Technology-supported shared decision-making in chronic conditions: A systematic review of randomized controlled trials.

OBJECTIVES: To describe the role of patients with a chronic disease, healthcare professionals (HCPs) and technology in shared decision making (SDM) and the use of clinical decision support systems (CDSSs), and to evaluate the effectiveness of SDM and CDSSs interventions. METHODS: Randomized controlled studies published between 2011 and 2021 were identified and screened independently by two reviewers, followed by data extraction and analysis. SDM elements and interactive styles were identified to shape the roles of patients, HCPs and technology. RESULTS: Forty-three articles were identified and reported on 21 SDM-studies, 15 CDSS-studies, 2 studies containing both an SDM-tool and a CDSS, and 5 studies with other decision support components. SDM elements were mostly identified in SDM-tools and interactions styles were least common in the other decision support components. CONCLUSIONS: Patients within the included RCTs mainly received information from SDM-tools and occasionally CDSSs when it concerns treatment strategies. HCPs provide and clarify information using SDM-tools and CDSSs. Technology provides interactions, which can support more active SDM. SDM-tools mostly showed evidence for positive effects on SDM outcomes, while CDSSs mostly demonstrated positive effects on clinical outcomes. PRACTICE IMPLICATIONS: Technology-supported SDM has potential to optimize SDM when patients, HCPs and technology collaborate well together.

Exploratory application of machine learning methods on patient reported data in the development of supervised models for predicting outcomes.

BACKGROUND: Patient-reported outcome measurements (PROMs) are commonly used in clinical practice to support clinical decision making. However, few studies have investigated machine learning methods for predicting PROMs outcomes and thereby support clinical decision making. OBJECTIVE: This study investigates to what extent different machine learning methods, applied to two different PROMs datasets, can predict outcomes among patients with non-specific neck and/or low back pain. METHODS: Using two datasets consisting of PROMs from (1) care-seeking low back pain patients in primary care who participated in a randomized controlled trial, and (2) patients with neck and/or low back pain referred to multidisciplinary biopsychosocial rehabilitation, we present data science methods for data prepossessing and evaluate selected regression and classification methods for predicting patient outcomes. RESULTS: The results show that there is a potential for machine learning to predict and classify PROMs. The prediction models based on baseline measurements perform well, and the number of predictors can be reduced, which is an advantage for implementation in decision support scenarios. The classification task shows that the dataset does not contain all necessary predictors for the care type classification. Overall, the work presents generalizable machine learning pipelines that can be adapted to other PROMs datasets. CONCLUSION: This study demonstrates the potential of PROMs in predicting short-term patient outcomes. Our results indicate that machine learning methods can be used to exploit the predictive value of PROMs and thereby support clinical decision making, given that the PROMs hold enough predictive power.

Factors used by general practitioners for referring patients with chronic musculoskeletal pain: a qualitative study.

Around 20% of the Dutch population is living with chronic musculoskeletal pain (CMP), which is a complex and multifactorial problem. This complexity makes it hard to define a classification system, which results in non-satisfactory referring from the general practitioner (GP). CMP is often explained using the biopsychosocial model in which biological, psychological and social factors cause and maintain the pain. The presented study investigated the factors related to the GPs' referral for patients with CMP to further treatment.Using convenience sampling, semi-structured interviews and a focus group were conducted among 14 GPs. The interviews were iteratively analyzed using inductive conventional content analysis.Analysis of the interviews demonstrated that there were 28 referral factors that were mentioned by more than 50% of the interviewed GPs. The results showed that the GPs were mostly focussing on the physical (e.g. pain location) and psychological (e.g. acceptation of pain) factors, indicating that they lack focus on the social factors. Furthermore, unfamiliarity of GPs with treatment options was a noteworthy finding.The referral of patients with CMP by GPs is complex and based on multiple factors. To improve referral, it is recommended to include social factors in the decision-making process and to increase the familiarity of the GPs with available treatments.