Evaluating the Efficacy of ChatGPT in Navigating the Spanish Medical Residency Entrance Examination (MIR): Promising Horizons for AI in Clinical Medicine

The rapid progress in artificial intelligence, machine learning, and natural language processing has led to increasingly sophisticated large language models (LLMs) for use in healthcare. This study assesses the performance of two LLMs, the GPT-3.5 and GPT-4 models, in passing the MIR medical examination for access to medical specialist training in Spain. Our objectives included gauging the model's overall performance, analyzing discrepancies across different medical specialties, discerning between theoretical and practical questions, estimating error proportions, and assessing the hypothetical severity of errors committed by a physician.

MATERIAL AND METHODS

We studied the 2022 Spanish MIR examination results after excluding those questions requiring image evaluations or having acknowledged errors. The remaining 182 questions were presented to the LLM GPT-4 and GPT-3.5 in Spanish and English. Logistic regression models analyzed the relationships between question length, sequence, and performance. We also analyzed the 23 questions with images, using GPT-4's new image analysis capability.

RESULTS

GPT-4 outperformed GPT-3.5, scoring 86.81% in Spanish (p < 0.001). English translations had a slightly enhanced performance. GPT-4 scored 26.1% of the questions with images in English. The results were worse when the questions were in Spanish, 13.0%, although the differences were not statistically significant (p = 0.250). Among medical specialties, GPT-4 achieved a 100% correct response rate in several areas, and the Pharmacology, Critical Care, and Infectious Diseases specialties showed lower performance. The error analysis revealed that while a 13.2% error rate existed, the gravest categories, such as "error requiring intervention to sustain life" and "error resulting in death", had a 0% rate.

CONCLUSIONS

GPT-4 performs robustly on the Spanish MIR examination, with varying capabilities to discriminate knowledge across specialties. While the model's high success rate is commendable, understanding the error severity is critical, especially when considering AI's potential role in real-world medical practice and its implications for patient safety.

Trends, Projections, and Regional Disparities of Maternal Mortality in Africa (1990-2030): An ARIMA Forecasting Approach

With the United Nations Sustainable Development Goals (SDG) (2015-2030) focused on the reduction in maternal mortality, monitoring and forecasting maternal mortality rates (MMRs) in regions like Africa is crucial for health strategy planning by policymakers, international organizations, and NGOs. We collected maternal mortality rates per 100,000 births from the World Bank database between 1990 and 2015. Joinpoint regression was applied to assess trends, and the autoregressive integrated moving average (ARIMA) model was used on 1990-2015 data to forecast the MMRs for the next 15 years. We also used the Holt method and the machine-learning Prophet Forecasting Model. The study found a decline in MMRs in Africa with an average annual percentage change (APC) of -2.6% (95% CI -2.7; -2.5). North Africa reported the lowest MMR, while East Africa experienced the sharpest decline. The region-specific ARIMA models predict that the maternal mortality rate (MMR) in 2030 will vary across regions, ranging from 161 deaths per 100,000 births in North Africa to 302 deaths per 100,000 births in Central Africa, averaging 182 per 100,000 births for the continent. Despite the observed decreasing trend in maternal mortality rate (MMR), the MMR in Africa remains relatively high. The results indicate that MMR in Africa will continue to decrease by 2030. However, no region of Africa will likely reach the SDG target.

COVID-19 Impact on DTP Vaccination Trends in Africa: A Joinpoint Regression Analysis

BACKGROUND

Deaths due to vaccine-preventable diseases are one of the leading causes of death among African children. Vaccine coverage is an essential measure to decrease infant mortality. The COVID-19 pandemic has affected the healthcare system and may have disrupted vaccine coverage.

METHODS

DTP third doses (DTP3) Vaccine Coverage was extracted from UNICEF databases from 2012 to 2021 (the last available date). Joinpoint regression was performed to detect the point where the trend changed. The annual percentage change (APC) with 95% confidence intervals (95% CI) was calculated for Africa and the regions. We compared DTP3 vaccination coverage in 2019-2021 in each country using the Chi-square test.

RESULT

During the whole period, the vaccine coverage in Africa increased with an Annual Percent change of 1.2% (IC 95% 0.9-1.5): We detected one joinpoint in 2019. In 2019-2021, there was a decrease in DTP3 coverage with an APC of -3.5 (95% -6.0; -0,9). (p < 0.001). Vaccination rates decreased in many regions of Sub-Saharan Africa, especially in Eastern and Southern Africa. There were 26 countries (Angola, Cabo Verde, Comoros, Congo, Côte d'Ivoire, Democratic Republic of the Congo, Djibouti, Ethiopia, Eswatini, The Gambia, Guinea-Bissau, Liberia, Madagascar, Malawi, Mauritania, Mauritius, Mozambique, Rwanda, Senegal, Seychelles, Sierra Leone, Sudan, Tanzania, Togo, Tunisia, Uganda, and Zimbabwe) where the vaccine coverage during the two years decreased. There were 10 countries (Angola, Cabo Verde, Comoros, Democratic Republic of the Congo, Eswatini, The Gambia, Mozambique, Rwanda, Senegal, and Sudan) where the joinpoint regression detected a change in the trend.

CONCLUSIONS

COVID-19 has disrupted vaccine coverage, decreasing it all over Africa.

Updosing nonsedating antihistamines in patients with chronic spontaneous urticaria: a systematic review and meta-analysis

There is a lack of large, randomized, double-blind studies that address antihistamine updosing for chronic spontaneous urticaria (CSU). The objective of this systematic review is to explore and analyse available data to provide clinical evidence for the efficacy of antihistamine updosing. We searched the literature in Medline, Scopus, Google Scholar, Embase, Web of Science and Cochrane databases using the keywords 'chronic, urticaria, antihistamines' to identify studies published between January 1990 and November 2014. We assessed quality using the Jadad score that evaluates quality of randomization, double-blinding and losses to follow-up. We identified 1042 articles and 15 articles were included in the final evaluation. We performed two meta-analyses, one that included studies that analysed treatment response among groups receiving different antihistamine dosages vs. placebo, and another that analysed antihistamine updosing in those patients who did not respond to standard dosages. Only five articles obtained a high quality level score. We did not find significant differences in response rates or number of weals in those patients who received a standard dosage vs. a high dosage. We found a significant improvement only in the pruritus variable of the Urticaria Activity Score scale. The estimated relative risk for improvement by increasing the antihistamine dosage was 2·27 [95% confidence interval (CI) 1·68-3·06]; however, there was significant heterogeneity. The proportion of nonrespondent patients with CSU who responded to antihistamine updosing was 63·2% (95% CI 57-69·6). We found that updosing antihistamines significantly improved control of pruritus but not weal number. However, the relative weakness of the studies and the significant heterogeneity among them made it difficult to reach a final conclusion.

A low dose of subperiosteal anaesthesia injection versus a high dose of infiltration anaesthesia to minimise the risk of nerve damage at implant placement: A randomised controlled trial

PURPOSE

To evaluate whether a low-dose subperiosteal anaesthesia is effective in minimising risks of inferior alveolar nerve damage at implant placement when compared to high-dose infiltration anaesthesia.

MATERIAL AND METHODS

One hundred and twenty patients requiring the placement of a single implant in order to replace a missing first mandibular were randomly allocated to two groups: group A (awake hemilip) subperiosteal crestal injection equal to 0.9 ml of articaine with 0.5% epinephrine and group B (numb hemilip) infiltration equal to 7.2 ml of articaine with 0.5% epinephrine in the vestibular fundus. Intraoperative sensory control using sensory tests was carried out in all patients. Outcome measures were neurological complications, intraoperative and postoperative visual analogue scale (VAS) scores for pain and swelling, and a questionnaire evaluating patient satisfaction. Patients were followed for 1 week postoperatively.

RESULTS

There were no cases of nerve injury. Seven days after surgery the postoperative VAS score for pain and swelling was lower in group A in a statistically significant manner (difference = -3.41%; 95% CI: -5.57, -1.26; P = 0.002 and difference = -3.33%; 95% CI: -5.41, -1.25; P = 0.002, respectively).

CONCLUSIONS

No nerve damage occurred using either anaesthesia types, therefore the choice of type of anaesthesia is a subjective clinical decision, however it may be preferable to use a low dose (0.9 ml) of subperiosteal anaesthesia, since it is unnecessary to deliver 7.2 ml of articaine to anaesthetise a single mandibular molar implant site.

Bronchial asthma prevalence in childhood

BACKGROUND

at the moment there is a considerable interest in the international comparison of the bronchial asthma prevalence, stimulated by the growing evidence of an increment in the frequency. The comparison among studies of bronchial asthma prevalence present the problem of the varied and different methodologies used by the authors, as well as of the lack of a clear definition of what is considered bronchial asthma.

METHODS

in this work we try to offer a global vision of the studies published on childhood asthma prevalence in the world.

CONCLUSIONS

in the studies carried out at European level, they highlight the high figures found in the British Islands superiors to those referred in the countries from the continent to exception of those observed in Switzerland. In Spain the figures of the studies carried out at national level are something below the countries of our environment, although there have been higher figures in studies carried out in Barcelona and Huelva. In Oceannia they are observed from all over the world in general the highest figures, above the American countries and very above the countries of Asia and Africa to exception of South Africa. Although there are considerable geographical differences in its presentation, bronchial asthma is an illness in constant increase in the entire world. It is for this reason that efforts should be carried out in the search of the causes of this increment, as well as in the standardization of the instruments of measure of the appearance of the illness.