Journal of Patient-Reported Outcomes最新文献

Background: Artificial intelligence (AI) and machine learning (ML) are increasingly integrated into healthcare, offering potential advancements in patient-reported outcome measures (PROMs) for surgical populations. Improved PROMs can enhance patient-centered care by accurately capturing patient experiences with minimal burden.

Objective: In the context of surgery, where recovery trajectories vary widely, this study aims to systematically review the use of AI and ML in the development, application, and prediction capabilities of PROMs in surgical populations, with a focus on psychometric properties and the predictive accuracy of post-surgical outcomes.

Methods: A comprehensive search of the PubMed database was conducted from inception until August 2024. Studies were included if they utilized AI or ML in the development, application, or predicting PROMs for surgical patients. Methodological quality was assessed using COSMIN and PROBAST tools, depending on study design. A qualitative synthesis of findings was performed.

Results: Twenty-two studies met the inclusion criteria, with 19 rated as high quality. Six studies focused on developing computer adaptive tests (CAT) PROMs, seven on evaluating psychometric properties, and five on ML for post-surgical outcome prediction. CAT PROMs showed comparable measurement accuracy to traditional PROMs, good to excellent construct validity, and significantly reduced patient burden by reducing the length of questionnaires. ML algorithms, such as logistic regression, random forests, extreme gradient boosting, and neural networks, achieved similar predictive accuracy for post-surgical outcomes, with no single model demonstrating consistent superiority.

Conclusions: AI and ML have the potential to improve PROM utilization in surgical care by enhancing efficiency and personalization while maintaining data quality. Clinicians can use AI-driven PROMs to reduce patient burden and integrate ML models for accurate post-surgical outcome prediction, thereby optimizing patient-centered care.

Purpose: Patient-reported outcomes (PROs) assist patients and clinicians in assessing treatment effectiveness and enhancing healthcare quality. This study aims to explore and analyze the application and characteristics of PROs in clinical trials of Traditional Chinese Medicine (TCM).

Methods: This cross-sectional study was based on randomized clinical trials of TCM between January 1, 2010, and December 31, 2022 in International Clinical Trials Registry Platform. For each included trial, data including study phase, design, participant demographics, target diseases, PROs, and PRO measurements were extracted. Trials were categorized into three groups: (1) recorded specified patient-reported outcome tools, (2) referenced patient subjective outcomes without specified tools, and (3) did not mention any PROs. Further descriptive statistical analysis were conducted on the most commonly used PRO tools in different countries and for different diseases.

Results: Among a total of 7783 eligible trials, 4858 (62.4%) listed explicit PRO tools, and 850 (10.9%) referenced PROs without specified tools. The most common conditions evaluated by PRO tools were musculoskeletal diseases (935 trials, 19.2%), symptoms (714, 14.7%), and neurological diseases (500, 10.3%). Frequently used PRO tools included the Visual Analogue Scale (VAS), 36-item Short-Form Health Questionnaire, and Pittsburgh Sleep Quality Index. Regionally, most PRO-related trials were in the Western Pacific (3904, 68.4%) and fewest in Africa (8, 0.1%). Countries conducting the most PRO-related trials were China, Iran, the USA, South Korea, and Brazil, focusing on musculoskeletal, symptoms, neurological, genitourinary, and digestive diseases, with varying popular disease-specific PRO tools by country. Musculoskeletal diseases were the primary focus in China, Brazil, and South Korea.

Conclusions: The use of PROs in TCM clinical trials has grown during the study period. However, there was an uneven regional distribution of PRO application and a lack of standardized, reliable PRO tools tailored for TCM. Great efforts are needed to enhance the quality and promote the use of PRO tools in TCM clinical research.

Background: Accurate assessment of physical activity behaviors allows clinicians to target counseling more effectively. Single questions regarding habitual physical activity provide the most feasible option for rapid physical activity assessment. This study evaluated construct validity of a single item physical activity question to self-report habitual activity as simple categorical descriptors (not active, somewhat active, active, very active).

Methods: Women (N = 120) completed one laboratory session for measurement of self-reported physical activity (International Physical Activity Questionnaire-IPAQ), anthropometrics and body composition, and physical activity-related quality of life (SF-36 physical function subscale, Vitality Plus Scale). Construct validity was evaluated as convergent and known-groups validity.

Results: Mean age was 60 ± 16 years (range 25-89). The majority (62%) reported being active or very active. Age was not significantly related to activity levels. Correlation analysis demonstrated good convergent validity. Significant negative correlations were found with body weight, body mass index (BMI), waist circumference, and body fat (all p <0.001). Significant positive correlations were found with vigorous intensity activity (p <0.001), moderate intensity activity (p = .004), walking (p = .005), and quality of life (p <0.001). Good known-groups validity was demonstrated by significant differences between habitual physical activity levels for body weight, waist circumference, body fat, vigorous intensity activity (all p <0.001), moderate intensity activity (p = .038), walking (p = .049), and quality of life (p <0.001). Regression models confirmed known-groups validity.

Conclusions: A single question with categorical descriptors is valid for brief clinical assessment of habitual physical activity in women across a wide age range.