Psychiatric nurses often write “provide structure” in their nursing plans, but without further description of nursing actions or achieved goals. The Nursing Interventions Classification (NIC) (Bulechek, Butcher, & McCloskey Dochterman, 2008) also does not mention an intervention called “provide structure” or “providing structure.” It is thus questionable whether nurses know exactly what to do to provide structure for psychiatric patients.
A recent review of the research literature using NIC keywords closely related to the concept of “providing structure” (e.g., use of structure, structure, restrictiveness, limit setting, setting limits, therapeutic milieu) revealed 40 research or opinion articles that were based on predominantly qualitative research. The methodological quality of these studies was assessed using the criteria of Van Tulder, Assendelft, Koes, Bouter, and the Editorial Board of the Cochrane Collaboration Back Review Group (1997) and, despite sufficient methodological quality, it proved difficult to compare the data on providing structure. Research designs and means for providing structure differed widely across the studies reviewed.
Based on the literature review and closer inspection of the included studies, three elements of providing structure could be discerned, namely to impose and maintain rules and limits, to assess the condition of the patient, and to interact with the patient (Voogt, Nugter, Goossens, & Van Achterberg, in press). With regard to the imposing and maintenance of rules and limits, two continua could be distinguished: (a) the continuum from general to specific structure (Garritson, 1983; Ransohoff, Zachary, Gaynor, & Hargreaves, 1982; Sebastian, Kuntz, & Shocks, 1990; Silver Curran, 2007; Vatne & Fagermoen, 2007; Yonge, 2002), and (b) the continuum from least to most restrictive forms of structure (Caplan, 1993; Delaney, 2006; Kozub & Skidmore, 2001; Morales & Duphorne, 1995; O'Brien, Woods, & Palmer, 2001; O'Brien, 2000; Vatne & Fagermoen, 2007).
In relation to the assessment of the patient's condition, four aspects were mentioned. These were the assessment of ego functions (Benfer & Schroder, 1985; Kerr, 1990a, 1990b), responses of the patient to being limited or supported (Delaney, Rogers Pitula, & Perraud, 2000; Garritson, 1983; Lancee, McCay, & Toner, 1995; Ransohoff et al., 1982), ability of the patient to adequately respond to redirection (Kozub & Skidmore, 2001), and the individual need of the patient for autonomy and self-control (Garritson, 1983; Lowe, 1992; Lowe, Wellman, & Taylor, 2003; Mohr, Mahon, & Noone, 1998; Morales & Duphorne, 1995).
Through the interaction nurses tried to relate to each patient, and nurses tried to put themselves in the patient's shoes to recognize the patient's individuality (Björkdahl, Palmstierna, & Hansebo, 2010). Nurses also tried to exchange mutual expectations about what to expect during treatment. This led to involvement of the patient in a process of cooperation with nurses and it led to the formulation of clear treatment goals (Caplan, 1993; Hopkins, Loeb, & Fick, 2009; Sebastian et al., 1990; Vatne & Fagermoen, 2007; Walker, 1994; Yonge, 2002). The three elements of providing structure appeared to be interrelated.
On the basis of literature review, three goals and several effects of providing structure could be identified, which ranged from most to least restrictive: to attain external security for the patient, to make mutual expectations within the treatment relationship explicit, and to attain the feeling that the patient better fits into the world and is recovering from illness (Voogt et al., in press).
The attainment of external security entailed the creation of a climate of trust, an intrapersonal feeling of safety at the same time (Lowe, 1992; Mohr et al., 1998; Puskar et al., 1990; Yonge, 2002), and a physically controlled environment to allow the patient to express tensions in a safe manner (Benfer & Schroder, 1985; Björkdahl et al., 2010; Vatne & Fagermoen, 2007; Vatne & Holmes, 2006; Vrale & Steen, 2005). Providing structure seemed most effective when psychiatric nurses respected the patient's autonomy, were sensitive and responsive to signs of fear and anxiety, and adapted their actions to the needs of the patient. If not, providing structure led to resistance, anger, fear, and incomprehension of treatment policy (Johansson & Lundman, 2002; Lancee et al., 1995; Mohr et al., 1998; Olsen, 2001; Silver Curran, 2007; Vatne & Fagermoen, 2007; Vatne & Holmes, 2006; Walker, 1994).
When mutual expectations were made explicit, barriers between nurses and patients lowered. This occurred as nurses tried to put themselves in the shoes of the patient (Björkdahl et al., 2010; Hopkins et al., 2009). In addition, the occurrence of unacceptable behavior could be avoided (Lowe, 1992). When inadequate communication between nurse and patient existed, expectations became unclear, mutual misunderstandings arose (Johansson & Lundman, 2002; Olsen, 2001; Walker, 1994), patients became insecure about the course of treatment, developed feelings of loneliness (Johansson & Lundman, 2002), and patients felt deceived, dissatisfied, and showed resistance to providing structure (D'Antonio, 2004; Olsen, 2001; Vatne & Fagermoen, 2007).
The final goal of achieving a better fit in the world meant that the nurse helped the patient to develop from a patient into a person in society, to live independently, and to allow the patient to have autonomy and responsibility, in order to enable the patient to shape his own recovery (O'Brien et al., 2001; O'Brien, 2000; Vatne & Fagermoen, 2007; Vatne & Holmes, 2006). Effects in relation to this final goal were that the patient felt supported, was able to function independently both inside and outside the hospital (Greig, Miller, Rollo, & McGillvray, 1985), and that he/she developed from a patient in the community to a person in the community (O'Brien et al., 2001; O'Brien, 2000).
Despite being able to identify critical elements of providing structure and three main goals for providing structure, no studies were found on the effectiveness of providing structure as a psychiatric nursing intervention.
The literature review led to the following provisional definition of providing structure as a psychiatric nursing intervention:
The aim of providing structure is to create a workable, well-organized situation between nurse and patient in which both can work purposefully and effectively towards the strengthening of ego-functions, towards the attainment of external security for the patient, towards explicit mutual expectations within the treatment relationship, towards participation in different life areas and recovery on the part of the patient. In order to do this, the nurse uses interaction, assesses the patient's condition, and imposes and maintains rules and limits in a balanced manner. (Voogt et al., in press)
Although in this provisional definition clear interrelations between the interaction between patient and nurse, the assessment of the patient's condition, and the imposing and maintenance of rules and limits existed, it remained unknown how these three elements interrelated, how nurses assessed the condition of the patient, and how nurses determined the amount of structure needed. For actual nursing practice, a balanced use of the different elements of providing structure appeared to be necessary. It was also difficult to decide if all of the goals related to providing structure were discerned and how nurses used the three key elements of providing structure to attain the aforementioned goals.
However, providing structure can also be construed as a complex intervention due to the number of components (i.e., nursing activities) that it involves. According to the Medical Research Council (MRC, 2008), many nursing interventions can be classified as a complex intervention, which they define as an activity that contains a number of components with the potential for interactions between them and which—when applied to the intended target population—can produce a range of possible outcomes and thus variable outcomes. For example: Providing structure applied in one situation can result in a calmed patient (or group of patients) but, applied in a different situation, it can result in an escalation of the situation (Kozub & Skidmore, 2001; Lancee et al., 1995; Lowe, 1992; Lowe et al., 2003; Sebastian et al., 1990). The complexity of an intervention can thus be defined in terms of the number of components and possible interactions between these components, and the degree of flexibility and tailoring possible for the intervention.
With that preceding in mind, an observational study of actual nursing practice was undertaken with an eye to developing a useful framework and an evidence base for the complex nursing intervention of providing structure, which would help nurses to underpin their interventions to provide structure. At this moment, the practice-based knowledge and experiences of nurses in combination with their personal styles seem to determine the application of providing structure.
The aim of this observation was 2-fold. First, we wanted to determine what nurses do when providing structure in actual practice and thus describe the content of various “providing structure” interventions in detail. Second, we wanted to identify the impact (i.e., results) of different forms of providing structure. The following research questions were thus formulated.
- What do nurses do when providing structure in actual practice?
- What are the observed results of providing structure?
- Top of page
- Discussion and Conclusions
Based on the principles of grounded theory (Charmaz, 2006; Strauss & Corbin, 1998; Wester, 1995), a qualitative research design was adopted. Grounded theory aims to discover the perceptions and significance of people's behaviors, to reconstruct the ways in which people make sense of behavior, and to identify how people's interpretations of behavior influence their interactions. Grounded theory also aims to ground theory in empirical data, and thus the name.
To obtain a comprehensive description of behavior and interactions, we undertook participant observation. These observations resulted in an extensive set of field notes on the verbal and nonverbal behavior of nurses and patients during events in which structure was provided.
For the study of complex interventions, the MRC (2008) recommends a combined process of development, testing/piloting, evaluation, and implementation with a dynamic interchange between the different phases in the process. Each phase is important and can be quite lengthy in itself (see Figure 1).
The present observational study can be considered part of the development phase for the creation of a complex “providing structure” intervention. First, we identified the evidence base for such an intervention via the aforementioned review and analysis of the literature (Voogt et al., in press). In the present study, we initiated the development of a framework and theory for providing structure as a psychiatric nursing intervention. A qualitative research design and a participant observation approach were chosen for this purpose. Furthermore, on the basis of the analyses of the qualitative data obtained in such a manner, the process and outcomes of providing structure for psychiatric patients were modeled.
With the use of purposive sampling (Morse & Field, 1996), we selected patients from two intensive care wards at a mental health hospital for observation. The expectation was that events that require providing structure would occur rather frequently on such wards with explicit boundaries. The patients on these wards could be characterized as patients with acute mental illness or a combination of acute mental illness and addiction problems, long-term care and chronic treatment needs, and a profound need for structure to meet these needs. Both wards were closed, and there were signed agreements with the patients in their treatment plans. Each ward had two separate units with 24 patients and 15 nurses in each unit.
The observed events had to meet one or more of the following criteria for consideration to be included:
- The nurse had intervened because the patient had to participate more in certain life areas; the intervention was to ensure safety; or the intervention was to create a more habitable environment.
- The nurse wanted the patient to do something that the patient initially did not want to do.
- The patient wanted something the nurse could not provide immediately.
The end of an event was assumed to be reached when the nurse and patient parted and the verbal/nonverbal communication between them ceased. These criteria were developed in expert meetings with experienced psychiatric nurses. It took three meetings until consensus was reached.
Observations were conducted between August 2009 and January 2010. Intensive nurse–patient interactions were expected to occur at wake-up, breakfast, patient meetings (either group or individual), coffee breaks, lunch, evening meal, and bedtime. A total of 52 events met the aforementioned criteria and were included for analyses: 30 on the crisis ward and 22 on the ward for double diagnoses.
The observations were performed by the first author. The observer had to be unobtrusive, so as not to disturb the usual ward routines (Polit, Beck, & Hungler, 2001). In one case, that meant to sit in the corner of the central living room with an eye on the nursing office, and in the other case the observer joined dinner with patients and nurses. The data from the first five observations were used to evaluate the initial observation format. The use of a standard observation format proved impossible because the observed events did not follow the order of the format or a standardized order. The use of a voice recorder to document the interaction also proved impossible because we could not get close enough to the participants without interfering. Thus, in the end, the observer simply made short notes on the nurse–patient interaction and providing structure process during the event. Immediately following the event, these notes were elaborated to record the event in as much detail as possible. After completion of the first 25 observations, the observation protocol was again evaluated to check that the written descriptions of the events met the inclusion criteria. No further adaptations were necessary.
The observer noted the date and time of the event, the number of patients and nurses present, the atmosphere on the ward at the time, the event itself, the initial reaction of the patient or nurse, the follow-up interaction, the verbal and nonverbal communication, how the interaction ended, and the outcomes of the interaction. The names of the patients and nurses were not noted and, in such a manner, the anonymity of the patients and nurses was assured.
Both the patients and the nurses were informed individually and in group meetings about the aims of the study, the methods, the use of the data, and the possibility of withdrawing from observation or the study any time. This information was also provided in a written document. Consents of patients and nurses were recorded in the minutes of group meetings. Prior to the start of each period of observation on a unit, the observer asked the nurses if the patients were stable enough to be observed and if there were any patients who did not want to be observed. When this proved to be the case, those patients were indeed excluded from observation. The medical ethics committee of the mental healthcare hospital approved the study and provided written consent.
To develop a categorization scheme for the observed interactions and activities, we undertook open, selective, and axial coding (Charmaz, 2006; Strauss & Corbin, 1998; Wester, 1995). We used constant comparison to check the emerging codes and categories, to examine tentative ideas regarding the data, and to refine the categories.
After a detailed open coding of the events, a preliminary coding tree of categories and activities was developed with brief descriptions of the codes on memos by the researcher. Next, the initial coding tree was discussed with a research group, both individually and at a group meeting for which cases were selected at random to discuss. Among the initial activity codes were “Ask something from the patient” and “Confront the patient.” As a result of the constant comparison of the event data with the coding memos and the use of axial coding, the researcher decided the two codes of activities to become part of a new category code “Stop patient's current behaviour” because both activities were aimed to stop the patient's behavior. Furthermore, the former code “Ask something from the patient” transformed in “Request the patient to do something.” “To request to do something” specifies expected behavior better than “to ask something from the patient.” “Confront the patient” transformed in “Confront after action took place” because this emphasized the importance of reflection on activities the patient had performed.
With the use of selective coding, this resulted in the category code “Stop patient's current behaviour” together with the activity code “Request” being a less restrictive manner of stopping patient's behavior than together with the activity code “Confront” as a more restrictive way to respond.
Finally, a logbook was created to keep track of all discussions of the cases in the research group, the decisions made during the analysis of the data, and the changes made to the coding tree.
Data analysis was supported by the software tool for textual analysis, MaxQDA (Kuckartz, 2007). Data analysis stopped when no additional data, theoretical insights, or properties of the core categories were discovered (Charmaz, 2006).
Discussion and Conclusions
- Top of page
- Discussion and Conclusions
The aim of this study was to describe, via participant observation, the complex nursing intervention of providing structure. In our search of the research literature, we found that the intervention label “providing structure” does not exist in the Nursing Interventions Classification (Bulechek et al., 2008) although psychiatric nurses in the Netherlands often use such an intervention (i.e., provide structure).
According to the guidelines of the MRC (2008), providing structure can be considered a complex intervention. Both an examination of the literature and observation should thus be part of the development phase for the study of this complex nursing intervention (MRC, 2008). In doing this, we asked the following two research questions:
- What do nurses do when providing structure in actual practice?
- What are the observed results of providing structure?
The answers to these research questions contribute to constructing a framework and theory for “providing structure.” With regard to the first research question, three phases could be distinguished which cover the beginning and end of an event, and also cover the range of activities of nurses during an event that required providing structure:
- The start of the interaction
- The intervention phase
- The end of the interaction
A nurse or a patient initiated an interaction typically when they wanted something from the other party. The subsequent response of the patient or nurse then formed the start of the intervention phase. The first response of the patient to a nurse was often a turning point in the event, which could either escalate or remain peaceful. The initial responses of the patients could be categorized along a continuum from most to least cooperative. The first responses of the nurses to a patient initiative could be divided into four categories: ask the patient for more specific information, explain something to the patient, support and encourage the patient to follow treatment as planned, or stop a patient's current behavior.
With regard to the second research question concerning the end of an interaction and the observed results of providing structure, the last response of the patient could again be classified along a continuum from most to least cooperative. The last responses of the nurses could be divided into four categories: reflect with the patient on what has just happened, act according to what has been agreed upon, concede to the patient while staying in contact but without mutual agreement, or set a clear boundary in order to stop the patient's persistent behavior. Although the reason for the nurse to concede to the patient could not be discerned through observations, we assume that this was done to prevent further escalation.
The frequent call for providing structure in nursing plans suggests that providing structure is largely experience based rather than evidence based, and that providing structure thus depends upon the knowledge and experience of the nurses and nursing teams. The MRC (2008) has acknowledged these findings and therefore considers close examination of the experiences of nurses to be part of the development and evaluation phases for the establishment of complex interventions. In the introduction, it was mentioned that the outcomes of providing structure could vary dramatically. Some preliminary explanations for this variability can be derived from the continuum of possible patient's reactions, as described in Table 3.
Table 3. Continuum From Most to Least Cooperative Final Responses of Patients in Providing Structure Interactions
|Patient verbally confirms cooperation||Patient acts as nurse proposes or asks him to do||Patient cooperates, but remains restless or shows irritation||Patient seems not willing to cooperate, but stays in contact without an explicit and clear mutual satisfying result||The patient does not cooperate, shows he does not want to stay in contact with the nurse, or returns and shows the same undesirable behavior as before and provokes the nurse||Patient refuses|
Our observations showed the first response of the patient to a nurse initiative to stand out as a turning point in the interaction between nurse and patient. At this point in an interaction, the event can easily escalate or remain relatively stable. In only 8 of 52 events, nurses mention their expectations to patients which may itself influence a successful application of providing structure. This refers to the exchange of expectations we found in the literature review, that, for example, when inadequate communication between nurse and patient existed, expectations became unclear and misunderstandings arose (Johansson & Lundman, 2002; Olsen, 2001; Walker, 1994), which led to resistance (D'Antonio, 2004; Olsen, 2001; Vatne & Fagermoen, 2007).
It is also possible that following assessment of the patient's first response, the intuitive reasoning of the nurse, the personality of the nurse, and the experience of the nurse and the nursing team come to bear on the situation and shape the reactions of both patient and nurse during the subsequent interaction.
In our review of the literature, we discovered that the NIC (Bulechek et al., 2008) did not use the intervention label “providing structure” despite the frequent use of this term by psychiatric nurses. In this literature review, we further identified three elements of providing structure: to impose and maintain rules and limits, to assess the patient's condition, and to use interaction as patient and nurse.
In this observational study, we could discern the three elements of providing structure as distinguished in the literature review. During our observations, we recognized the continuum of general to specific agreements mentioned in the literature, in the continuum of explaining general to specific issues (see Table 2). But, it was impossible, with the use of observations, to understand just how the nurse assessed the condition of the patient, for example, in relation to the assessment of ego functions (Benfer & Schroder, 1985; Kerr, 1990a, 1990b), or the assessment of the patient's need for autonomy and self-control (Garritson, 1983; Lowe, 1992; Lowe et al., 2003; Mohr et al., 1998; Morales & Duphorne, 1995). On the basis of the observations, we assume the existence of a continuing assessment of the condition of the patient. The ability of the patient to adequately respond to redirection (Kozub & Skidmore, 2001) and the responses of the patient to being limited or supported (Delaney et al., 2000; Garritson, 1983; Lancee et al., 1995; Ransohoff et al., 1982) may be first assessed by using Table 1, and during following contacts by using Tables 2 and 3, which all contain a range of responses and nursing activities.
On the basis of our previous review of the literature and the present observational study, we conclude that a number of frequently mentioned and related concepts, keywords, and activities could be distinguished as part of providing structure as a psychiatric nursing intervention. The literature review yielded three elements, two continua, three goals, and a description of effects related to these goals. The observational study delivered a process of providing structure, where each phase of the three steps process activities of nurses and responses of patients are distinguished. On the basis of two studies, we might suggest that “providing structure” be introduced as a new intervention label within the NIC (Bulechek et al., 2008). To supplement this proposal, the nursing responses and activities described on the basis of our observations should be included as part of “providing structure” in the NIC.
We also conclude, on the basis of the present observations, that we can now elaborate upon the process of providing structure as a complex nursing intervention and identify a useful theoretical framework, model the process of providing structure, and evaluate the attained outcomes. In such a manner, we can better understand just why patients and nurses respond in a particular manner and make the decisions that they do during an interaction that calls for providing structure. And we can then try to gain further insight into their expectations with regard to their actions in future research.
The first limitation on the present study was the position of the observer and the incomparability of the events we observed. The position of the observer on the two wards had to be unobtrusive in order not to disturb usual ward routines; this required a certain distance from events and may have led to the incomplete description of some events. Therefore, we made adjustment to record the nurse–patient interaction. This is described in the Method section.
The second limitation on the basis of this observational study is that the data were collected in one country and are therefore culturally specific.
Implications for Nursing Practice
The present observations and classifications of the interactions between patients and nurses during events requiring the provision of structure give us a provisional framework for reflection and feedback on the performance of nursing activities. Psychiatric nurses and nursing teams should be encouraged to reflect upon exactly what they do when providing structure for a patient and the results that this yields. Such reflection can provide insight into why providing structure may sometimes—but not always—result in an escalation of events. Moreover, the results of this observational study will contribute to a formalization of the nursing intervention in the NIC.