Online data collection strategies used in qualitative research of the health field: a scoping review

Enhancing diagnostic accuracy in symptom-based health checkers: a comprehensive machine learning approach with clinical vignettes and benchmarking

Introduction

The development of machine learning models for symptom-based health checkers is a rapidly evolving area with significant implications for healthcare. Accurate and efficient diagnostic tools can enhance patient outcomes and optimize healthcare resources. This study focuses on evaluating and optimizing machine learning models using a dataset of 10 diseases and 9,572 samples.

Methods

The dataset was divided into training and testing sets to facilitate model training and evaluation. The following models were selected and optimized: Decision Tree, Random Forest, Naive Bayes, Logistic Regression and K-Nearest Neighbors. Evaluation metrics included accuracy, F1 scores, and 10-fold cross-validation. ROC-AUC and precision-recall curves were also utilized to assess model performance, particularly in scenarios with imbalanced datasets. Clinical vignettes were employed to gauge the real-world applicability of the models.

Results

The performance of the models was evaluated using accuracy, F1 scores, and 10-fold cross-validation. The use of ROC-AUC curves revealed that model performance improved with increasing complexity. Precision-recall curves were particularly useful in evaluating model sensitivity in imbalanced dataset scenarios. Clinical vignettes demonstrated the robustness of the models in providing accurate diagnoses.

Discussion

The study underscores the importance of comprehensive model evaluation techniques. The use of clinical vignette testing and analysis of ROC-AUC and precision-recall curves are crucial in ensuring the reliability and sensitivity of symptom-based health checkers. These techniques provide a more nuanced understanding of model performance and highlight areas for further improvement.

Conclusion

This study highlights the significance of employing diverse evaluation metrics and methods to ensure the robustness and accuracy of machine learning models in symptom-based health checkers. The integration of clinical vignettes and the analysis of ROC-AUC and precision-recall curves are essential steps in developing reliable and sensitive diagnostic tools.

Cross-cultural adaptation and validation of the Healthy Work Environment Assessment Tool in Brazilian culture

OBJECTIVES

to adapt and validate the content of the Healthy Work Environment Assessment Tool for Brazilian culture, and evaluate the practical aspects of its application.

METHODS

methodological study that followed six stages: translation; synthesis; back translation; content validation by a group of experts, pre-testing and approval of the process by the author of the original instrument.

RESULTS

the first three stages were carried out by contracted companies. In the committee, two items and the title of a subscale were evaluated in a second round, when consensus was reached among experts. In the pre-test, more than 93% of professionals agreed that the tool was easy to understand. The average completion time was 8.53 minutes. The American Association of Critical-Care Nurses authorized publication of the results.

CONCLUSIONS

the adaptation of the tool to Brazilian culture was completed following the adopted framework. In addition to the evidence of content validity, the tool appears promising for managerial use.

Guia de Enfermagem Escolar para promoção da saúde de jovens estudantes: construção e validação

RESUMO Objetivos: descrever o processo de construção e validação do Guia de Enfermagem Escolar para promoção da saúde de jovens estudantes. Métodos: estudo metodológico, realizado de fevereiro a dezembro de 2021, composto por Pesquisa Convergente-Assistencial fundamentada no Modelo de Promoção da Saúde de Pender. A partir da literatura e diálogo com 11 enfermeiros nos sete grupos focais online, foram construídas ações. Posteriormente, 24 juízes avaliaram o conteúdo e a aparência. Resultados: o guia propõe estratégias para o desenvolvimento das práticas de enfermagem escolar enfocando a promoção da saúde. O Índice de Validade de Aparência variou de 0,63 a 1,0, e o total foi de 0,84. O Índice de Validade de Conteúdo variou de 0,95 a 1,0, e o total foi de 0,997. Conclusões: o guia incorporou as necessidades dos jovens reconhecidas pelos profissionais, e a fase de avaliação confirma sua validação, podendo ser utilizado no contexto da prática com jovens.

School Nursing Guide for student health promotion: construction and validity

OBJECTIVES

to describe the process of construction and validity of a School Nursing Guide for student health promotion.

METHODS

a methodological study, carried out from February to December 2021, composed of Convergent Care Research based on Pender's Health Promotion Model. Based on the literature and dialogue with 11 nurses in the seven online focus groups, actions were constructed. Subsequently, 24 judges assessed content and appearance.

RESULTS

the guide proposes strategies for developing school nursing practices focusing on health promotion. The Appearance Validity Index ranged from 0.63 to 1.0, and the total was 0.84. The Content Validity Index ranged from 0.95 to 1.0, and the total was 0.997.

CONCLUSIONS

the guide incorporated the needs of young people recognized by professionals, and the assessment phase confirms its validity, and can be used in the context of practice with young people.

Breast cancer screening program for risk groups: facts and perspectives

OBJECTIVES

to measure the frequency and compliance of breast cancer screening, according to the risk for this disease.

METHODS

a cross-sectional study with 950 female users of 38 public Primary Health Care services in São Paulo, between October and December 2013. According to UHS criteria, participants were grouped into high risk and standard risk, and frequency, association (p≤0.05), and screening compliance were measured.

RESULTS

6.7% had high risk and 93.3% standard risk, respectively; in these groups, the frequency and compliance of clinical breast examination were 40.3% and 37.1%, and 43.5% and 43.0% (frequency p=0.631, compliance p=0.290). Mammograms were 67.7% and 35.5% for participants at high risk, and 57.4% and 25.4% for those at standard risk (frequency p=0.090, compliance p=0.000).

CONCLUSIONS

in the groups, attendance and conformity of the clinical breast exam were similar; for mammography, it was higher in those at high risk, with assertiveness lower than the 70% set in UHS.

Multiprofessional family health residency as a setting for education and interprofessional practices

OBJECTIVE

to know the experiences lived during the residency by graduates of a Multiprofessional Residency Program in Family Health that could contribute to the development of Interprofessional Education and/or Collaborative Practice.

METHOD

a qualitative study with residents who entered a Multiprofessional Residency Program of a Brazilian public university in 2017, a period in which the theme of interprofessionality was implemented in the activities of the residency. Data was collected using an electronic form built from the theoretical framework of interprofessional education. Content analysis was used to process the data.

RESULTS

nine residents participated, distributed among the professions of Physical Education, Nursing, Nutrition, Dentistry, Psychology and Social Work, five of whom were female and with a mean age of 28.4. Two categories emerged: the Residency as a setting for learning from the other, and the Residency as a setting for understanding the role of the other. Interprofessional education and practice provided opportunities for the development of collaborative skills, enhancing teamwork and interprofessional work.

CONCLUSION

the multiprofessional logic was evidenced in the resident's practice; and the gradual insertion of activities such as case discussions, shared services and inter-sectoral actions aligned with the theoretical-methodological framework of interprofessionality favored an approach to interprofessional work.

A survey of patient blood management for patients undergoing cardiac surgery in nine European countries

STUDY OBJECTIVE

To describe and compare patient blood management (PBM) practices in cardiac surgery in nine European countries and identify the main risk factors for bleeding or transfusion according to the surveyed centres.

DESIGN

We set up an online survey to evaluate PBM practices in two clinical scenarios, risk factors for bleeding or transfusion, and previous experience with antifibrinolytics.

SETTING

This survey was completed by European anesthesiologists in 2019.

PATIENTS

No patients were included in the survey.

INTERVENTION

None.

MEASUREMENTS

We evaluated the degree of implementation of PBM practices in patients undergoing cardiac surgery.

MAIN RESULTS

Ninety-eight of 177 responses (38%) were complete with variable response rates by country. In a non-emergent situation, no respondents would transfuse red cells preoperatively in an anaemic patient, while cell salvage (89%) and antifibrinolytics (82%) would almost always be used. Optimization of Hemoglobin level (36%) and use of off-pump techniques (34%), minimally invasive surgery (25%) and relatively recently-developed CPB technologies such as mini-bypass (32%) and autologous priming (38%), varied greatly across countries. In an emergent clinical situation, topical haemostatic agents would frequently be used (61%). Tranexamic acid (72%) and aprotinin (20%) were the main antifibrinolytics used, with method of administration and dose varying markedly across countries. Five factors were considered to increase risk of bleeding or transfusion by at least 90% of respondents: pre-operative anaemia, prior cardiac surgery, clopidogrel 5 days or less before surgery, use of other P2Y12 inhibitors at any point, and thrombocytopenia <100.10⁹ platelets/mm³.

CONCLUSION

PBM guidelines are not universally implemented in European cardiac surgery centres or countries, resulting in discrepancies in techniques and products used for a given clinical situation.