The rapid development of AI solutions reveals opportunities to address the underdiagnosis and poor management of chronic conditions in developing settings. Using the method of simulated patients and experimental designs, we evaluate the quality, safety, and disparity of medical consultation with ERNIE Bot in China among 384 patient-AI trials. ERNIE Bot reached a diagnostic accuracy of 77.3%, correct drug prescriptions of 94.3%, but prescribed high rates of unnecessary medical tests (91.9%) and unnecessary medications (57.8%). Disparities were observed based on patient age and household economic status, with older and wealthier patients receiving more intensive care. Under standardized conditions, ERNIE Bot, ChatGPT, and DeepSeek demonstrated higher diagnostic accuracy but a greater tendency toward overprescription than human physicians. The results suggest the great potential of ERNIE Bot in empowering quality, accessibility, and affordability of healthcare provision in developing contexts, but also highlight critical risks related to safety and amplification of sociodemographic disparities.

Healthcare provider dearth, particularly in Low and Middle-income countries (LMICs), reduces the development towards improved healthcare outcomes and accomplish the Sustainable Development Goals (SDGs-3). This study aims to investigate the association between Professional and Non-professional healthcare provider distribution and neonatal mortality through the mediating role of maternal and child healthcare (MCH) services in Pakistan. The present research used data from the 2018 Pakistan Demographic and Health Survey (PDHS) total sample that comprises (N = 8022). The dependent variable includes the risk of neonatal mortality 0–30 days. Bivariate and Multivariate analyses were conducted by utilizing the Cox proportional hazards and regression models. The estimation of mortality rates was achieved through generalized structural equation modelling (GSEM) and the 5000 bootstrap technique. All the investigations were carried out in STATA v. 16.1. Our results revealed that around 2.46% of neonatal deaths occur. The average healthcare provider distribution type was professional, 83.45%, non-professional, 1.45%, and no care, 15.11%. 0.88% of females are less likely to die within the first thirty days of their birth compared to males. The regression outcomes demonstrate that the healthcare providers were positive and significantly linked to neonatal mortality. Additionally, a unit increase in healthcare providers has a 1.20% Higher likelihood of increasing the MCH and overall reducing neonatal mortality (HR = 0.60). The outcomes illustrate that healthcare provider distribution and MCH play a vital role in minimizing neonatal mortality. To ensure well-distributed healthcare provider distribution, promoting maternal education, empowering women’s decision-making in acquiring MCH services utilization, and expanding access to free MCH services are the essentials for better maternal and healthcare outcomes.

Background: Health care workers (HCWs) played a key role in preventing and controlling COVID-19. Higher infection risks and intensive work led to occupational burnout for many HCWs, which may affect their lifestyle behaviors and weight. Objective: This study aimed to assess HCWs’ self-rated health status, overweight and obesity rates, lifestyle behaviors, and psychoemotional changes from 2019 to 2022 across China and to analyze the factors associated with changes from underweight or normal weight in 2019 to overweight or obese in 2022. Methods: In this retrospective study, 100 health care institutions were randomly selected from 5 provinces or regions in China. All HCWs who worked in the institutions for at ≥3 years were invited to complete the electronic questionnaire and participate in the online survey from August 1, 2022, to August 31, 2022. Collected data included changes in lifestyle behaviors (dietary habits, physical activity, sleep quality, smoking, alcohol consumption), psychoemotional conditions (persistent stress or recurrent anxiety or depressed mood), health status, and chronic disease control from December 2019 to August 2022. Height and weight in 2019 and 2022 were retrieved from annual physical examination records. Overweight and obesity were defined as 24.0 kg/m2≤BMI<28.0 kg/m2 (overweight) and BMI≥28.0 kg/m2 (obesity). Chi square tests and ANOVAs were used to assess the associations between groups. Logistic regression models were used to analyze the factors associated with HCWs becoming overweight or obese from 2019 to 2022. Results: The questionnaire was submitted by 23,234 HCWs. Of the underweight or normal weight HCWs in 2019, 12.67% (1486/23,234) became overweight or obese in 2022; this change was associated with the following factors: 34-43 years old (OR 0.843, 95% CI 0.740-0.960), 44-53 years old (OR 0.738, 95% CI 0.635-0.960), and 54-63 years old (OR 0.503, 95% CI 0.368-0.685; reference: 24-33 years old), reduction in or never or rarely engaging in physical activity (OR 1.201, 95% CI 1.055-1.368; reference: increase in physical activity; P=.006), increased appetite (OR 2.043, 95% CI 1.788-2.034; reference: reduction or no change in appetite; P<.001). From 2019 to 2022, 51.29% (11,917/23,234) of the respondents experienced increased persistent stress or recurrent anxiety or depressed mood; 44.38% (10,311/23,234) stayed up late more often. Increased persistent stress or recurrent anxiety or depressed mood was associated with physical activity (OR 0.421, 95% CI 0.398-0.447; P<.001) and appetite (OR 1.601, 95% CI 1.483-1.728; P<.001). Conclusions: The pandemic was associated with overweight and obesity for HCWs due to changes in lifestyle behaviors, especially reduced physical activity and increased appetite related to increased persistent stress or recurrent anxiety or depressed mood caused by excessive workload. An integrated approach is needed to address overweight and obesity and lifestyle changes among HCWs by releasing negative psychoemotional conditions through workload reduction in future stressful events.

Background: Pakistan has the third highest under-five mortality rate globally and minimal progress in reducing it, obstructing its achievement of Sustainable Development Goals (SDGs) 3.2 targets. Despite a decline, Pakistan remains one of the top ten contributors to worldwide under-five mortality. Proper antenatal care (ANC) is vital for improving maternal and child health outcomes. This study aims to assess the association between ANC visits and under-five mortality. Moreover, this study also measures the association between the timing of ANC visits and under-five mortality. Methods: A difference-in-difference (DID) study design with propensity score matching (PSM) was employed to investigate the contributing impact of ANC on the estimation of under-five mortality rates. Statistics from two datasets, 2012 and 2018, of Pakistan demographic and health surveys (PDHS) have been utilized for analyses. This research sample consists of 15698 women aged 15–49 who attended ANC visits at varying times before childbirth. Results: This study has reported 730 cases of under-five mortality. 98% involved Women who had less than eight or more ANC visits, whereas 2% of under-five mortality cases occurred among those women who had eight or more ANC visits. Overall, our results showed that women who had 1–3 ANC visits reduced the likelihood of under-five mortality in Pakistan by 36% (CI=0.459–0.909, P-value 0.012), the women who had 4–7 ANC visits decreased the likelihood of under-five mortality by 45% (CI=0.364–0.843, P-value 0.006). Moreover, the females with eight or more ANC visits also reduced the likelihood of under-five mortality by 98% (CI=0.512–1.896, P-value 0.966). Conclusions: Policies and programs focusing more on ANC visits should be implemented to reduce under-five mortality rates significantly. By emphasizing timely and frequent ANC visits, these insights support the development of targeted interventions that can significantly improve child survival rates and support the achievement of SDG targets. Moreover, based on our DID analysis, the implementation of the free maternal and child healthcare (MCH) policy in Pakistan has led to a significant increase in ANC utilization and a consistent reduction in the under-five mortality rates.

Background: Digital health technologies (DHTs) are transforming global health care delivery, yet physician adoption remains highly variable and influenced by a complex interplay of individual, institutional, and technological factors. In China, despite national initiatives such as the “Healthy China 2030” strategy promoting DHT integration, understanding physicians' heterogeneous perceptions is essential for effective implementation. Objective: This study aimed to identify distinct latent profiles of Chinese physicians based on their perceptions of DHT benefits, barriers, and behavioral intention, and to examine the demographic and occupational factors associated with profile membership. Methods: A cross-sectional survey was conducted among 4851 physicians (female, n=2994, 60.69% ; mean age 38.4, SD 8.7 years; 51.58% (n=2502) with more than 10 y working experience) from 46 hospitals in Shaanxi Province, China, between October and December 2023. Assessment included nine indicators across three domains: Perceived Benefits (4 items), Adoption Barriers (4 items), and Behavioral Intention (1 item). Latent profile analysis was used to identify distinct subgroups of physicians based on their response patterns. Multinomial logistic regression examined predictors of profile membership, and results were reported as odds ratios (ORs) with 95% CIs. Results: The latent profile analysis identified 5 distinct profiles: Reform-Adaptable (n=516, 10.64%; 95% CI 9.76%-11.52%), Negative (n=1003, 20.68%; 95% CI 19.50%-21.86%), Neutral (n=2276, 46.92%; 95% CI 45.50%-48.34%), Reform-Conservative (n=545, 11.23%; 95% CI 10.33%-12.13%), and Positive (n=511, 10.53%; 95% CI 9.66%-11.40%). Significant intergroup differences were observed in demographic and occupational characteristics. For instance, compared with the Negative profile, male physicians were less likely to belong to the Neutral (OR 0.76, 95% CI 0.64‐0.90; P =.001) and Reform-Conservative (OR 0.67, 95% CI 0.54‐0.84; P =.001) profiles. Compared to the Neutral profile, physicians with a master’s degree or above were less likely to be in the Reform-Conservative profile (OR 0.75, 95% CI 0.59‐0.96; P =.052). Those working in tertiary hospitals were less likely to belong to the Positive group (OR range 0.56‐0.66, P =.001). High-income physicians were more likely to be in the Reform-Conservative group (OR range 1.83‐2.38, P =.001). In addition, higher occupational stress was associated with a greater likelihood of Positive profile membership (OR range 1.12‐1.26, P =.001), while better work satisfaction predicted higher odds of Positive profile membership (OR range 1.04‐1.16, P =.001). Conclusions: This study introduces a novel, person-centered approach by identifying five distinct perceptual typologies among physicians, moving beyond traditional variable-centered analyses. This typology provides an evidence-based foundation for tailored interventions. For instance, the Reform-Adaptable group may need barrier reduction, while the Reform-Conservative group may require clearer value demonstrations. This nuanced understanding can help healthcare systems enhance the impact and scalability of digital health technologies in real-world clinical practice.

Background: Overuse stands as an important factor in the unreasonable growth of healthcare expenditure in many countries, increasing the burden of both patients and national healthcare insurance. The heterogeneity of diseases results in significant differences in clinical manifestations and treatment sensitivity among different patients with the same illness, making it urgently necessary to explore and resolve the objective assessment and measurement of overuse. This study aims to investigate the current status of overuse in primary healthcare services in China, analyze and discuss the effect of overuse on medical expenses, and draw attention to the issue of overuse.Method: The study adopted unannounced standardized patients to conduct 242 effective visits to 156 primary medical institutions in western Chinese province. The number of unnecessary medical service items reported after the visits is used to determine whether overuse exists. We set up a multiple linear regression model and used statistical methods to analyze the effect of different types of overuse on various medical expenses incurred by patients.Results: (1) Among the 242 visits, overtreatment occurred 139 times, accounting for 57.44% of the total visits. Among them, over-examination accounted for 37.41%, and over-medication accounted for 62.58%. (2) Regarding medical expenditure, the average total medical cost per capita for the 242 valid visits was 32.73 yuan, with examination fees amounting to 22.96 yuan and drug costs totaling 13.60 yuan. (3) Compared to the non-overuse group, patients in the overuse group incurred a total cost of 106.57 yuan, which was 54.33 yuan higher. The examination fees for the overuse group were 69.74 yuan, representing an additional expenditure of 18.83 yuan compared to the non-overuse patients. Drug costs for the overuse group were 39.27 yuan, an increase of 35.74 yuan compared to the non-overuse patients. Thus, overuse led to an increase of 104, 27, and 91% in total medical costs, examination fees, and drug costs, respectively. (4) For every additional unnecessary medical service item, the total medical cost will increase by 18.66 yuan, the examination cost will increase by 3.65 yuan, and the drug cost will increase by 14.48 yuan, with an increase rate of 13, 17, and 20%, respectively. For every additional unnecessary medical examination, the examination cost will increase by 11.28 yuan, with an increase rate of 13%, while there is no significant impact on total medical costs and drug costs. For every additional unnecessary drug item, the total medical cost will increase by 19.71 yuan, and the drug cost will increase by 9.81 yuan, with an increase rate of 29 and 27%, respectively, while there is no significant impact on examination costs.Conclusion: The study found that overuse is highly prevalent in primary healthcare institutions in the western provinces of China, including two types: over-examination and over-medication, with the problem of excessive medication being more prominent. Overuse has led to an increase in medical expenses, which is specifically reflected in the rise of total medical costs and drug costs. The growth in the number of unnecessary medical items is the main cause of the unreasonable rise in medical expenses.