A Systematic Review of the Effectiveness of Communication Interventions for Health Care Providers Caring for Patients in Residential Care Settings Katherine S. McGilton, RN, PhD, Veronique Boscart, RN, MScN, Mary Fox, RN, PhD, Souraya Sidani, RN, PhD, Elizabeth Rochon, PhD, Riva Sorin-Peters, PhD, Reg. CASLPO, CCC(Sp)
ABSTRACT Objectives: This systematic review will describe the theoretical grounding, components, duration, mode of delivery, and outcomes of communication interventions for health care providers delivering care in residential care settings and will evaluate the effectiveness of these interventions. Methods: We conducted a comprehensive literature search of multiple databases published from January 1985 to the first week of December 2007, supplemented by a hand search of the references in all relevant articles, to find studies that met the inclusion criteria. Intervention details were extracted, and the studies’ validity was evaluated independently by two researchers using a standardized data collection form based on Cooper and Hedges’ (1994) approach to quality assessment. Results: Of the six studies that met the inclusion criteria (three randomized controlled trials, three quasi-experimental designs), three used a theoretical framework to guide intervention design. Across the six studies, the most commonly used components were (1) cognitive (to teach staff about communication), (2) behavioral (including practice at the bedside), and (3) psychological (involving individualized feedback). Despite the studies’ variability in methodological quality, their results indicated that communication interventions have a positive effect on staffs’ knowledge and communication skills and on residents’ agitation and challenging behaviors. However, none of the studies provided sufficient information on the duration of the intervention and on determining which interventions were most effective. This made it difficult to draw conclusions about the effectiveness of the interventions’ different components. Conclusion: Although communication training has been shown to have positive effects on staffs’ communication knowledge and skills as well as on resident outcomes, future controlled intervention research is needed to assess the effectiveness of individual intervention components. KEYWORDS communication, intervention studies, systematic review, nurse–patient interactions, long-term care, nurse–patient communication Worldviews on Evidence-Based Nursing 2009; 6(3):149–159. Copyright ©2009 Sigma Theta Tau International
Katherine S. McGilton, Research Scientist and Career Scientist MOH & LTC, Toronto Rehabilitation Institute, Toronto, ON, Canada; Veronique Boscart, Research Associate, Toronto Rehabilitation Institute, Toronto, ON, Canada; Mary Fox, Assistant Professor, Career Scientist MOH & LTC, York University, Toronto, ON, Canada; Souraya Sidani, Professor and Canada Research Chair Health Interventions, Daphne Cockwell School of Nursing, Ryerson University, Toronto, ON, Canada; Elizabeth Rochon, Associate Professor, Graduate Department of Speech-Language Pathology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; Riva Sorin-Peters, Speech-Language Pathologist, Regional Stroke Program—North & East GTA, Sunnybrook Centre for Independent Living, Sunnybrook Health Sciences Centre, Toronto, ON, Canada. Support for this research was provided by Toronto Rehabilitation Institute and the Ministry of Health and Long-Term Care, Canada. Address correspondence to Katherine S. McGilton, Toronto Rehabilitation Institute, 130 Dunn Avenue, Toronto, ON M6K 2R7, Canada; [email protected]
Accepted 26 January 2009 Copyright ©2009 Sigma Theta Tau International 1545-102X1/09
Worldviews on Evidence-Based Nursing r Third Quarter 2009 149
Communication Interventions Systematic Review
t is well known that interactions between health care providers (HCPs) and persons living in residential settings are crucial (Clark 1996; Nay 1998; Keller & Baker 2000; Burgio et al. 2002). Effective communication allows HCPs to assess the individual needs of residents (Farrell 1993; de Lucio et al. 2000) and provide nursing care that is tailored to the individual (Grypdonck 1993a, 1993b). Effective communication between HCPs and persons living in long-term care (LTC) has been linked to improved quality of life and well-being for residents in general (Brannon et al. 1992; Maas et al. 1994; Caris-Verhallen et al. 1999) and less agitation for residents with cognitive impairment (Burgio et al. 2002). Yet, although the importance of communication in providing care is widely recognized (Nay 1998), both residents and HCPs are often dissatisfied with this aspect of care. Residents perceive that staff is not readily available to respond to their questions or requests (Henderson et al. 2007) and often feel disempowered, dehumanized, and devalued (Coyle 1999). To communicate effectively, one should master a defined knowledge base and set of skills (Johnson 1994; Poskiparta et al. 1999). Therefore, providing HCPs with training in communication skills has been cited extensively in the literature as a means of remedying communication problems (Kruijver et al. 2000; Chant et al. 2002). Communication training has been shown to have positive effects on HCPs’ communication (Caris-Verhallen et al. 1999; Chant et al. 2002), yet the content of many communication skills training programs has been criticized for only focusing on outcomes (Fielding & Llewelyn 1987). Outcomes-focused evaluation provides a narrow and sometimes distorted understanding of the effects of an intervention (Chen & Rossi 1989). Not only have such studies failed to assess and analyze the mechanisms linking the intervention to its effects, but they also have not taken into consideration the potential influence of extraneous factors on the specified outcomes. Clinically, such evaluations fall short of providing specific information to guide practice (Sidani & Braden 1998). This view has been supported by others who state that current training in communication skills is inadequate to address the HCPs’ and residents’ needs, and that a greater range of communication skills should be provided within a coherent theoretical framework (Crute et al. 1989; Gijbels 1993; Heaven & Maguire 1996), where various components of the intervention are clearly articulated. Research examining the effectiveness of focused communication skills training is lacking in several areas (Chant et al. 2002). Many studies are limited in their description of the outcomes and have inadequate research designs (Hulsman et al. 1999). Although reviews and critical ap-
150 Third Quarter 2009 r Worldviews on Evidence-Based Nursing
praisals of the literature on this subject have been published (Chant et al. 2002), none of these reviews has systematically described the intervention components, their duration and mode of delivery, their theoretical grounding, and their outcomes. Nor have they evaluated their effectiveness while delivered to residents by HCP in residential care settings. Therefore, we conducted a systematic review to describe communication interventions targeting HCPs and to assess their effectiveness. Our aims were (1) to describe the sample and setting, theoretical grounding of the communication intervention, intervention components, duration and mode of delivery, and outcomes of communication interventions for HCPs working in residential care settings and (2) to evaluate the effectiveness of the interventions. In this article, we report the findings of our review.
METHODS The approach to the systematic review followed The Cochrane Handbook for Systematic Reviews of Interventions (version 5.0.1), the official document of the Cochrane Collaboration, which describes the process of preparing and maintaining systematic reviews on the effects of health care interventions. The specific study identification, inclusion criteria, screening process, quality assessment, and data synthesis will be described in detail. Study Identification The search strategy sought to compile the most extensive list possible of all published, relevant randomized and nonrandomized controlled clinical trials that may fit the inclusion criteria (Counsell 1997). The authors defined the term “communication intervention” to describe different aspects of interventions focused on enhancing HCPs’ approach to communication and/or interaction with residents. To guide this review, communication intervention was conceptualized as an informational educational, behavioral, or organizational intervention aimed at changing knowledge, beliefs, and/or behaviors. HCPs were defined as registered nurses, registered practical nurses, health care aides, personal support workers, occupational therapists, occupational assistants, physiotherapists, physiotherapist assistants, dieticians, social workers, speech language pathologists, recreational therapists, recreational therapist assistants, and/or physicians. Health care consumers were defined as adult patients, clients, or residents. Residential care settings were defined as settings where nursing care was delivered to a person residing there for more than 3 months. The keywords to search the literature included the following: communication, health personnel, residential facilities, randomized controlled trial (RCT), and
Communication Interventions Systematic Review TABLE 1 Detailed search strategy DATABASES
SEARCH KEYWORDS USED
EBM Reviews—Cochrane Central Register of Controlled Trials, 4th Quarter 2007 MEDLINE (1950 to November, Week 2, 2007) EMBASE (1980 to November, Week 2, 2007) PsycINFO (1985 to December, Week 1, 2007) CINAHL (1982 to December, Week 1, 2007)
Communication, health personnel, residential facilities, controlled clinical trial, and randomized controlled trial
quasi-experimental design. These terms were first searched independently and then combined. Relevant studies were identified in two steps. First, one of the authors, in collaboration with a professional librarian, conducted a computerized search of studies published from January 1985 to the first week of December 2007 in the following electronic databases: EBM Reviews— Cochrane Central Register of Controlled Trials, MEDLINE, R . The detailed search EMBASE, PsycINFO, and CINAHL strategy is shown in Table 1. The same keywords were used in the search strategy completed for all databases. In the second step, the reference list of each publication identified from the database search was examined for additional relevant studies. Inclusion criteria. Studies were included if they met the following criteria: (1) primary research; (2) evaluated a communication intervention aimed at improving HCPs’ communication and/or interaction with the resident, resident health and well-being and/or resident agitation, or HCPs’ communication skills and knowledge; (3) used a RCT or a quasi-experimental (i.e., nonrandomized cohort) design; (4) included HCPs and/or consumers as research participants; (5) took place in institutional care settings such as LTC, chronic care, or nursing homes; and (6) published in English. Studies were excluded if they implemented an intervention without a specific focus on communication outcomes, consisted of posttest outcome data collection, and/or aimed to enhance communication between HCPs/resident and family dyads. Screening process of study eligibility. The methods we adopted for eligibility screening in this review are those described by Cooper and Hedges (1994) and the Cochrane Collaboration Centre (1996). Two reviewers independently conducted the screening process. Retrieved abstracts from the two-step identification process were reviewed independently to assess for eligibility. Disagree-
ment on an abstract’s eligibility was resolved by inviting a third reviewer who assessed the abstract and discussed it with the two original reviewers. A meeting of all reviewers was called to reach a consensus on the final list of eligible abstracts. If any doubts or disagreements about particular abstracts were raised, a discussion took place. Decisions for including a study for full review were made based on a consensus among the entire review group. Quality Assessment Upon completion of data extraction, the first reviewer independently assessed each of the six selected studies that were considered eligible for methodological quality using the Amsterdam–Maastricht Consensus List for Quality Assessment’s (van Tulder et al. 2003). It has 11 criteria that assess the risk of bias (i.e., specification of eligibility criteria for participants, specification of treatment application, blinded care provider, etc.). The reviewer scored each criterion as a “yes” (criterion was met) or “no” (criterion was not met or not clearly stated) as per the guidelines (van Tulder et al. 2003). The second reviewer then independently examined the extracted data for accuracy and performed a second quality assessment. Based on the reviewer’s judgment, a decision was made on the risk of bias. Cochrane (2002) indicates that assuming that all risks are equal, and there are no serious flaws to the study’s design, a study is methodologically sound if both reviewers rated six out of 12 criteria as having been met. The definition of serious flaws included items such as concealment of allocation, more than 50% dropout, or no pretest measures. Any disagreements in study quality between the two reviewers were resolved through consensus. Table 2 presents the agreed upon rating of the six selected studies’ quality. Data Synthesis and Analysis Each reviewer extracted data from the full articles that were deemed eligible during the screening process. Data from the six selected studies were entered in tables and summarized using a qualitative narrative approach (Chalmers & Altman 1995). A second researcher independently examined the extracted data for accuracy. No quantitative analysis was carried out for this review. Rather, a detailed account of the characteristics and effectiveness of the intervention studies with comments on study quality was presented. The characteristics of the interventions that were described included its theoretical grounding, duration and mode of delivery, outcomes, and three intervention components: cognitive, behavioral, and psychological (Sidani & Braden 1998). Cognitive components are designed to support and contribute to an existing cognitive structure that facilitates understanding communication. Behavioral components are aimed at promoting Worldviews on Evidence-Based Nursing r Third Quarter 2009 151
Communication Interventions Systematic Review TABLE 2 Quality assessment of internal validity of the selected clinical and RCTs with the Amsterdam–Maastricht consensus list for quality assessment BURGIO CARIS-VERHALLEN DIJKSTRA GOLDEN ET AL. 2002 ET AL. 2000 ET AL. 2002 & REESE 1996
ITEMS 1a. Randomization 1b. Concealment of allocation 2. Comparable subgroups at baseline 3. Blinded care provider 4. Correction for attention; same treatment (dose), co-intervention 5. Acceptable compliance 6. Blinded patient 7. Acceptable withdrawals during intervention period 8. Blinded outcome assessor 9. Relevance measures 10. Timing assessment 11. Intention to treat analysis Total
+ + +
TAPPEN ET AL. 2001
VAN WEERT ET AL. 2005
− + + − +
+ + + − +
+ + + − +
+ + + (except age) − +
+ + + + +
+ − ?
+ − +
+ − +
+ − +
+ − +
+ + +
− + + ?
− + + ?
− + + ?
− + + ?
− + + +
+ + + –
Note. Range = 0–11 points; + = yes, criterion was met; − = no, criterion was not met or not clearly stated.
communication skills and managing the continuing challenges that HCPs encounter when communicating with residents. HCPs are taught new skills, including how to apply them in practice. Psychological components provide HCPs with individual mentorship and feedback, which encourage them to reflect on their practice and express their feelings, with the goal of regulating their emotional response. The intervention’s effectiveness was examined in relation to the following outcomes: frequency and duration of interaction between HCPs and residents, new skills being applied, behavior of HCPs, and behavior of residents.
RESULTS Selected Studies The combination of the search terms, that is, communication, health personnel, health care consumers, and residential facilities, yielded 598 manuscripts. When the search terms intervention study, RCT, and quasi-experimental design were added, only 43 studies were identified. After identical duplicate titles selected through different databases were excluded, 33 studies were left. No additional relevant studies were found in the reference list of the publications identified from the database search. All abstracts were reviewed for eligibility. If the information in the abstract did not provide sufficient data to determine the study’s eligibility, the full manuscript was reviewed. Twenty-seven studies did not meet the eligibility criteria. Fourteen of them did not have a defined intervention. In another eight, communication was not the main focus of 152 Third Quarter 2009 r Worldviews on Evidence-Based Nursing
the intervention and outcomes. Three studies focused on improving HCPs’ communication with residents’ families. Two articles reported on the same study (Caris-Verhallen et al. 1997; Roth et al. 2002). In total, six studies (Golden & Reese 1996; Caris-Verhallen et al. 2000; Tappen et al. 2001; Burgio et al. 2002; Dijkstra et al. 2002; Van Weert et al. 2005) met the inclusion criteria for the systematic review (see Figure 1). A summary of the included studies describing study design, sample size, interventions, and outcome measures is presented in Table 3. Of the included studies, three were RCTs and three were quasi-experimental with a control group. Sample and Settings All of the studies took place in LTC settings, primarily nursing homes; the number of sites included in the studies ranged from one to six nursing homes. All studies collected data from residents. Additionally, Burgio et al. (2002), Caris-Verhallen et al. (2000), Dijkstra et al. (2002), and Golden & Reese (1996) collected data from certified nursing assistants (CNAs) and/or nurses. The other studies collected data from a variety of HCPs. The residents’ average age was 82.5 (±2.8) years. The sample sizes ranged from 16 to 120 resident participants and from 16 to 106 HCP participants. Four studies involved resident participants with dementia (Tappen et al. 2001; Burgio et al. 2002; Dijkstra et al. 2002; Van Weert et al. 2005). Three of these studies showed that the residents had behavioral disturbances in addition to dementia (Burgio et al. 2002; Dijkstra et al. 2002) and communicative disabilities
Communication Interventions Systematic Review
Manuscripts excluded Potentially relevant search terms identified and screened on title and abstract for retrieval (N = 598)
Main reasons: -Type of study did not meet criteria -Type of participants did not meet criteria
Manuscripts retrieved for further screening (n = 43)
-Type of outcome measures did not meet criteria (n = 555) Identical manuscripts deleted (n = 10)
Additional reviews retrieved for further screening (n = 33)
Manuscripts excluded (n = 27) -Type of study or intervention did not meet criteria (n = 14)
Included in best evidence synthesis (n = 6)
-Communication was not focus (n = 8) -Outcomes focused on family (n = 3) -Two papers reported on the same study
Figure 1. Flowchart of selection of studies. (Tappen et al. 2001). For the HCP sample, only information about job category was provided, that is, CNAs (Burgio et al. 2002; Dijkstra et al. 2002) or regulated nurses (Caris-Verhallen et al. 2000). Theoretical Grounding Of the six studies that met the selection criteria, three were based on one of the following theoretical frameworks: (1) behavior management in which motivation was adapted from the Behavioral Supervisory Model (Burgio et al. 2002); (2) the Nursing Child Assessment Satellite Training Model (NCAST) developed by Golden and Reese (1996); and (3) Snoezelen Philosophy (i.e., principles similar to patient-centered care, Van Weert et al. 2005). Only one study (Burgio et al. 2002) was judged to be sufficiently theoretically developed, based on criteria to assess theoretical quality that has been used in other systematic reviews (McKevitt et al. 2004). Intervention Components All interventions were reviewed based on the previously discussed cognitive, behavioral, and psychological components (Sidani & Braden 1998). The six studies used different combinations of intervention components, yet all interventions started with a cognitive (i.e., educational) component aimed at increasing staff general knowledge
about communication. Three studies added education to deal with challenging behaviors from residents with dementia (Golden & Reese 1996; Tappen et al. 2001; Burgio et al. 2002). In two of the three RCTs (Golden & Reese 1996; Burgio et al. 2002) and two of the quasiexperimental studies (Caris-Verhallen et al. 2000; Dijkstra et al. 2002), the educational component was followed by a behavioral component to enhance the HCPs’ ability to manage residents’ behavior and implement effective communication strategies. These researchers taught the new skills to HCPs, including how to apply them in practice. In two studies (Golden & Reese 1996; Burgio et al. 2002), the intervention was supplemented with a psychological component to help HCPs regulate their emotional responses. In these two studies, HCPs were provided individual mentorship and feedback from the supervisors on their units, who encouraged the HCPs to reflect on their new practice. Intervention Delivery and Duration All cognitive intervention components were didactic, that is, the intervener instructed the HCP participants in a lecture format. Some interventions involved a workbook or other didactic materials, such as presentations or additional educational reading materials. For the behavioral training, case studies and demonstration, some form of mentoring (direct supervision), and bedside training were employed. Psychological support consisted of the intervener teaching the supervisors or managers to use the feedback observation tool and to deliver support to participating HCPs. This component also included follow-up support and encouragement for the participating HCPs. No information was available on the size of the groups. Only two studies reported information about the intervener’s education and experience. The training was provided by the first author who was skilled in the use of a specific scale (Golden & Reese 1996). The intervener was a qualified and experienced professional trainer in the intervention techniques (Van Weert et al. 2005). The duration of the cognitive component for most of the interventions was limited to one or two group sessions, varying from 30 minutes to 4 hours. Exceptions were two studies, Caris-Verhallen et al. (2000) and Golden & Reese (1996), in which cognitive educational sessions were delivered to HCPs for 2 or 3 days. The behavioral component took place over 3–4 weeks in the reviewed studies. Two researchers provided a psychological component lasting from 3 (Golden & Reese 1996) to 6 months (Burgio et al. 2002) post workshop. Outcome Assessment The outcomes measured in these six studies included knowledge and behaviors of HCPs and behaviors of Worldviews on Evidence-Based Nursing r Third Quarter 2009 153
47 nurses; 109 patients
33 NAs–resident pairs (treatment group); 33 NAs– resident pairs (control group)
Experimental-control, pre- and posttest design
CarisVerhallen et al. 2000
Dijkstra et al. 2002
88 residents with behavior disturbances; 106 CNAs
Randomized clinical trial
Burgio et al. 2002
TABLE 3 Characteristics of included studies
154 Third Quarter 2009 r Worldviews on Evidence-Based Nursing Nontrained group
Nontraining in communication techniques: 24 nurses in the community 81 patients in the community
Use of effective communication techniques; personalized memory books; 1-hour didactic in-service & a 2- to 4-week criterionbased, hands-on training during care routines
Training in communications: 23 nurses in the home for the elderly 28 patients in the home for the elderly
One group of supervisory nurses on the Formal Staff Management (FSM) units were instructed to implement the FSM system following the CNAs’ 4 weeks of training on behavior management with knowledge- and performance-based assessments of skill acquisition
One group of supervisory nurses on the Conventional Staff Management (CSM) units were instructed to continue with their normal supervisory routine following the CNAs’ 4 weeks of training on behavior management with knowledge- and performance-based assessments of skill acquisition
BMSC; self report test; CABOS; Discourse categories (utterances, initiating [questions, prompts], responsive [facilitators–repetitions, encouragements, cues])
Nurses: Total amount of utterances during: Home care—hygienic care, technical nursing care, psychosocial care; Institutional care—hygienic care, technical nursing care, psychosocial care Patients: Verbal behaviors, e.g., social behavior and jokes, affective behavior, behavior that structures communication, communication of nursing and medical topics, communication of lifestyle and emotional topics
Cohen–Mansfield Agitation Inventory (CMAI) scale to assess residents’ aggressive behaviors, physically nonaggressive behaviors, and verbally aggressive behaviors; Behavior Management Skills Checklist (BMSC) Computer-Assisted Behavioral Observation Systems (CABOS)
Communication Interventions Systematic Review
Multiple-probeacross-group design with random selection of staff
Repeated measures three-group design with random assignment to treatment; raters were blinded to treatment group assignment
Quasi-experimental pre- and posttest design
Tappen et al. 2001
Van Weert et al. 2005
Golden & Reese 1996
TABLE 3 (Continued)
120 residents (60 treatments and 60 controls); 120 CNAs (60 treatments and 60 controls)
16 patients with severe and profound mental retardation; 16 direct-care staff
No application of snoezelen
23 patients engaged in assisted walking (6-minute walk)
Nontrained group: 12 staff–resident pairs have no training
Application of snoezelen, integrated in 24-hour dementia care
Trained Group 1: two resident–staff pairs were trained in the adapted MR/DD NCAFS; following the 1st treatment, staff were probed up to four times during mealtimes. Trained Group 2: six more resident–staff pairs underwent 2nd treatment phase and were probed up to four times during mealtimes. 22 patients engaged in conversation intervention; 20 patients engaged in combined walking (modified 6-minute walk) and conversation intervention
CNAs’ positive affective or socioemotional verbal communication (showing empathy, social talk and validation); CNAs’ instrumental communication (questions about facts and cognitive knowledge); CNAs’ nonverbal behaviors (affective touch); Residents’ nonverbal communication (CNA-directed gazing and smiling)
Treatment fidelity; functional mobility
Staff’s verbal and nonverbal behaviors during mealtime (training task) and during teaching (generalization task); Resident behaviors, e.g., engaging people, engaging objects, noncompliance, and self-stimulation
Communication Interventions Systematic Review
Worldviews on Evidence-Based Nursing r Third Quarter 2009 155
Communication Interventions Systematic Review
residents. All reviewed studies reported that enhanced knowledge and behaviors for the staff were the primary outcome variables. No self-report data were collected from the resident participants in these studies because most had cognitive impairment or were not able to reliably selfreport. Outcomes assessed for the HCPs included selfreport knowledge gained about communication strategies (Golden & Reese 1996; Caris-Verhallen et al. 2000; Van Weert et al. 2005) and self-report knowledge of behavioral management (Burgio et al. 2002; Dijkstra et al. 2002; Van Weert et al. 2005). No psychometrics were reported for the self-report scales. All studies included outcomes for staff derived from observational behavioral measures. The studies used different measures to evaluate the impact of the intervention on HCPs’ behaviors and focused on staffs’ use of behavioral management strategies and/or communication strategies with their assigned residents. Burgio et al. (2002) used the Behavior Management Skills Checklist to observe the number of effective and noneffective communication and behavioral strategies used (alpha coefficient = 74%). Caris-Verhallen et al. (2000) used a video interaction analysis checklist, which has shown good interrater reliability (0.82) in previous studies. Tappen et al. (2001) collected data on the HCPs’ communication skills by using a Norman Rockwell picture without religious significance and a picture description test. Golden & Reese (1996) employed the adapted Nursing Child Assessment Feeding Scale (NCAFS) and the Nursing Child Assessment Teaching Scale (NCATS) to evaluate staff skills in communicating but did not provide any reliability or validity information of these two scales. Outcomes assessed for the residents included their observed behaviors, such as agitation (Burgio et al. 2002), social engagement (Golden & Reese 1996), conversation (Tappen et al. 2001), nonverbal communication (Van Weert et al. 2005), and change in mobility status (Tappen et al. 2001). None of the reviewed studies used similar outcomes to measure the success of the intervention on HCP or resident outcomes. Effectiveness of the Intervention All reviewed studies indicated a positive change in HCPs’ communicative behavior, skills, and knowledge, upon completion of the intervention. Findings revealed significant changes in the HCPs’ communication skills, that is, they used more positive statements, gave more information to residents, used more open-ended questions, and were rated as more involved, warmer, and less patronizing. HCPs’ negative communication/interaction strategies, such as showing disapproval, decreased. Resident outcomes included an increase in responsiveness and eye 156 Third Quarter 2009 r Worldviews on Evidence-Based Nursing
contact with the HCPs, as well as a decrease in residents’ verbal disapproval, anger, and agitation. All studies had a cognitive and/or educational component in their intervention. Therefore, it was difficult to clearly identify the contributions of the specific components to HCP and/or resident outcomes. Yet, the four studies (Golden & Reese 1996; Caris-Verhallen et al. 2000; Burgio et al. 2002; Dijkstra et al. 2002) that incorporated a cognitive and a behavioral component to the intervention, demonstrated clear, yet nonsignificant, changes in HCP skills and some improvement in resident outcomes. Two teams of researchers also included a psychological component (positive feedback to staff by the supervisors on the units and encouraged reflection on their new practice), which they claimed helped to sustain the change in HCPs’ communication behaviors (Golden & Reese 1996; Burgio et al. 2002). Both studies had positive outcomes for HCPs and residents. Duration of the intervention also may have influenced the positive effects of the intervention in Burgio et al.’s (2002) study. Its 6-month duration allowed time for staff for trial and to become familiar with the new communication and behavioral strategies (Burgio et al. 2002).
LIMITATIONS OF THE REVIEW One limitation of this review was that only English language studies were included in the search strategies, which may have led to the omission of some relevant published articles. Another was that merging results was not possible due to the studies’ diverse intervention and concepts measured in the study. Therefore, the authors could provide only a narrative account of the six studies. In addition, while we acknowledge that nurse–patient communication involves a complex interplay of psychosocial, cultural, and person specific dynamics, none of these factors were taken into account for this review, which may be viewed as a limitation. However, the purpose of this systematic review was to focus on specific protocols that have been developed to enhance staffs’ communication efforts, and the review demonstrated these practical approaches were efficacious. Future communication interventions should include protocols that are individualized to the patients’ particularities. In addition, factors influencing the complex interplay between patient and nurse at the system level (i.e., workload, cultural norms on the unit) should be investigated to determine the effect on the specific protocols that were found to enhance successful nurse– patient interactions.
Communication Interventions Systematic Review
DISCUSSION Overall, this review supports findings of individual studies examining a communication intervention for staff employed in LTC settings. Although a relatively large amount of research has been conducted in the field of communication, very few studies have used controlled trials, as is expected when using complex interventions. Most research, which has relied on observational or descriptive designs, has indicated that effective staff–patient communication is necessary to provide quality care, and that training staff for these communication skills and strategies can have an impact on several patient and staff outcomes. Yet, for this current review, a considerable variation found in the reviewed studies between their intervention content, the components, the duration and the mode of delivery makes it difficult to pinpoint which interventions are more effective compared with others. The results of this review highlighted several gaps in identifying an effective intervention. There were also weaknesses in the reviewed studies. Only half of them used a theoretical framework to develop their intervention; moreover, it is not clear how the frameworks guided those studies. Developing a study based on a sound theoretical framework would provide details about additional factors that could influence the intervention’s effectiveness (Chen & Rossi 1989). These factors are essential to building new knowledge for clinical practice and enhancing quality of care (Sidani & Braden 1998). The review also revealed that the communication interventions themselves were not always well described. This lack of information makes it impossible to replicate a given study. Several flaws in methodological quality were noted in regard to sample and setting. For example, some studies reported a very small sample; that, combined with inadequate reporting of sampling strategies, response rate or attrition rate, made it difficult to determine the representativeness of the samples. Convenience sampling and purposeful sampling caused the risk of respondent bias. Most outcome measures were not described in detail, lacked information on psychometric properties, and were developed and used to evaluate an intervention without pretesting the instrument before the intervention. Psychometrically sound measures and explicitly defining what constitutes changes in communication skills and knowledge were not used, making it difficult to understand the precise communication skills that changed over the course of the intervention. Only two studies (Golden & Reese 1996; Burgio et al. 2002) evaluated long-term outcomes (6month postintervention), which limits the ability to draw conclusions about the sustainability of the intervention’s effectiveness.
Despite these methodological limitations, some key findings of the reviewed studies are worth mentioning. All intervention studies did find a positive change in staff’s communicative behavior and skills and a decreased level of agitation and anger for the resident. Participating staff indicated that the training improved their level of knowledge related to communication strategies. However, it is important to note that cognitive components of an intervention are designed to support and contribute to an existing cognitive structure, yet this cognitive content is not retained when it is not supplemented by a behavioral component aimed at applying and reinforcing the new skills in the clinical setting (Sidani & Braden 1998; Grimshaw et al. 2003). Furthermore, this review presented some preliminary evidence that cognitive, behavioral, and psychological components seem to be necessary to influence HCPs’ behaviors. Implications for Practice and Future Research HCP–resident communication is of utmost importance to provide high-quality care, especially in residential settings, where care providers are often the only source of social interaction with residents. Therefore, strong interpersonal communication skills are a prerequisite for HCPs working in these settings (Bowling et al. 1993). Based on the review, we recommend that to enhance communication skills of HCPs, the intervention should be multilevel and comprise of three components: educational training, practice, and support. Educational training is directed at providing staff with increased knowledge about effective communication techniques that are specific to their residents’ communication disabilities. The training could encompass didactic, learner focused, and interactive strategies. While focusing on communication strategies, behavioral management strategies also need to be addressed. How nursing staff communicate with patients when a responsive behavior occurs can influence patients’ behavioral disturbances. Incorporating practice sessions, using techniques focused on changing staffs’ communication behaviors, will assist HCPs to have more successful interactions. The behavioral training could include a demonstration of effective communication and interaction strategies, staff practicing their skills with each other, leading to bedside mentorship with resident–nurse dyads. Finally, psychological support for HCPs to use the new skills with their residents requires opportunities for reflection on their new practice, encouragement, and feedback from managers and clinical support staff. Future research includes the need to focus on rigorous and controlled intervention studies based on a theoretical framework, where the specific components, duration and content are made explicit. Thus, replication of the study Worldviews on Evidence-Based Nursing r Third Quarter 2009 157
Communication Interventions Systematic Review
will be possible. Experimental methods could be used to test the efficacy of the communication intervention and use similar reliable and valid measures of HCP and resident outcomes across the studies to advance the science in this area of inquiry. Qualitative methods could be employed to better understand processes that facilitate implementing new evidence into practice to determine contextual factors necessary to address prior to implementing the new practice change. Controlled intervention studies are needed to examine the tendency observed in individual studies that interventions with an educational, behavioral, and psychological component have a positive effect on staff and patient outcomes.
CONCLUSIONS Although researchers and health care organizations have expressed substantial interest in improving patient care by enhancing HCP communication, this review highlighted an important gap in knowledge, showing that there is no clear evidence showing which are the most effective interventions to adopt. Despite findings that communication training has positive effects on staffs’ communication knowledge and skills and on resident outcomes, limited information about the content, duration and mode of delivering intervention components makes it unclear which components are associated with beneficial outcomes. The research conducted to date provides a foundation for further research. In the future, communication interventions will need to be designed that account for the complex interplay between the resident, the nurse, and their environment. References Bowling A., Ebrahim S., Brannon D., Smyer M. & Cohn M. (1993). Strategies for improving nursing home care: A research roundtable. Journal of Long-Term Care Administration, 20(4), 18–20. Brannon D., Smyer M. & Cohn M. (1992). Organizational strategies for improving nursing home care. Journal of Long-Term Care Administration, 20(4), 118–129. ∗Burgio L.D., Stevens A., Burgio K.L., Roth D.L., Paul P. & Gerstle J. (2002). Teaching and maintaining behavior management skills in the nursing home. Gerontology, 42, 487–496. Caris-Verhallen W.M., de Gruijter I.M., Kerkstra A. & Bensing J.M. (1999). Factors related to nurse communication with elderly people. Journal of Advanced Nursing, 30, 1106–1117. Caris-Verhallen W.M.C.M, Kerkstra A. & Bensing J.M. (1997). The role of communication in nursing care for 158 Third Quarter 2009 r Worldviews on Evidence-Based Nursing
elderly people: A review of the literature. Journal of Advanced Nursing, 25, 915–933. ∗ Caris-Verhallen W.M., Kerkstra A., Bensing J.M. & Grypdonck M.H. (2000). Effects of video interaction analysis training on nurse-patient communication in the care of the elderly. Patient Education and Counseling, 39(1), 91–103. Chalmers I. & Altman D.G. (Eds.). (1995). Systematic reviews. London: BMJ Publishing Group. Chant S., Jenkinson T., Randle J., Russell G. & Webb C. (2002). Communication skills training in healthcare: A review of the literature. Nurse Education Today, 22, 189–202. Chen H.T. & Rossi P.H. (1989). Issues in the theory-driven perspective. Evaluation and Program Planning, 12, 299– 306. Clark P.G. (1996). Communication between provider and patient: Values, biography, and empowerment in clinical practice. Ageing and Society, 16, 747–774. Cochrane Collaboration Centre. (1996). Centre for reviews and dissemination. Retrieved September 17, 2007, from www.cochrane.org. Cooper H. & Hedges L.V. (1994). The handbook of research synthesis. New York, N.Y.: Russell Sage Foundation. Counsell C. (1997). Formulating questions and locating primary studies for inclusion in systematic reviews. Annals of Internal Medicine, 127, 380–387. Coyle J. (1999). Exploring the meaning of dissatisfaction with health care: The importance of “personal identity threat.” Sociology of Health and Illness, 21, 95– 124. Crute V.C., Hargie O.D.W. & Ellis R.A.F. (1989). An evaluation of a communication skills course for health visitor students. Journal of Advanced Nursing, 14, 546– 552. de Lucio L.G., Lopez F.J.G., Lopez M.T.M., Hesse B.M. & Caamano M.D.C. (2000). Training programme in techniques of self-control and communication skills to improve nurses’ relationships with relatives of seriously ill patients: A randomized controlled study. Journal of Advanced Nursing, 32, 425–431. ∗ Dijkstra K., Bourgeois M., Burgio L. & Allen R. (2002). Effects of a communication intervention on the discourse of nursing home residents with dementia and their nursing assistants. Journal of Medical Speech Language Pathology, 10(2), 143–157. Farrell E. (1993). Communication: A critical part of nursing. American Nurse, 25(7), 40–41. Fielding R.G. & Llewelyn S.P. (1987). Communication training in nursing may damage your health and enthusiasm: Some warnings. Journal of Advanced Nursing, 12, 281–290.
Communication Interventions Systematic Review
Gijbels H. (1993). Interpersonal skills training in nursing education: Some theoretical and curricular considerations. Nurse Education Today, 13, 458–465. ∗ Golden J. & Reese M. (1996). Focus on communication: Improving interaction between staff and residents who have severe or profound mental retardation. Research in Developmental Disabilities, 17, 363–382. Grimshaw J.M., McAuley L., Bero L.A., Grilli R., Oxman A.D., Ramsay C., Vale L. & Zwarenstein M. (2003). Systematic reviews of the effectiveness of quality improvement strategies and programmes. Quality Safety in Health Care, 12, 298–303. Grypdonck M. (1993a). Informeren en communiceren, Part 1 [Giving information and communicating, Part 1]. Tijdschrift voor Ziekenverpleging, 19, 639–643. Grypdonck M. (1993b). Informeren en communiceren, Part 2 [Giving information and communicating, Part 2]. Tijdschrift voor Ziekenverpleging, 20, 683–688. Heaven C.M. & Maguire P. (1996). Training hospice nurses to elicit patient concerns. Journal of Advanced Nursing, 23, 280–286. Henderson A., Van Eps M.A., Pearson K., James C. & Henderson P. (2007). “Caring for” behaviors that indicate to patients that nurses “care about” them. Journal of Advanced Nursing, 60, 146–153. Hulsman R.L., Ros W.J., Winnubst J.A. & Bensing J.M. (1999). Teaching clinically experienced physicians communication skills: A review of evaluation studies. Medical Education, 33, 655–668. Johnson J.R. (1994). The communication training needs of registered nurses. Journal of Continuing Education in Nursing, 25, 213–218. Keller V. & Baker L. (2000). Communicate with care. RN, 63(1), 32–34. Kruijver I.P., Kerkstra A., Francke A.L., Bensing J.M. & van de Wiel H.B. (2000). Evaluation of communication training programs in nursing care: A review of the literature. Patient Education and Counseling, 39(1), 129–145.
Maas M., Buckwalter K., Swanson E. & Mobily P. (1994). Training key to job satisfaction. The Journal of LongTerm Care Administration, 22(1), 23–26. McKevitt C., Redfern J., Mold F. & Wolfe C. (2004). Qualitative studies of stroke: A systematic review. Stroke, 35, 1499. Nay R. (1998). Contradictions between perceptions and practices of caring in long-term care of elderly people. Journal of Clinical Nursing, 7, 401–408. Poskiparta M., Liimatainen L. & Kettunen T. (1999). Nurses’ self-reflection via videotaping to improve communication skills in health counseling. Patient Education and Counseling, 36, 3–11. Roth D.L., Stevens A.B., Burgio L.D. & Burgio K.L. (2002). Timed-event sequential analysis of agitation in nursing home residents during personal care interactions with nursing assistants. Journals of Gerontology: Series B. Psychological Sciences and Social Sciences, 5, 461– 468. Sidani S. & Braden C.J. (1998). Evaluating nursing interventions: A theory-driven approach. Thousand Oaks, CA: Sage. ∗ Tappen R.M., Williams C.L., Berry C. & DiSesa D. (2001). Conversation intervention with Alzheimer’s patients: Increasing the relevance of communication. Clinical Gerontologist, 24(3/4), 63–75. van Tulder M., Furlan A., Bombardier C. & Bouter L. (2003). Updated method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group. Spine, 28, 1290–1299. ∗ Van Weert J.C.M., van Dulmen A.M., Spreeuwenberg P.M.M., Ribbe M.W. & Bensing J.M. (2005). Effects of “snoezelen,” integrated in 24-hour dementia care, on nurse-patient communication during morning care. Patient Education and Counseling, 58(1), 312– 326. ∗
Manuscripts used for the systematic review.
Worldviews on Evidence-Based Nursing r Third Quarter 2009 159