Forum Replies Created
-
Posts
-
-
1. In my experience, the biggest challenge to achieving sustainability in health information systems is the lack of user-centered design; some applications are not tailored to user needs, making them difficult to use and leading to reluctance or disinterest in their adoption.
2. EHIS should designed to adapt to changing needs and technologies by incorporating modular architectures, customizable interfaces, and flexible data input methods to address the biggest challenge of user-centered design. Ensuring ease of use and relevance to different user groups fosters better adoption and sustainability by making the systems more intuitive and aligned with user needs.
-
1. Perceived ease of use and usefulness of a personal health record system may vary among different demographics, with younger individuals generally finding it easier to use, men potentially perceiving it as more useful for technical tasks, and higher education levels correlating with a greater appreciation of its usefulness and ease of adoption.
2. In my experience with e-health applications, other factors such as technology self-efficacy, user interface design, user experience, and the perceived usefulness of the application are crucial for assessing the intention to use the system. Additionally, the quality of system design, the reliability and accuracy of the data provided, the level of technical support available, privacy and security concerns, and the integration with other health information systems are significant variables that should be considered.
-
1. Implementing AI in epidemic surveillance can be enhanced by integrating advanced data analytics, such as natural language processing and machine learning, to process vast multi-dimensional open-source data, including social media, news reports, and satellite data, for early detection of epidemic signals. Additionally, employing AI to filter, sort, and curate unstructured data will reduce noise and provide more accurate and actionable intelligence for public health officials.
2. Utilizing AI for public preparedness offers significant benefits, including early detection of epidemic signals, improved data accuracy through advanced filtering and curation, and predictive modeling for effective intervention strategies. AI can enhance real-time monitoring and response, potentially preventing the spread of diseases. However, challenges include ensuring data privacy, overcoming resistance to technology adoption among public health authorities, addressing ethical concerns, and securing sufficient funding and resources for implementation and maintenance. Additionally, the accuracy of AI models depends on the quality of the data they are trained on, which may vary across regions.
-
1. To identify barriers and unmet needs in health information seeking among youth for HIV/STI and reproductive health beyond those discussed in the paper, additional factors to consider include the role of digital literacy and access to online health resources, cultural and societal norms affecting openness to discussing sexual health, peer influence and support networks, availability and accessibility of youth-friendly healthcare services, confidentiality concerns, and the impact of socioeconomic status on healthcare access. Understanding these factors can provide a more comprehensive view of the challenges faced by youth in seeking necessary health information and services.
2. In many communities, vulnerable groups such as homeless individuals, undocumented immigrants, LGBTQ+ youth, and those living in remote areas often miss out on receiving necessary health information due to lack of access, fear of stigma, language barriers, and limited digital connectivity. To reach these individuals, we can deploy mobile health clinics, partner with local organizations for community outreach, utilize digital platforms for accessible information, and create safe spaces for seeking care. Measuring the impact can be done through pre- and post-intervention surveys, tracking health service utilization, conducting focus groups, and monitoring health outcomes like HIV/STI testing rates to ensure the effectiveness of these efforts.
-
1. The model can select the 10 most influential factors, making it concise and effective. Surgeons can use a user-friendly interface to input patient data and receive alerts or recommendations. Additionally, integrating the model into a mobile app ensures accessibility and convenience. Training sessions for clinical staff and updated clinical guidelines will ensure proper usage. The model helps stratify patients into risk categories, aiding in resource allocation, patient counseling, and tailored surgical approaches. Continuous monitoring and updates will refine the model, enhancing its accuracy and effectiveness in preoperative planning and decision-making.
2. The potential benefits of using this model in a real-world clinical setting include improved preoperative planning, enhanced risk stratification, and better resource allocation, reducing the incidence of massive intraoperative blood loss (IBL) and better patient outcomes. Additionally, integrating the model into a mobile app can provide surgeons with convenient, real-time access. However, limitations include the need for extensive validation with larger, diverse datasets, the risk of over-reliance on the model, and the necessity for continuous updates and monitoring to maintain accuracy and relevance.
-
1. In Thailand, Subdistrict Health Promotion Hospitals in rural areas use the JHIS program as their Hospital Information System to record NCD patient data and prescribe drugs electronically. The JHIS program is easy to use and allows for efficient data transfer to a central server. However, it lacks interoperability with other hospital systems, which can hinder comprehensive patient care coordination across different healthcare facilities.
2. In Thailand, Village Health Volunteers (VHVs) play a crucial role in managing NCD patients at home, performing tasks such as screening blood pressure or blood glucose, and delivering medications. While they help mitigate the shortage of healthcare workers, VHVs face a significant workload, including activities like mosquito larva elimination and redundant data entry into multiple mobile apps mandated by the Ministry of Public Health. Therefore, while nurse or community health worker-facilitated tools are beneficial, streamlining VHV tasks and enhancing digital tool integration could further improve NCD management and care quality in remote areas.
-
1. To improve the safety of medical AI systems, an approach from development to implementation is essential.
Input Phase: A Data Controller should audit the data used for training and testing AI to ensure that it is legally acquired and ethically sourced, avoiding any infringement of data privacy laws. This includes verifying the quality and integrity of data, ensuring it is representative and free from biases that could affect the AI’s performance.
Process Phase: AI researchers should adopt transparent model development practices, ensuring their methods are original and do not infringe on the intellectual property of others. Open documentation throughout the development process is crucial to maintaining scientific integrity and reproducibility.
Implementation Phase: Once deployed, the AI system’s performance must be evaluated continuously, especially among minor populations and underdeveloped regions where training data may be scarce. This helps in ensuring the AI does not favor resource-rich populations, thereby maintaining equity in healthcare delivery. Additionally, the AI system should be designed to assist rather than replace human judgment to prevent misdiagnosis in critical cases. Ethical training for doctors on using AI as a supportive tool rather than a definitive diagnostic source is imperative to ensure they understand the AI’s limitations and maintain responsibility for patient care.
2. Transparency: Clear and understandable explanations of how the AI system works, including how it uses data in training and important features, are essential.
– Regulation: Compliance with all relevant healthcare regulations and standards is necessary to ensure the AI system is legally and ethically sound. -
1. Body temperature as an additional predictor could enhance the accuracy of predicting PIH. Body temperature is a critical vital sign that reflects the patient’s overall physiological state and can influence hemodynamic stability. Perioperative hypothermia or hyperthermia can impact vascular tone and blood pressure regulation, potentially contributing to the risk of hypotension.
2. When developing predictive models for clinical purposes, it’s essential to balance explainability and predictive accuracy. Traditional statistical models, such as logistic regression, Poisson regression, and Cox Hazard regression, offer valuable interpretability by providing odds ratios (OR), relative risks (RR), and hazard ratios (HR), which quantify the impact of predictors on outcomes in a straightforward manner. These metrics are crucial for clinicians to understand and communicate the rationale behind their decisions to patients and colleagues. While machine learning models often achieve higher predictive accuracy, they typically function as “black boxes,” offering less transparency about the underlying factors influencing predictions. In my opinion, clinicians would prefer models that provide clear explanations, as this transparency is vital for informed clinical decision-making and effective patient communication. Therefore, future research should explore ways to enhance the interpretability of machine learning models, perhaps through hybrid approaches that combine the strengths of both traditional and machine learning methodologies.
-
1. Oncologists and medical doctors, as well as nurses and health professionals, can leverage ML-based symptom predictors to enhance patient care. These tools should incorporate the duration of symptoms to improve accuracy, and the prediction outcomes should indicate the probability of specific organ-related cancers. However, it is essential to communicate to patients that cancer is a rare disease, and any predictive results should be followed by thorough evaluations with a healthcare provider to avoid unnecessary panic and ensure accurate diagnosis and treatment.
2. When developing machine learning models to predict cancer symptoms, researchers must consider that cancer is a rare disease, which can lead to class imbalance issues in the dataset. To address this, the outcome of the ML models should include predictions for more common diseases in addition to cancer. This approach helps balance the dataset and ensures that the model can provide meaningful predictions across a range of potential diagnoses, thereby improving the overall accuracy and utility of the symptom predictor.
-
Wow, Thitikan, your case report is detailed and clearly presents the questions under consideration. You did an excellent job on thoroughness and clarity. Here are my opinions on how to improve your work.
Screening Form
Date Format: Ensure the date format is consistent and clear. To avoid confusion, consider using “DDMMMYYYY” with the month in text format (e.g., 15JAN2024).
– Checkbox in question 7: There is a missing checkbox.
The gender of the participant may need to be logically consistent with a Pregnancy test and age to check if they met the inclusion criteria.
– Urine Pregnancy Test: This test should only be indicated for women. Ensure this is specified in the form.
– Screening Outcome Placement: It might be more logical to place the screening outcome at the end of the form.Enrollment Form
– Visit Number: Since the enrollment form is used only once, there might not need to include a visit number.
– Redundancy: The sections for SCREENING OUTCOME and ELIGIBILITY CRITERIA are redundant, as these are already addressed in the screening form. Consider removing these to streamline the form.Your organized form and comprehensive detail inspired me. Keep up the good work 😀
-
-
I think it can facilitate comparability and meta-analysis. By standardizing data collection, coding, and storage practices across different studies and settings, researchers can more easily compare results and aggregate data. This comparability is crucial for conducting meta-analyses, where data from multiple studies are combined to draw broader conclusions. Data standards ensure that the data is consistent and reliable, enhancing the quality and robustness of research findings across various clinical studies.
-
I conducted a stroke study querying data from a Health Data Center, which is implemented:
– Authentication and Access Control: Access to the data is secured through authenticated accounts. Each access and data query is logged with a timestamp, ensuring a comprehensive audit trail.
– Data Encryption: To protect sensitive information, such as Thai identification numbers, data is encrypted using SHA-256, a strong cryptographic hash function. This helps prevent unauthorized access to personal data.
– Data Backup and Recovery: Daily backups are performed on physical servers to ensure data can be recovered during hardware failure or other data loss scenarios. This step is crucial for maintaining data availability and integrity over time.
– Data Query Software: Apache Hive is utilized to query data. This software supports data summarization, querying, and analysis, making it suitable for handling large datasets. -
Reflecting on my experience in data collection and management, I believe it would be better if I could improve on a data management plan and implement data quality control measures. Specifically, I would have emphasized developing a comprehensive Case Report Form (CRF) to ensure standardized data collection across all sources. Additionally, implementing more robust Data Quality Control measures from the outset would have helped identify and address issues earlier in the process, ultimately improving the overall quality and reliability of the data for my project.
-
I have experience with data collection for my HT-controlled and stroke outcome research:
– Purpose: I collected data for research purposes, specifically focusing on hypertension control and stroke outcomes.
Primary or secondary data: I gathered both primary and secondary data. Primary data was collected from patients, while secondary data was obtained from the Health Data Center.
– Methods used: For primary data, I conducted a qualitative in-depth questionnaire with patients. For secondary data, I used SQL queries to extract relevant information from the Health Data Center’s database.
Challenges faced: One major challenge was the extensive data cleaning required for the text-format data obtained from the Health Data Center, which took approximately three months of effort. Additionally, categorizing the data based on defined criteria and automating the process through coding posed further challenges. -
-
-
-
For my colleagues, I guess the problem is your image doesn’t show up because you should type
by yourself. If you copy, the “” may change to another symbol.
-
-
-
I would discussion point 1 in page 340. I used to think the P value showed the chance that the null hypothesis (a default assumption that there’s no effect or difference) might be true just by random luck. For example, I believed a P value of 0.05 meant the null hypothesis could randomly happen 5% of the time. But, I got it wrong. What the P value really does is; if we assume the null hypothesis is right, how likely it is to see the data we collected . So, when we get a P value of 0.05, it doesn’t tell us about the 5% chance of the null hypothesis itself being true. Instead, it means there’s a 5% chance of finding data like ours if the null hypothesis were true. It’s basically saying, ‘Given the null hypothesis is true, our data seem pretty unlikely. It’s about the data’s fit with the null hypothesis, not the hypothesis’s own likelihood.
-
4. Maternal Mortality Rate
Definition: The rate of death from any cause related to or aggravated by pregnancy or its management, excluding accidental or incidental causes, during pregnancy and childbirth or within a specified period of postpartum per XXX live births.Calculation:
Maternal Mortality Rate
=(Number of maternal deaths / Number of live births) × XXXUsefulness: Highlights the safety and quality of maternity care, which is vital for improving maternal health services.
5. Infant Mortality Rate
Definition: The number of deaths of infants under one year old per XXX live births in the same year.Calculation:
Infant Mortality Rate
=(Number of infant deaths / Number of live births) × XXXUsefulness: Indicates the overall health of a society, reflecting the societal conditions affecting children’s survival.
6. Neonatal Mortality Rate
Definition: The number of deaths of infants aged 0-28 days per XXX live births in a given year.Calculation:
Neonatal Mortality Rate
=(
Number of neonatal deaths / Number of live births) × XXXUsefulness: Important for assessing the quality of antenatal and perinatal care and for planning interventions to reduce neonatal deaths.
-
Hello everyone! My name’s Mumi. I’m a field epidemiologist and doctor working at Chonburi Hospital in Thailand. Sometimes, when outbreaks or public health issues occur, I collect data, describe to see insights, and use analytics to compare and find risk factors. I have learned some statistics from FETP Thailand and do some analysis to finish the job before my teacher gets angry 😀
-
As both an epidemiologist and a clinical doctor, adhering to ethical principles and good practices is paramount in contributing to COVID-19 control policies. Key among these are:
– Equity and Fairness: Ensuring equitable access to testing, treatment, and vaccines, regardless of socioeconomic status, is crucial. This involves advocating for resource allocation that prioritizes vulnerable populations.
– Transparency: Sharing accurate, timely information about the virus, its spread, and prevention measures with the public and within the healthcare community fosters trust and compliance with control measures.
– Privacy and Confidentiality: While conducting contact tracing and reporting cases, it’s essential to protect individuals’ privacy and confidentiality to maintain public trust and encourage cooperation.
– Beneficence and Non-maleficence: In clinical and public health decisions, balancing doing good and avoiding harm is vital. This includes making evidence-based decisions on lockdowns, quarantine measures, and treatment protocols to maximize benefits and minimize harm.
– Solidarity and Cooperation: Collaborating with local and international bodies, sharing knowledge and resources, and working together towards a common goal are essential for an effective response to the pandemic.
-
In Thailand, the UHC scheme has made significant strides in ensuring healthcare availability, capacity, funding, and management. However, a notable weakness, from my experience, lies in the imbalance between primary prevention and tertiary care. The Department of Disease Control (DDC) allocates a disproportionately low budget—about 10-20%—to primary prevention compared to what is spent on tertiary prevention, such as hospital treatments. This trend is not unique to Thailand but is observed globally, where the emphasis often shifts to the cost of health only after falling ill, despite primary prevention being more cost-effective. Addressing this imbalance is crucial for a more sustainable and effective healthcare system.
-
I strongly believe in the importance of data sharing, especially regarding anonymized or encrypted datasets. Sharing such data can immensely benefit public health research and policy-making, allowing for a broader analysis and understanding of health trends and outcomes. This approach fosters collaboration and innovation in the field and ensures that individual privacy is maintained. Responsibly sharing data can optimize resources, enhance research capabilities, and contribute to more effective disease prevention and control strategies. It’s a balance between advancing public health knowledge and safeguarding personal information, a principle that aligns closely with my professional ethos and the ethical standards of my field.
-
In my hospital setting, where a 20-year-old, outsourced EMR system is in use, there are advantages and challenges to consider. The EMR system enhances efficiency and accessibility, allowing quick access to patient records and facilitating continuity of care. It’s beneficial for data analysis and research, enabling the aggregation of large datasets. Additionally, EMRs significantly reduce the need for physical storage of paper records. However, the challenges with my current system are notable. Its user-unfriendly interface can impede efficient data access and entry, impacting overall productivity. The need for financial resources for any updates or customization due to the system being outsourced adds a financial burden. Also, the potential for outdated technology raises concerns about integrating new technologies and data security. Lastly, the limited customization for local needs may result in inefficiencies, suggesting a need for continual updates and a user-centered design approach to maximize the benefits of EMRs in my hospital. 🏥💡🖥️
-
Missing Data: Understanding the pattern of missingness is crucial. Linear interpolation can be an effective imputation method for continuous data like weight, especially if the data points are time series.
Selection Bias: Using the entire dataset is ideal if the dataset is well-collected and representative. In cases where representativeness is a concern, proportional sampling helps ensure each subgroup is adequately represented, reducing the risk of bias.
Data Analysis and Training: Class imbalance is a significant challenge. Techniques like SMOTE (Synthetic Minority Over-sampling Technique) or adjusting class weights in model training can help. Ensuring balanced representation in your training data is key to building robust models.
Interpretation and Translational Applicability of Results: Collaborating with frontline stakeholders who understand the local context is vital for meaningful interpretation. For translational applicability, especially in app development, employing user-centered design principles ensures the end product meets its users’ actual needs and preferences.
Privacy and Ethical Issues: Hashing identifiers like names and phone numbers is a great way to protect individual privacy. Ensure that the hashing method is robust and irreversible. Always align with data privacy laws and ethical guidelines in handling sensitive information.
-
I agree with the recommendation to define “Corruption” as it encompasses a broad spectrum of unethical practices, ranging from minor instances like accepting gifts at academic meetings to major systemic issues like government fiscal budget corruption. Identifying specific types of corruption to target is crucial. Taking a holistic view to understand the patterns and structures of these corrupt practices is essential.
In terms of establishing research, I believe qualitative research focusing on firsthand experiences would be highly beneficial. This approach should ensure confidentiality to encourage openness and honesty among participants.
Lastly, prioritizing actions based on the research findings and the identified patterns of corruption can lead to more effective strategies in combating these unethical practices.
-
Right now, I’m working on my research, and the lessons learned from the course will help me.
Define Project Goals: Complete data cleaning and analysis for hypertension and stroke research; write and finalize the manuscript.
Objectives: Ensure data quality, comprehensively analyze and write a clear, informative manuscript.
Success Criteria: High-quality research data, insightful analysis results, and a well-written manuscript ready for submission.List All Tasks
Break down the project into specific tasks: data cleaning, data analysis, and manuscript writing (introduction, methods, results, discussion).Create a Gantt Chart
Visualize the timeline for each task, including data analysis phases and manuscript drafting, with clear deadlines.Add a Safety Margin
Include buffer time for unexpected data analysis or writing delays, especially considering potential technical challenges or revisions.Monitor Progress
Regularly review data analysis and manuscript writing progress, adjusting the plan as needed.Readjust Plan
Be flexible to modify the plan in response to analytical findings, peer feedback, or personal time constraints.Project Review
After manuscript submission, reflect on the project’s successes and areas for improvement, applying these insights to future research projects. -
-
To ensure robust disaster recovery in my hospital, begin with a thorough Risk Assessment and Business Impact Analysis. This involves identifying and prioritizing critical systems and data and assessing potential risks such as natural disasters and cyber-attacks. For data backup solutions, implement incremental backups to regularly capture only the data that has changed since the last backup. This approach is both cost-effective and efficient for large datasets. Additionally, utilize cloud services like AWS S3 or Google Cloud Storage for off-site data backups, leveraging their scalability and cost-efficiency. It’s also crucial to conduct regular testing and drills of the disaster recovery procedures to ensure their effectiveness, making necessary updates based on these test outcomes. Ongoing training for staff on disaster recovery protocols and the importance of data backups is essential. Finally, continuously review and update the disaster recovery plan to accommodate technological changes, such as Windows patch updates, organizational shifts, or emerging threats, ensuring the plan remains current and effective.
-
I want to improve my concentration in listening, a skill critical in my work as a field epidemiologist and medical doctor, and I plan to adopt several strategies. These include practicing mindful listening to stay present, minimizing distractions by turning off notifications and choosing quiet environments, actively participating in discussions through questions and summaries, and taking notes to maintain focus. Additionally, I will take regular breaks during long sessions to refresh my attention and seek feedback from colleagues to identify missed points. These steps will help enhance my listening skills, ensuring I fully grasp complex information crucial in healthcare settings.
-
Implementing High Availability technology in an HIS primarily enhances patient care and hospital operations through reduced downtime, improved data integrity and reliability, and consequently, better patient care. Reduced downtime ensures that critical patient information and hospital services are always accessible, eliminating medical records and other vital data delays. This continuous availability is crucial for providing timely and effective healthcare. Enhanced data integrity and reliability are another significant benefit, as the system ensures patient data is consistently accurate and secure, which is vital for correct diagnosis and treatment plans. This reliability in data management directly contributes to improved patient care, as healthcare providers can make informed decisions based on complete and up-to-date patient information, leading to better health outcomes and patient satisfaction.
-
My strongest EQ component is my social skills. I find it quite natural to communicate with others, empathize with their situations, and build trust. I enjoy motivating people around me, which fosters a collaborative and supportive environment. This strength helps me in personal interactions and significantly in my professional life, where teamwork and networking are essential.
On the flip side, I recognize self-regulation as a weakness of EQ. I sometimes struggle with staying engaged in long-term projects, and I can get distracted by playing video games, which might offer immediate fun but ultimately disrupt my productivity and focus.
To improve my self-regulation, my action plan includes:
Setting Clear Goals: Define specific, measurable goals for my projects to give my work a clear direction and purpose.
Structured Breaks: Incorporate scheduled breaks for gaming as a reward for achieving certain milestones in my projects, using it as a positive reinforcement rather than a distraction.
Mindfulness Practices: Use mindfulness exercises to enhance focus and reduce impulsiveness, such as short meditation breathing exercises on the TIDE app.
Accountability: Share my project timelines with a colleague or friend, adding an external layer of accountability.
Reflective Journaling: Keep a journal to reflect on my daily activities, note when I succumb to distractions, and identify patterns or triggers that I can address. -
What happened?
Last week, a hacker used my credit card details to make unauthorized purchases on a website. I became aware of this through my banking app, which notified me of repetitive suspicious online transactions.Impact on the System/Users:
This breach had a direct financial impact, leading to my bank canceling my credit card and issuing a new one. It also impacted my trust in and the reliability of my digital systems.Preventive Measures:
To prevent such incidents in the future, I have taken several steps:Enhanced Security Measures: I have updated all passwords for my online accounts, especially those where my credit card information is stored.
Regular Monitoring: I now closely monitor my financial statements to quickly detect any unauthorized activity.
Using Secure Networks: I have become more vigilant about using secure networks for transactions and avoid using public Wi-Fi for financial operations.
Two-Factor Authentication: I have enabled two-factor authentication on all accounts that offer this feature, adding an extra layer of security. -
-
-
Thanks, Nichcha. Audition is crucial to maintaining a good system. Unfortunately, it rarely happens in many surveillance systems. But once it happen, it could benefit by many.
Your study appears to be a retrospective cohort study, if I’m not mistaken. It can also provide valuable recommendations for improving the report or data collection systems. It’s not necessary to only conduct surveillance evaluations to improve systems. Any methodological study can contribute to system improvements.
-
-
-
-
-
Thanks, Dr.Teeraboon, for your insight and feedback. It’s good to systematically pinpoint the critical points of the system, which is essential for improving the surveillance system. Also, military surveillance evaluation would be very interesting as it would have a more reliable report system than civilian surveillance.
-
-
-
I understand the complex challenges you’re facing in Myanmar regarding health informatics. Despite these difficulties, I believe there are simple, pioneering steps you can take to make a positive impact. One such step could be adopting free, open-source programs to transition from paper-based records to EHR in areas within your influence. This approach can be a quick win, offering a practical start towards broader transformation. Remember, as a pioneer in this field, there’s nothing wrong with trying new methods. Your efforts could set a precedent for others to follow. I encourage you to embrace this opportunity – every small step can lead to significant changes. All the power to you in this!
-
It shocked me that a developed place like Hong Kong still faces a workforce shortage issue. I think we encounter the same problem in public health: limited resources and unlimited demand. One key selling point of Health Informatics is its ability to improve healthcare while reducing service costs. I hope that policymakers will recognize this core value and hire more health informaticists for organizations in need.
-
Your observation about the potential shift in service care priorities due to leadership changes is indeed a significant concern. The example you provided, one of the Health Promoting Hospitals in my area discontinued peripheral nerve screening for diabetic patients because it was no longer a Key Performance Indicator (KPI) under the new executive, highlights a critical issue.
When administrative changes occur, especially with non-medical executives taking the lead, there’s a risk that the primary goals of healthcare services might shift away from patient-centric outcomes to more administrative or financially driven targets. This can lead to neglecting essential health services that are crucial for patient care but may not align with the new leadership’s KPIs.
-
