Regarding organizational outcomes as related to the SEIPS 2.0 model, improvements were found in two patient safety culture dimensions after the first six months of this comprehensive intervention. No improvement was found in professional outcome after the first six months, as measured by perceptions of teamwork. After the full 12 months, however, improvements were found in both professional and organizational outcomes. Improvement in professional outcomes were shown in three out of four perceptions of teamwork dimensions. Regarding organizational outcomes, improvements were found in three patient safety culture dimensions. These results indicate that the team training program had an effect after 12 months of implementation. The GLMM estimates demonstrated an effect of time on the patient safety culture dimensions (organizational outcome) “Organizational Learning and Continuous Improvement” and “Communication Openness” after 6 months, and the estimates also demonstrated that physicians had an effect on the patient safety culture dimensions “Frequency of Events Reported” and “Patient Safety Grade”. Furthermore, the teamwork dimension “Mutual Support” was associated with “Patient Safety Grade” after 12 months of intervention.
No significant improvement in T-TPQ measures after six months may be explained by the fact that few of the TeamSTEPPS tools had been implemented by that point. We expected to find improvement in “Communication” after 6 months since the tools Closed-loop and SBAR (Situation, Background, Assessment, Request or Recommendation) were implemented in the work system in an early phase of the intervention. After 12 months of intervention, however, the results showed improvement in three teamwork dimensions (“Situation Monitoring”, “Mutual Support”, and “Communication”). The Cross monitoring strategy was implemented after five months, and the STEP tool was implemented after nine months [40], so the improvement in “Situation Monitoring” may be due to the implementation of these tools. “Situation Monitoring” is about continuously scanning the environment for important information, watching out for other team members, exchanging relevant information, and jointly reevaluating patient goals [41]. The improved scores in “Mutual Support” may be a result of the “Task Assistance” and “Two Challenge Rule” strategies that were implemented in the work system during the study period [40]. “Mutual Support” is about cautioning each other about potentially risky patient safety situations and about assisting one another during high workloads [41]. When seeing these improvements in teamwork dimensions from a system perspective, they are seen as improved professional outcomes (see Fig. 1). Previous studies from the context of surgical wards that have measured self-reported teamwork have produced ambiguous results [55–57]. Paull et. al [57] found improvement in all scores in their multicenter study when the scores were measured immediately after the training. Study results collected a short time after a team training may benefit from the positive experience the participants have just had and can be seen to reflect a strong Hawthorne effect [58]. The reason why we did not see improvements in team decision making in our study may be due to the time points for measuring. Previous studies that showed enhanced scores in decision making measured two weeks and two months after simulation training [59, 60 ]. Our results for team decision making may also be explained by the fact that the TeamSTEPPS program does not emphasize decision making, and therefore, there was not a focus on this important aspect of teamwork in the intervention.
The organizational outcome measured by the patient safety culture results (HSOPS) showed improvement in “Organizational Learning & Continuous Improvement” and “Communication Openness” after six months of intervention, and the improvement in the latter was also sustained after 12 months, both of which are interesting results. “Communication Openness” is a measure of whether staff freely speak up if they see something that may negatively affect a patient and if they feel free to question those with more authority than themselves [61]. This result is therefore of importance regarding the patient safety culture in the ward, as it may contribute to catching an adverse events before it reaches a patient. Regarding whether the healthcare professionals reported diverse types of adverse events in our study, the average answer was “sometimes” at all data collection times, while the registered adverse events increased during the study period. An increase in adverse events is not desirable but may be seen as an improvement in the reporting culture. The main purpose of reporting is to learn from adverse events [62], and learning is an important part of the human factors approach to patient safety. After six months, improvements were found in organizational outcomes (in two patient safety dimensions). After the full 12 months, improvements were found in both organizational outcomes (three patient safety culture dimensions) and professional outcomes (three teamwork dimensions). The mixed model estimates demonstrated that physicians had effect on two patient safety culture measures. Furthermore, results showed that teamwork was associated with Patient Safety Grade [63]. The improvement in the HSOPS dimension “Organizational Learning – Continuous Improvement” (organizational outcome) may indicate that the healthcare professionals perceived their ward as a learning unit. This result also supports the mixed model estimate, which demonstrated that the time had an effect on “Organizational Learning & Continuous Improvement” after six months. The estimates also demonstrated that the healthcare professional`s perceptions of “Communication Openness” were affected by time (six months), which corresponds with the results from the t-test analyses, where “Communication Openness” showed significant improvements after both 6 and 12 months. The “Frequency of Events Reported” and “Patient Safety Grade” were affected by the physicians which is an interesting finding since it is often challenging to involve physicians in interprofessional interventions in wards [64]. All professions were trained together, which may have influenced the professional and organizational outcomes in a positive way. In addition to the sustained improvement in “Communication Openness”, two more dimensions of HSOPS were improved after 12 months: “Teamwork Within Unit” and “Manager’s Expectations & Actions Promoting Patient Safety”. Management and leadership are important enablers in achieving effective teamwork and patient safety in complex organizations [65], and improvement in these three dimensions of the patient safety culture may enable further work and future improvement in the other patient safety culture dimensions in the surgical ward.
Our improved patient safety culture results in three dimensions of the HSOPS (organizational outcome) are in line with those from previous research in diverse hospital contexts. Two multicenter studies found improvement in three HSOPS dimensions when measured after 12 months [66, 67], and Thomas and Galla [64] found improvements in three HSOPS dimensions after 2 years. Schwartz, Welsh [67] found a decrease from 6 to 12 months in their multicenter study, a decrease they explained with a need for early refresher training.
The professional outcome “Mutual Support” was associated with “Patient Safety Grade” at the end of the study period, which is interesting from a human factors perspective since this T-TPQ dimension encompasses items focusing on patient safety and emphasizes the strong patient safety aspect of the TeamSTEPPS program.
The use of the conceptual framework contributed to an enhanced understanding of the system approach in our study, which is important to implement and sustain innovations [68]. When implementing teamwork tools such as SBAR, Closed-loop, and Cross-monitoring [40] in the work system, the use of the tools and strategies in the clinical work processes have influenced professional outcomes indicating that the teamwork competencies of the healthcare professionals improved during the study period. Transfer of the learning from team training is crucial to patient safety and interesting from a human factors perspective, as outcomes are influenced by the learning-to-transfer pathway [69]. The improvement in organizational outcomes (patient safety culture) may be due to the TeamSTEPPS intervention in the work system (see Fig. 1). In this study, the implementation was conducted by the master trained leaders and the champions on the change team, which may have contributed to the transfer and sustainment of this human factors innovation initiative.
Study limitations
The study has some limitations. The lack of randomization and controls may have threatened the internal validity, although a pre-post design is useful where there are practical barriers to a randomized design [70]. The study samples were small, but the response rates were satisfying, without risk of response bias. Because of the uncontrolled design, we cannot conclude that the improvements were due to the intervention. There are always secular trends that might be occurring at the same time in a surgical ward, and which may have influenced our results [71]. Because of the study limitations, caution must be taken in generalizing the results.