What characterizes the work culture at a hospital unit that successfully implements change – a correlation study

This study was designed as a correlation study. One of the basic assumptions in this study are that predominant behaviors are an artifact of the typical work culture in the unit. We compared the results of two different units at the same hospital [20, 21]. The findings in the present study will be a comparison between the two units where successful and unsuccessful implementation had taken place. The units were chosen because the implementation were connected to the electronic patient record, and each unit had an introductory program related to the implementation. The implementation process was examined and characterized as either successful or unsuccessful [21‐24]. Researchers who examined the sustainability of the implementation characterized units as either unsuccessful or successful in the implementation. In both cases, the implementations were regarding electronic patient records. In the study from 2004, a computer based tool for report assessment of symptoms and functioning were attempted to be implemented; in this case, the tool was not used at all [22, 23]. In the 2013 study, the use of nursing diagnoses in the electronic patient record were implemented [21, 24]. In both studies, using or not using the tool could be observed via the electronic patient record. During the implementation process in both units, a program with information, education and training was completed by all the health care personnel at the units [21‐24]. In both cases, we followed the implementation while measuring the work culture at the two different units using the same questionnaire.

In the spring of 2013, healthcare personnel working at the successful unit (succ) filled in and returned the questionnaire. Of the 101 healthcare workers working at the unit, 70 participated (69%). The sample of healthcare personnel consisted of 63 nurses, 6 assistant nurses, and 2 nursing managers; there were 69 females and 2 males. The questionnaires were distributed and filled in at seminars about the implementation. Only healthcare personnel working more than half time were invited to participate in the study. These findings were compared with findings from an earlier study conducted in 2004. Of the 36 healthcare personnel working at the unsuccessful unit (unsucc), 25 (70%) filled in and returned a questionnaire. The sample consisted of 17 nurses, 2 physicians, 2 physiotherapists, and 4 other professions. There were 24 females and 1 male in the sample. The questionnaires were distributed and filled in at morning meetings (taking approximately 10 min and with researcher present to answer questions) or delivered to the mailboxes of those who were not present at the meeting. Two follow- ups were conducted.

The sociodemographic data were equal for the personnel who participated and those who did not. The personnel who worked night shifts were more likely to not participate in the study.

In total, this present study consisted of 106 healthcare personnel. The university hospital involved in the study had 993 beds and 59,016 hospitalizations in 2013.

The Systematizing Person-Group Relations Instrument (SPGR) was used for data gathering and investigation [14, 25‐27]. The SPGR process is based on the “Semantic Differential scaling technique” established by Osgood [28]. Earlier studies [29‐31] have described the validity and reliability of the SPGR tool and the instrument has been used in different settings [32‐34]. The subsequent detailed appearance of the SPGR tool is presented similarly to the methodological descriptions in an earlier study [32].

The SPGR scale consists of 24 items describing organizational behavior. Each item is rated on a scale of the behavior described as occurring never or seldom (1), sometimes (2) or often (3). The organizational behaviors are described along dimensions labeled as; Control versus Nurture (C-N), Opposition versus Dependence (O-D), and Withdrawal versus Synergy (W-S), where the value for the poles of the three dimensions’ results from the ratings of four of the 24 items. A brief description of the behavior describing each dimension is given in Table 1.

The “Control” pole of the C-N dimension includes analytical, task-oriented, or autocratic behavior and the “Nurture” pole includes caring, empathic, or spontaneous behavior. Along the O-D dimension, “Opposition” includes critical, assertive, or self-sufficient behavior and “Dependence” is passive and obedient behavior. Along the S-W dimension, “Synergy” includes engagement and constructive goal-orientated behavior, and “Withdrawal” is passive rebellion and resignation [14, 27, 32]. The theoretical foundation for SPGR and psychometrics has been elaborated in the work of Sjøvold [14, 27]. A further discussion of the psychometrics of the SPGR methodology can be found in the SPGR manual [31]. In Table 2, the results of the survey are presented by two facets for each pole of the SPGR dimensions (eg C1 and C2 for C) and transposed to a nine-point scale (Fig. 1).

Independent samples of students’ t-tests were conducted to look for differences between the two selected perspectives based on the findings and the two samples were correlated. The relevant data was analyzed using the Statistical Package for Social Sciences (SPSS Inc., Chicago, IL, USA) version 21.0 for Windows. All research hypotheses were tested at the 0.05 and 0.01 significance level for the two-tailed test. Checking for normality gave acceptable results in our sample for the test used.

The ethical guidelines of voluntary participation and the possibility of withdrawal at any point were followed. All gathered data was anonymized. The Regional Committees for Medical and Health Research Ethics in Norway assessed the study as a quality assessment project of the actual hospital. Based on that the actual department management and the privacy ombudsman at the hospital sanctioned the study.

It seems like the work environment at succ unit are characterized by high influences in ruling (C1), task-orientation (C2), caring (N1), criticism (O1), loyalty (D1), acceptance (D2), engagement (S1), empathy (S2) and less by resignation (W1). Task-orientation (C2), caring (N1), engagement (S1), and empathy (S2) can be characterized as positive qualities in the work culture, so long as they do not contribute to imbalance related to the other vectors. Both empathy and engagement belong to the “Synergy” dimension, which is important in organizations for developing a higher level of maturity in both independent work and collaboration [26]. However, resignation (W1) represents negative qualities in the work culture; findings reported in Table 2 indicate that the respondents at the unsucc unit experience higher levels of self-sacrificing and resignation than the succ group. The high mean scores on the self-sacrificing vector, may be associated with employees who are not always doing their tasks with joy, and can yield fertile conditions for a culture of complaining, dissatisfaction, and passivity. Both the vectors ruling (C1) and task orientation (C2) can be positive qualities in a changing process because the properties of the vectors can develop into a more efficient way of dealing with the changes, and can also contributed to a feeling of autonomy and having decision-making opportunities in the work culture [35]. The vectors loyalty (D1) and acceptance (D2) can also further the implementation process. When healthcare workers feel loyal and accepting of changes, they accept the task of working towards a common goal. Also, when making changes and implementing them it is important to be critical in a constructive way. The high scores in task-orientation (C2), caring (N1), and engagement (S1) indicate that this is present in the succ unit. This can lead to both collaborative relationships and promote decision making [36].

Task-orientation can be viewed as a more “high tech” approach to the patients’ situation, especially if this includes implementation of new technology [1, 22]. Earlier findings show that computer technology influenced the communication between the patient and healthcare personnel and lead to an “artificial way of communication” [22].

To facilitate changes in healthcare, it is important to influence the behavior and intentions of the healthcare personnel [1]. Both behavior intentions and behavior are influenced by several factors such as attitudes, norms, and motivation, and are well-described [19, 37]. Influencing values and norms is generally difficult, whereas motivation and attitudes are more susceptible to influence and may be influenced by the healthcare personnel’s present life situation [1].

As found in another study [3] user input is important to a successful implementation process. In this study, we found that the respondents at the succ unit had higher scores on the vectors task-orientation (C2) and engagement (S1), which may indicate a work culture that promotes discussions where the health care personnel may express their opinions. Together with high scores on the vectors loyalty (D1) and acceptance (D2), these attributes can promote general agreement, which has been found to be an important factor in obtaining successful implementation [4]. Furthermore, the high scores on the vectors task-orientation (C2), loyalty (D1), and acceptance (D2) from the respondents shows a work culture at the succ unit that could point toward the use of structured decision support procedures in the implementation process at this unit, as found to be crucial in another study [5]. The rising research - practice gap described by Ramsey [6], may be bridged by supporting both task-oriented and engagement behavior among healthcare personnel when changes are introduced and implementation is begun. This may also contribute to a more positive implementation climate and a focus on the existing constraints in the work culture, described by other studies as a manager’s responsibility to foster [7, 8].

This study also has limitations. The two units had different implementation programs, but both units participated in an implementation program before the implementations. Many years passed between the two studies in this article, and other aspects may have an influence on the work culture. However, the present findings can give an indication as to the direction that research ought to follow in subsequent studies. In subsequent studies, a control group could be used to see if the results differ in a unit without an implementation process. The response rate in both units were high with 69% and 70% participation respectively. Nevertheless, the sampling process may have influenced the results if only the most enthusiastic healthcare personnel participated; and we have little knowledge about the non-respondents. This study was conducted in Norway with a Norwegian population. In Norway, work conditions are usually favorable for workers, so the results of this research could not be generalizable to other contexts without taking that into consideration [38]. Furthermore, the study has been carried out in a field where this focus has not been thoroughly described previously.