low frequency, which may limit some analyses; and completing the NASA-TLX after the procedure during a busy workflow may result in delayed or incomplete forms. Limitations: The single-center, observational design limits causal inference and generalizability to other institutions. Conversely, prospective, procedure-level direct observation, combined with multilevel analysis of clustered repeated observations of the same nurse, and the pairing of self-reported workload with directly observed objective process indicators, strengthens the methodological rigor of the study. The Missed Safety-Critical Care Observation Form is a behavioral, procedure-level observation tool rather than a psychometric self-report scale; therefore, internal consistency statistics such as Cronbach's alpha will not be used as a primary quality indicator. Quality will instead be supported through clear operational definitions, observer training, pilot standardization, and inter-rater agreement in a subsample. This approach is intended as a methodological complement to the self-report-weighted emphasis of the existing perioperative missed-care literature. Ethical Principles: The study will be conducted only after obtaining the required written approvals from the Agri Ibrahim Cecen University Scientific Research Ethics Committee and from Agri Training and Research Hospital. As the study is observational, routine perioperative care, task allocation, and clinical decision-making processes will not be modified, and observation records will not be used for quality or disciplinary monitoring purposes. Written informed consent will be obtained from participating perioperative nurses; participation is entirely voluntary and non-participation or withdrawal will not affect working conditions or institutional rights. Nurses' names or registry numbers will not be entered into the analytic dataset, and individual observation results will not be disclosed to supervisors, coworkers, or third parties. Patients are not direct participants in this research: patients will not undergo any additional procedure, will not be interviewed, and no data determining patient identity will be collected; procedures will be recorded using an anonymous research code only. The study will be conducted in accordance with the relevant principles of the Declaration of Helsinki and applicable legislation on the protection of personal data.