The construction of a conceptual framework explaining the relation between barriers to change of management of neuropsychiatric symptoms in nursing homes: a qualitative study using focus groups

In the Netherlands 80.000 people with dementia reside in care facilities such as nursing homes [1]. Of these residents, approximately 27% use antipsychotic drugs as a treatment for neuropsychiatric symptoms (NPS) and 40% use antidepressants [2]. Guidelines advise a restricted use of psychotropic drugs in the treatment of NPS and advocate the use of psychosocial interventions [3]. The analysis and treatment of NPS is a multidisciplinary process, wherein, among others, the physician, psychologist and nursing staff play an important role [4]. Proper treatment of NPS is important, due to the negative influence of improper treatment of NPS on the quality of life of residents and on nursing staff. For example, nursing staff might experience anxiety and burnout as a result of NPS in residents [5, 6]. Therefore, various multidisciplinary interventions have been developed to reduce the frequency of psychotropic drug use and/or NPS, or to improve residents’ quality of life [4, 7‐12]. Unfortunately, the (long-term) effectiveness of many of these interventions in terms of reduction of psychotropic drug use was shown to be limited [7, 8, 12, 13].

The small effects of interventions for psychotropic drug use may be a result of difficulties to implement interventions and induce change in nursing homes [14]. In that respect, a number of studies has been conducted to identify specific barriers towards implementation of (complex) interventions in nursing homes, often by means of a process evaluation. A major barrier – which has been reported on multiple occasions – is the complexity of the guideline or intervention to be implemented [15, 16]. These interventions are frequently complex due to a multidisciplinary approach, in which each discipline (i.e. nurses, physicians, and psychologists) applies different types of interventions [4]. In addition, a major barrier reported was the high turnover of the nursing home workforce [14‐19]. Moreover, reorganizations, other innovations running at the time of the intervention, no relevance felt by the staff [8, 17, 18, 20], and the culture of the care unit - including attitude towards change -, are barriers towards changing current practice [14].

In the past, research has been conducted to identify barriers and to classify these into categories i.e., themes. For example, Mentes & Tripp-Reimer [19] provide an overview of barrier themes encountered in nursing home research: residents, staff, administrative and organizational issues, attitudes, research protocols and research assistants. Furthermore, Corazzini et al. [21] studied challenges (barriers) encountered while implementing ‘culture change in nursing home staff’ and ‘leadership behaviors’ that facilitated this change. They found six key themes, which described these challenges and leadership behaviors: ‘relationships’, ‘standards and expectations’, ‘motivation and vision’, ‘workload’, ‘respect of personhood’ and ‘physical environment’ [21]. Finally, the English National Institute for Health and Clinical Excellence (NHS) carried out a systematic review, which offered five types of barriers to change in healthcare and ideas to overcome these barriers. The five types of barriers identified were: ‘awareness and knowledge’, ‘motivation’, ‘practicalities’, ‘acceptance and beliefs’ and ‘skills’ [22]. Identifying themes of barriers is important as these can assist in understanding the causes of barriers and how to address these. Yet, insight into the relationships between themes of barriers is even more helpful in effectively addressing barriers, as it allows for a better determination of the magnitude of the barrier and of strategies to resolve it [23]. There is evidence indicating that assessment of barriers – before attempting implementation of an intervention or attempting to change current practice – will increase the chance of success [24].

In relation, Van Bokhoven et al. [23] mention a modified ‘model of barriers and facilitators’ based on the PRECEDE-PROCEED concept and theory of planned behavior, to provide a foundation for a structured quality improvement intervention. This model focuses especially on improvement of quality of life of residents and pertains to health care practice in general. However, the model does not include specific barriers encountered in nursing home practice, nor does it address the barriers encountered in improvement of the quality of care. Although many barriers to change and overarching themes have been identified in previous research, there is no framework available that explains the relationship between these perceived barriers in nursing homes. In a nursing home, the residents, professionals and provided care have specific characteristics that may result in encountering other barriers to change than in other settings, warranting a study that is specific for nursing homes [25]. Therefore, the aim of this study is to explore the perceived barriers to change regarding management of NPS in nursing homes and to construct a conceptual framework providing insight into the importance of and relationships between these barriers.

In this study we focused on the identification of perceived barriers to change in nursing homes and we aimed to construct a conceptual framework explaining the relation between these different barriers. We extracted eight themes of barriers that impede the extent to which change is likely. Some are direct barriers and some are indirect barriers. For example, ‘suboptimal communication’, ‘inadequate (multidisciplinary) collaboration’ and ‘reactive coping & resilience of organization’ are indirect barriers. These themes do not necessarily influence the extent of change directly, but do so via another theme or route. All identified themes are hierarchically related, wherein ‘organizational barriers and ‘deficiency of staff knowledge’ were the foundation of all other themes. Hereafter, ‘suboptimal communication’ and ‘inadequate (multidisciplinary) collaboration’ may cause ‘differences in perception’, which in turn can lead to disorganization. In addition, ‘personal barriers’ may influence ‘reactive coping & resilience of organization’ and via that route influence ‘disorganization of processes’. Moreover, ‘personal barriers’ influence these interacting layers. Especially ‘reduced staff motivation and effort’ and ‘negative staff emotions’ play an important role herein. The extent to which possible change is impaired can be determined by identifying existing barriers and categorizing them according to the conceptual framework. Subsequently addressing these barriers could enhance the possibility to change.

Various barriers, found in this research, are known from previous research. For example, ‘organizational barriers’, ‘personal barriers’ (such as a reduced motivation and effort) and ‘deficiency of staff knowledge’ are well-known categories of barriers to change [8, 14‐22]. Our study adds that these categories may be the fundament to achieving change; without proper knowledge, organizational support and personal factors there will only be a small extent of change possible. Similarly, Zwijsen et al. [14] and the National Institute for Clinical Studies [37], among others, have identified issues in ‘inadequate (multidisciplinary) collaboration’ and ‘suboptimal communication’ before. We found two additional interrelated themes that influenced the possibility of impaired change, which were not identified before; ‘Differences in perception’ and ‘Reactive coping & resilience of organization’.

Although many studies have identified themes of barriers, only one elaborated on the relations between the different themes [23]. Whereas van Bokhoven et al. [23] constructed a framework wherein the barriers are split into external factors and professional factors influencing and explaining professional behavior, our framework focuses on the explanation of the extent to which change can be hindered. Due to the similar organizational nature of nursing homes and the fact that we recruited participants from both a small-scale living facility and a large scale care unit, this framework might be transferable to other nursing homes. However, first our framework should be confirmed and refined in additional research. Thereafter, our framework could be a tool for classifying barriers and identifying which problems might arise in the process of change. Some researchers have already tried to take ‘known barriers to implementation’ into account when implementing an intervention [38, 39]. Others actually identified the local barriers towards implementation before starting the implementation, to allow for optimal implementation of interventions [40]. Furthermore, approaches are available to assess the readiness to change in an organization, among others ‘the nursing home working conditions survey’ [41, 42]. Future implementation research could focus on identifying local barriers and classifying them with our framework to allow assessing the impact on the extent of change. After classification, a specific strategy for implementation could be chosen to enhance the effectiveness of implementation.

Some of the major studies included nursing staff and treatment staff in their focus groups [16, 21], acknowledging the importance of the influence of relatives and their perception [20]. Our study is one of the first to include family members in the focus groups to allow for a 360 degrees view of the barriers to change experienced in a nursing home. Furthermore, we underlined the importance of including nursing staff in the focus groups, because they form the bridge between treatment staff and patients and their relatives.

Although our study resulted in a novel framework explaining the relationships between barriers to change in nursing homes, it had some possible drawbacks. First, the study was carried out in preparation of selecting and implementing an intervention for reducing inappropriate psychotropic drug use. The focus of the focus group questions was therefore on management and treatment of NPS in combination with the prescription of psychotropic drugs. We asked concrete questions about suboptimal care and did not use the more abstract terminology of barriers to change. Due to this strategy we hope to have facilitated the conversation and to have elicited specific information about everyday practice. However, there is a possibility that we missed some of the barriers encountered. Secondly, the fact that the manager invited focus group participants and was present in the focus groups of the nursing staff and that the moderator sometimes asked provoking questions, could have negatively influenced participants to speak frankly. Next, there was no physician present in the focus groups since the attending physician was newly employed in this nursing home at the time of the research and was therefore unable to reflect on processes and change in this nursing home. Yet, a nurse practitioner functioning at the level of a physician was present, as was described in our methods section. Nevertheless, although many barriers mentioned concern actions of the physician, it is uncertain if the physician or nurse practitioner was meant. This might lead to a skewed interpretation of barriers. Finally, some barriers found in other research did not emerge in the focus groups in this study, such as culture on the care unit and complexity of the change or intervention trying to be achieved [14‐16]. This might be a result of exploring barriers independent from implementing an intervention, including a solitary nursing home, not being able to work according to the principle of data saturation or simply a difference in perspective on the definition of the barrier. The two latter aspects are limitations to this study implying that it is too early to generalize the results. Nevertheless, it prompts investigation whether culture on the unit should be added to the model or whether it is reflected in barriers already present in the model, such as the ‘organizational barriers’, ‘inadequate (multidisciplinary) collaboration’ and ‘personal barriers’. Therefore, we suggest to broaden the scope to other nursing homes and to look into all barriers encountered in nursing home research, not only barriers related to NPS and psychotropic drugs use. Furthermore, we suggest to repeat our method of organizing different mono-disciplinary focus groups and analyze the data deductively, according to our framework, next to performing inductive analyses. In this way it can be assessed if our conceptual framework is complete or if some other (known) barriers or themes arise during the new analysis, complementing the framework. Lastly, we suggest research into facilitators to change. Although it is possible that the facilitators are the opposite of the barriers found, there is no certainty on these findings yet. This will result in a more complete picture of the possible extent to change in nursing homes and will provide practitioners with tools to implement changes and overcome barriers.

We can conclude that we have provided a basic, conceptual framework explaining the relationships between different overarching themes of barriers towards achieving change in management of NPS in nursing homes. The framework may be used as a fundament to assess and to classify barriers to change after it has been confirmed and refined in additional research. It can assist in future research in the determination of steps to be taken when wanting to either improve the extent of change possible, or to establish the current extent to which change may be hindered. Future research could focus on the classification of local barriers and try to resolve and address these barriers. Specifically, the ranks of suboptimal communication, inadequate (multidisciplinary) collaboration and personal barriers call for action into resolving the barriers before attempting implementation of an intervention, to provide optimal implementation.