A safe work environment for nurses is characterised by, among other factors, good professional relations, a supportive management style, a balanced work schedule, concordance between nurses' increased workload and nurses' skill-mix, adequate time to meet patients' needs, professional autonomy, adequacy of resources, and opportunities for professional advancement. However, nurses often assess their work environment as stressful (Hinno, 2012) and complex, while they recognise the staff shortages, limited resources and the pressured conditions as the most important issues (Milisen et al, 2006). The stressful nature of nursing often leads to burnout, low productivity, absenteeism (Coomber and Barriball, 2007; Garrett, 2008), and, in the long term, contributes to understaffing, thus further aggravating the problem (Shirey, 2006).
In several studies it has been reported that a positive work environment is associated with fewer occupational injuries (Gershon et al, 2007; Hanrahan et al, 2010), less burnout (Gershon et al, 2007), and increased job satisfaction (Coomber and Barriball, 2007; Garrett, 2008). The negative effects of a negative nurses' work environment may also lead to poor patient outcomes, such as increased mortality and complications (Mark and Harless, 2010; Penoyer, 2010; Trinkoff et al, 2011), readmissions (Ma et al, 2014), and increased healthcare costs (Lewis and Malecha, 2011). Regarding the effects of the nurses' work environment on patients' health outcomes, it has been shown that good professional relations are associated with a reduction in the frequency of pressure ulcers, pneumonia and medicine errors (Manojlovich and DeCicco, 2007; Manojlovich et al, 2009).
To the authors' knowledge, the only relevant systematic review about the effects of nurses' work environment on both patients' and nurses' outcomes is the one by de Cordova et al (2012), in which the effects of the nurses' work schedule were examined (with outcomes such as patients' length of hospital stay and nurses' injuries and wellbeing). The purpose of the current systematic review was to identify and analyse the studies in which the effects of the nurses' work environment on both patients' and nurses' outcomes are examined.
Methods
Literature search strategy
An electronic search was carried out in the Medline database during May 2016, by using the key words ‘working conditions’, ‘working environment’, ‘nurses’, ‘nursing staff’, ‘patients’, and ‘outcomes’ in various combinations.
Selection criteria
The articles were included if they described findings of research studies about the effects of at least one characteristic of nurses' work environment on both patients' and nurses' outcomes. Nursing staffing was not used as a work environment characteristic, because its correlation to patients' and nurses' outcomes has been extensively investigated and presented in the literature (Hodgkinson et al, 2011). Furthermore, poor staffing has an effect on all the other factors of a work environment, so its use as a characteristic would mean a significant bias.
The criteria for the inclusion of the studies in the present review were:
- Studies published in English
- Research studies (prospective, cross-sectional or retrospective) Studies examining the effects of nurses' work environment on both patients' and nurses' outcomes
- Studies in which both patients and nurses participated
- Studies in which only questionnaires were used for the self-assessments of outcomes.
The authors chose to use only these studies because of their similarity (surveys), in order to gather more comparable results (that is satisfaction, burnout), as there are very many different outcomes in the literature (too many complications for the patients and outcomes for the nurses, from injuries to musculoskeletal disorders). The authors decided that this approach would strengthen their results as they would be more specific.
Literature screening and data extraction
Prior to possible inclusion, all the retrieved studies were evaluated by two reviewers independently, in order to define their relevance to the purpose. Finally, a manual search was conducted on the included articles' references, to find other similar articles. The data that were extracted included: type of study, duration of data collection, instruments for data collection, country in which the study was conducted, study population, and study topic.
Results
In total, 109 articles were yielded from the initial database search, of which 41 articles were excluded after the review of their title. Following review of their abstracts, a further 31 studies were excluded. Finally, 30 studies were excluded after reviewing their full text. The number of articles included in this systematic review was initially 7, but 3 more articles were retrieved by searching the references of the articles already included, resulting in 10 articles (Figure 1).
This systematic review includes four cross-sectional studies (Vahey et al, 2004; Tervo-Heikkinen et al, 2008; You et al, 2013; Shang et al, 2014), while the remaining were descriptive correlational studies (Kangas et al, 1999; McNeese-Smith, 1999; Mark et al, 2003; Donahue et al, 2008; Bacon and Mark, 2009; Purdy et al, 2010). Six studies were conducted in the USA (Kangas et al, 1999; McNeese-Smith, 1999; Mark et al, 2003; Vahey et al, 2004; Donahue et al, 2008; Bacon and Mark, 2009), two in China (You et al, 2013; Shang et al, 2014), one in Canada (Purdy et al, 2010), and one in Finland (Tervo-Heikkinen et al, 2008). In two studies (Kangas et al, 1999; McNeese-Smith, 1999) the sampling of the nurses was performed randomly, and in two studies (Mark et al, 2003; Bacon and Mark, 2009) the sampling of the patients was performed randomly. In the studies of Shang et al (2014) and You et al (2013) the sampling of the hospitals was performed in a purposive manner. The main characteristics of the studies are summarised in Table 1.
Discussion
The stressful nature of the nursing profession often leads to poor nurse outcomes (Shirey, 2006), as well as to poor patient outcomes (Penoyer, 2010). The purpose of this systematic review was to analyse the studies in which the effects of the nurses' work environment on both patients' and nurses' outcomes were examined. The main outcome measure, both for patients and nurses, was satisfaction from care or from work respectively.
In the majority of the studies, patient satisfaction with nursing care was affected by several factors of the nurses' work environment; for example, patients receiving nursing care in the primary care delivery model expressed more satisfaction than patients in the case management delivery model (Kangas et al, 1999), and were more satisfied if they knew that one specific nurse was in charge of each patient's care (McNeese-Smith, 1999). On the contrary, larger ward sizes contributed to lower patients' satisfaction (Mark et al, 2003). In the study of Vahey et al (2004) nurses' work environment had an effect on patients' outcomes both directly (with patients hospitalised in units with good environments being more satisfied in comparison with those in units with poor environments), as well as indirectly, as nurses' emotional exhaustion and personal accomplishment had significant effects on patients' satisfaction. Similar results were found in the studies by Donahue et al (2008) and Tervo-Heikkinen et al (2008), where patients were more satisfied when the nurses reported higher empowerment status. However, there are contradictory results in two studies; in the study by McNeese-Smith (1999), there seemed to be no significant connections between managerial motivation and patients' satisfaction, or between nurse outcomes and patients' satisfaction, while in the study by Purdy et al (2010), no significant connections between nurses' outcomes and patient satisfaction or patient self-care were reported.
In general, patients who were hospitalised on units with good work environments appeared to be more satisfied with the nursing care than the patients on units with poor work environments (Vahey et al, 2004; Donahue et al, 2008; Bacon and Mark, 2009; You et al, 2013). In the scientific literature, the negative effects of the nurses' work environment have been extensively reported to result in poor patient outcomes, such as increased mortality and complications (Mark and Harless, 2010; Penoyer, 2010; Trinkoff et al, 2011), although patients' satisfaction with the care in connection with the nurses' work environment is not so thoroughly addressed; therefore, these findings need to be researched in more depth in the future. Patients participating in the studies that were included in this review were more satisfied when they knew that one nurse was in charge of their care (McNeese-Smith, 1999), and when they were hospitalised in small nursing units (Mark et al, 2003). Better nurse-to-patient ratios are also recognised as contributing positively to dimensions of care such as respect for the patient's personal values and needs (Bolton et al, 2003).
Regarding nurses' outcomes, in the study by Kangas et al (1999), no significant differences were found in nurses' job satisfaction with different organisational structures; however, perceiving the environment as supportive was a predictor of nurses' job satisfaction. Other factors of the nurses' work environment that affected job satisfaction were the manager's motivation for achievement (McNeese-Smith, 1999), professional nursing practice, availability of support services, smaller ward sizes, and lower complexity of care (Mark et al, 2003). Also, when the possibilities for professional development were better, nurses felt that nursing had respect, good professional relations (Tervo-Heikkinen et al, 2008), and empowerment (Purdy et al, 2010). There are numerous studies in which it has been shown that good professional relations for the nurses are associated with a reduction in the frequency of patients' pressure ulcers, pneumonias, and medicine errors (Manojlovich and DeCicco, 2007; Manojlovich et al, 2009), although the effect of the environment on these outcomes was not the objective of this systematic review.
According to the findings of the present review, both nurses' burnout and intention to leave are affected by nurses' work environment, with the nurses who work in units with good environments also presenting with less burnout (McNeese-Smith, 1999; Vahey et al, 2004; You et al, 2013; Shang et al, 2014), lower intention to leave (McNeese-Smith, 1999; Vahey et al, 2004), and less dissatisfaction (You et al, 2013; Shang et al, 2014). An interesting finding is that when nurses' evaluations of standards of professional nursing practice increased, staffing adequacy and professional relations were evaluated as better (Tervo-Heikkinen et al, 2008). This is important because the problem of understaffing is perhaps the most critical worldwide, and relates to a safer environment for the nurses (Mark et al, 2007), and the patients (Mark and Harless, 2010; Trinkoff et al, 2011). Aiken et al (2012) also stated that even a very important factor such as the nurse-to-patient ratio improves outcomes only in hospitals with a good work environment, while it has no effects on outcomes in hospitals with inappropriate environments.
Work environment factors that already have been recognised through research as enhancing nurses' decision to remain in the profession include the ability to exert control over their work, to have greater participation in decision-making, a good balance in professional relations (Coomber and Barriball, 2007; Ritter, 2011), and professional autonomy (Kangas et al, 1999; Iliopoulou and While, 2010), all of which are characteristics that at the same time are associated with patients' satisfaction with nursing care (Tzeng et al, 2002). Furthermore, in the study by McNeese-Smith (1999), nurses' motivation for achievement led to increased nurse productivity. Bacon and Mark (2009) showed that the only characteristic of the nurses' work environment negatively associated with nurses' professional engagement was work complexity, which in turn was associated with poorer work conditions. Similar findings have also been recognised in the literature, as different characteristics of the nursing environment, such as the stressful nature of the profession, are reported to lead to low productivity (Shirey, 2006), while increased nurses' workload leads to increased absenteeism (Coomber and Barriball, 2007; Garrett, 2008). It has also been argued that supporting nurses through clinical nursing supervision may have a positive influence on their perceptions of wellbeing, anxiety, and sense of being in control, while psychosocial work has an influence on nurses' professional engagement and motivation (Bégat et al, 2005).
To summarise, in the majority of the studies included in the present systematic review, nurses' burnout and job satisfaction, as well as patients' satisfaction with care, are affected by the nurses' work environment. Patients were more satisfied when they were hospitalised in small wards, when they knew that one nurse was in charge of their care, and when nurses did not experience time pressure and were able to provide information. Thus, the work environment constitutes a determinant factor, besides staffing and nurses' education, for the improvement of quality of health care and, at the same time, relates to improved outcomes for the nurses. However, there are findings—for example regarding the connection between patients' satisfaction, perceived empathy or self-care and the nurses' work environment—that are contradictory and therefore make generalisation of these results impossible. Further research is needed in order to examine how the nurses' environment affects both patients and nurses.
Limitations
The study's limitations include the age of two articles, which were published in 1999. However, the remaining studies were published after 2003, up to 2014. The fact that only in two of the studies was the sampling of the nurses performed randomly and only in three of the studies was the sampling of the patients random constitutes a limitation for the quality of the included studies. However, all of the studies were prospective, had a large sample (of both patients and nurses), and the sampling was undertaken in a consecutive or purposive manner. Furthermore, the fact that only studies written in English were included may constitute a bias. Finally, if other patient or nurse outcomes were examined (patient falls, nosocomial infections, needle-stick injuries, low back pain), the results might perhaps reveal a stronger connection between nurses' work environment and patients' and nurses' outcomes, since all the included studies were surveys and were carried out only with the use of questionnaires.
As discussed in the selection criteria, nursing staffing was not used as a work environment characteristic in this systematic review, because the correlation with patients' and nurses' outcomes has been extensively investigated and presented in the literature (no one can deny the detrimental effects of poor staffing on patients' outcomes), and poor staffing has an effect on all the other factors of a work environment, so to use it as a characteristic would lead to a significant bias. Although politics in different healthcare systems around staffing levels are different in different parts of the world, nursing is always stressful wherever it is practised. This is one of the reasons that this systematic review's findings are important.
Conclusions
The implications for practice derived from the current systematic review include that efforts for the provision of a good work environment for nurses leads to improved patient satisfaction from the nursing care, as well as to nurses' increased job satisfaction and low burnout. Thus, the work environment constitutes a determinant factor for the improvement of the provided healthcare quality and, at the same time, relates to improved outcomes for the nurses.
A supportive environment for the nurses, motivation from management, opportunities for professional development, education about communication skills and ways of improving professional relationships and resolving conflict, as well as smaller unit sizes, are all ways to improve nurses' working environment that are supported from this review's findings. More randomised studies are needed in the future, in order to explore the effects of the nurses' work environment on more patients' and nurses' outcomes. In addition, qualitative analysis and/or mixed methods analysis would provide a more in-depth understanding of the situation.
Key Points
- In the majority of the included studies, patient satisfaction seems to be affected by several factors of the nurses' work environment, such as the nursing care delivery model, the ward size, and the nurses' empowerment
- In two studies no significant connections between nurses' work environment and patients' satisfaction were found
- For the nurses, perceiving their work environment to be supportive, the manager's motivation for achievement, professional nursing practice, availability of support services, smaller ward sizes, lower complexity of care, and the possibilities for professional advancement were predictors of job satisfaction
- Nurses working in units with good environments presented higher job satisfaction, lower burnout, and lower intention to leave