The findings clearly indicated an association between how the nursing staff rated the ward atmosphere and how they perceived their psychosocial work environment, although the correlations were small to moderate in size . This indicates that the ward atmosphere and the psychosocial work environment were related but separate phenomena. The data also revealed that there were no differences between nurses and nurse assistants concerning perceptions of the psychosocial work environment and the ward atmosphere.
Two of the ward atmosphere subscales, Personal Problem Orientation and Program Clarity, were important for the participants' perception of Empowering Leadership. High ratings on those two aspects of ward atmosphere were associated with perceiving a high level of Empowering Leadership. The latter assesses the degree to which the leadership stimulates encouragement and empowerment to take part in decision-making and developments. The importance of leadership for the staff's working conditions is well recognized. Empowering leadership may increase an individual's feelings of organizational justice, respect, and trust in management , and leadership empowerment has also been related to job satisfaction . Furthermore, Kanter  suggests that employees' attitudes would improve along with an increase in organizational effectiveness if structured empowering is used. This indicates that it is important to increase leadership empowerment and a possible way to facilitate this could be to improve the ward atmosphere factors of Personal Problem Orientation and Program Clarity. Personal Problem Orientation assesses the degree of encouragement for the patients to freely and openly express and talk about feelings and personal problems. Program Clarity involves how stabile, evident and clear the treatment structure, rules and expectations are at the ward . It might be that the encouraging nature of Personal Problem Orientation is also important to the nursing staff, in terms of increasing the leadership empowerment. However, no previous research seems to exist concerning the relationship of these aspects, and there are thus no studies that can confirm that assumption. Furthermore, the clarity aspect regarding treatment structure, rules and expectations embedded in Program Clarity may stimulate an encouraging and empowering leadership style by enabling managers to involve the staff in decisions and developments concerning ward and treatment issues. The relationship may, however, be circular. The ward atmosphere may be important for empowering leadership, but improvements in the ward atmosphere may also, through for example staff training , be initiated by an empowering leadership.
The findings of the present study also suggest that Program Clarity was related to the nursing staff's perceived Role Clarity. It seems reasonable to assume that a high level of Program Clarity, with a clear treatment structure and clarity of expectations, would increase the Role Clarity as appraised by the nursing staff. Role clarity has been found to be an important aspect in relation to perceived stress among psychiatric staff [13, 14, 38]. To find ways to increase the Role Clarity, as perceived by the nursing staff, thus seems to be important. A possible way to accomplish this would be to concentrate on both the psychosocial work environmental aspects and the ward atmosphere and thus work on two fronts in order to make improvements.
Practical Orientation and Order and Organization explained almost one fourth of the variance in Organizational Climate. The Organizational Climate, as measured in the present study, taps aspects such as perceived levels of encouragement, support and comfort in the organization. It also involves levels of appraised communication within teams and between group members. Social support from colleagues is an important factor of the psychosocial work environment . In the present study, ratings of the ward atmosphere subscale of Practical Orientation and Order and Organization were positively related to nursing staff perceptions of the Organizational Climate. Practical Orientation denotes the degree to which the staff and the ward activities guide the patients towards problem solving, both during the latter's admission to the ward and outside of it. Order and Organization concerns the staff's ability to keep scheduled appointments with patients and the patient's ability to follow daily schedules and treatment plans . No previous studies seem to have focused on the relationship between these aspects, but one could speculate that the identified association between ward atmosphere factors and the nursing staff's perception of the psychosocial work environment might indicate that high levels of Practical Orientation and Order and Organization could stimulate staff communication and increase support between co-workers. Furthermore, with high levels of Order and Organization the staff might be more likely to keep arrangements and appointments with the patients, and do this on time. It is possible that this may lead to a reduction in tension between colleagues, as well as to improved communication and support. It may also be the other way around, i.e. that a beneficial Organizational Climate promotes Practical Orientation and Order and Organization. Either way, the relationships between these phenomena indicate that also involving ward atmosphere issues could be a more effective approach, when addressing a psychosocial work environment problem in this area.
The study did not identify any differences between nurses and nurse assistants in terms of perceived psychosocial work environment or ward atmosphere. No previous studies seem to have addressed those phenomena, however, studies have found similar results concerning perceived stress and burnout among nurses and nurse assistants [25, 26]. Although these staff groups tend to have somewhat different work responsibilities, they seem to perceive the psychosocial work environment and ward atmosphere in similar ways. It can thus also be argued that any improvements in the psychosocial work environment and the ward atmosphere would be beneficial to both staff groups.
The descriptive results from the present study indicate that the participants appeared to rate the psychosocial work environment and ward atmosphere as moderately high for most of the subscales. However, the nursing staff perceived the Support from Superiors as rather low. In addition, no ward atmosphere variables were found to be related to this psychosocial work environment aspect according to the logistic regression. The significant bivariate association found in the correlational analysis disappeared after dichotomizing for the logistic regression, in which some of the variation in the variables was lost. It is possible that Support from Superiors is mainly contingent on other factors, not addressed in this study. It is also possible, that the leadership style on the wards in the present study emphasized empowering aspects and role clarity, and that leadership support was less emphasized.
There are some reservations about these findings. The response rate was lower than desired (52.3%), despite several reminders, and it is possible that there was a selection bias, in turn possibly influencing the results. Still, the present sample seemed to be representative of the staff working at the 12 units in terms of mean age and the distribution of nurses/nurse assistants. The reliability of the instruments used for the data collection was another problem of this study, where five of the WAS subscales had to be excluded from the analysis due to low Cronbach's alpha values (<0.5). Similar problems have been reported previously, especially concerning the factor structure of WAS [40, 41]. These methodological issues have also been reported for the revised version [32, 33]. According to Røssberg and Friis , a Cronbach's alpha ≥0.5 is acceptable for comparing ward means , and since the present study included only aggregated data, the WAS subscales were kept for analysis if they reached that value. The instrument used to assess psychological and social factors at work, QPSNordic 34+, has to our knowledge not been formally tested. In the present study variables were formed on the basis of areas and scales identified in the full version of QPSNordic. With this in mind, further studies are warranted in order to establish subscales of the QPSNordic 34+. Although four of the ward atmosphere variables contributed significantly to the psychosocial work environment variables of Empowering Leadership, Role Clarity, and Organizational Climate, the amount of explained variance was limited. Especially concerning Empowering Leadership (10.4%) and Role Clarity (14.7%), the findings indicate that contextual variables not included in this study, such as the work load and the physical milieu, may be of importance for the psychosocial work environment. Besides, personal characteristics, such as self-efficacy and self-mastery, might mediate the relationship between the ward atmosphere and the way the psychosocial work environment is perceived, but this remains to be studied. This study fulfilled the purpose of specifically focusing on how ward atmosphere factors and the psychosocial work environment were related, which opens up for studies of more complex relationships in future research.