Nurses’ experience with the clinical application of a research-based nursing protocol in a long-term care setting


  • Michelle R Simpson BSN, RN,

  • Patricia Stevens PhD, RN,

  • Christine R Kovach PhD, RN

Michelle R Simpson
College of Nursing
University of Wisconsin-Milwaukee
Cunningham Hall
1921 East Hartford Avenue
WI 53201-0413
Telephone: +1 414-229-6233


Aim.  To describe nurses’ experience with the clinical application of a research-based nursing protocol (The Serial Trial Intervention) within a long-term care setting.

Design.  A descriptive, qualitative study was conducted with a convenience sample of eight nurses from three nursing homes, who assessed and treated residents with dementia according to the Serial Trial Intervention protocol.

Methods.  Each nurse participated in a semi-structured interview between September 2003 and May 2004. Interviews were audiotaped and transcribed verbatim. Qualitative content analysis of the data, including thematic analysis, was used to identify patterns of experience.

Results.  Three themes emerged that offer insight into the factors contributing to implementation of research-based practice in a clinical setting. These include determining to intervene, pertinent steps of the protocol and facilitators and barriers.

Conclusions.  The interplay between the protocol, the residents receiving care, the nurses providing care and the setting in which the care is provided, are interacting to affect the outcomes expected.

Relevance to clinical practice.  Nurses who recognize the research process, the need for continual improvement in patient care and who possess competency in comprehensive physical assessment are needed to implement this evidence-based protocol successfully. The regulatory atmosphere, workload structure and interdisciplinary collaboration are additional factors contributing to the successful use of the Serial Trial Intervention.


Research reveals North Americans are not receiving the highest quality of health care that meets their needs and which is based on the best scientific knowledge (Chassin et al. 1998, Harrington et al. 2000, Kohn et al. 2000). In accordance with these findings, The President's Advisory Commission on Consumer Protection and Quality in the Health Care Industry (1998) has set as national priorities the reduction of variation in health care practices and the communication of processes for producing optimal patient outcomes.

Nursing intervention science has the potential to contribute to improved healthcare delivery (Valente 2003). Evidenced-based practice, based upon interventional research, can improve the efficiency and effectiveness of healthcare delivery and improve health outcomes (Harris 1997, Woolf et al. 1999). However, the estimated 17 years it takes for science generated by randomized clinical trials to be translated into daily practice poses an alarming problem (Committee on Quality Health Care in America 2001).

Various approaches to changing practice based on current evidence exist (Grol 1997). Within the healthcare industry, the educational approach – that of educating the individual healthcare provider – predominates (Grol 1997, Anderson et al. 2002). However, Anderson et al. argue a healthcare provider's knowledge of current evidence may not be the single most important factor to consider when attempting to change practice. Accordingly, the purpose of this paper is to report the findings of a descriptive study in which nurses’ experiences with implementing The Serial Trial Intervention (STI), a research-based nursing protocol, were explored. The research question guiding this study was: What is the experience of nurses in implementing the STI, a research-based nursing protocol, in a clinical setting? Findings offer unique implications for nurses and the systems in which they practice.


The integration of best-known empirically tested practices into the clinical milieu has been attempted using various approaches including educational, epidemiological, marketing, behavioural, social interaction, organizational and coercive approaches (Grol 1997). In studies examining the implementation of research findings to change clinical practice, the results are conflicting. Grol argues that no single method is superior. Moreover, various groups of healthcare providers may meet with different barriers to changing practice. Lack of time to read research and implement ideas, lack of independence at work and lack of support from co-workers are among some of the barriers to implementing evidence-based practice that have been identified (Lewis et al. 1998, Kajermo 2000). Reports of limited access to literature, lack of training in critical thinking and information seeking skills and the emphasis on practical vs. intellectual knowledge are also contributing barriers (Royle et al. 1996).

Despite the uncertainty about which method most effectively facilitates change from current practice to best known practice, and the various barriers potentially impeding such change, there is the need to continue building programmes of research that can be translated into practice. In this paper, we examine one such case of translating research into practice – the implementation of the STI.

The serial trial intervention

The American Geriatrics Society Panel on Chronic Pain in Older Persons (1998) recommends that the assessment of physical discomfort in cognitively impaired older adults include non-verbal behaviours, vocalizations, changes in function and caregiver reports. However, research suggests that current models of late stage dementia care fail to guide recognition and treatment of a severely demented person's remaining potential for comfort and quality of life (Algase et al. 1996, Kovach et al. 1999).

Studies have demonstrated that behaviours such as aggression, fidgeting, screaming and tensing muscles are symptoms of physical and affective discomfort and other unmet needs (Hurley et al. 1992, Mahoney et al. 1999, Kovach et al. 2001). However, physical discomfort is untreated and under treated, both pharmacologically and non-pharmacologically in patients with dementia (Ferrell et al. 1990, 1992, Carroll 1997, Davis 1997, Feldt et al. 1998). The STI offers a combined physiological, psychosocial and environmental treatment approach – as it guides examination of behaviours associated with dementia from both the physical and affective domains.

The STI is a nursing intervention which incorporates systematic serial assessment and sequential trials of treatments designed to improve comfort of people with dementia who can no longer verbally make their needs known (Kovach et al. 2006b). Assessments are linked to treatment and subsequent evaluation of patients’ responses. The STI delineates that the nurse begin with a physical assessment and proceed to an affective needs assessment. Targeted treatments and trials of treatments are provided by the nurse. The STI is comprised of five steps. These include (i) physical assessment with critical appraisal of patient's history; (ii) affective assessment to include psychosocial and environmental factors; (iii) trial of non-pharmacological comfort treatments; (iv) trial of analgesics; and (v) consultation and possible trial of psychotropic medication. The nurse proceeds sequentially through the steps until the problem is treated and behavioural symptoms have returned to baseline or substantially improved. In the randomized double-blinded experiment investigating the effectiveness of the STI nursing protocol, the STI was found to be more effective in decreasing discomfort and returning behaviours to baseline than current care. The group of nurses using the STI also demonstrated more persistence in assessing and intervening than the control group nurses (Kovach et al. 2006b).


Setting and sample

In the randomized double-blinded clinical trial that evaluated the effectiveness of the STI, a total of 54 nurses employed within 14 long-term care facilities in one Midwestern state, were involved in the assessment and treatment intervention for individuals with dementia. Inclusion criteria for nurse participants were (i) at least six months of experience caring for individuals with dementia; and (ii) employed on the dayshift 32 hours or more per week.

Nurses agreeing to participate in the STI study received an initial 45-minute introduction to the study, seven hours of classroom education and educational reinforcement at the site twice a week. Education was provided by two advanced practice nurses. The training included enhanced physical assessment skills; focusing on common sources of physical discomfort in older people, assessment of high arousal and low arousal, environmental stress, meaningful human interaction, geriatric pain management, review of non-pharmacological comfort treatments and appropriate and inappropriate use of psychotropic medications. Nurses used daily logs to record data for the variable measurements and were maintained by the nurses for one month following the first change in the patient's behaviour.

The study reported in this paper includes descriptive data from eight of the 54 nurses who participated in the clinical trial. At the time, the interviews were conducted for the descriptive study, the nurses who had the most recent experience with implementation of the STI were asked to participate. All nurses who implemented the STI at these sites agreed to the interview. This yielded a sample size of eight. These eight nurses were employed at three of the seven facilities in the treatment aim. Seven of the participants were Registered Nurses (RNs) and one was a Licensed Practical Nurse (LPN).


A descriptive, qualitative design was used to elicit nurses’ experiences using the STI nursing protocol (Janesick 2000). Nurses were interviewed for 30 minutes to one hour in a private office at their work site by a single interviewer. A semi-structured interview guide was used to elicit nurses’ answers to open-ended questions about their experience with implementing the STI protocol (see Appendix A). Probing questions were asked to expand upon specific topics. Interviews were audiotaped and transcribed verbatim. All procedures were approved by the appropriate institutional review board.


Qualitative content analysis of the data, including thematic analysis, was used to identify patterns of experience (Lofland & Lofland 1995). To begin uncovering themes, the nurses’ interviews were read several times and notations were made of key concepts, paraphrases and quotations. Initial themes were generated from each nurse's interview. Key concepts, paraphrases and quotations were identified and extracted from the transcripts. These key concepts, paraphrases and quotations were then shared and discussed by members of the research team. Adequate paraphrases that sufficiently captured the theme were constructed and compared across individual nurse interviews to identify similarities and differences. The themes that were consistently supported across all of the nurse interviews are reported.


Three significant themes emerged in these nurses’ interviews. The first theme is the identification of behaviours and determining to intervene. The nurses describe their thought processes in determining what they consider to be a triggering behaviour in a particular resident. A second theme explains the steps of the protocol perceived to have changed the nurses’ practice the most. The third theme is the facilitators and barriers to successful implementation of the protocol. Each theme is discussed in turn and quotes from the nurses are used to illustrate analytic arguments.

Determining to intervene

Individuals with dementia often exhibit verbal or physical behaviours that are disruptive, problematic and challenging to manage (Allen-Burge et al. 1999, Burgio et al. 2001). The nurses in this study describe their decision-making process in determining what behaviours require intervention. The behaviours these nurses indicate require intervention fell within three categories: (i) the resident demonstrates a behaviour, i.e. outside their norm, signalling some type of discomfort; (ii) the resident is a physical threat to self; and (iii) the resident is disruptive to others. Each of the nurses interviewed expressed they likely missed behaviours. In addition, the consistent theme of determining to intervene was not solely based on the resident's demonstration of a behaviour outside their norm, but was often based on whether that behaviour was disruptive to others. Here, the behaviours that aid the nurse in determining to intervene are described. Several nurses described observing a change in behaviour or functioning that triggered the nurse's attention:

Changes in behavior, changes in mood. Sometimes it's a weight loss – that's a big one. Changes in their ADL functioning.

Some of the behaviors, like occasionally acting out or slapping aren't as urgent as somebody who actually looks like they're in physical distress- those who start falling immediately.

Another nurse states:

It depends on what the behavior is, the person, what the impact is on them, and what the impact is on others. I see the impact on the resident, the impact on the other residents, and the impact on the caregivers. You have to look at the whole circle of people involved. I think that's the biggest thing.

Determining to intervene, based on how disruptive a dementia resident's behaviour is to others, is described by several nurses:

The deciding factor that aids in determining to intervene is whether there is a possibility for harm to themselves, to others, to staff - or how much the staff complains:

I think specifically a resident that needs attention is when they are acting out; disrupting other residents and disrupting the routine on the unit.

Acting out. If someone gets angry and hits a person. If the noise the person is making is totally disruptive to other people, then we need to intervene and remove that person from the area.

Individuals with dementia who display behaviours, but are described by nurses as not requiring intervention, are explicated:

Probably the ones that sit there quietly; ones that really don't act out and are not disruptive. We have a resident that continually talks to herself. She's constantly saying things to herself, but not disrupting anyone. That's just a behavior of hers. We don't necessarily always intervene when she's talking to herself. We implement a plan of action if the behavior is disruptive, but if it's not, then we just let them be.

Whether or not the person is being disruptive either to the whole unit or just to himself. That's when you decide that that's the one that needs attention.

Looking back, some of the behaviors the residents were doing, they no longer do - because the dementia had progressed and they just gotten over it. Or they continue to do it (the behavior) and it's harmless. So, we just…it's easier to let them do it since they're not hurting anybody than trying to say ‘no you can't do that’ because then that just increases the agitation.

If that person is exhibiting a behavior and it's not a problem to someone else. If a person wants to go down the hall picking at the wall, they can do that as long as they're not picking at another person.

It is compelling to recognize that six out of eight of the nurses interviewed reported that the most prominent factor provoking intervention was not based solely on the behaviour the individual with dementia displayed, but how disruptive the behaviour was – either to himself (the individual displayed physically unsafe behaviours) or to others (residents or staff).

Pertinent steps of protocol most influential in changing practice

Each of the steps of the STI is significant to the successful assessment and treatment of physical and/or affective discomfort experienced by the dementia resident. However, when asked what step or steps of the protocol were the most influential in changing their resident care, these nurses consistently indicated step one. This step involves performing a comprehensive physical assessment and checking the resident's history. The nurses describe how this step was not consistently performed prior to implementing the STI. In addition, the steps which least influenced the care provided to subjects in this study included the need for consultation and administration of psychotropic. One nurse stated, ‘We never got that far’. Another indicated, ‘I was able to find out what the problem was before using a psychotropic’.

Here nurses describe what they perceived to be the steps that had most impact on and changed the way they provided care:

I think we have to start with step one all the time, but we do not do that automatically. Once you have step one figured out (physical exam and history), we need to stop and think about step two (the affective assessment), because we automatically go to five (consultation and psychotropic) and we're used to going to four now (trial of analgesic).

We need to be taking that step back and actually look at their physical condition first.

Or as another nurse stated:

It is the combination of the steps that has changed the delivery of care.

The process of a comprehensive physical assessment by the nurse and the actual critical appraisal of the resident's medical history was not usual care. Current practice allows for assumptions to be made regarding dementia and behaviours associated with this disease. The steps in the STI offer a process by which to place these assumptions in the background and consider the physical aspects along with the medical history of the resident first.

Facilitators to successful clinical implementation of the serial trial intervention

Physician collaboration

The impact of collaborative efforts between the nurse and the physician can positively effect the implementation of the STI. The nurse, through the systematic use of this protocol, may determine an analgesic or other prescribed medication could decrease a resident's discomfort. Through open communication and effective collaboration, the best outcome for the resident can be achieved:

When the physician came in and we were using the tool, I went back and showed it to him. I said, ‘this is what we keep doing and he's not getting any better’. The physician found this very interesting and gave us an order for a drug for the itching that we could give at bedtime. The resident's behavior improved dramatically. The thing is - you need to have a doctor who is willing to sit down and look at this tool. This was a wonderful physician.

Seeing the intervention as a challenge

The various roles in which the nurse functions may either be seen as a burden or as a challenge. It is when the skill or responsibility is seen as a challenge that the care provided will more likely positively impact a resident. Here a nurse describes:

She (the nurse) liked doing this; she saw it as a challenge. I think you have to get someone that really would want to be involved in this type of protocol, really stick with it, and not see it as a meaningless task or a burden. You need people who are willing to take the time to go through the process, rather than just seeing it as a meaningless task. There was some staff that were really involved. Another nurse of mine was not. You have people that jump on the bandwagon and say, ‘oh yeah, I'm all for seeing if we can improve this’ and others who are just not interested.

Educational reinforcement

When a new method or process is taught, the need for continuing feedback and clarification is necessary. The idea that once the classroom education is complete and subsequent successful implementation of the learned process will follow is not necessary true. The nurses who received education on the use of the STI comment on the benefits of continuing support and learning:

I think what made it the easiest was the training and the weekly contact with you and your staff to clarify issues and keep the ball rolling.

It takes a little time and effort to use all of the steps and to really think about them. I think we are a little bit better than when we started, but we still need constant reminders.

Barriers to successful clinical implementation of the serial trial intervention

Lack of physician collaboration

As nurses described how they proceeded through the five steps of the protocol, the inability to implement the step involving administration of an analgesic was noted. This was largely because of lack of physician collaboration. Here, nurses describe their experiences with attempting to facilitate the use of this protocol step:

We tried Tylenol and it really didn't make any difference. The doctor didn't want to go any further.

We never got any further than step two. We did the physical assessment, found the problem, did the treatment. We tried some Tylenol for his knee, but we couldn't get any further. The Tylenol wasn't really effective and we couldn't get physician cooperation.

Lack of time

While the process of following the protocol was deemed useful, the lack of time available to do so successfully was identified. The regulatory environment of long-term care and financial constraints experienced in these settings leads to tremendous amounts of documentation – and their often exists limited staff to complete this work. Here nurses describe how the time factor impacted on their care:

I think to really do the resident credit and do a service to them, you have to be consistent. But do we have the time and energy? No.

I'm sure this happens at other facilities too…distracted by other things that we had to do. We would get interrupted for a time going through step one, two, and three. We'd have to wait for awhile and come back to it. Because I'm sure you know in a nursing home everything is not perfect, and it would be nice if we could have sat there for four hours a day and just monitor five residents without any distractions or interruptions. Like I said, on my unit, it was quite difficult.

The most difficult part of using the protocol? I would say time – not time to do it, time to sit down and write it down. Time to write down what you were doing.

Disengaged staff

The participation in the use of this protocol was voluntary. Nurse managers of units along with other staff nurses were asked to participate. Some nurses did not see the value in learning this protocol and subsequently did not receive training or participate in the study. There were those nurses, who when approached with the protocol, were simply not interested. Clearly, nurses who are disengaged from evidenced-based practice may have an impact on the quality of care provided to dementia patients in these settings:

I talked to both of my nurses about the protocol and what we needed to do. One really wasn't interested in following these guidelines, and I think you'd see that anywhere. Just didn't think it would benefit our unit to do this.


Limitations of this qualitative study are many. It is important to understand the issues that have an impact on nurses’ clinical application of research-based protocols. However, in this case, we are limited to nurses’ experience with one research-based protocol within one type of clinical setting. Additionally, it is not possible to know if the experiences these nurses described might have been different if the protocol had not been part of a clinical trial. Future research may be expanded to investigate nurses’ experience with the clinical application of a variety of research-based protocols, across a range of clinical settings. In addition, comparing the perceptions of using and implementing research-based practice between nurses, based on educational preparation and varying levels of expertise, may also be of value. Lastly, investigating the implementation of research-based protocols within different facilities, while considering key organizational characteristics, is also recommended.

Despite these limitations, these experiential descriptions help us to understand nurses’ perspective of the clinical application of evidence-based practice. More specifically, these findings are significant to improving the STI. The process of successful implementation of the STI began with nurses’ willingness to consider new knowledge. Nurses are increasingly expected to integrate the best clinical evidence into their patient care (Polit & Beck 2004). However, as described in this investigation, there are nurses who are not even willing to consider an alternative to their current practice. While this demonstrates lack of professional accountability, factors such as individual nurse characteristics (educational preparation and expertise level), the RN–MD relationship and subjective and objective organizational factors (job satisfaction, staffing, structure and organizational leadership) that may be contributing to disengaged nursing practice warrants further exploration.

Nurses described the manner in which they determine which behaviours exhibited by dementia residents require intervention and which behaviours do not. The residents with behaviours who are ‘quiet’ and non-disruptive are left alone. They are felt not to require interventions. This finding is troubling, as it conflicts with recommendations in the literature, the Nurse Practice Act and the education and instructions given with the STI (American Geriatrics Society Panel on Chronic Pain in Older Persons 1998, Wisconsin Statutes and Administrative Code 2005, Kovach et al. 2006). Nurses are responsible for reducing discomfort based on an individual resident's need, not on how disruptive they are. While the disruptiveness could be understood to gauge the resident's level of behaviour change, it should not be the determining factor. The finding that nurses were unlikely to intervene when behaviours were not disruptive to others is consistent with findings in the main study. Subjects who had vocal symptoms received significantly more assessment [t (54) = −1·68, P = 0·049] and received more new pharmacological interventions (φ = 0·241, P = 0·035) than those subjects who were non-vocal (Kovach et al. 2006a).

For nurses to conceptualize fully what should be considered behaviour, behaviour change in the resident with dementia and, moreover, what behaviours require intervention, the need for on-going education and consistent reinforcement will likely be needed. Support from clinical educators or patient care facilitators, whose primary function is to lead nurses in their application of best-known health care practice, could well serve patients in long-term care settings. Nurses providing direct patient care should be involved in the continuous quality improvement process employed by the organization and this process should include reading and interpreting research findings.

Lack of physician collaboration and lack of time were among the barriers to successful implementation of the STI identified by the nurses. Included among the 10 Institute of Medicine proposals for redesign of the United States healthcare system is cooperation among clinicians (Committee on Quality Health Care in America 2001). Providing physicians with the evidence of research-based nursing practice and the education and incentive to support nursing practice that requires collaborative efforts (such as is the case with the STI) is necessary. Collegial and collaborative relationships are imperative to delivering the highest quality of care to individuals with dementia. In addition, long-term care organizations need to assess the RN/LPN–MD relationships that presently exist within their setting and create mechanisms by which collaboration is fostered. Lastly, nurses caring for individuals with dementia should consider the time it takes to manage disruptive behaviour [between 5·7 and 201·5 minutes (Souder & O'Sullivan 2003)] could be far better spent effectively assessing and treating the behaviour through an evidenced-based practice (the STI) found to decrease discomfort significantly and return behaviours to baseline.


As stated by Hegyvary (1993), the world of practice is comprised of a complex system, which includes multiple factors, multiple effects and mutual causation. The interplay between the protocol, the residents receiving care, the nurses providing care and the setting in which the care is provided, are interacting to affect the outcomes expected (Sidani & Braden 1998). The factors contributing to the successful implementation of the nursing intervention highlighted in this paper are indeed multiple.

How can we facilitate the use of an empirically supported protocol in the real world of practice? Nurses who recognize the research process and the need for continual improvement in patient care, which involves changing practice when indicated, are needed within every practice setting. Specific to this protocol, nurses who possess competency in physical assessment are critical. The regulatory atmosphere, workload structure and interdisciplinary collaboration are additional elements having an impact on the implementation of research-based nursing practice in this study. As Grol (1997) indicates, no single strategy predominates. This inquiry does not point to one strategy but does demonstrate the use of multiple strategies concurrently may be necessary to facilitate the use of the critical body of evidence-based nursing practice.


The authors wish to thank Patrician E. Noonan MSN, RN, CS, APNP for her contributions to this study.


Study design: MS; data analysis: MS and manuscript preparation: MS.


Appendix A: interview guide

How do you recognize behaviours in people with dementia?
How do you determine what behaviours need your attention?
Are there some behaviours that you think you miss?
Can you give me an example of when you identified a behaviour and did not intervene?
Tell me a time when you used the STI protocol and you were pleased with how it worked?
Tell me a time when you used the STI protocol and you were not pleased with how it worked?
How is the STI useful in your practice?
What processes or steps in the protocol are the most critical?
Do you think any one step of the protocol has changed how you deliver care more than another?
What motivates you to go through the protocol?
What makes you move through the steps of the protocol easier?
How persistent do you feel you have to be in using the protocol on your residents?