Computer‐based patient decision aid in interventions for patients with chronic disease

In conjunction with the demographic shift towards an aging population, there is a rising global prevalence of chronic diseases such as cardiovascular disease, cancers (i.e., lung, colorectal, thyroid, liver, gastric, and breast cancers), chronic respiratory disease, chronic kidney disease, and diabetes (Kanbay et al., 2024). These conditions are significantly adding to the escalating medical burden and represent some of the most prevalent and costly health issues. Individuals with chronic diseases must navigate a multitude of medical decisions over their lifetimes as the progressive nature of these conditions necessitates consideration of diverse treatment modalities (Schongut-Grollmus et al., 2021). Computer-based patient decision aids (CPDAs) have been found to offer additional benefits compared to traditional paper-based or video tools by providing personalized evidence-based care that respects patients' autonomy and preferences (Owens et al., 2023). By using CPDAs, patients can access relevant information that enhances their satisfaction and decrease decisional conflict (DC) without requiring the involvement of extra healthcare personnel (Chair et al., 2023). Wu et al. (2024) conducted a systematic review and meta-analysis of 22 studies to quantify the effectiveness of CPDAs for individuals with chronic diseases in shared decision-making. The results of this meta-analysis demonstrated that CPDAs could significantly reduce DC and strengthen knowledge as compared to conventional PDAs when making medical decisions. However, CPDAs showed a non-significant effect on decision regret, satisfaction, self-efficacy, anxiety, and quality of life (all p > .05). Even though this was a comprehensive review presenting multifaceted data synthesized through meta-analysis, a few limitations exist. We would like to propose some suggestions to improve this meta-analysis. First, substantial heterogeneities were observed in several analyses of this meta-analysis. The authors conducted a sensitivity analysis to detect the sources of heterogeneity. However, this method can only identify the great varieties among the included studies. In fact, all the included RCTs met the inclusion criteria of this meta-analysis. The relative risk (RR), odds ratio (OR), or standardized mean differences (SMD) derived from the relevant included studies should be collectively pooled when these studies belong to the same topic. This fact indicates that omitting a study is not recommended, especially in the light of limited included studies, which may affect the real effect size. In this circumstance, the authors are recommended to conduct a subgroup analysis (stratification analysis) and meta-regression analysis which may contribute to identifying the critical factors in influencing the overall effect size or main sources of heterogeneity. Based on the characteristics of included studies in Wu et al.'s study, the influencing factors include geographic region, specific chronic diseases (different disease types, status, duration, and severity), sample size, mean age, and education level. Therefore, stratification analysis and meta-regression analysis may be conducive to detecting the relation between these covariates and the outcomes as well as identifying the sources of heterogeneity. Second, it does not seem reasonable to combine all the data from the related included studies that provided specific outcomes. In the 22 included studies, 16 out of them were studies of cancers, while the remainder 4 studies were non-cancer chronic diseases. Therefore, the outcomes of this meta-analysis can be divided into a cancer group and a non-cancer group. For example, most of the included studies in the '1.1.1 Usual care' and '1.1.2 Paper-based' subgroups of Figure 1a,b were the studies on cancers. Therefore, the overall pooled SMD is suggested to split into two parts (cancer and non-cancer). Third, a random-effects model was applied to evaluate the effects of CPDAs throughout Wu et al.'s study. Commonly, I2 > 50% or p < .05 indicates substantial heterogeneity between studies, and a random-effects model is recommended; otherwise, a fixed-effects model is employed. The authors should give an essential explanation on this choice. Last, most of the included studies (21/22, 95%) were the countries from Europe, America, or Australia, which may induce a significant geographical bias. The authors should describe why such research in the Asia area is sparse and present an outlook on future research in Asian countries. CPDAs are effective approaches for chronic disease management, especially in reducing DC and enhancing the knowledge of the patients in shared decision-making. However, as mentioned by Wu et al., future studies should focus on the detailed compositions, intervention measures, and frequencies of CPDAs. Besides, more high-quality randomized controlled trials with large-sample sizes are needed to investigate the effects of CPDAs on those patients with lower education levels and other chronic conditions. This work was supported by the grants from the Science and Technology Program of Traditional Chinese Medicine in Zhejiang Province (No. 2024ZL1272 for Tianyao Yang; No. 2023ZR134 for Xiangjing Wang). The authors declare that they have no conflict of interest. None.