Forum Replies Created
-
Posts
-
-
I understood it as same as Anawat. In addition, I would like to suggest this site for self-learning. https://www.kdnuggets.com/2016/04/association-rules-apriori-algorithm-tutorial.html
I think this site provides content that is easy to understand. Hope it helps and pleases correct me if I misunderstood too!!
-
Since my objective for the HBV modeling from week 2 is to study HBV transmission dynamic only and I didn’t include any intervention in my model. However, I would like to slightly modify HBV focus and restructure the modeling, by focusing on “HBV infection received from mother-to-child route” in Thailand. The intervention is antiviral prophylaxis. Therefore, the modified modeling is SEIR model which S= susceptible, IA = acute HBV infected from mother-to-child, Ic = Progress to Chronic HBV infection, and R = Recovered – considered as children having sufficient immune, E = exposure to antiviral (mothers and infants receiving antiviral). The similar research is https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-022-13594-y
WHO recommends the use of antiviral prophylaxis for the prevention of hepatitis B transmission from mother-to-child. (https://www.who.int/news-room/fact-sheets/detail/hepatitis-b). Based on the meta-analysis, the efficacy of peripartum antiviral prophylaxis to reduce the risk of HBV mother-to-child transmission (tenofovir disoproxil fumarate, lamivudine, telbivudine) were ORs around 0·10 to 0·17 which could calculate as lower risk of infection for 73-90%. (https://hal.archives-ouvertes.fr/pasteur-03697722)
Please see my full report (the modified model and variables) here: https://tinyurl.com/mrxpyey9
-
I will use SIR model for the mathematical modeling of HBV transmission dynamic in Thailand regardless of vaccination and transmission route. The variables include S = susceptible, I = HBV infected, and R = recovered people (as suggested by https://science.buu.ac.th/ojs246/index.php/sci/article/view/1623/1551 ). Also, I is subclassified between acute infection and chronic infection; IA= early HBV infected and IC= Chronic/late HBV infected. As HBV infection is permanent, recovered individuals will define as having non-symptoms related to HBV (controlled symptoms).
I can not attach the table and model . Please access the full report here https://tinyurl.com/3yjcs4bn
Thank you.
-
I am interested in using mathematical modeling to estimate hepatitis B disease in Thailand and probably the impact of the HBV vaccine. There are a few related articles:
https://academic.oup.com/ije/article/34/6/1329/707548?login=false
https://www.hindawi.com/journals/cmmm/2014/475451/.There is one study using Thai data: https://science.buu.ac.th/ojs246/index.php/sci/article/view/1623/1551. However, it is outdated (2017), it would be nice if I can put newer data and re-estimate the disease dynamics and impact of vaccination.
I am also interested in Hepatitis C among injecting drug users, but there is limited data on this topics (https://journals.lww.com/aidsonline/Fulltext/2008/08200/New_challenges_for_mathematical_and_statistical.1.aspx, https://www.sciencedirect.com/science/article/abs/pii/S0022519318305666)
-
Please find my dashboard here https://tinyurl.com/2m42769d
I prefer white background, and I try to use simple charts/graphs as my expected users are both healthcare professionals/public health and non-healthcare professionals.
The first page of the dashboard provides an overview of distributed cases (confirmed cases, recovered cases, and deaths) across the globe. It can help public health professionals to understand the situation of COVID-19 going on across regions/countries. The user can use the maps or slicers to filter the data belonging to specific country/capital and date. Even though the maps show only confirmed cases, the recovered/deaths data were linked to the maps. Therefore, the user can also get the cumulative recovered/deaths. There is a graph showing the trend of cases over time and a table for complete data. Moreover, The second page shows the cases by continent. Additionally, the user can select country-level data from the slicers.
-
Please find my dashboard here https://tinyurl.com/y3ua7hee
The first page shows the distribution of cases across the location using maps, cards/rows, and slicers.
The second page displays the stacked bar, column, and donut chart for confirmed, recovered, and deaths cases.
The third page shows the line graphs demonstrating the confirmed case over time, forecast deaths, and the combined confirmed case (column) with the line of deaths to see the trend.
The fourth page shows a sparkline chart of daily confirmed cases in the top 10 countries.
The fifth page displays a funnel chart and TreeMap of the number of deaths/recovered by country.
The last page shows the scatter plots between each country’s daily confirmed cases (x-axis and size) and sum deaths (y-axis) over time (1-month bins). -
Please find my dashboard here: http://tinyurl.com/2p9446td
My dashboard mainly focuses on confirmed COVID-19 cases across the continent, country, and capital. Using a map and a line graph for data visualization makes it easier to understand the trend of confirmed cases/ongoing transmission over time, which is useful for disease surveillance and public health. The user can review the full data in the matrix and table, and also filter data for specific locations, GDP, and the number of confirmed cases/deaths/recovered based on their preference.
-
https://pandemic.internationalsos.com/2019-ncov/covid-19-data-visualisation
The dashboard is informative and consists of several components/data. It would be nice for professionals such as public health specialists, data analysts, healthcare professionals, and statisticians. However, it might be a bot in-details for the general population.
(I will focus on the main graph)
1) The main graph is the line graph that demonstrates the COVID-19 cases over time. It is nice, simple, and easy to understand. The developer also adds several metrics, such as confirmed death and people vaccinated, to be selected for data visualization. It allows adding lines based on country or population of interest. However, It can create confusion if adding too many lines to the graph (e.g., select several countries at the same time). The background is minimal, contrasting with the line color, making it easy to read. There is no connecting dot; however, I think it is fine because it prevents the over-crowded symbol, and the value will appear once we place the mouse arrow at the line.
2) The developer also presents map data visualizations demonstrating case density over locations. I like the color gradation in the data display that can indicate the severity and larger number of cases (light yellow = low no.of cases, red = high no.of cases) -
-
Your CRF is comprehended and well-organized. I really like the brief informed consent part. However, I have a few questions and comments below.
– I am not sure whether we should separate the form for each visit or not – one for screening and one for enrollment. Therefore, this is still a doubt for me, and I can’t comment on any.
– I like the consistency of “Yes” being the first choice. Nice pattern. but I found the last question with No being the first choice.
– I would suggest adding the general instruction, such as check in the checkbox or filling the information before beginning any form/section.
– Do you consider adding race to the form as well?
– I agree with Andrew that subject initials might affect data privacy.Well done!!
-
-
-
I don’t have any experience in clinical trial conducting but I did data collection in pharmacovigilance electronic database called ARISg system in my formal organization. There are audit trails, data backup and recovery plan in place. Every users will have a specific account and every actions made would be collect on real-time basis. Data is archived in the cloud database. I also believe the system has edit check/logic check as well as an optional workflow.
-
I worked as a pharmacovigilance specialist before dealing with adverse events and pharmacovigilance data systems. I am most involved in project implementation and closer parts, such as data entry, QC, data structure, study report, and SAE reconciliation. I do not have experience in project initiation, especially for CRF, DMP, and database setup and programming. If you have a chance to go back, I think I should have done better in data structure coding/management to prevent any messy data and easy to analyze later.
-
1. Purpose of data collection: For research, for public health surveillance, or others
I collected data for my research project when I was in the last year of my pharmacy program
2. Was it primary or secondary data collection?
It is the secondary data collection from existing data in the published papers
3. Methods used for data collection
I used the excel form for data extraction and collection
4. Were there any problems that occurred regarding data collection?
As it is secondary data and sometimes the data are diverse across papers/research questions. Therefore, there are several unstructured data and the data type/variables are different. I needed to clean the data and group the data several times. -
Please see my wrap-up here. https://drive.google.com/file/d/135ALLhQlpHBaDtTRR01RToi5exHqIZXW/view?usp=sharing
Thank you -
Please find my wrap up here https://drive.google.com/file/d/1XGD0cKry2xph0Zv_R_Y8ZmC_XZt8962L/view?usp=sharing
Thank you
-
Please find my summary here: https://drive.google.com/file/d/1zFPsR5VQdGOCUWX0sazuS6IXLfSweT5C/view?usp=sharing
Thank you. -
Hello. please find my warp-up attached.
Or access here: https://drive.google.com/file/d/19f836bf1VgSABjLgJbtQKFfEEAkLHhT4/view?usp=sharing
-
I would say that I am more familiar with Frequentist than Bayesian because I’ve learned it before, and it was used in several clinical data and publications. I also heard about Bayesian from epidemiological research which is interesting to me. However, I haven’t used it myself. It seems like the probability is assigned ahead to a hypothesis in the Bayesian view but not for a frequentist view. I also agree with Auswin and Arwin that Bayesian tends to rely on prior knowledge, and we should learn both approaches to understand the difference.
-
My name is Napisa Freya Sawamiphak. I am currently working as a Medical Science Liaison in a pharmaceutical company. I have been involved in several medical research and medical information. I read new articles every day and use this information for training and scientific communication. In my work, I don’t analyze the data by myself but use the result mostly. However, I still need to understand the methods and statistical analysis for critical appraisal before using the data. I also took a course in clinical statistics previously.
-
-
I think new technology should be more useful, easier to use, and have advantages over the old technology in some ways such as affordable, practical to the actual practice, higher precision, better capacity to detect small specimens, or personalized treatment – highly effective with low adverse event occurrence.
-
Factors related to learning ability, experience, attitude toward use, frequency of new tech use might influence perceived ease of use and usefulness.
For example,
1) Elderly may feel challenged to learn new things and have low frequency using systems, 2) socioeconomic – people with low income may not have an appropriate device/facility to assess new tech, which leads to inconvenience and low experience, 3) some people think that the new tech does not have a relative advantage over the current practice resulting in their low perceived usefulness -
As we are curious about the reason why they are not using bednets, I think it is better to perform additional qualitative research/data collection, perhaps by conducting a focus group interview. After gathering insights and understanding their perspectives, we can also conduct quantitative research later to analyze patterns, predict trends/averages, verify causal association and generalize results in larger groups.
-
-
-
Mortality rate is the ratio of death events in a particular population during a specific time period. The rate was usually calculated per 1000 or 100,000 individuals.
Calculation: Mortality Rate = (number of deaths during a given time interval) / (total population in that interval) x (expressed unit – usually use per 1000 – 100,000 individuals)
Usefulness: It is the last and worst outcome that can occur in health problems. It can indicate the health status of a population and assess the severity of receiving exposure leading to death. It is also useful to public health authorities to prioritize the health issue.
-
Thai UHC covers treatment costs within the public hospitals and one person/patient is required to pay only 30 Baht for all treatment, the hospital will manage the rest using fixed funding from the government.
In my opinion, the strength is it supports healthcare costs to all people without considering the socio-economic status. It really values life and equality. However, the weaknesses are the dissimilarity of funding and the patients’ density in each hospital, leading to the difficulty of resources distribution. Also, the available treatment options are limited to drugs under the National list of essential medicine only, to save cost (cost-effectiveness); however, it might not be the best choice in some severe cases.
In order to make this UHC scheme works, I agree with Auswin that the distribution of resources must be planned and distributed properly for both funding and the health workforce.
-
I have the opportunity to work on the medical information system related to the COVID-19 vaccine. Therefore, I would say I supported the COVID-19 best practices by providing scientific data rather than providing direct healthcare. I think my role meets the ethics principle on beneficence and non-maleficence because we provided clear, correct, and unbias data that would be helpful for COVID-19 patients’ care and maximize patient safety. Also, my organization encourages employees to work remotely, keep social distancing, wear face masks and face shields, and provide vaccine/health check-up/ATK tests to ensure a safe place to work and prevent the COVID-19 spread.
-
I agree with Auswin and Ashara that Health informatics is still new in Thailand; therefore, several organizations didn’t understand the roles and responsibilities of health informatics workforce clearly. Although the accessibility of health IT systems is a gap and it is not fully adopted in some hospitals/areas, the shortage of health informatics workforce is still presented compared with the workload and emerging of health informatics systems, in my opinion.
-
I will consider data sharing. It will provide several benefits such as data exchange across settings, supporting research, providing accurate patients history, decreasing duplicate medications or procedures, and reducing medication errors. However, we need to ensure confidentiality and data security by setting well-controlled regulations/protocols and having supportive teams to handle the system or any errors.
-
My workplace is not a healthcare setting; therefore, we don’t use EMR (I have experience using it while internships only) but I agree with all opinions here that EMR is easy to access and convenient for healthcare professionals to retrieve data on real-time basis. It can prevent data loss and can be used across departments, good for patients who visit several sites. However, we need to be concerned about data confidentiality and data security. EMR also affects healthcare professionals’ workflow because it takes time to get trained on how to use the new system. Therefore, it might create a burden to the users during the first launch.
-
I think we can reduce missing data by
1) Encouraging to use of terms in EHR standards and providing protocol/procedure of required data that need to be filled in the system. Training and make alignment within cross-functional teams and users is also crucial.
2) Set up the mandatory field and fixed answers, e.g. choices Female/male, if possible (for structured data)For data analysis, it would be easier for statisticians to analyze data if they fully get necessary and clean data. However, I also agree with several opinions by classmates that good collaboration is important. Additionally, the team should consult an expert on analysis workflow/any possible errors and also training once initiating any innovative technology as suggested by Arwin.
-
I agree with the recommendations stated in the article. I also agree with several opinions from classmates that it is difficult to do in reality. Hierarchy and seniority are presented in some Thai organizations. Also, organizational culture is very important, which could affect people’s courage to speak up and make changes.
I think the role model and creating an organization’s culture for anti-corruption is crucial. It would be great to provide a hotline for reporting suspected events anonymously, creating a clear workflow/SOP, and recording evidence clearly while performing tasks. Therefore, it would be easier to retrieve and track the action.
-
An example of a health system improvement based on my experience is the Online Adverse event reporting system implemented by the Thai FDA. Previously, all hospitals and pharmaceutical companies needed to report adverse events to the FDA using the paper-based form and send it to the FDA. Then, FDA officers would document it manually. It took time and workforce. It could also delay the response and management plan when serious AEs have occurred after the Online AE reporting system is implemented. All sites can report spontaneous AE individually and save time intaking data in the AE system. Additionally, the FDA can analyze the data on a real-time basis and respond to any critical events on time.
Several factors could affect the process of implementation. First, healthcare professionals might not be familiar with the system when it first launched. Due to the high workload of healthcare professionals, reporting AEs to the AE system would create an extra workload and take time for them to learn how to use the system. In addition, all data filled in the AE system should be standardized. Therefore, the users need to understand the structured data in order to complete the reporting process. Moreover, creating a new online system would require financing for initiation and maintenance costs. Also, the hospital leaders need to agree and advocate for their employees to use the AE system.
-
In my organization, I think data replication and backup are very important, especially in cloud computing (we usually use Onedrive now). We rarely use hard copies, so I think it is better to use technology to store and backup softcopies of the documents. Also, we should have a business continuity plan in place, to make sure all employees understand the procedures and are aware of the disaster.
-
Implementing the High Availablity technology in the hospital information system (HIS) would provide benefits in many ways. It will provide instant data availability, therefore no missing data during the process, and every department can monitor and link data on a real-time basis. It can also lower medication errors and prevent duplicated orders/therapy that help for patient safety. With the well-organized system, it will also reduce waiting time and workload for workers, provide better performance.
-
I haven’t experienced any CIA issues but I have a similar situation with Karina. Sometimes, we need to collect personal data or CV/Resume for project registration and for internal processes. Therefore, we redacted all unnecessary sensitive data before using the documents and also protect confidentiality by limiting access to the data.
-
-
I love it!!. I am not actually a fan of a black background, but you designed it well. It looks nice with the light text!!. The table and map/graph are well-designed and capture all necessary information. I have only one comment. I think the recovered/deaths data in the table represented cumulative numbers rather than daily cases (except I filter only 1-day-data). Therefore, It would be nice to adjust the table headlines for recovered/deaths. However, please let me know if I misunderstood.
Well done!!
-
-
Good point!!!. I found several questionnaires and other data collection forms providing choices only yes/no, negative/positive without “not applicable: option. The good point is it is mandatory for users to choose either choice but the disadvantage is like you mention. For someone who actually don’t know about that or no data, the section is likely to leave blank.
-
-
-
-
-
-
-
-