Development and implementation of a multilevel intervention to improve pain management in Swiss nursing homes

Pain is a highly prevalent symptom in older people and particularly among nursing home residents. Under- or untreated pain can lead to severe physical and psychological consequences, such as impaired mobility, sleep disturbances and an increase of depressive symptoms. Despite the availability of evidence-based recommendations for the management of geriatric pain, pain management in nursing homes remains a persistent issue. Various barriers on the level of residents, care workers and the organizations have been reported to hinder adequate pain management. One critical aspect in this regard are common negative beliefs about pain and its management in older people, particularly in residents with cognitive impairment. Previous studies to improve the adoption of evidence-based pain management in nursing homes showed mixed results. Yet, few studies have systematically investigated the factors contributing to the difficulties in implementation. Implementation science highlights the critical role of contextual factors in implementing new practices. Knowledge of the local hindering and facilitating factors towards the new practice, hence, is crucial to develop and select appropriate strategies to facilitate implementation and to help interpretation of varying implementation success. The overall aim of this dissertation was to develop and implement a multilevel intervention to improve pain management in a group of Swiss nursing homes. Therefore, in the first phase, a comprehensive analysis of the implementation context, incorporating perspectives of care workers and residents was conducted. Based on these insights, in the second phase, a multilevel intervention and implementation strategies were developed and evaluated in a subsample of nursing homes. This dissertation is embedded in the ProQuaS (Identification and development of interfaces and Processes to improve Quality of life of residents at Senevita) study, a three- years mixed-methods study focusing on potentially avoidable hospitalizations and pain management in nursing home residents. The ProQuaS project is conducted in the institutions of Senevita AG, a privately-owned nursing home group in Switzerland, with currently 26 institutions providing assisted-living and long-term care. This dissertation focuses on the pain management aspects of the overall project. Overall, this dissertation is structured in eight chapters. Chapters 1 and 2 provide an overview of the topic and state the aims of this dissertation.11 Chapter 1 provides a general introduction to the field of pain management in nursing homes, including an overview of the state of the science in pain management practice and interventions to improve pain management in NHs. Furthermore, a short introduction to principles of implementation science, the Swiss NH context in general and the ProQuaS project in specific, are provided. Chapter 2 describes the aims of this dissertation. The articles presented in Chapters 3 and 4 constitute the findings of the contextual analysis, which has been conducted in the first phase of ProQuaS. Chapter 3 reports a mixed-methods study, which integrated data of care worker surveys and focus group discussions to identify facilitating and hindering factors for pain management in nursing homes. The capability-opportunitymotivation determine behavior (COM-B) model was used to discuss implications for the selection and development of implementation strategies. Results from the questionnaire survey indicate, that lacking availability (60.9%) and application of non-pharmacological treatment (53.6%); reluctance of residents to report pain (51.1%) and lack of time for a comprehensive pain assessment (50.5%) are perceived to be the major barriers to pain management. Focus group discussions corroborated and extended the quantitative findings with facilitators, e.g., joint ward rounds with physicians, good knowledge of the resident, and further barriers, including high turnover and lacking established routines. The findings suggest that implementation strategies should aim at increasing pain management knowledge and foster motivational aspects to sustainable change in pain management practice, while at the same time factoring in contextual factors, such as high turnover. Chapter 4 reports a qualitative study describing nursing home residents’ perceptions of barriers and facilitators and their needs regarding pain management. We identified three central themes from the data: dealing with major life changes, managing pain, and using formal care. Overall, the interviews highlighted the multidimensionality of NH residents’ pain experience and their corresponding needs. Participants perceived that care workers were not always able to respond adequately to these needs. Instead, participants indicated to have learned to cope with their pain using self-developed strategies and by directly consulting their physicians. These findings emphasize our findings of Chapter 3. Amongst others, strategies to improve pain management in nursing homes need to address prevailing attitudes and shortages in knowledge concerning pain12 management in older people. Creating a common understanding of the biopsychosocial and spiritual dimensions of pain is crucial to enable the provision of person- centered pain management. Chapter 5 comprises the study protocol of the second part of the ProQuaS study. It provides a detailed overview of the intervention and implementation strategies and the evaluation plan for the outlined hybrid II effectiveness- implementation study. Furthermore, the conceptual framework of ProQuaS is presented- highlighting potential associations between the intervention, implementation strategies, contextual/individual factors and outcomes. The conceptual model is based on the Consolidated Framework of Implementation Research (CFIR) and the Theoretical Domains Framework to provide a comprehensive understanding of facilitating and hindering factors. The intervention consists of a pain management guideline which has been derived from international recommendations to geriatric pain management and was adapted to the nursing homes’ context in collaboration with a group of stakeholder from the participating nursing homes. The central implementation strategies are interactive training workshops for care workers of all levels and the training and introduction of pain champions on the participating wards. Additionally supporting strategies encompass preparatory and ongoing meetings with the nursing homes’ leadership, adaptions of the residents’ documentation software and the provision of cardboard pain assessment scales and printed copies of the pain management guidelines on the wards. The articles presented in Chapters 6 and 7 constitute the evaluation of the intervention and its implementation with regard to the residents and care workers. Chapter 6 reports the effectiveness of the intervention on pain- related resident outcomes and the utility of implementation strategies with regard to the intervention’s reach and acceptability. Using a quasi-experimental pre-post design with one pre- and two post- measures at three and six months after baseline, we could show that all pain-related resident outcomes improved over time. In a self-report subsample (n= 43), changes in intensity of average pain were significant at three months, and changes in intensity of worst pain were significant at three and six months. In addition, we looked at reach and acceptability of the intervention as perceived by care workers and found that about 76% of care workers were familiar with the guideline, 70.4% answered that the guideline is practical and correlates with their ideas of good pain assessment (75.9 %) and treatment (79.7 %). These findings13 suggest, that our approach implementing a pain management guideline using interactive training workshops and introducing trained pain champions could partially improve residents’ pain-related outcomes. However, a process evaluation will provide further insights into the underlying mechanisms and potential barriers of the implementation process. Chapter 7 reports a quantitative process evaluation that explores the mechanisms of change related to our implementation strategies. Based on the conceptual framework of ProQuaS presented in Chapter 5 we hypothesized that by conducting training workshops and introducing pain champions on the wards, we would increase care workers’ self- efficacy in pain management. Furthermore, we hypothesized that self- efficacy in pain management is associated with adoption of the implemented pain management guideline. To test our hypotheses, we used data of the care worker questionnaire survey which was conducted at baseline, after three and six months. Our findings show that care workers’ self- efficacy in pain management at three and six months was significantly higher compared to baseline. Self-reported adoption of the pain management guidelines was mixed, depending on the core component between 44% and 73%. We found significant associations between care workers’ self-efficacy and adoption of two core components: ‘conducting a comprehensive assessment’ and ‘use of PAINAD’. There were no significant associations with the components ‘documentation’ and ‘re-evaluation’, though. These findings suggest that our strategies might have increased care workers’ self-efficacy, however with regard to adoption, self- efficacy might not be the only contributing factor. Documentation and reevaluation of residents’ pain after (non-)pharmacological treatment might rather depend on memory processes. Our findings emphasize the importance of continuous commitment of internal facilitators, e.g., pain champions, to drive implementation efforts. Chapter 8, ultimately synthesizes and discusses major findings of all studies in the context of the literature. Furthermore, strengths and weaknesses of the studies are discussed, and implications for future research and practice are presented. This dissertation contributes to the current literature in the field of pain management in nursing homes by using implementation science principles and behavioral theory to improve understanding of implementation challenges. Translation of this knowledge is facilitated by the detailed description of the local context, including determinants of change in pain management, on the one hand and the implementation strategies on the other hand