Organization Culture of Safety-Comparison of Two Periods, before and after Intervention

Introduction

Since 1999, when the Institute of Medicine declared that “to err is human” and recommended to avoid “blaming and shaming”, our culture of safety changed tremendously [1]. Nowadays we try to investigate errors and adverse events, to learn from our mistakes, and prevent future sentinel events and patients’ suffering. There is a significant advance in patients’ safety in hospitals all over the world. Quality

and safety indicators of processes and outcome are very popular and help us keep the high level of treatment quality and safety. Hospitals’ accreditation by international institute (such as the JCI), regular investigations, check lists before operations and important invasive procedure, measuring quality indicators and annual quality improvement plans, helped medical centers to achieve high level of safety and prevented many preventable deaths [2, 3].

In 2015 a questionnaire regarding staff positions and knowledge of patients’ safety culture in Israeli hospitals was circulated by the Ministry of Health. The questionnaire was based on a Canadian paper [4], and was offered to employees (physicians, nurses and other health workers) in every public general hospital. Very few staff members in our hospital answered the questionnaire. Four main issues were addressed: the understanding of patient’s safety, the attitude of the direct manager to safety problems, the number of reporting and communication.

After an intervention and intensive quality improvement plan we repeated the questionnaire in our medical center. This time we approached 1000 employees, and had compliance of 641 participants.

The aim of our study is to assess the change in staff members’ attitude towards the hospital culture of safety in 2017 compared to 2015, and to associate this attitude to improvement in quality indicators and reports regarding adverse events and near miss cases.

Methods

We used Can-PSCS [4] (previously called the Patient

Safety Culture Tool) data collected from staff in a large, cross- sectional sample of Canadian healthcare organizations. Survey data was collected in 2011 by Accreditation Canada as part of the Qmentum accreditation programme [5].

A questionnaire of 15 items was introduced to 500 staff members in Rabin Medical Center in 2015. After an intervention plan performed in 2016, the same questionnaire was introduced to 1000 staff members in 2017. We compared the results of the questionnaires of 2015 and 2017 looking for change in staff members’ attitude towards patient’s safety in the medical center. We also compared the results of quality indicators and number of adverse events reported before and after the intervention.

Results

The results of both questionnaires, 2015 and 2017, are shown in Figure 1. The compliance was much better in 2017 than in 2015, 64.1% versus 20.6%, respectively P<0.001. In 2017 only 33% of the participants believe that there are patient’s safety problems in their unit, and only 14% think that only by chance more severe adverse events do not happen. The vast majority believes that staff members behave with dignity and respect to each other and that they learn from their mistakes and act on a daily basis to prevent future mistake as well. Eighty-one percent think that their superior does not ignore safety issues, 73% think that the manager seriously considers staff suggestions for the improving patient safety, and 60% testify that their superior them on a job well done, according to an established patient safety procedure. Thirty eight percent of the participants reported “near miss” cases in the last year, 3.8 in average per employee, and 69% said that a “near miss” is always reported; 76% said that they have “no blame nor shame” policy in their department and an atmosphere that encourage reporting adverse events and “near miss” events, and only 16% thought that after reporting the staff member is blamed, instead of looking for system fault. Eighty percent declared that staff will freely speak up if they see something that may negatively affect patient safety, and 62% thought that a staff member will freely speak up if they see a wrong decision made by their superior.

Comparing the results of the questionnaires between 2015 to 2017 there is a significant improvement in 13 out of 15 items. In only 2 items the questionnaire of 2015 had better results than 2017, but the results were not statistically significant (Figure 1). The RR of 2017 versus 2015 was 1.555 with 95%CI 1.370-1.766 (P<0.0001). The differences were statistically significant in 7 items and with a trend for significance in 2 items.

Discussion

When getting the results of our safety culture survey in 2015, measuring providers’ perceptions across our health settings, we were very disappointed, even though our results were very similar to other hospitals in Israel with no significant differences. Yet, we thought that the small number of participants and the very low compliance, do not allow a meaningful interpretation of this important questionnaire.

Following the survey of 2015 we performed a comprehensive quality intervention project and repeated the questionnaire in 2017. This time the compliance was much higher as were the improvement in the staff perceptions regarding safety climate and interpretation of the health providers regarding safety issues.

Ginsburg LR, et al. [4] confirm the validity of the Can-PSCS in a comprehensive study, initial and cross-validation samples of 13126 and 6324 direct care providers from 119 and 35 health settings across Canada, respectively. They described 6 dimensions of patient’s safety: organizational leadership support, incident follow-up, supervisory leadership, unit learning culture, enabling open communication, judgment- free environment and enabling open communication.

It is no easy or feasible to claim that our proactive activity changed the opinions of our staff, physicians, nurse, pharmacists, secretaries and other people involved in patients care. Yet, we also found a significant improvement in almost every measure that directly related to the safety culture. For example most of the quality indicators and “near miss” reports were significantly improved. It’s possible that engaging in all of the changes mentioned might have led to real changes in the answers to the survey questions. But, with such a large change in the response rate, it’s also just as likely that the different pattern reflects inclusion of a broader range of staff. Sometimes one likes to argue that the people with better attitudes are usually the ones who respond to surveys, but it can also be that the unhappy staff wants to express their unhappiness. It’s hard to know. So, the paper does not work well as a robust demonstration of an intervention that improves culture-it’s too hard to know if the results represent real improvement or just reaching a broader population of staff.

The relationship between intervention and improvement in medicine has been an issue for debate for many years [6, 7]. The correlation between process and outcome is not always clear, and sometimes can be misleading. We use many quality indicators of process such as duplex in transient ischemic attack (TIA), deep vein thrombosis (DVT) risk assessment, and prophylactic antibiotics before colectomy, without assessing outcome of TIA, DVT or colectomy in our hospital, relying on the literature benchmark that might be very different in our hospital or community. We are not the first to find a positive correlation between change in the culture of safety and staff attitude and better outcome of quality measures. Haynes AB et al demonstrated significant changes in safety attitude and relationship to decreased postoperative morbidity and mortality following implementation of a checklist-based surgical safety intervention [8]. Thus, we believe that quality improvement plans will really change the staff approach and behavior towards a better safety culture in the hospital.

Conflict of Interest

The authors disclose no conflicts