Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet

Background: Clinical decision support systems (CDSSs) based on routine care data, using artificial intelligence (AI), are increasingly being developed. Previous studies focused largely on the technical aspects of using AI, but the acceptability of these technologies by patients remains unclear. Objective: We aimed to investigate whether patient-physician trust is affected when medical decision-making is supported by a CDSS. Methods: We conducted a vignette study among the patient panel (N=860) of the University Medical Center Utrecht, the Netherlands. Patients were randomly assigned into 4 groups—either the intervention or control groups of the high-risk or low-risk cases. In both the high-risk and low-risk case groups, a physician made a treatment decision with (intervention groups) or without (control groups) the support of a CDSS. Using a questionnaire with a 7-point Likert scale, with 1 indicating “strongly disagree” and 7 indicating “strongly agree,” we collected data on patient-physician trust in 3 dimensions: competence, integrity, and benevolence. We assessed differences in patient-physician trust between the control and intervention groups per case using Mann-Whitney U tests and potential effect modification by the participant’s sex, age, education level, general trust in health care, and general trust in technology using multivariate analyses of (co)variance. Results: In total, 398 patients participated. In the high-risk case, median perceived competence and integrity were lower in the intervention group compared to the control group but not statistically significant (5.8 vs 5.6; P=.16 and 6.3 vs 6.0; P=.06, respectively). However, the effect of a CDSS application on the perceived competence of the physician depended on the participant’s sex (P=.03). Although no between-group differences were found in men, in women, the perception of the physician’s competence and integrity was significantly lower in the intervention compared to the control group (P=.009 and P=.01, respectively). In the low-risk case, no differences in trust between the groups were found. However, increased trust in technology positively influenced the perceived benevolence and integrity in the low-risk case (P=.009 and P=.04, respectively). Conclusions: We found that, in general, patient-physician trust was high. However, our findings indicate a potentially negative effect of AI applications on the patient-physician relationship, especially among women and in high-risk situations. Trust in technology, in general, might increase the likelihood of embracing the use of CDSSs by treating professionals.

Background: Acute kidney injury (AKI) is associated with increased risks of incidence or progression of chronic kidney disease (CKD), kidney failure (KF), or major adverse kidney events (MAKE), however, it remains unclear which individuals have higher risks. Hence, we systematically reviewed the literature to explore differences in kidney dysfunction risks between AKI stages, AKI durations, and clinical settings. Methods: We performed a systematic search in PubMed and Embase to identify studies that examined at least one outcome of interest in individuals with AKI versus without AKI, with a minimum follow-up of one year. Hazard/odds ratios (HR/OR) were pooled using random effects models. Heterogeneity across patient and disease characteristics was examined using subgroup and meta-regression analyses. Results: We searched 70 studies, encompassing 1 838 668 individuals, including 165 715 with AKI. All studies were of moderate to high quality. Individuals with AKI had higher risks of CKD incidence [AKI 25.8%/no AKI 8.7%; HR 2.36 [95% confidence interval (CI) 1.77–2.94)]], CKD progression [AKI 43.1%/no AKI 35.6%; HR 1.83 (95%CI 1.26–2.40)], KF [AKI 2.9%/no AKI 0.5%; HR 2.64 (95%CI 2.03–3.25)], and MAKE [AKI 59.0%/no AKI 32.7%; OR 2.77 (95%CI 2.01–3.53)]. The pooled effect estimates for CKD incidence after AKI lasting <3 days remained significant [OR 2.37 (95%CI 1.68–3.07)], even in individuals with AKI stage 1 only [HR 1.49 (95%CI 1.44–1.55)]. Diabetes mellitus, hypertension, requiring acute dialysis, cardiovascular surgery, or coronary artery disease were associated with higher CKD incidence or progression risks. Conclusions: Risks for kidney dysfunction were higher for all individuals with AKI. Risk estimates were heterogeneous between patient subgroups, based on AKI stage, AKI duration, and clinical setting, yet even individuals with the lowest stage or shortest duration of AKI remained at higher risk for CKD incidence or progression. This highlights the need to develop tailored follow-up strategies to recognize kidney function decline post-AKI and initiate kidney protective measures in a timely fashion.

Background Severe postpartum hemorrhage (SPPH) is the leading cause of maternal mortality and morbidity worldwide. Platelet anomalies frequently occur during pregnancy. However, their role in the etiology of SPPH is largely unknown. Objective To study the relation between platelet parameters and SPPH. Methods This retrospective single-center cohort included deliveries between 2009 and 2017. SPPH was defined as ≥1000 ml blood loss within 24 h after delivery. Platelet parameters were measured within 72 h before delivery. Multiple imputation was performed for missing data. Odds ratios were adjusted (aORs) for maternal age, multiple gestation, macrosomia, induction of labor, preeclampsia, and hemolysis, elevated liver enzymes, and low platelets syndrome. Results A total of 23 205 deliveries were included. Of the 2402 (10.4%) women with thrombocytopenia (<150 × 10⁹/L), 10.3% developed SPPH, compared with 7.6% of women with a normal platelet count (aOR: 1.34, 95% CI: 1.14–1.59). Women with a platelet count of <50 × 10⁹/L were most at risk (aOR of 2.24 [1.01–4.94]) compared with the reference group with normal platelet counts; the aOR was 1.22 (0.77–1.93) for the 50–99 × 10⁹/L platelet count group and 1.31 (1.10–1.56) for the 100–149 × 10⁹/L platelet count group. Plateletcrit was associated with SPPH (aOR 1.15 [1.08–1.21] per 0.05% decrease), and, although rarely present, a platelet distribution width (PDW) ≥23% (n = 22) also increased the odds of SPPH (aOR 6.05 [2.29–16.20]). Conclusion Different degrees of thrombocytopenia were independently associated with the occurrence of SPPH. Despite their relation to SPPH, plateletcrit and a PDW of ≥23% have limited additional value in addition to platelet count.

Background: The new concept of difficult-to-treat rheumatoid arthritis (D2T RA) refers to RA patients who remain symptomatic after several lines of treatment, resulting in a high patient and economic burden. During a hackathon, we aimed to identify and predict D2T RA patients in structured and unstructured routine care data. Methods: Routine care data of 1873 RA patients were extracted from the Utrecht Patient Oriented Database. Data from a previous cross-sectional study, in which 152 RA patients were clinically classified as either D2T or non-D2T, served as a validation set. Machine learning techniques, text mining, and feature importance analyses were performed to identify and predict D2T RA patients based on structured and unstructured routine care data. Results: We identified 123 potentially new D2T RA patients by applying the D2T RA definition in structured and unstructured routine care data. Additionally, we developed a D2T RA identification model derived from a feature importance analysis of all available structured data (AUC-ROC 0.88 (95% CI 0.82–0.94)), and we demonstrated the potential of longitudinal hematological data to differentiate D2T from non-D2T RA patients using supervised dimension reduction. Lastly, using data up to the time of starting the first biological treatment, we predicted future development of D2TRA (AUC-ROC 0.73 (95% CI 0.71–0.75)). Conclusions: During this hackathon, we have demonstrated the potential of different techniques for the identification and prediction of D2T RA patients in structured as well as unstructured routine care data. The results are promising and should be optimized and validated in future research.