1. Introduction
Pain is a common and often overlooked issue in critically ill patients, stemming from both the underlying illness and necessary medical interventions. This is especially true for those unable to self-report due to sedation or impaired consciousness. As such, effective pain management remains a crucial aspect of intensive care [
1]. Modern Intensive Care Unit (ICU) pain management emphasises frequent, precise assessment and a multimodal treatment approach. This involves combining pharmacologic and non-pharmacologic strategies to ensure adequate analgesia while minimising sedation [
1].
Pain is a deeply individual and multifaceted phenomenon. If not properly identified, treated, or controlled, it can have serious repercussions for the patient, both immediately and over time. The persistence of this problem is largely due to the subjective character of pain, the wide variation in how it is experienced and communicated among individuals, and the resulting difficulties in achieving accurate evaluation and diagnosis [
2]. Pain has multiple contributing factors and may arise from various triggers such as critical illness, invasive treatments, and routine medical procedures [
3].
Critically ill surgical patients often face poorly managed acute pain, which can lead to chronic pain and other complications [
4]. Pain is a significant concern in these patients, affecting both short-term recovery and long-term health [
4,
5]. Effective management of pain, agitation, and delirium (PAD) is essential in the ICU, especially for surgical patients who may have unique needs [
5]. Insufficient pain control can lead to chronic pain and post-traumatic stress disorder in ICU survivors [
4]. Although opioids are commonly used to manage non-neuropathic pain in these patients, a multimodal approach is recommended to minimise side effects [
5].
As part of the multimodal approach to pain management, clinical supervision in nursing plays a vital role. By providing continuous support and guidance, it helps ensure timely and effective pain assessment and interventions, thereby enhancing patient care and safety in the ICU. Clinical supervision models in nursing play a vital role in enhancing nurses’ competencies and overall performance. Various approaches have been identified, including coaching, mentorship, and individualised learning [
6]. Clinical supervision has proven to be effective in fostering peer support, strengthening professional accountability, and improving nurses’ skills and knowledge [
7]. It is a process of learning and professional support aimed at developing nursing practice through regular discussions with experienced and skilled colleagues [
7].
Among the more recent context-centred clinical supervision models is the SafeCare Model, developed in Portugal. This innovative model emphasises flexibility and contextual adaptation, aiming to improve the quality and safety of care, while also enhancing satisfaction for patients, nurses, and healthcare organisations [
8,
9]. SafeCare promotes a supervisory culture by fostering supportive environments for nursing practice and professional development, identifying key areas where clinical supervision can have a significant impact [
10]. The model is structured around four core phases—situational diagnosis, identification of clinical supervision needs, model implementation, and outcome evaluation—and is built upon four foundational pillars: context, nursing care, professional development, and supervision. Notably, it introduces the concept of the Clinical Supervision-Sensitive Indicator, which serves as a tool to monitor and evaluate the effectiveness of supervision in promoting safer and higher-quality nursing care [
9,
10].
In recent years, context-centred clinical supervision models, such as the SafeCare Model developed in Portugal, have gained increasing relevance. This flexible and adaptive model aims to enhance the quality and safety of nursing care by fostering a supervisory culture and supporting professional development through structured, reflective processes [
9]. Given its focus on improving care environments and outcomes, SafeCare is particularly applicable to ICU—complex clinical settings characterised by advanced technology, critically ill patients, and the need for continuous, specialised nursing care [
11]. Within this context, effective pain assessment in critically ill surgical patients remains a clinical priority and a challenge in polyfunctional ICU settings. Clinical supervision may serve as a valuable strategy to reinforce best practices in this area.
This study aims to: (1) identify how pain assessment is currently performed in critically ill surgical patients in a polyvalent ICU in a hospital in the Porto district; (2) identify the training needs to be addressed in clinical supervision sessions for nurses regarding the Clinical Supervision-Sensitive Indicator—Pain; and (3) compare pain assessment practices before and after the implementation of clinical supervision sessions.
2. Materials and Methods
2.1. Study Design and Settings
A quantitative, quasi-experimental, cross-sectional study was conducted using a pre- and post-intervention design. Based on the SafeCare model, the study followed three key phases: situational diagnosis, implementation of clinical supervision sessions, and outcome evaluation.
The setting was a polyvalent ICU in a hospital in the Porto district, focusing on pain assessment in critically ill surgical patients. The pre-intervention phase identified existing pain assessment practices and training needs related to the Clinical Supervision-Sensitive Indicator—Pain. These findings guided the content of the clinical supervision sessions delivered to ICU nurses.
2.2. Population and Sample
The target population for this study consisted of 31 nurses working in the polyvalent Intensive Care Unit (ICU) of a hospital in the district of Porto, Portugal. All nurses were available and met the inclusion criteria, so the sample was selected using a non-probability total convenience sampling method, based on their availability and willingness to participate during the study period. All eligible nurses were invited, and all agreed to participate, resulting in the inclusion of the entire accessible population. This approach was considered appropriate given the specific operational context of the ICU, as it ensured the feasibility of data collection without disrupting patient care. While this sampling method may limit the generalizability of the findings, the full participation rate contributed to the internal validity of the study.
2.3. Data Collection and Analysis
Data collection was carried out through audits of the Clinical Supervision-Sensitive Indicator—Pain in critically ill surgical patients, using nursing records documented in the B-Simple clinical information system (B-ICU module).
The sensitive clinical supervision indicator, pain, was analysed in terms of its characterisation, assessment and treatment of pain in the critically ill surgical patient.
When characterising the Clinical Supervision-Sensitive Indicator—Pain, an analysis was made of the pain present, at 24 h, 48 h, 72 h, and after 72 h, the type of pain, its location, duration, intensity and the factors exacerbating the pain.
The data analysis was based on nursing documentation related to the characterisation, treatment, and assessment of pain, as defined by the Clinical Supervision-Sensitive Indicator—Pain.
The data collection tool was based on a literature review of national and international intensive care guidelines and the institutional sedation, analgesia, and delirium protocol implemented in the ICU.
This unit, pain assessment for critically ill surgical patients, is conducted according to a standardised schedule: at admission, every 2 h during the first 6 h of hospitalisation, every 4 h from 6 to 48 h, and every 6 h after 48 h of ICU stays.
Pain was assessed using the Numeric Rating Scale (NRS) for spontaneously ventilated patients and the Behavioural Pain Scale (BPS) for those under invasive mechanical ventilation. Reassessment occurred one hour after analgesia and during invasive procedures, when pain was present.
Data collection was conducted in two phases: pre-intervention (November 2022) and post-intervention (May 2023), through audits of Clinical Supervision-Sensitive Indicator—Pain records in 31 total critical ill patients. Between these phases, targeted 6 clinical supervision sessions were held in February 2023, delivered by the principal investigator, involving all ICU nursing teams. Each session lasted 60 min and was considered working time.
Data were analysed using Software Statistical Package for the Social Sciences (SPSS) version 25.0.
2.4. Statistical Treatment
Descriptive statistics were performed by analysing the measures of central tendency in the nurses’ records and assessments and in the non-recording and non-assessment of pain in surgical patients before and after the intervention.
The t-test for paired samples, parametric difference test, was used to confirm that there are statistical differences in the non-assessment before and after the supervisory intervention.
2.5. Ethical Considerations
This study respected the ethical and deontological principles of the nursing profession, as well as the ethical standards for health research outlined in the Declaration of Helsinki and the Personal Data Protection Law. Authorization for data collection was obtained from the hospital’s Board of Directors and the Health Ethics Committee under file number 53/2022, with approval granted by all relevant bodies on 24 November 2022.
3. Results
3.1. Participant Characteristics
The study included 31 ICU nurses, mostly female (55%), aged predominantly between 31 and 36 years. Most held permanent contracts (81%), and 81% were general nurses, with 36% holding a postgraduate nursing specialisation and 19% recognised as nurse specialists. In terms of education, 36% had a postgraduate specialisation, 35% a bachelor’s degree, 19% a master’s, and 10% a postgraduate diploma. The most frequent area of specialisation was medical-surgical nursing.
Regarding experience, 32% had 11–15 years in nursing, and 32% had up to 5 years of ICU experience. In the current ICU, 61% had worked there for less than 5 years. Only 19% had training in clinical supervision, including 13% with advanced competence and 6% with a postgraduate diploma in Clinical Supervision (CS).
3.2. Nursing Records and Audit Scope
The nursing records analysed at both times (pre-intervention and post-intervention) refer to the same nurses. At each time point, the records made by the intensive care unit nurses for surgical critically ill patients in the B-Simple programme were examined, totalling 32 surgical critically ill patients.
In order to adequately analyse the sensitive clinical supervision indicator, pain, we considered the characterisation, treatment and evaluation of the sensitive clinical supervision indicator, pain. In the pre-intervention characterisation of the sensitive clinical supervision indicator, pain, we analysed the pain present, at 24 h, 48 h, 72 h, and after 72 h, the type of pain, its location and duration, its intensity and the factors that exacerbated the pain.
3.3. Pre-Intervention Findings
In the pre-intervention phase, the pain assessment audit focused on three key components: pain characterisation, treatment, and evaluation. Characterisation included the presence of pain at specific time points (24 h, 48 h, 72 h, and after 72 h), as well as its type, location, duration, intensity, and aggravating factors. Treatment and evaluation were assessed based on nursing documentation and adherence to the ICU’s sedation, analgesia, and delirium protocol.
A total of 16 patients had completed 24 h of hospitalisation, 12 patients had completed 48 h of hospitalisation, 7 patients had completed 72 h of hospitalisation, and 5 patients had completed more than 72 h of hospitalisation. In the first 24 h of hospitalisation, eight patients (50%) were identified as having pain, at 48 h, pain was present in two patients (16.7%), at 72 h, one patient (14.3%) had pain, and after 72 h, two patients (40%) had pain (
Table 1).
In critically ill surgical patients, the predominant type of pain is acute generalised pain (56.3%). In terms of location, the abdominal region (43.8%) is the most frequent location of pain, and its duration is predominantly intermittent pain (56.3%). The intensity of generalised pain is predominantly moderate (43.8%), and the main pain exacerbating factor identified is mobilisation/positioning (37.5%).
With regard to the intensity of pain according to the type of ventilation and the pain scale used, it can be seen that in critically ill surgical patients, whether on spontaneous ventilation (
n = 3) or invasive mechanical ventilation (
n = 4), the predominant intensity of pain is moderate (
Table 2).
Regarding the treatment of the sensitive clinical supervision indicator, Pain, (
Table 3), in the pre-intervention, it can be seen that pharmacological treatment and the combination of treatments, pharmacological and non-pharmacological, were used equally in the treatment of pain (
n = 5).
According to the nursing records (
Table 4), the sensitive indicator of clinical supervision in Nursing Pain was assessed and documented in all surgical patients on admission (16 times), every 2 h (18 times), every 4 h (60 times) and every 6 h (43 times). Instances where pain assessment was not recorded were identified during invasive procedures (4 cases) and in the follow-up assessment one hour after analgesic administration (8 cases).
3.4. Post-Intervention Findings
After the clinical supervision sessions, a second audit was conducted three months later, using the same criteria and focusing on the same Clinical Supervision-Sensitive Indicator—Pain components. This allowed for a direct comparison with the pre-intervention results to assess improvements in documentation and clinical practice.
In the post-intervention period, the Clinical Supervision-Sensitive Indicator—Pain was again characterised through statistical analysis, including the presence of pain at specific time points (24 h, 48 h, 72 h, after 72 h), the type of pain, location, duration, intensity and exacerbated factors of pain.
The 15 patients had completed 24 h of hospitalisation, 12 patients had completed 48 h of hospitalisation, 9 patients had completed 72 h of hospitalisation, and 7 patients had completed more than 72 h of hospitalisation.
Within the first 24 h of hospitalisation, pain was documented in 6 patients (40%), at 48 h, pain was present in 3 patients (25%), at 72 h, 2 patients (22.2%) had pain, and after 72 h, 1 patient (14.3%) continued to report pain (
Table 5).
In critically ill surgical patients, the predominant type of pain is generalised acute pain (33.3%). In terms of location, the abdominal region (20%) is the most frequent location of pain, and its duration is predominantly intermittent pain (46.7%). The intensity of generalised pain is predominantly moderate (26.7%), and the main pain exacerbating factors identified are mobilisation/positioning (13.3%) and wound treatment (20%).
Regarding the intensity of pain according to the type of ventilation and the pain scale used, it can be seen that critically ill surgical patients on spontaneous ventilation had pain of severe intensity (
n = 1), and critically ill surgical patients on invasive mechanical ventilation had pain of moderate intensity (
n = 4) (
Table 6).
As shown in
Table 7, the most frequently recorded approach was the combined use of pharmacological and non-pharmacological treatments (
n = 7).
In the post-intervention period, according to the nursing records (
Table 8), the sensitive indicator of clinical nursing supervision, pain, was assessed in all surgical patients on admission (15), every 2 h (15), every 4 h (56), and every 6 h (81). As for the absence of nursing records, this occurred at the time of invasive procedures (1) and in the assessment of pain one hour after the administration of analgesia (1).
According to the analysis of the results obtained for the surgical patient, there are statistical differences in the assessment of the sensitive indicator of clinical nursing supervision, pain, before (Mean = 5.61; Median = 5.00; Standard Deviation = 2.290) and after the supervisory intervention (Mean = 7.26; Median = 6.00; Standard Deviation = 2.543), with an increase in the minimum (4) and maximum (12) value of pain assessments in the post-intervention compared to the pre-intervention.
As documented in the nursing records, there were statistically significant differences in the non-recording and non-assessment of pain in surgical patients before (Mean = 0.39; Median = 0.00; Standard Deviation = 0.761) and after the supervisory intervention (Mean = 0.06; Median = 0.00; Standard Deviation = 0.359), with a decrease in the maximum value (2) of non-assessments of pain in the post-intervention compared to the pre-intervention (3). The results obtained are confirmed by the
t-test for paired samples (
Table 9).
Table 9 shows that the t-value is 2.270, with 30 degrees of freedom and an associated significance of less than
p = 0.031. As such, it can be concluded that the nurses significantly reduced the number of non-assessments of the sensitive indicator of clinical supervision in Nursing Pain in the post-intervention period with surgical patients, since the
t-test (t (30) =2.270;
p < 0.031) indicates statistically significant differences.
These findings suggest that clinical supervision sessions had a positive impact on nurses’ documentation practices and adherence to pain assessment protocols, potentially contributing to improved patient outcomes and safer, more consistent care.
4. Discussion
In order to promote continuous improvement in the quality and safety of care provided to critically ill surgical patients with pain, a systematic audit of nursing records was carried out. This analysis was conducted to identify priority areas for intervention associated with the sensitive indicator of clinical supervision, pain, particularly its management.
The analysis of the nurses’ assessment and records, carried out at two different times (pre- and post-intervention), revealed that, in both cases, the presence of pain decreases between 24 and 72 h, and that patients hospitalised for more than 72 h experience pain more frequently. In the previous study by López-Alfaro et al. [
12], critically ill surgical patients experienced pain during the first 24 h after admission to intensive care.
Although the prevalence of pain remained significant after the intervention, there was a slight decrease in its frequency over the time of hospitalisation. This may indicate an improvement in the suitability and effectiveness of therapeutic interventions, as well as earlier identification and pain control. Uncontrolled pain can contribute to longer post-operative stays in the intensive care unit and readmissions to hospital after discharge [
13].
The presence of pain is related not only to the pathology and complications associated with the critical condition, but also to immobilisation and constant invasive procedures during hospitalisation [
14]. In this context, Herlianita et al. [
15] emphasise that pain can be present even at rest, exacerbated by environmental and psychological factors, with its intensity increasing during invasive procedures.
The type and location of the pain recorded was directly associated with the reason for admission, type of surgery, underlying pathology and comorbidities, with generalised acute pain in the abdominal region, with an intermittent pattern, being the most frequently described [
12]. Pain intensity was influenced by invasive procedures, surgical or traumatic diagnoses, as well as demographic variables [
16].
Pain assessment and management in critically ill patients remain challenging due to communication barriers. Patients with spontaneous ventilation and no sedation who are able to self-report, tend to communicate pain more accurately and report higher levels of pain, compared to patients with invasive mechanical ventilation and sedation [
17]. Nurses face various barriers to effective pain management, including lack of knowledge about assessment tools, patients’ inability to communicate, and absence of standardised protocols [
18]. Facilitators include ongoing education, patient self-reporting ability, and effective collaboration between healthcare professionals [
18].
Among the factors that most exacerbated pain were surgical wound care and mobilisations/positioning, which were often associated with moderate to severe pain. In the previous studies, concluded that nursing procedures such as wound care and positioning increased pain intensity [
19,
20]. Gardner et al. [
21] report that procedures such as dressing changes and wound cleaning are responsible for moderate to severe pain in 74 per cent of surgical patients, 36 per cent of whom experience severe pain according to the numerical scale.
Pain management in intensive care should combine pharmacological and non-pharmacological treatment in order to maximise analgesia and minimise sedation [
1]. The non-pharmacological treatment used was analgesic positioning, therapeutic massage, and cryotherapy. Pre-intervention, significant gaps were identified in pain assessment, namely the absence of records at critical moments, such as after the administration of analgesia or during invasive procedures. Post-intervention, there was an improvement in the frequency and consistency of records, with an increase in the average number of pain assessments (from Mean = 5.61 to Mean = 7.26) and a significant reduction in the number of non-assessments (from Mean = 0.39 to Mean = 0.06), as confirmed by the
t-test for paired samples (t (30) = 2.270;
p = 0.031). These data point to increased adherence to pain assessment and treatment protocols and greater systematisation of clinical practice, the importance of reassessing pain at key moments in care practice, such as after the administration of analgesia and during invasive procedures [
14].
Effective pain assessment and management are associated with better outcomes in critically ill patients, and its correct assessment and documentation is essential. This shows that nurses need to be retrained in pain assessment and management [
3].
The results obtained show that the clinical supervision intervention had a positive and statistically significant impact on the assessment and recording of pain in surgical patients by nurses in the ICU, which indicates an improvement in the adequacy and effectiveness of therapeutic interventions, as well as earlier identification and control of pain. This evidence supports the importance of clinical supervision as a strategy for training and continuous improvement in the quality and safety of care, particularly in the context of intensive care, where pain assessment is a sensitive indicator of the quality of nursing care.
Clinical supervision should also be emphasised as an essential resource for professional development, promoting critical reflection, informed decision-making and adherence to good, evidence-based practices [
10,
22]. In this study, clinical supervision helped to make professionals more aware of the importance of systematic pain assessment, encouraging compliance with institutional protocols, such as the sedation, analgesia, and delirium protocol in this ICU.
In clinical practice, the regular implementation of supervision programmes focusing on sensitive areas such as pain is recommended. In the field of research, it is essential to analyse the impact of clinical supervision on other indicators that are sensitive to nursing practice and to assess the sustainability of these interventions in the medium and long term.
At the level of health policies, the data supports the structured integration of clinical supervision into organisational models and continuing education programmes, promoting patient safety, and more effective and patient-centred care.
This study provides quantitative evidence of the positive impact of clinical supervision on improving pain assessment in critically ill surgical patients, contributing to the advancement of knowledge in this field. However, limitations are acknowledged, namely the small number of participants and the fact that the study was carried out in a single hospital unit, which could compromise the generalisability of the results.