Clinical Decision Support Systems in Adults with Chronic Respiratory Disease: A Systematic Review

Background Clinical decision support systems (CDSS) are systems which analyse and present data so that users can make decisions more easily and are suggested as part of the solution to poor adoption of guidelines.Aims The aim of this review was to describe the literature on CDSS in chronic respiratory disease management, their function and the features that may impact their effectiveness.Methods The electronic databases PubMed, MEDLINE, EMBASE, Web of Science, the trial registries Cochrane Central Register of Controlled Trials, EU Clinical Trial Register, ClinicalTrials.gov and World Health Organisation (WHO) trial registry were searched with the aid of a medical librarian. All searches began from September 2019 to March 2021. Data from relevant studies were extracted independently by 2 review authors using Covidence software and Microsoft Excel. Study quality was assessed. CDSS description and indication, function and the theory underpinning the CDSS (e.g., guidelines, expert opinion, etc.) were themed and summarised. The presence or absence of the important CDSS features were recorded i.e., (i) computer-based decision support, (ii) automatic provision of decision support as part of clinician workflow, (iiii) provision of decision support at the time and location of decision-making and (iv) provision of recommendations rather than just assessments.Methods The electronic databases PubMed, MEDLINE, EMBASE, Web of Science, the trial registries Cochrane Central Register of Controlled Trials, EU Clinical Trial Register, ClinicalTrials.gov and World Health Organisation (WHO) trial registry were searched with the aid of a medical librarian. All searches began from September 2019 to March 2021. Data from relevant studies were extracted independently by 2 review authors using Covidence software and Microsoft Excel. Study quality was assessed. CDSS description and indication, function and the theory underpinning the CDSS (e.g., guidelines, expert opinion, etc.) were themed and summarised. The presence or absence of the important CDSS features were recorded i.e., (i) computer-based decision support, (ii) automatic provision of decision support as part of clinician workflow, (iiii) provision of decision support at the time and location of decision-making and (iv) provision of recommendations rather than just assessments.Results Of the 2647 articles screened, 27 papers or abstracts (n = 22 full text and n = 5 abstracts) describing 26 studies met all inclusion and exclusion criteria. The 26 studies evaluated asthma care (n = 16), COPD care (n = 9) and both asthma and COPD care (n = 1). The highest level of evidence was generated by RCTs (n = 7). The 26 studies described 24 different CDSS. Most were based solely or in part on clinical and best practice guidelines. Whilst most CDSS had multiple functions, which were wide ranging, medicines management was the most prevalent function. Most systems had 3 out of 4 of the specific system features that have been correlated with improvements in clinical practice. Generally, in both controlled studies and in other study designs, patient and process outcome measures were commonly reported with performance outcomes measures less commonly reported. Fewer positive effects were reported in the higher quality studies (i.e., controlled studies). The success rate of the CDSS was higher in those studies that had 3 out of the 4 important CDSS features.Results Of the 2647 articles screened, 27 papers or abstracts (n = 22 full text and n = 5 abstracts) describing 26 studies met all inclusion and exclusion criteria. The 26 studies evaluated asthma care (n = 16), COPD care (n = 9) and both asthma and COPD care (n = 1). The highest level of evidence was generated by RCTs (n = 7). The 26 studies described 24 different CDSS. Most were based solely or in part on clinical and best practice guidelines. Whilst most CDSS had multiple functions, which were wide ranging, medicines management was the most prevalent function. Most systems had 3 out of 4 of the specific system features that have been correlated with improvements in clinical practice. Generally, in both controlled studies and in other study designs, patient and process outcome measures were commonly reported with performance outcomes measures less commonly reported. Fewer positive effects were reported in the higher quality studies (i.e., controlled studies). The success rate of the CDSS was higher in those studies that had 3 out of the 4 important CDSS features.Results Of the 2647 articles screened, 27 papers or abstracts (n = 22 full text and n = 5 abstracts) describing 26 studies met all inclusion and exclusion criteria. The 26 studies evaluated asthma care (n = 16), COPD care (n = 9) and both asthma and COPD care (n = 1). The highest level of evidence was generated by RCTs (n = 7). The 26 studies described 24 different CDSS. Most were based solely or in part on clinical and best practice guidelines. Whilst most CDSS had multiple functions, which were wide ranging, medicines management was the most prevalent function. Most systems had 3 out of 4 of the specific system features that have been correlated with improvements in clinical practice. Generally, in both controlled studies and in other study designs, patient and process outcome measures were commonly reported with performance outcomes measures less commonly reported. Fewer positive effects were reported in the higher quality studies (i.e., controlled studies). The success rate of the CDSS was higher in those studies that had 3 out of the 4 important CDSS features.Results Of the 2647 articles screened, 27 papers or abstracts (n = 22 full text and n = 5 abstracts) describing 26 studies met all inclusion and exclusion criteria. The 26 studies evaluated asthma care (n = 16), COPD care (n = 9) and both asthma and COPD care (n = 1). The highest level of evidence was generated by RCTs (n = 7). The 26 studies described 24 different CDSS. Most were based solely or in part on clinical and best practice guidelines. Whilst most CDSS had multiple functions, which were wide ranging, medicines management was the most prevalent function. Most systems had 3 out of 4 of the specific system features that have been correlated with improvements in clinical practice. Generally, in both controlled studies and in other study designs, patient and process outcome measures were commonly reported with performance outcomes measures less commonly reported. Fewer positive effects were reported in the higher quality studies (i.e., controlled studies). The success rate of the CDSS was higher in those studies that had 3 out of the 4 important CDSS features.Conclusion This review demonstrates that CDSS can improve chronic respiratory disease management processes, performance and clinical outcomes in adult patients with asthma and COPD. Results from non-controlled study designs provided valuable information on important process and performance outcomes, including healthcare utility.