Influence of Burnout on Patient Safety: Systematic Review and Meta-Analysis

1Patient safety is one of the main challenges for quality of care, as adverse events are common in health services. In the United States, medical errors account for more than 250,000 deaths a year, making it the third leading cause of death. A study conducted in Norway has identified that patients who die in hospitals experience seven times the rate of serious adverse events.

2Patient safety is highlighted as one of the main challenges nowadays, as in many cases, patient care—in addition to system failures, poor organizational processes, and faulty management—also depends on health professionals. Changing habits requires a better training of these professionals based on the assumption that several studies indicate that failures in patient safety are also a consequence of human failures regarding communication, teamwork, and psychological health among health professionals.

3Based on evidence, training would be on clinical practice guidelines, adverse events, technologies used in the service, better working conditions, continuous guidance on infection prevention methods, and offering better psychological and emotional support to health professionals.

4Professional well-being, depression, anxiety, and burnout syndrome are determining factors that influence the care provided to patients, although they have different characteristics among them. It is suggested that there is a proven relationship between poor well-being and moderate to high levels of burnout and poor patient safety resulting in assistance errors.

5Burnout is a disorder that is directly related to workplace conditions due to the occupational stress in which health and education workers are constantly under, mainly associated with interaction with other people; as such, this condition is deeply studied in these workers. Regarding their usual risks, health professionals are exhausted during work hours, and therefore will not be able to perform an effective care practice and may cause harm to both patients and health service in general.

6A study by Roteinstein et al. found a large variability (0% to 80.5%) of this disorder among physicians. Among nurses, a prevalence of 30% was found in at least one of the three main domains (depersonalization, emotional exhaustion, and low professional achievement) of the disturbance.

7Patient safety permeates through organizational, social, and individual factors that depend on physical and human resources. Burnout also depends on these characteristics, such as physical organizational factors and human interactions; therefore there is an intrinsic link between these.

8Adverse events are defined as possible complications that may arise from patient care, emerging from errors that do not belong to the natural history of the disease. When these are due to errors, they are called avoidable adverse events.

9Thus, because of the high prevalence of burnout among health workers due to the increasing number of adverse events, the following question arises: What is the relationship between burnout and patient safety?

10The exposed problem justifies the need for studies to elucidate the associations between patient safety and burnout since to the relationship between them is frequently evidenced. The findings of this research will contribute toward a better understanding of this issue and provide visibility to the subject.

11Thus, the objective of this study is to analyze the relationship between burnout and patient safety.

12This systematic review and meta-analysis was performed according the Preferential Reports for Systematic Reviews and Meta-Analysis (PRISMA). The study was conducted in January 2018.

13Following a standardized Population, Intervention, Comparison and Outcome (PICO) format, this study investigates the relationship between burnout syndrome (population) in the context of patient safety (intervention) by observing the influence that this relationship can exert on the assistance (comparison) and analyzing the results to understand the problem (outcome).

14The articles were searched through the following databases, which have a huge scope in the research topic: Pubmed (http://www.ncbi.nlm.nih.gov/pubmed) and Web of Science (https://webofknowledge.com/). In both databases, two refining searches were carried out regarding human studies. The first used the following combination of the descriptors: patient safety AND burnout professional safety AND organizational culture. The second search was performed with the following combination: patient safety AND burnout professional safety AND safety management.

15The burnout definition was taken from the Maslach Burnout Inventory (MBI) form published in 1986, which is the main burnout assessment tool to date. For patient safety, we used the term applied in the Hospital Survey on Patient Safety Culture (HSOPSC) instrument. The rest of terms were used to refine the search regarding professional practice.

16Initially, a period for filtering articles was not restricted, given the need to search all available articles.

17The selection process began by reading the titles and summaries of each article. Second, after excluding abstracts that were not adequate, the articles were read in full.

18Data on method, participants, interventions, and outcomes were analyzed and extracted by two independent reviewers. In order to assess the risk of bias and the quality of the study, the Chochrane Manual for Systematic Intervention Reviews was used, where any discrepancies were resolved by a third researcher.

19Only studies with human beings, in English language, evaluating the relationship between health professionals and patient safety were included. No study protocols or qualitative research were included as they do not present quantitative results. Author letters, theses, dissertations, and research on animals were excluded.

20The effect size of continuous outcomes was evaluated using the weighted mean difference (WMD) and dichotomous outcomes assessed using the risk ratio (RR) with a 95% confidence interval (CI). Heterogeneity was evaluated with I2 statistics. A random-effects model was applied regardless of the heterogeneity of results.

21The meta-analysis was performed with the Stata 14.0 program (StataCorp LLC, College Station, TX, USA) using the random-effects model. A meta-analysis of proportion and metabias was also performed for effect analysis based on the standard error.

22A data quality assessment was performed, and the screening process was also executed by other researchers, as well as the statistical analysis of the data by a third statistician.

23The first search generated a total of 124 articles. After filtering by reading titles, abstracts, and duplicates, 95 studies were excluded and 29 articles were selected for a reading of the full text. Of these, 21 articles met the inclusion criteria, as is illustrated in Figure 1.

24The selected studies were carried out mostly with nurses and physicians, mainly in the United States, followed by European countries; 90% were cross-sectional studies, 5% were prospective, and 5% longitudinal, with studies between the years 2006 and 2018 (Table 1).

25Table 2 illustrates the purpose of the selected studies that met the inclusion criteria and their main findings.

26The analysis demonstrated a relationship between burnout and patient safety actions. There was an effect of 2.67 with a confidence interval of 2.3 to 3.0, which represents a probability of superiority of 96.7% (Figure 2).

27The analysis of proportions demonstrated an effect on patient safety actions of 0.60, evidencing a probability of superiority of 66.4%. It represents an average effect based on effect interpretation and associates the development of burnout in health professionals with patient safety actions (Table 3).

28Most of the studies were within an acceptable standard error in health with representativeness in the range of 95%, reinforcing the congruence with the meta-analysis and assertiveness in the general interpretation of results (Figure 3).

29The results revealed an association between burnout and patient safety permeating the work process, personal characteristics, and teamwork. Although some studies indicate this association in a less significant way, the meta-analysis shows an association of more than 60%.

30Studies corroborate that the presence of burnout leads to a decline in patient safety. In units with higher burnout scores, there was a deterioration of teamwork climate, safety, and job satisfaction. Professional tiredness, one of the exhaustion criteria, implies a lower ability for effective teamwork, which negatively affects patient safety.

31Higher levels of burnout were also associated with unfavorable outcomes, patient dissatisfaction, and increased patient and family complaints. This can be explained by emotional fatigue and depersonalization, which trigger in the health professional the feeling of exhaustion and cynicism, becoming distant and cold in front of the patients’ needs, which compromises the quality of care.

32Although most studies confirm this association, Garrouste-Orgeas et al. found no influence of burnout on the occurrence of medical errors, nor did they identify an association between the disease and patient safety culture scores. Tawfik et al. observed that general healthcare associated infections (HAI) rates were not associated with burnout, but the feeling of working hard.

33Rodrigues, Santos, and Sousa specified that the work environment as a source of stress, excessive workload, and lack of organizational support are the main contributing factors for this association. This was confirmed by Holden et al. who pointed out that external efforts negatively impacted burnout and patient safety, but personal or internal efforts were not associated with the occurrence of adverse events and exhaustion.

34The workload is indicated as a determining factor for professional fatigue. Specifically, for health professionals with high burnout rates, high hourly loads had a strong negative impact on patient safety.

35This information can be evidenced when trying to understand the elements that predicted hospital infections, since when directing the research instrument to nurses, it was identified that a third of them had burnout, and this variable was associated with urinary tract infection; if a nurse has burnout, each additional patient included in their workload will be exposed to a higher risk of infection.

36Among the factors identified by the professionals, the problems related to changes in their respective health services, the imbalance between personal and professional life, and their consequences for the quality of care and patient safety stand out. More specifically, physicians identify that work pressure and hierarchy are detrimental aspects, while nurses are primarily concerned with their quality of life and patient safety at work.

37Carayon et al. pointed out 58 risk categories that involve working conditions in a hospital environment and the influence of human and systemic factors on the occurrence of errors, where some of them are lack of professionalism, high workload, poor use of tools and technologies, messy work space, and the hierarchical culture, which in many cases still prevails, contributing to a disjointed service; these circumstances were also evidenced in other results.

38On the other hand, on nurses’ complaints related to the hierarchy of medical level, Oliveira et al. noticed that the main generators of the problems of personnel conflict, besides doctors, are the nurses, managers, or professionals who have more years of experience in the ward or have a leadership position.

39In this line, teamwork is evidenced as one of the important points that prevent the development of burnout in professionals, and consequently, behaviors that may harm the patient’s safety. However, the disarticulation of teamwork can also lead to the opposite, as Welp, Meier, and Mansen showed that emotional and professional exhaustion has an undesirable effect on teamwork.

40It can be observed that one of the major issues for the development of burnout focuses on the organizational flow of work that consequently influences personal satisfaction, patient safety culture, and psychological aspects such as mental exhaustion.

41A positive safety culture toward the patient was associated with the absence of burnout and a high capacity to deal with stressful situations. Open communication, management support, professional suitability, mutual learning, and teamwork are considered dimensions of safety culture; it is likely that the professional inserted in this scenario feels productive and satisfied, and therefore exposed to a lower chance to develop burnout.

42In view of this, the use of leadership walk rounds is a routine that aims to periodically establish a relationship between leaders (heads of health services) and professionals who work in patient care, and to identify and solve problems involving the care and patient safety. Sexton et al. confirm in different dimensions and moments the positive influence that this routine exerts on the negative perception of burnout, improving patient safety results.

43Another point of conflict, especially for nurses, is autonomy and leadership at work. Spence and Leiter see nursing leadership as a positive point to effectively affect patient safety outcomes, in addition to having good interpersonal relationships, reduces the chances of exhaustion and prevents professional fatigue.

44In view of the above, it is possible to highlight the emerging need for strategies aimed at controlling burnout and especially its association with patient safety. The protocol for training implementation called TeamSTEPPS (Team Strategies and Tools to Improve Patient Safety and Performance) aims to improve the psychological safety of the team, reduce burnout among professionals, and extend the report of errors.

45These results highlight the importance of working under the precepts of health promotion, as this will help with positively affecting the patient’s safety and the quality of life of professionals and the population.

46This study, as well as routines of improvement of organizational workflow in health, strengthens the need for interventions for the improvement of health services given the large number of burnout cases observed and their direct association with patient safety. The development of new studies aimed at identifying this association more closely in all health professionals, regardless of their line of work, is extremely relevant for health interventions.

47The presence of burnout among health professionals is associated with worsening patient safety. High levels of burnout is related to external factors, such as high workload, long hours, and interpersonal relationship.

48Avoiding professional exhaustion is an important strategy for improving patient safety.