Soe Htike

[Soe Htike]

I’m also stuck with an error message in Exceedence Probability.

> exc1 <- sapply(mod.eco.reg$marginals.fitted.values,FUN = function(marg){1 – inla.pmarginal(q = 1, marginal = marg)})
> summary(exc1)
Length Class Mode
0 list list
> exc1.cutoff <- c(0, 0.8, 0.9, 0.95, 1)
> cat.exc1<- cut(unlist(exc1),breaks=exc1.cutoff,include.lowest=TRUE)
Error in cut.default(unlist(exc1), breaks = exc1.cutoff, include.lowest = TRUE) :
‘x’ must be numeric

I can move on the the later steps, but because of this error, I cannot combine plots in the final step.

[Soe Htike]

Thanks for your discussion points, friends. I do agree with all of you about the enhanced predictive accuracy, timeliness of predictions and even the challenges in data alignment.
When I was preparing this article to present to you, I had a few concerns about the scalability and adaptability of the system if my country were to adapt and implement the same system in different environmental and epidemiological contexts.
It’s also related to the second question, which is that the role of stakeholders in implementing an innovative system will be one of the key challenges in a context that historically relied on paper-based reporting and has limited digital literacy of healthcare workers

[Soe Htike]

The author’s interest in investigating the suicide problem in Thailand during the specified time stems from the increasing recognition of suicide as a significant public health issue. Thailand, like many other countries, faces complex social and economic challenges that may influence mental health and suicide rates. By examining the determinants of suicide rates across different provinces, the author aimed to identify specific risk factors and their spatial distribution, thereby providing a deeper understanding of the underlying causes. This research is crucial for informing targeted public health interventions and policies to mitigate suicide risks and improve overall mental health outcomes in Thailand.

One of the potential risk factors mentioned in the paper is household income. Interestingly, the study found that higher household income is associated with increased suicide rates in Thailand. This counterintuitive finding suggests that economic affluence might contribute to social isolation, a factor that can significantly impact mental health. Higher-income individuals may experience greater pressure to maintain their economic status, leading to stress and anxiety. Additionally, wealthier individuals might have weaker community ties compared to those in lower-income brackets, where social cohesion is often stronger due to shared economic struggles. This lack of strong social support networks can leave individuals more vulnerable to mental health issues and suicidal behaviors. Therefore, understanding the role of income as a risk factor highlights the need for mental health support services that are accessible to all economic groups and the importance of fostering community connections regardless of economic status.

Statistical modeling plays a vital role in investigating the epidemiology and spatial aspects of the suicide problem in Thailand. By applying advanced statistical techniques, researchers can analyze complex data sets to identify significant risk factors and their interactions. Models allow for the control of confounding variables and the assessment of potential effect modifiers, providing a clearer picture of the relationships between different determinants and suicide rates. Spatial epidemiology, in particular, enables the examination of geographical patterns and clustering of suicides, highlighting areas with higher risks and facilitating targeted interventions. Bayesian models and spatio-temporal analyses can incorporate prior knowledge and account for spatial dependencies, improving the accuracy and reliability of the findings.

[Soe Htike]

In mental health settings, ePROs are fundamental for clinicians to effectively monitor patients’ moods, anxiety levels, and treatment progress. This real-time data collection is crucial for timely interventions and adjustments to treatment plans. For patients with chronic illnesses, the use of ePROs is vital to improve patient engagement and self-management by allowing them to regularly report their symptoms and health status. This continuous feedback loop is imperative for healthcare providers to identify any issues early and tailor treatments to individual needs.
Patients require a user-friendly interface with a simple design, customization options, interactive elements such as gamification, educational content, and automated reminders to enhance engagement and usability. Multilingual support is essential to accommodate diverse populations. Healthcare providers will need seamless integration with EHR systems, automated reporting, advanced analytics, and predictive modeling tools to support clinical decision-making. Customizable templates and dynamic questionnaires are necessary to ensure relevance and minimize response burden.

[Soe Htike]

Ko Phyo downloaded maptools package from CRAN archives. We manually installed it in our updated R Studio. It worked fine but showed warning messages like “it won’t be available for so long in the future” or something like that.
I searched for updated packages for maptools and installed “sp” and “terra,” but none of them worked with the codes. So, only manually installing map tools finally saved all four of us. Kudos to Ko Phyo for his initiative.

[Soe Htike]

In our country, one main challenge to achieving sustainability in HIS is the reliance on external funding. Donors initially fund many HIS projects, but their project lifespan is too short for the local agencies and health systems to maintain these systems. I think countries like ours need to consider developing strategies for integrating HIS into their national budgets and seeking diverse funding sources, if possible.

To be adaptable, the EHIS should be able to work with other health information systems (interoperability) and grow as more users or data are added (scalability). It’s also important for the system to allow for flexible data entry and reporting formats to meet changing health data requirements.

[Soe Htike]

1. Differences in Perceived Ease of Use and Usefulness Among Different Demographics

Age
Younger users generally find e-health applications easier to use due to greater familiarity with technology, while older users may struggle due to less experience and potential physical limitations. Younger users might value features that offer convenience, whereas older users might find systems more useful for managing chronic conditions and communicating with healthcare providers.

Gender
Men may find e-health applications easier to use if the interface is straightforward and task-oriented, whereas women may prefer clear instructions and user-friendly design. However, I don’t believe that gender can significantly impact technology use unless the study population is at a community level with low socio-economic status.

Education Level
Users with higher education levels are likely to find e-health applications easier to navigate due to better digital literacy, while those with lower education might need simpler interfaces. Higher education users might value advanced features for detailed health tracking, whereas lower education users might appreciate basic functionalities like appointment reminders.

2. External Factors to Consider for Extending the Proposed Model

It’s important to make health information affordable and easy to understand. Offering affordable or subsidized options to help all users, no matter their income, is key. Information about the model needs to be clear and simple, and providing educational resources can help people understand it better. Accessible technical support and training are also essential. Tutorials, help desks, and user guides can help users feel more confident using the system.

[Soe Htike]

1. Reasons for Limited Incorporation of Locations in Epidemiological Research and the Interdisciplinary Nature of Spatial Epidemiology
The lack of attention to location in epidemiological research has historical reasons. Collecting and analyzing spatial data has been challenging, and early studies were limited by the lack of tools and methods to capture and interpret geographic information accurately. It was easier to focus on individual attributes (person) and temporal trends (time) as these aspects were simpler to quantify and analyze with the available technology.
Spatial components have not been used much because it’s hard to take into account differences in space and the influence of many environmental factors. Traditional ways of studying illnesses were not good at dealing with the complicated patterns of space data, so people preferred to work with easier information. Also, there were problems with keeping space data private and secret, which made it harder for researchers to study geography.
Spatial epidemiology is a mix of different fields that helps us understand how health issues are connected to specific places. It combines geography, statistics, environmental science, sociology, and public health. This approach allows us to study how diseases are spread out in different areas and what causes them. We use GIS and spatial statistics to visualize and analyze the data effectively. By using knowledge from different areas, spatial epidemiology can deal with complex health problems that result from many different factors, such as environmental exposures and economic conditions. This approach not only improves the depth of health research but also helps in creating targeted public health plans and policies.

2. Importance of Place as a Disease Determinant
It’s widely known that where a person lives or works can affect their health. The physical environment, like air and water quality, access to green spaces, and exposure to pollutants, can directly affect a person’s health. For example, people in areas with a lot of air pollution are more likely to have breathing and heart problems. Living near factories or places that handle hazardous waste can mean being around harmful substances, which can lead to higher rates of certain cancers and other health problems.
Living in low-income neighborhoods may lead to higher rates of chronic diseases because people might not have easy access to healthy food or medical care. Also, communities where people get along well and support each other may help reduce health risks. On the other hand, communities with weak support systems could make health risks worse.
Cultural norms and practices specific to different regions can influence health behaviors and outcomes. For instance, dietary habits, physical activity levels, and smoking rates can vary significantly across regions, leading to differences in disease prevalence. The design of urban spaces and transportation systems can also impact lifestyle choices and physical activity levels, affecting the incidence of obesity, diabetes, and other lifestyle-related diseases.

[Soe Htike]

We can increase awareness and provide training to healthcare providers for accurate and timely reporting. Regular refresher training sessions can help maintain high data collection and reporting standards. Improve communication and coordination between different hospital units, such as the antenatal care clinic, STI/HIV clinic, and health screening units. Implementing electronic health records (EHR) that integrate with the national surveillance system can streamline data entry and reduce errors. Address data completeness and accuracy issues, particularly in coding nationality and diagnosis, to improve the overall reliability of the surveillance system.

My experience with disease surveillance systems has shown that one of their strengths is their ability to provide real-time data on disease incidence. This real-time data is very important for quick public health responses. These systems can quickly identify emerging outbreaks, allowing fast intervention and control measures. Additionally, advanced surveillance systems often combine many data sources, such as lab results, clinical reports, and social media, improving their ability to detect and monitor disease trends. However, there are weaknesses as well. One major challenge is ensuring that the data is of good quality and complete since surveillance systems rely heavily on accurate reporting by healthcare providers. Another issue is integrating different data sources, which can be technically complex and require many resources. Lastly, keeping data privacy and security in mind is critical, especially when dealing with sensitive health information. To tackle these weaknesses, continual investment in technology, training, and strong data governance practices is needed.

[Soe Htike]

To improve epidemic surveillance using AI, we can integrate it with current public health systems. We should update our systems to handle big data sets and ensure AI tools can easily share data with public health databases. We should also train healthcare workers to use AI effectively and create user-friendly interfaces. Using advanced machine learning algorithms that keep learning from new data can help us detect outbreaks more accurately and quickly. By working together, tech companies, healthcare providers, and government agencies can use these AI systems, ensuring they meet public health surveillance’s specific needs.

Using AI for public health readiness has many advantages. AI can quickly analyze large amounts of data, finding patterns and irregularities that traditional methods might miss. This can help predict disease spread and improve resource allocation and emergency planning. However, implementing AI presents challenges, such as ensuring data privacy and quality and addressing potential biases. Setting up AI systems and training personnel can also be costly. Overcoming these challenges requires a strategic approach, including strong data governance, ongoing training, and investment in reliable AI technologies.

[Soe Htike]

Apart from the points mentioned above, I’d like to add the level of awareness and education about HIV/STI and RH issues can determine how and when youth seek information. We can evaluate the effectiveness of school-based sex education programs and the overall knowledge levels among young people regarding HIV/STI prevention and RH. The other issue we should consider is the mental health issues and psychological barriers that can impede the willingness of the youth to seek health information and care. We should investigate the prevalence of mental health issues among youth and how these affect their health-seeking behaviors.

People with low socioeconomic status usually have financial constraints and lack of education can magnify access to health information and services. They often have to prioritize basic needs over health and the social stigma deep-rooted in society makes them distrust healthcare institutions. We should tailor our communication channels to target these vulnerable populations.

[Soe Htike]

The decision tree model developed to predict massive intraoperative blood loss (IBL) in pancreatic surgery is very effective. It can accurately identify high-risk patients and help plan surgeries. The model has over 80% accuracy in training and testing sets, showing that it can reliably predict which patients are at higher risk of significant blood loss during surgery. This accuracy makes the decision tree model strong and valuable in clinical settings, providing surgeons with important information to improve patient outcomes.
Despite its effectiveness, the decision tree model is not without limitations. Its reliance on the quality and completeness of input data means that any inaccuracies or missing information can compromise the model’s predictions. Furthermore, there is a risk of overreliance on the model, potentially overshadowing healthcare providers’ clinical judgment and experience.
The practical implications of the decision tree model in clinical use are substantial. By accurately predicting which patients are likely to experience massive IBL, the model allows for better preoperative planning and resource allocation. Surgeons can ensure that the necessary preparations, such as having additional blood products on hand and scheduling surgeries during times when more staff are available, are made for high-risk patients. This improves the safety and outcomes of the surgeries and enhances the efficiency of hospital operations.

[Soe Htike]

A tool to manage non-communicable diseases (NCDs) led by a nurse or community health worker can effectively help in areas with few healthcare workers. With the right training and support, these professionals can handle NCDs well. Projects like the mPower Heart Project have proven these interventions can improve patient outcomes in places with limited resources. By using local health workers who are already part of the community, these programs can ensure steady care, involve patients more, and provide culturally sensitive health education. These aspects are crucial for managing chronic conditions like hypertension and diabetes.

However, just relying on nurse- or community health worker-led tools may not fully solve the challenges caused by the lack of healthcare workers. It’s important to think about other methods to complement this approach and improve the care of NCD patients. Telemedicine can play a big role in this by allowing remote consultations with specialists and making expert advice available in remote areas. Also, task-shifting strategies, where certain medical tasks are given to less specialized health workers under proper supervision, can make the most of the workforce available and ensure more patients get timely care. Additionally, educating and empowering patients with self-management and digital health tools can help them stick to their treatment and lifestyle changes, which can lessen the load on healthcare systems.

[Soe Htike]

Some additional ways to improve the safety of medical AI
In the first point, I would like to emphasize the importance of having human experts review AI recommendations before making clinical decisions. For instance, PathAI offers diagnostic suggestions that are reviewed and confirmed by pathologists. This approach combines AI accuracy with human expertise to enhance diagnostic reliability. Secondly, it is crucial to use diverse datasets representing different demographics to train AI models. This helps to reduce bias and ensures that the models are applicable to various patient groups.

[Soe Htike]

Hi, everyone. Here is the link to my dashboard.
https://lookerstudio.google.com/reporting/aa872c88-914d-463a-a12e-a1855bf403de

It includes score cards, control buttons, drill down table, area chart, bubble map and line map. I tried to keep it simple and hope the filters work. Please let me know if there’s any error in my dashboard. Thank you all.

[Soe Htike]

To better predict PIH, we can gather more data. We can start by monitoring the patient’s heart rate variability (HRV) and cardiac output to get real-time insights into their cardiovascular status. This can help us spot early signs of hypotension. We should also keep an eye on oxygen saturation levels during induction to catch any sudden drops that could cause low blood pressure. Detailed information on anesthetic drug dosing can show how specific drugs affect blood pressure changes. Understanding the type and duration of surgery is important since different surgeries pose varying risks for hypotension. Lastly, being aware of operating room conditions helps us consider environmental and circadian factors, which can influence a patient’s physiological response.

To improve predictive models for PIH and address current limitations, future research should focus on a few key areas. First, studies should be conducted at multiple sites and include a variety of surgical procedures, such as orthopedic, cardiovascular, and neurosurgical operations, to make the models relevant to different types of patients and surgeries. Long-term studies and real-time monitoring are important for understanding outcomes over time and continuously updating the models during surgeries. To ensure these models are effectively integrated into clinical workflows, we need to include clinical expertise through Clinical Decision Support Systems (CDSS) and create user-friendly interfaces. Ethical and regulatory considerations, including obtaining patient consent and ensuring data privacy, must be strictly followed. Continuous model improvement can be achieved by establishing feedback loops with clinicians and developing adaptive learning models that evolve with new data. By focusing on these future research directions, we can make PIH predictive models more accurate, reliable, and applicable across various surgical procedures and patient populations.

Thanks for your presentation, Teeraboon. You explained clearly on things that I didn’t understand when I read the article.

[Soe Htike]

Thanks for your wonderful explanation of a complex and sophisticated research field, Toby. I want to participate in your second discussion point.
Using machine learning to predict symptoms in cancer patients can be a game-changer. It can offer personalized care and timely interventions. However, we must be careful to avoid potential harm from incorrect predictions that might lead to inappropriate treatments. Keeping a close eye on how these ML models perform in real-world settings is crucial. Regularly evaluating and tweaking them ensures they remain reliable and beneficial. This ongoing monitoring helps ensure we’re helping patients rather than causing unintended issues. Balancing the scales between the fantastic benefits and the possible risks means we always aim to provide the best care. So, while these ML models hold great promise, staying vigilant is key to their successful and safe application in healthcare.

[Soe Htike]

Hi, everyone. The links below are the screenshots of my data visualization for this week’s assignment, and here is the link to my looker studio project: https://lookerstudio.google.com/s/l-ocRayIxzU

Running Sum and Comparison
https://snipboard.io/or4GIT.jpg

Running Delta
https://snipboard.io/T4WxQk.jpg

Drill Down and Date
https://snipboard.io/VowJhT.jpg

Pivot Table
https://snipboard.io/tpjhR3.jpg

Score Card
https://snipboard.io/e2lvDb.jpg

Time Series
https://snipboard.io/K35Ggx.jpg

Bar Chart
https://snipboard.io/fgZ91n.jpg

[Soe Htike]

Hi Kansiri,

I’m using 3.30.1 version. I agree with you about some interface changes in different versions, but I think most of the core functions are more or less the same. To make it easier to load dataset or shape file layers, I keep opening the browser panel and layers panel on the left side of the canvas. If they are not visible at the start, I click view, go down to panel and check these two boxes. The other way round is we can right-click on the screen anywhere except the canvas so that we can check the necessary panels and toolbars.

[Soe Htike]

I want to choose the WHO COVID-19 Dashboard. Please kindly see the link below for your reference.
https://data.who.int/dashboards/covid19/cases?n=o&m49=104

The dashboard contains a wide range of data, such as circulating variants, case counts, deaths, and vaccinations. It also offers global-scale data, invaluable for understanding the pandemic’s impact worldwide. The dashboard’s data is regularly updated to keep the information current. Additionally, cautionary notes and explanations about the interpretation and correlation of data are provided, which helps users understand the limitations and nuances of the data. Furthermore, the dashboard includes a glossary and definitions for the variables used, aiding users in understanding the terms and interpreting the data more effectively.

The data is extensive, but the user interface could be more user-friendly. For example, it’s a bit challenging to navigate through different sections. It would be great if the dashboard could provide more granular data, such as province or district-level data. Presenting the data as weekly statistics might not be intuitive for everyone. It might be more user-friendly to have an option for daily data or to choose between daily and weekly data. Additionally, the dashboard doesn’t have a direct option to download the data visualizations, which could be limiting for users who want to include the visualizations in reports or presentations.

[Soe Htike]

Thank you very much for your suggestions, Teerawat. It’s really insightful. I understand that my CRF is way too simple and does not cover all the necessary variables after reviewing Nichcha’s CRF and reading your comments. I will try to get better in the future, friend.

[Soe Htike]

Wow! I was amazed by your thorough CRF, Nichcha! You’ve shown your expertise and experience in clinical data management. I feel proud to review your CRF.

Your study ID naming system is awesome, and the skip logic and branching questions are also brilliant. In a constructive way, I’d like to suggest you include the current status of the underlying disease (if yes, in 6d). If I were the guy responsible for data entry, numbers in the parentheses would confuse me a bit. I understand these numbers are essential for coding later in the data analysis step. But for the data entry guy, all they have to do is fill in the exact information in the relevant data field, right? The last thing is to include questions about the known allergy of the participant (yes/no/don’t know) and specify the type of allergy in screening or enrollment.

It’s such an honor to learn from you.

[Soe Htike]

To avoid redundancy, we can exclude BMI from the CRF. It can be calculated using height and weight later if necessary. Calculating BMI separately also allows for flexibility in analysis, as researchers can apply different BMI categories or thresholds based on study objectives or guidelines without being constrained by a fixed variable in the CRF.

[Soe Htike]

Implementing data standards in clinical research offers numerous benefits, including the ease of cross-study comparisons and meta-analyses. Dengue research, for example, involves conducting several studies across different regions and settings, each collecting data on various aspects of the disease, such as symptoms, disease severity, and treatment outcomes. Standardized data formats and terminology are crucial in this scenario. Without them, comparing findings across these studies becomes daunting due to inconsistencies in how data is collected, recorded, and reported.
By implementing data standards, such as CDISC standards, researchers studying dengue can ensure that data is captured and documented consistently across different studies. Standardized data collection forms and terminology ensure that variables such as fever duration, platelet counts, and disease classification are uniformly defined and recorded across studies. This helps researchers to combine data from multiple studies with more confidence, perform meta-analyses, and derive meaningful insights into various aspects of dengue epidemiology, pathogenesis, and treatment efficacy.
This standardized data collection and reporting approach significantly improves the reliability and robustness of research findings in dengue. Researchers can now gather data from various sources, identify patterns and trends that may not have been apparent in individual studies, and draw more accurate conclusions about the effectiveness of interventions such as vaccines or antiviral treatments. This approach enables comprehensive cross-study comparisons and meta-analyses, leading to a stronger evidence base for dengue research. It also helps develop clinical practice guidelines and contributes to more effective disease prevention, diagnosis, and management strategies.

[Soe Htike]

While I don’t have direct experience conducting clinical trials or similar study projects, I have managed data from my research endeavors, primarily involving quantitative and qualitative data analysis. Unlike the comprehensive processes described in my friends’ experiences, my data collection and management have been relatively straightforward. Most of my data were collected using paper forms and audio recordings, which were later transformed into digital formats using quantitative and qualitative data analysis software, respectively. As this data’s sole user and custodian, I indirectly managed access control. I stored the data on my laptop and backed it up on Google Drive, which is two-factor authentication for added security.
Compared to the sophisticated data management processes outlined in the above responses, my approach may seem almost nothing. Nonetheless, I recognize the importance of robust data management practices, particularly in larger-scale studies involving multiple stakeholders and sensitive information.
As I engage in more complex research experiences or collaborate on larger projects, I aim to adopt more rigorous data management processes similar to those described by my friends. Additionally, I will explore utilizing specialized data management software tailored to the needs of the research projects I am involved in, further enhancing efficiency and compliance with best practices in data management.

[Soe Htike]

I have completed several steps in managing data, such as selecting methods for data collection, conducting interviews and surveys to collect quantitative data, recording qualitative data and translating it for thematic analysis, and encoding quantitative data into statistical software for analysis.
However, there are still some critical steps that I have yet to address, including ensuring that data cleaning and validation procedures are comprehensive, establishing robust protocols for data storage and backup, and thoroughly documenting procedures for data processing and analysis. It is important to implement these steps to ensure data integrity, reliability, and reproducibility, enhancing the project’s quality and rigor. Doing so would also improve the credibility of the findings and make it easier for future research to build upon this dataset.

[Soe Htike]

I collected data to investigate the factors hindering or promoting compliance with the standard procedures for eliminating malaria among healthcare service providers in a township working towards malaria elimination. My objective was to provide insights that can be used to develop effective strategies for enhancing malaria elimination efforts and healthcare delivery in the area.

I collected primary data. For the quantitative aspect, I interviewed all the HSPs in the townships using a self-administered structured questionnaire. I conducted IDI and KII with the selected participants for the qualitative aspect.

In my research, I used mixed methods to collect data. For the quantitative component, I employed a self-administered structured questionnaire. At the same time, for the qualitative aspect, I conducted In-Depth Interviews (IDI) and Key Informant Interviews (KII) with selected participants to gain an in-depth understanding of their perspectives and experiences. The data collection process involved using paper-based tools. The structured questionnaire was administered to participants, and their responses were later encoded into SPSS for quantitative data analysis to ensure accuracy and consistency. For the qualitative aspect, I used paper-based guides for IDI and KII, and recordings were translated into English verbatim. Thematic analysis was then conducted using NVivo to explore themes in the qualitative data.

I faced several challenges during the process of collecting data. These included traveling extensively to different healthcare facilities in the study area to meet with participants, difficulty recruiting participants due to travel restrictions imposed by the COVID-19 pandemic, and some missing data in the quantitative data collection. Furthermore, for qualitative data collection, I encountered challenges with travel logistics, increased costs, and potential interviewer bias, as some participants knew me as a public health officer working in the Ministry of Health.

[Soe Htike]

This is my summary of what I’ve learned in week 4. Please see in the following link –> https://snipboard.io/tBJngY.jpg

[Soe Htike]

I’d like to share my summary of this week.

https://snipboard.io/aKq2mT.jpg

[Soe Htike]

Please kindly see the attached link for what I’ve learned this week.

[Soe Htike]

I would like to share a link with you all about AI and ethics in health that I have learned this week.

[Soe Htike]

As a leader of a contact tracing team during the COVID-19 pandemic in Myanmar, I upheld various ethical principles and good practices to make a meaningful contribution to the control policy. First and foremost, maintaining confidentiality was of utmost importance. I ensured that all sensitive information regarding positive cases and their contacts was handled with the utmost privacy and disclosed only to authorized personnel involved in the public health response.
Secondly, I recognized the significance of transparency in this critical situation. I communicated openly with individuals identified through contact tracing, offering clear and accurate information about their situation, the necessity of isolation or quarantine, and the public health measures in place. This transparency fostered trust and cooperation, which are essential for successful contact tracing efforts.
In addition, I ensured empathy and sensitivity in my interactions. I acknowledged the anxiety and concerns of individuals affected by the virus and approached them with empathy, addressing their questions and providing support. This approach not only enhanced the overall experience for those involved but also contributed to increased compliance with recommended measures.
Furthermore, collaboration was an essential ethical principle that I adhered to. I worked closely with healthcare professionals, local authorities, and relevant stakeholders to ensure a coordinated and efficient response. This collaborative approach helped streamline processes, share information effectively, and address emerging challenges in real-time. Continuous self-reflection and adaptation were integral to my role. The dynamic nature of the pandemic demanded constant evaluation of contact tracing strategies. I engaged in regular assessments of our procedures, identified areas for improvement, and implemented necessary adjustments to enhance the effectiveness of our efforts in controlling the spread of COVID-19 in the Myanmar public health setting.

[Soe Htike]

Myanmar has set a target of achieving Universal Health Coverage (UHC) by 2030 and has made progress towards this goal. The National Health Plan (2017-2021) was a key policy document of the former Ministry of Health and Sports. The plan has identified several critical areas that need attention to achieve UHC, including financial protection, strengthening the health workforce, and improving the township health system.

Defining the essential package of health services, improving primary healthcare infrastructure and engaging communities are positive steps towards improving health service delivery. The mobilization of resources for health and the integration of health programs within broader development agendas indicate a comprehensive approach to UHC in Myanmar. Another strength is the focus on financial protection to promote equitable access to health. The government has recognized the need to increase spending on health to reduce out-of-pocket expenditures and has decentralized health workforce allocation decision-making to states/regions.

However, there are still weaknesses that need to be addressed. Healthcare infrastructure in rural areas is limited, which hinders equitable access to services. Myanmar needs sustained investments in health systems, workforce development, and technological integration for UHC initiatives to scale up. Financial constraints and insufficient budget allocations for health could further impede progress towards UHC. A multi-faceted strategy involving increased funding, targeted capacity building, and innovative solutions is required to enhance healthcare accessibility, especially in underserved regions.

To make UHC work effectively in Myanmar, there must be a focus on strengthening health governance, improving data systems for better decision-making, and fostering partnerships with international organizations and stakeholders. Emphasizing preventive and primary healthcare measures, along with robust community participation, can enhance the preventive aspect of healthcare, reducing the burden on the system.

[Soe Htike]

We have a shortage of qualified health informaticians in our country, and not many people acknowledge the importance of health informatics as a crucial component of the health system. The health informatics workforce in Myanmar faces significant challenges, such as awareness issues, high implementation costs, gaps in training programs, and collaboration challenges.

Additionally, political instability and violence have disrupted the health information system, leading to a lack of availability and quality of health data and information. Insufficient technological infrastructure, such as outdated hardware and insufficient connectivity, further hinders the effective deployment and utilization of health informatics solutions.

Furthermore, the absence of robust data security measures and frameworks is a significant concern regarding data security. Ensuring the privacy and integrity of health data is crucial for building trust in health informatics systems.

The lack of standardized protocols and interoperability among different health information systems hinders seamless data exchange, which may lead to fragmented health records and reduced overall system efficiency.

Cultural and organizational resistance to adopting new technologies and workflows can hinder the successful implementation of health informatics. Overcoming resistance requires a comprehensive change management strategy. Additionally, a shortage of research initiatives and limited investment in innovation in health informatics may result in a slow pace of technological advancements and the adoption of best practices in Myanmar.

To address these challenges, we need a comprehensive approach that includes infrastructure development, cybersecurity measures, standardization efforts, change management strategies, and fostering a culture of research and innovation in health informatics. Above all, Myanmar needs a sustainable solution for peace to bring need-based, equitable and meaningful changes to the health system and all stakeholders involved in the system.

[Soe Htike]

As someone who has been struggling to assess public health data at the country level, I personally prefer transparent and easy data-sharing practices. However, if I were the custodian of a dataset from my country, I would have to consider various factors before sharing the dataset. I would need to strike a balance between the benefits of data sharing and the privacy and ethical considerations of the people in the dataset.

It is important to remember that data involves people, and there is a lot of personal and sensitive information behind the numbers. Therefore, the benefits of data sharing must outweigh the potential harmful effects on the people in the dataset. As many friends discuss the advantages of data sharing, I want to focus on the disadvantages. Apart from the risks to individuals’ privacy and confidentiality, there is a serious issue I want to highlight. Inappropriate use and misinterpretation of shared data could lead to erroneous conclusions.

If I decide to share a dataset, I will prioritize obtaining informed consent from the individuals who contribute to the data. I will also implement anonymization and aggregate data to mitigate privacy risks while maintaining data utility. Moreover, I will establish a clear data use agreement outlining the level of permission given for data use.

[Soe Htike]

EMRs are changing the way primary healthcare operates by improving data accessibility, promoting patient-centered care, and facilitating better communication among healthcare professionals. They simplify workflows by providing instant access to patient data, enabling informed decisions, personalized care plans, and continuity of care across different settings. EMRs also enhance patient safety and care quality by reducing errors and enabling comprehensive medication management. Additionally, they contribute to improvements in the healthcare system by increasing transparency, accountability, and the potential for analysis and audits. Overall, EMRs represent a significant advancement over traditional paper-based records, offering numerous benefits that enhance patient outcomes and healthcare efficiency.
Although EMRs offer various advantages, they also pose some significant challenges. The transition to EMRs requires a considerable initial investment in technology infrastructure, staff training, and system customization, which can be expensive and disruptive. Data security and privacy are crucial issues as electronic records are vulnerable to unauthorized access, necessitating robust cybersecurity measures. Overreliance on EMRs can lead to vulnerabilities in the event of technical glitches or outages, and the design of the user interface and ease of use directly affects successful adoption. Furthermore, the digital divide, which refers to unequal access to technology and resources, can worsen existing healthcare disparities. Therefore, strategic planning, robust security protocols, and equitable solutions are essential for successful EMR implementation and adoption.

[Soe Htike]

The article discusses a number of challenges related to data-driven healthcare research. These challenges include disease definition, source availability, data sharing, data quality and missing data, translational applicability of results, dependence problem, data linkage, data inconsistency, interpretation of results, unstructured data, data integrity, training, legal and ethical issues, and data security. However, I have chosen the top three challenges that I believe are most significant:
1. Data quality and missing data – incomplete or inaccurate data can distort results and lead to incorrect conclusions.
2. Data security – safeguarding data from breaches is a major concern, especially when the healthcare system is not fully digitalized.
3. Legal and ethical issues – it is vital to comply with data protection laws and ethical guidelines.

To address these challenges, I’d suggest the following:
1. Implement rigorous data validation and cleaning processes and use advanced statistical techniques to handle missing data.
2. Employ robust encryption methods, secure data storage solutions, and regular security audits to protect data.
3. Stay up-to-date with the latest data protection laws and ethical guidelines and ensure transparent and ethical practices in data collection, processing, and usage.

[Soe Htike]

I agree with the four recommended points for public health professionals to fight corruption in health systems. They provide a strong foundation for tackling this complex and widespread issue.
Public health professionals can help break the silence on corruption by acknowledging its existence, raising awareness, and promoting transparency and ethical conduct within the healthcare system. Open discussions with various stakeholders are crucial to combat corruption. Public health professionals can lead these dialogues and establish channels for productive communication, building trust and a shared commitment to solving the issue.

In addition to the four points mentioned earlier, some additional elements could further strengthen the fight against corruption. “Empowering communities” to identify, report, and demand accountability for corruption can promote transparency and ensure that resources reach those who need them most. Community-based monitoring initiatives and raising awareness about corruption can play a crucial role in combating corruption.
“Strengthening governance and institutions” is essential to prevent corruption. Healthcare professionals can advocate for reforms that promote transparency, accountability, and ethical conduct within healthcare institutions. Investing in technology like open data platforms, digital payment systems, and online complaint mechanisms can increase transparency and reduce opportunities for corruption.
“Addressing underlying drivers” such as poverty, inequality, and weak rule of law is equally important. Public health professionals can advocate for broader societal reforms that address these underlying issues and create a more equitable and just environment. They can also advocate and support the implementation and enforcement of legal and regulatory frameworks that prevent and sanction corruption while protecting the rights and interests of the people.

Overall, fighting corruption in health systems requires a sustained and multi-pronged approach, especially in healthcare systems in third-world countries.

[Soe Htike]

I would like to discuss an example of efforts to improve Myanmar’s Primary Healthcare system. The project aims to improve the accessibility and quality of basic health services for disadvantaged individuals in rural and conflict-affected areas by working with both the Ministry of Health (MOH) and Ethnic Health Organizations (EHOs). The project also seeks to strengthen coordination and trust between the two sides.
However, there are some barriers encountered in this process of improving the system, including inadequate and inequitable health financing, which limits the availability and affordability of health services, particularly for poor and marginalized groups. The shortage and uneven distribution of the health workforce, especially midwives, also affect the quality and coverage of maternal and child health services. The centralization and low responsiveness of the health system hinder the feedback and participation of local stakeholders, reducing the flexibility and efficiency of service delivery. The political instability and ongoing conflicts also pose security risks and challenges to the implementation and sustainability of the project.

[Soe Htike]

Please let me give an example of a personal project management example using these 12 steps.

Step 1: Define the project clearly.
Project Goals: Finish DBHI diploma and Google Data Analytics course within six months.
Objectives: Complete all DBHI modules and Google Data Analytics assessments.
Success criteria: Diploma from DBHI and a certificate for the Google Data Analytics course while spending quality time with family.

Step 2: List all tasks.
I’ll break down each course into modules so that I can catch up with these daily.
I’ll list personal tasks related to the return home and family vacation (booking travel, accommodation, etc.).

Step 3: Get tasks into the right order.
I’ll identify dependencies between courses and tasks (e.g., specific DBHI modules needed for Google Data Analytics).
I’ll consider workload fluctuations and prioritize based on deadlines and complexity.

Step 4: Add a safety margin.
I’ll allocate buffer time for unexpected delays or increased workload in each course.
I’ll consider including an extra week for my visit home to account for travel uncertainties.

Step 5: Crash the project plan (if necessary).
I’ll analyze if any tasks can be combined or streamlined to shorten the overall duration.
I’ll explore online resources or study groups to potentially accelerate learning.

Step 6: Create a Gantt chart.
The Gantt chart will visually represent the tasks, deadlines, and buffer time for both courses and the visit home.
I’ll use different colors or sections to distinguish course and personal tasks.

Step 7: Look at resources.
I’ll assess my personal time availability and energy levels throughout the six months.
I’ll identify potential support (teachers, friends and family) who can offer assistance with studies or personal tasks.

Step 8: Think about what might go wrong.
I’ll consider potential risks like illness, political issues, or family emergencies.
I’ll develop contingency plans for each risk to minimize disruptions.

Step 9: Monitor progress.
I’ll regularly track my progress in both courses and my preparations for the visit home.
I’ll use my Gantt chart to identify any deviations from the plan and adjust accordingly.

Step 10: Monitor cost (time and resources).
I’ll be mindful of my energy levels and adjust my workload if needed to avoid burnout.
I’ll track time spent on each task and course to ensure efficient use of my resources.

Step 11: Readjust plan.
I’ll try to be flexible and adapt my plan if necessary due to unforeseen circumstances.
I might need to consider adjusting the timing of my visit home for optimal work-life balance.

Step 12: Project review.
After completing both courses and my visit home, I’ll reflect on what went well and what could be improved.
I’ll use this knowledge to refine my approach for future personal projects.

[Soe Htike]

As a team leader, I believe it is important to create a safe and supportive environment for my team members to communicate and share their views, ideas, and worries. I’d like to establish respectful rules for our interactions. I show positive and inclusive behaviors, such as listening carefully, asking open-ended questions, acknowledging and appreciating contributions, providing feedback and coaching, and resolving conflicts. I also seek input from people across a wide variety of backgrounds, and I show genuine interest and curiosity in their perspectives and experiences. I encourage my team members to be respectful of each other and to value our differences. I believe these behaviors help us create a more inclusive workplace culture where everyone feels valued.

[Soe Htike]

To ensure an uninterrupted malaria surveillance system, a robust disaster recovery plan is essential. This plan should prioritize the protection of critical data, including case reports, lab results, and mapping data. Assessing potential risks and vulnerabilities, from natural disasters to cyberattacks, helps determine acceptable downtime and data loss limits (RTO and RPO).

Data backup and replication are crucial safeguards. Utilizing cloud storage or geographically dispersed data centers ensures data availability even in the event of local disruptions. Step-by-step recovery procedures outlining the restoration of data and systems provide a clear roadmap for quick action.

Regularly testing and updating the plan is vital to ensure its effectiveness against evolving threats and technological advancements—open-source solutions like DHIS2 for reporting offer cost-effective options for resource-constrained organizations.

Cloud-based platforms like AWS and Azure offer scalable data storage and recovery solutions, particularly valuable for large datasets. Mobile data collection applications with offline capabilities enable data capture in remote areas with limited connectivity, ensuring continuous surveillance even in challenging circumstances.

Finally, training personnel responsible for data management and recovery equips them with the necessary skills and knowledge to implement the plan effectively in a disaster. By prioritizing critical data, assessing risks, implementing safeguards, and ensuring plan effectiveness, organizations can ensure the resilience of their malaria surveillance systems and continue their vital work in combating the disease.

[Soe Htike]

Benefits for the patients
Patients will experience improved accessibility to their medical records and timely delivery of healthcare services. With HA ensuring system uptime and minimizing disruptions, healthcare providers can access critical patient information promptly, leading to quicker diagnosis and treatment. Reduced downtime also means fewer delays in scheduling appointments and obtaining test results, contributing to a smoother patient experience. Additionally, HA increases the security and privacy of patient data, instilling confidence in the confidentiality and integrity of their health information. Patients can be reassured knowing that their medical records are secure and accessible, even in the event of system disruptions.
Benefits for the hospital
With minimized system disruptions, healthcare professionals can optimize their workflow, leading to increased productivity and streamlined processes. HA ensures that essential hospital services, such as electronic health records and diagnostic tools, remain consistently available, contributing to better-informed decision-making by medical staff. The reliability of HA also enhances the hospital’s reputation, attracting more patients and fostering trust within the community. Furthermore, the improved data protection and reduced risk of downtime align with regulatory compliance, safeguarding the hospital against potential legal and financial implications.

[Soe Htike]

My weakness as a listener lies in my tendency to focus on the overall meaning and flow of conversations, often causing me to skim over specific details. This leads to difficulties in remembering names, dates, and intricate instructions, impacting my ability to follow complex directions, retain factual information, and build strong relationships.

I can improve my detail recall by actively engaging in strategies like setting pre-listening intentions, activating prior knowledge, paying close attention to key phrases, chunking complex information, utilizing notetaking and visualization techniques, regularly reviewing and summarizing key points, and testing my recall. This commitment to deliberate practice will enhance my listening skills and improve my overall interactions with others.

[Soe Htike]

I don’t have an experience regarding the discussion topic as our health system is not well-established. The public sector is still rolling out eHMIS, and the private sector is using its own EHR systems. But when it comes to training the public health staff in digital literacy, I use this hypothetical example, which can be related to the discussion. Suppose a healthcare organization was using an outdated software system to manage its ePHI. An employee unknowingly opened a phishing email that installed malware on the system. This malware exploited a vulnerability in the outdated software and gained unauthorized access to the ePHI.
The malware was able to access and alter patient records, violating the integrity principle of the CIA triad. This resulted in incorrect patient information, such as wrong medication dosages or misdiagnoses, being recorded. Furthermore, the malware disrupted the system’s operation, violating the availability principle. This caused delays in patient care as healthcare providers were unable to access patient records when needed. Lastly, there was a potential breach of confidentiality if the malware was able to exfiltrate the ePHI to the attacker.
So, to prevent this from happening again, the organization must do the following safeguards.
Regular Software Updates and Patching
Employee Training
Access Control
Regular Backups
Encryption
Intrusion Detection System
Regular Audits

[Soe Htike]

My strongest component of emotional intelligence is empathy. I consistently prioritize understanding and appreciating the perspectives of others, avoiding judgment and embracing positive intent. When a colleague faced a challenging situation, instead of focusing solely on the outcomes, I took the time to understand the pressure they were under and offered support, creating a more collaborative work environment. To further enhance my empathy, I will actively seek out diverse perspectives, engage in more one-on-one conversations to deepen my understanding of others and continue practicing active listening to strengthen my connection with colleagues.
While I excel in understanding others, my weakest component is motivation. I’m not good at motivating others. There have been instances where I struggled to inspire and motivate myself, hindering my ability to maintain a consistently optimistic outlook, especially in challenging situations.
There can be many ways to improve my motivation. But I’d like to highlight the line that I remember from the course, “Walk the talk.” I can only make promises that I can realistically keep. When I make a mistake, I apologize and take steps to rectify the situation. I’d make sure my actions are consistent with the values I hold. Walking the talk is an ongoing process, not a one-time event. By walking the talk, I can not only motivate myself and others but also create a more positive and productive environment for everyone involved.

[Soe Htike]

Thank you for choosing a very important disease that is becoming a public health threat in Myanmar, Ko Myat. The mortality rate of drug-resistant TB is presumably higher in our country than the regional benchmark. Personally speaking, my father was one of the few survivors of pre-XDR TB.
I learned the indicators you proposed from three perspectives: whether they are practical, effective, and achievable. Let me break down some indicators that particularly are interesting to me.
Sensitivity, PVP rate, and evaluation of missing data are practical measures to assess data quality. They are effective in ensuring that the surveillance system provides reliable and timely information about MDR-TB cases. With proper protocols and training, achieving accuracy, completeness, and timeliness in data reporting is feasible, especially with technologies like GeneXpert and digital reporting tools.
User satisfaction, response rates, and feedback mechanisms are practical measures. Quick response times, stakeholder feedback incorporation, and modification actions contribute to system effectiveness. But in our context, I think the sustainability of the surveillance systems mostly relies on concrete response mechanisms and accountable, responsive feedback systems. If we can ensure these mechanisms, I believe the service quality indicators are achievable.

[Soe Htike]

Thank you for sharing, Ko Pyae. I’d like to congratulate you for having a very clear and comprehensive case definition for your surveillance system. The involvement of the stakeholders, such as healthcare providers, public health agencies, laboratories, key populations, and community organizations, ensures a multi-faceted approach. I believe it facilitates a more effective response to the HIV epidemic. One thing I’m considering is whether the system proposed will integrate or replace the existing surveillance system. If it’s intended to integrate, I’d like to know how you will ensure the compatibility and interoperability between the systems to avoid redundancy and enhance overall efficiency. If it’s intended to replace the existing open MRS, I’d like to know how you will advocate for the nationwide stakeholders and implementing partners to use the new system. Thank you, Ko Pyae.

[Soe Htike]

Thanks for your suggestion, Ajarn Chawarat. After discussing with Ko Phyo, I used maptools and lettice packages. It went well in the end.

[Soe Htike]

Thanks for your suggestion, Ko Phyo.

I’m stuck in the command while adding a new column in the matched table.

> data.suicides$ID <- seq(1, Nareas)

Error message is “replacement has 32 rows, data has 0”. So I think the two tables matched are not matched in the earlier steps. But I’m not sure and I dont’t know how to fix the error. So I’m shouting for help so that I can finish the assignment before the deadline.

[Author]

Posts