This study employed an analytical cross-sectional design.

National District Hospital is a 197-bed facility in Bloemfontein, Free State Province, South Africa. The facility provides district-level services to the greater Bloemfontein region, serving a community of about 567 000.¹³

The study population consisted of 256 healthcare professionals delivering direct professional patient care at NDH. No sampling was done, as the entire study population was targeted for participation.

The inclusion criteria:

The exclusion criteria:

The Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS) 2.0 Teamwork Perceptions questionnaire (T-TPQ) was used. This is a validated tool that categorises core components of effective teamwork into five domains: team structure, leadership, mutual support, communication, and situational monitoring (see Table 1 for domain definitions).^{8,14,15,16,17,18,19} The T-TPQ assesses these domains with seven statements per domain on a 5-point Likert scale, where 1 = ‘Strongly disagree’ and 5 = ‘Strongly agree’.^10,12

The questionnaire, available only in English as per the hospital policy on communication, consisted of statements confirming participants’ eligibility to participate, two relevant demographic questions, and the T-TPQ. Questionnaires were distributed to all healthcare professionals over a 5-day period (08 April 2024 – 12 April 2024).

To maximise response and avoid disrupting service delivery, time slots covering both day and night shifts were arranged with each department in advance. Departmental heads or supervisors indicated the number of staff available to participate. Questionnaires were completed in departmental tearooms to ensure adequate privacy, anonymity, and voluntary participation. Complete questionnaires were folded and placed into a sealed box, thereby implying informed consent.

A pilot study was conducted 3 months before the main study with 10 participants (including nursing staff, allied health professionals, and doctors) at NDH to test the questionnaire. The biostatistician from the Department of Biostatistics, University of the Free State (UFS), confirmed that these responses could be included in the results as no adjustments were made to the questionnaire content.

The principal researcher transferred the data from the paper-based questionnaires to the Protection of Personal Information Act (POPIA)-compliant REDCap platform hosted on the UFS’s secure server. Incomplete questionnaires were discussed individually with the allocated biostatistician to determine usability. The biostatistician accessed and analysed the data using SAS statistical software (version 9.4). The numerical variables were summarised by means, standard deviations (s.d.), minimum, and maximum, and the categorical variables were summarised by frequencies and percentages. Differences between numerical variables were evaluated using the Wilcoxon Two-Sample test for unpaired data. A p-value of p < 0.05 was taken to be statistically significant.

Full ethical approval was obtained from the Health Sciences Research Ethics Committee (HSREC) of the University of the Free State. The ethical approval number is UFS-HSD2023/1664/2811. Permission was also obtained from the Head of the Free State Department of Health. The Agency for Healthcare Research and Quality (AHRQ), granted permission to use the TeamSTEPPSO 2.0 Teamwork Perceptions Questionnaire (T-TPQ) and the TeamSTEPPS^® 2.0 Teamwork Perceptions Questionnaire (T-TPQ) Manual. Implied informed consent was obtained from all participants to promote anonymity and confidentiality.

The overall mean teamwork perception score among healthcare professionals were 3.7 (74%). Table 2 shows the overall teamwork perception results for the five teamwork domains.

All domains scored relatively similarly, between 3.6 (72%) and 3.9 (79%). The communication domain scored the highest (3.9; 79%). The situational monitoring domain scored the lowest (3.6; 72%), closely followed by the leadership domain (3.6; 73%).

The total analysed sample was n = 129; however, for inter-disciplinary analysis, one questionnaire was excluded due to the participant not indicating their healthcare discipline, resulting in an n = 128.

Table 3 shows the difference in perception of teamwork among healthcare disciplines. There was no significant difference in the perception of teamwork among the healthcare disciplines (p = 0.434). The highest mean score for the perception (4.1; 81%) was reported by the dieticians (n = 2) and occupational therapists (n = 2), and the lowest scores were recorded by doctors (3.5; 70%) and pharmacy staff (3.6; 71%).

Table 4 shows the differences in the perception of teamwork among healthcare departments. There was no significant difference in the perception of teamwork among the different healthcare departments (p = 0.066). The Thusong trauma centre, NDH’s unit for survivors of abuse, scored the highest (4.7; 94%), while the lowest scores were recorded by the outpatient department (3.3; 66%) and casualties (3.5; 70%).

Table 5 and Table 6 summarise the mean scores, highlighting specific areas of strength and weakness across disciplines and departments.

The biggest variance was seen in the leadership domain (Table 6), which ranged from 4.5 (90%) for occupational therapists to a low score of 2.2 (44%) for physiotherapists. Doctors consistently reported the lowest scores across most domains (team structure, situational monitoring, mutual support, and communication), verifying their overall score seen in Table 3.

The Thusong trauma centre recorded the highest scores, achieving two perfect means of 5.0 (100%) in the leadership and communication domains. The lowest scores recorded in the leadership domain were recorded by the outpatient (2.3; 47%) and physiotherapy (2.2; 44%) departments. Thus, leadership appears to be a vulnerable domain for these areas.

Looking closer at the average scores for the different teamwork domains at NDH (Table 2), communication was identified as the strongest domain at NDH (3.9; 79%). This indicates staff are generally capable of interacting with one another and with patients in a timely and understandable manner. In comparison, NDH’s casualty communication score of 73% contrasts strongly with 52% in an emergency department reported in a similar study by Weaver et al.²², emphasising that results cannot be generalised.

Two areas for potential growth were identified: situational monitoring (72%) and leadership (73%). This implies staff members may have difficulty recognising important changes in their work environment, such as bed availability, changes in a patient’s condition, or a decrease in staff performance. This finding may also reflect a generalised lack of attention to detail and thoroughness within the clinical environment, which the 66 incomplete questionnaires may support. While the leadership domain speaks to a leader who is present, approachable, and leads by example, the 73% suggests a systemic opportunity to improve role clarity, task assignment, and psychological safety. These findings contrast with the diverse domain strengths and weaknesses found in other healthcare environments.^9,20,21,22 For example, NDH’s Theatre team rated situational monitoring at 86%, far exceeding the 41% reported for an operating room in a study by Jonas et al.²¹

Despite the variations in the mean scores among the healthcare disciplines at NDH (n = 128), no statistically significant difference was found (p = 0.434), possibly indicating that the observed differences are due to chance rather than systematic inter-disciplinary issues.

The lowest mean scores were recorded by doctors (70%) and pharmacy staff (71%). These are arguably the two disciplines with the most dynamic changes in the number and variety of patient contact. Appreciating this, both professions rated situational monitoring as their lowest teamwork domain (Table 5), stressing the importance of communicating changes in workload among team members. Conversely, the highest mean scores were reported by occupational therapists and dieticians. However, these results should be interpreted with caution due to the small sample sizes (n = 2). It was interesting to note that the two highest-scored domains for the two disciplines were leadership and team structure (Table 5), which may reflect the functional simplicity and clear roles typical of small, self-contained units.

A critical, localised finding emerged from the domain breakdown: physiotherapists (n = 2) reported the lowest score across the entire study in the domain leadership (2.2; 44%). This finding, despite their high communication score, suggests failure in perceived leadership within that specific unit, warranting investigation. This specific finding shows precisely why measuring teamwork perception is so valuable: it directs managers to critical areas needing immediate action.

By comparing departments, while not achieving statistical significance (p = 0.066), a trend became clear that departmental context may influence teamwork perceptions.

Once again, the high-patient-turnover departments, the outpatient department and casualties, reported the lowest mean scores. Despite high communication scores (85%), the low leadership score (47%) in the outpatient department suggests staff feel they can communicate effectively but lack guidance or support from leadership. On the other hand, casualties displayed moderate leadership (74%) but appeared to lose sight of situational monitoring (67%), likely due to inherent patient pressure and flow in an emergency setting.

The Thusong trauma centre provided an example of positive deviance, recording the highest mean score (94%) and perfect scores in leadership and communication (100%, Table 6). This result may be attributed to the department’s small size, highly specialised purpose, and sense of mutual support necessary for dealing with sensitive patient issues.

Firstly, the high volume of excluded data must be acknowledged. Of the 195 returned questionnaires, 66 (33.8%) were excluded due to incompleteness (n = 62) or being returned empty (n = 4). The final analysis represents only 50.4% of the total target population (N = 256). The representativeness of the result may, therefore, have been affected by non-response bias. In the healthcare setting, individuals with high workloads or those who feel less engaged with their teams are often less likely to complete surveys. The findings might over-represent the views of those more motivated or satisfied, potentially skewing the teamwork perception scores. However, the diversity of the final sample suggests a broad range of professional perspectives.

Secondly, sample size and anonymity were concerns. Several disciplines and departments have only a few staff members. While measures were taken to protect confidentiality, this limited size may have introduced response bias and inhibited honest reporting, as participants may have been wary of identification. Thus, the outlier results observed in small groups (e.g. Thusong trauma centre, physiotherapy) must be interpreted with caution.

Thirdly, instrumental limitations also exist. The T-TPQ has not been formally validated in the South African healthcare context, potentially affecting reliability of these results.

Finally, the results are specific to the NDH context. The findings cannot be generalised to all hospitals in South Africa, as teamwork perceptions are known to vary across different levels of care (e.g. primary and secondary) and sectors (e.g. private facilities versus public facilities).

From this study, the recommended focus for quality improvement must be directed towards the two weakest domains. To address the low leadership scores, training focusing on functions such as task assignment and effective delegation should be organised for managerial staff in the two lowest-scoring departments, specifically.²³ A confidential management review should be conducted simultaneously to understand the structural factors contributing to these lower scores. To improve situational monitoring hospital-wide, NDH should introduce formal communication tools, such as TeamSTEPPS ‘Check-Backs’ and ‘Handoffs’, to verify information during high-risk transitions.²⁴ This improvement in formal process links directly to the need for better data accuracy and thoroughness.

Additionally, the hospital should leverage its areas of excellence by using the Thusong trauma centre as an internal benchmark for positive teamwork. This could involve facilitating workshops led by Thusong staff to teach other departments the organisational factors that contribute to their effective teamwork.

Finally, we recommend that future research projects formally validate the T-TPQ tool for the South African context. Qualitative studies (focus groups or interviews) should also be conducted to explore the reasons for specific low scores reported by high-volume disciplines, and the extreme deficit observed in one specific discipline.