Background: The recommended first-line treatment for unspecific and degenerative back pain consists of movement exercises and patient education. Objective: Using a pragmatic, randomized controlled trial, we evaluated the effectiveness of a digital home exercise program on self-reported pain intensity compared with the standard of care for physiotherapy. Methods: Participant recruitment was based on newspaper advertisements and a consecutive on-site assessment for eligibility and enrollment. Participants with unspecific and degenerative back pain aged ≥18 years were randomly assigned in a 1:1 ratio to receive a 12-week stand-alone digital home exercise program or physiotherapy. The digital home exercise program included 4 exercises daily, while physiotherapy included 6 to 12 sessions, depending on the severity of symptoms. The primary outcome was pain, which was assessed using a verbal numerical rating scale. The clinical relevance of pain reduction was assessed using the following thresholds: improvement of at least 1.4 points on the verbal numerical rating scale and a pain reduction of at least 30%. Results: During the study period, 108 participants were assigned to the intervention group and 105 participants to the control group. The mean difference in pain scores between the 2 groups at 12 weeks was −2.44 (95% CI −2.92 to −1.95; P<.01) in favor of the intervention group. The group receiving the digital therapeutic achieved a clinically relevant reduction in pain over the course of the study (baseline vs 12 weeks), with a mean change of −3.35 (SD 2.05) score points or −53.1% (SD 29.5). By contrast, this change did not reach clinical relevance in the control group (mean −0.91, SD 1.5; −14.6%, SD 25.3). Retention rates of 89.9% in the intervention group and 97.3% in the control group were maintained throughout the study. Conclusions: The use of the app-based home exercise program led to a significant and clinically relevant reduction in pain intensity throughout the 12-week duration of the program. The intervention studied showed superior improvement in self-reported pain intensity when compared with the standard of care. Given the great demand for standard physiotherapy for unspecific and degenerative back pain, digital therapeutics are evolving into a suitable therapeutic option that can overcome the limitations of access and availability of conventional modes of health care delivery into this spectrum of indications. However, further independent evaluations are required to support the growing body of evidence on the effectiveness of digital therapeutics in real-world care settings. Trial Registration: German Clinical Trials Register DRKS00022781; https://tinyurl.com/hpdraa89

Background: Healthcare systems around the world experience increasing pressure to control future growth of healthcare expenditures. Among other initiatives, quality and value-based benchmarking has become an important field to inform clinical evaluation and reimbursement questions. The International Consortium for Health Outcomes Measurement (ICHOM) has become one of the driving forces to translate scientific evidence into standardized assessments that are routinely applicable in day-to-day care settings. These aim to provide a benchmarking tool that allows the comparison and competition of health care delivery on the basis of value-based health care principles. Methods: This work focuses on the consolidation of the ICHOM methodology and presents insights from 27 routinely implemented Standard Sets. The analysis is based on a literature review of the ICHOM literature repository, a process document review and key informant interviews with ICHOM’s outcomes research and development team. Results: Key findings are that the scope of ICHOM Standard Sets shifted from a more static focus on burden of disease and poorly standardized care pathways to a more dynamic approach that also takes into account questions about the setting of care, feasibility of implementing a benchmarking tool and compatibility of different Standard Sets. Although certain overlaps exist with other initiatives in the field of patient reported outcomes (PRO), their scopes differ significantly and they hence rather complement each other. ICHOM pursues a pragmatic approach to enable the benchmarking and the analysis of healthcare delivery following the principles of value-based healthcare. Conclusion: The ICHOM Standard Sets complement other initiatives in the field of patient-reported outcomes (PRO) and functional reporting by placing a particular focus on healthcare delivery, while other initiatives primarily focus on evaluation of academic endpoints. Although ICHOM promotes a pragmatic approach towards developing and devising its Standard Sets, the definition of standardized decision making processes emerged as one of the key challenges. Furthermore, the consolidation of core metrics across number of disease areas to enable the parallel implementation of different Standard Sets in the same care setting is an important goal that will enable the widespread implementation of patient-reported outcome measures (PROM).

Despite the efforts of the past decades, cancer is still among the key drivers of global mortality. To increase the detection rates, screening programs and other efforts to improve early detection were initiated to cover the populations at a particular risk for developing a specific malignant condition. These diagnostic approaches have, so far, mostly relied on conventional diagnostic methods and have made little use of the vast amounts of clinical and diagnostic data that are routinely being collected along the diagnostic pathway. Practitioners have lacked the tools to handle this ever-increasing flood of data. Only recently, the clinical field has opened up more for the opportunities that come with the systematic utilisation of high-dimensional computational data analysis. We aim to introduce the reader to the theoretical background of machine learning (ML) and elaborate on the established and potential use cases of ML algorithms in screening and early detection. Furthermore, we assess and comment on the relevant challenges and misconceptions of the applicability of ML-based diagnostic approaches. Lastly, we emphasise the need for a clear regulatory framework to responsibly introduce ML-based diagnostics in clinical practice and routine care.

Background: The GPT-4 is a large language model (LLM) trained and fine-tuned on an extensive dataset. After the public release of its predecessor in November 2022, the use of LLMs has seen a significant spike in interest, and a multitude of potential use cases have been proposed. In parallel, however, important limitations have been outlined. Particularly, current LLMs encounter limitations, especially in symbolic representation and accessing contemporary data. The recent version of GPT-4, alongside newly released plugin features, has been introduced to mitigate some of these limitations. Objective: Before this background, this work aims to investigate the performance of GPT-3.5, GPT-4, GPT-4 with plugins, and GPT-4 with plugins using pretranslated English text on the German medical board examination. Recognizing the critical importance of quantifying uncertainty for LLM applications in medicine, we furthermore assess this ability and develop a new metric termed “confidence accuracy” to evaluate it. Methods: We used GPT-3.5, GPT-4, GPT-4 with plugins, and GPT-4 with plugins and translation to answer questions from the German medical board examination. Additionally, we conducted an analysis to assess how the models justify their answers, the accuracy of their responses, and the error structure of their answers. Bootstrapping and CIs were used to evaluate the statistical significance of our findings. Results: This study demonstrated that available GPT models, as LLM examples, exceeded the minimum competency threshold established by the German medical board for medical students to obtain board certification to practice medicine. Moreover, the models could assess the uncertainty in their responses, albeit exhibiting overconfidence. Additionally, this work unraveled certain justification and reasoning structures that emerge when GPT generates answers. Conclusions: The high performance of GPTs in answering medical questions positions it well for applications in academia and, potentially, clinical practice. Its capability to quantify uncertainty in answers suggests it could be a valuable artificial intelligence agent within the clinical decision-making loop. Nevertheless, significant challenges must be addressed before artificial intelligence agents can be robustly and safely implemented in the medical domain.

Background: Musculoskeletal conditions are the main drivers of global disease burden and cause significant direct and indirect health care costs. Digital health applications improve the availability of and access to adequate care. The German health care system established a pathway for the approval of “Digitale Gesundheitsanwendungen” (DiGAs; Digital Health Applications) as collectively funded medical services through the “Digitale-Versorgung-Gesetz” (Digital Health Care Act) in 2019. Objective: This article presents real-world prescription data collected through the smartphone-based home exercise program “Vivira,” a fully approved DiGA, regarding its effect on self-reported pain intensity and physical inability in patients with unspecific and degenerative pain in the back, hip, and knee. Methods: This study included 3629 patients (71.8% [2607/3629] female; mean age 47 years, SD 14.2 years). The primary outcome was the self-reported pain score, which was assessed with a verbal numerical rating scale. The secondary outcomes were self-reported function scores. To analyze the primary outcome, we used a 2-sided Skillings-Mack test. For function scores, a time analysis was not feasible; therefore, we calculated matched pairs using the Wilcoxon signed-rank test. Results: Our results showed significant reductions in self-reported pain intensity after 2, 4, 8, and 12 weeks in the Skillings-Mack test (T3628=5308; P<.001). The changes were within the range of a clinically relevant improvement. Function scores showed a generally positive yet more variable response across the pain areas (back, hip, and knee). Conclusions: This study presents postmarketing observational data from one of the first DiGAs for unspecific and degenerative musculoskeletal pain. We noted significant improvements in self-reported pain intensity throughout the observation period of 12 weeks, which reached clinical relevance. Additionally, we identified a complex response pattern of the function scores assessed. Lastly, we highlighted the challenges of relevant attrition at follow-up and the potential opportunities for evaluating digital health applications. Although our findings do not have confirmatory power, they illustrate the potential benefits of digital health applications to improve the availability of and access to medical care. Trial Registration: German Clinical Trials Register DRKS00024051; https://drks.de/search/en/trial/DRKS00024051

Background: Musculoskeletal conditions are among the main contributors to the global burden of disease. International guidelines consider patient education and movement exercises as the preferred therapeutic option for unspecific and degenerative musculoskeletal conditions. Innovative and decentralized therapeutic means are required to provide access to and availability of such care to meet the increasing therapeutic demand for this spectrum of conditions. Objective: This retrospective observational study of preliminary use and outcome data explores the clinical outcomes of Vivira (hereafter referred to as “program”), a smartphone-based program for unspecific and degenerative pain in the back, hip, and knee before it received regulatory approval for use in the German statutory health insurance system. Methods: An incomplete matched block design was employed to assess pain score changes over the intended 12-week duration of the program. Post hoc analyses were performed. In addition, a matched comparison of self-reported functional scores and adherence rates is presented. Results: A total of 2517 participants met the inclusion criteria and provided sufficient data to be included in the analyses. Overall, initial self-reported pain scores decreased significantly from an average of 5.19 out of 10 (SD 1.96) to an average of 3.35 out of 10 (SD 2.38) after 12 weeks. Post hoc analyses indicate a particularly emphasized pain score reduction over the early use phases. Additionally, participants with back pain showed significant improvements in strength and mobility scores, whereas participants with hip or knee pain demonstrated significant improvements in their coordination scores. Across all pain areas and pain durations, a high yet expected attrition rate could be observed. Conclusions: This observational study provides the first insights into the clinical outcomes of an exercise program for unspecific and degenerative back, hip, and knee pain. Furthermore, it demonstrates a potential secondary benefit of improved functionality (ie, strength, mobility, coordination). However, as this study lacks confirmatory power, further research is required to substantiate the clinical outcomes of the program assessed. Trial Registration: German Clinical Trials Register DRKS00021785; https://drks.de/search/en/trial/DRKS00021785