Forum Replies Created
-
Posts
-
-
-
The qualitative research is the tools to finding in-depth answer in some question like this one. The traditional quantitative tells only how many household did not use bednets, but no reason was told. Sometimes, transformation of qualitative attributes to quantitative was not enough to tell the true answer. In-depth interviewing, observation, participant observation, or shadowing might be useful to finding the truth.
-
-
-
-
-
specific attack rate (Actually “specific incidence proportion”)
Definition – the incidence proportion (attack rate) in specific, interested population/sample
Calculation – number of cases group by specific attribute / number of population group by specific attribute
Usefulness – specific attack rate give more information about occurrence of disease such as the attack rate of influenza-like illness among students and teachers in the school, is the the different between students and teachers? that could lead to hypothesis generation and testting.
-
page 342
12. P values are properly reported as inequalities (e.g.,report ‘‘P< 0.02’’ when P= 0.015 or report ‘‘P> 0.05’’ when P= 0.06 or P= 0.70)there was a big mistake, it was easier to report p-value directly. Another reason is it could be differ if we use different dataset. However, the very small p-value should report in inequality form.
-
My background was the epidemiologist trainee. So, every assignment in the course always use statistics. From descriptive statistics like counting, frequency, central tendencies measurement to inferential like survey, sample and population, estimation or hypothesis testing. We also enrolled in the basic stat courses and advance courses depends on demand for each trainee.
-
-
My dashboard is here.
Please feel free to review and leave some comments.
-
I try to make some of graph using provided variables. My dashboard divides into three tabs: cases, deaths, and recovered. That’s could reflect the situation around the world.
-
My dashboard contains daily cases and total accumulated case through the time, also selecting by country and range of GDP
-
Here are my wrap up image.
-
I would like to present this dashboard from WHO
Likes:
– The fancy dashboard that contains overall situation and by countries situation
– Interactive data presentation
– Present more than case number, they are death number transmission status and vaccination status.
– Also presenting trends of disease progression in the world and countries
– All graph and chart are already decluttering.
Dislikes:
– It is not containing in 1 page, need to scroll down for more information.
– Sometimes provide too much information.
– Use too many colors in the same page. -
-
-
The image is not show.
You can see the image by follow THIS link. -
-
-
Actually, I never conduct the full RCT or research in traditional way, but I will be shared some of my experiences. To prevent any accident, the data backup was created as soon as full data set was acquired in my recently work. I stored my data in three places; my laptop, external hard drive, and google drive platform. But I did not establish the data recovery plan yet. For checking and validating data, I started these processes as the exploratory data analysis step. It did not follow the best practices because there must be check during data entry, but I have not well planned enough at this moment.
-
According to the processes, I found that I missed many processes to ensure the validity, accuracy and reliability of data. At first, I extracted raw data from EMR instead of providing CRF to study site. It greatly affected to data quality also main result. Then, DMP was not develop before study was initiated, it made some confusion during the study period. Database access control was not set up either because I thought only me could access this database and it located in my laptop. One day my laptop broke, and I became furious. However, I have done many ways to ensure data quality, manipulated and analyzed my data to get final result of my research such as EDA, checking validity, read and understand my data over and over again. This experiences and lectures made me clearly understand about data manage process.
-
I would like to share the public health surveillance evaluation. According to the objective, we would like to measure quantitative (sensitivity, PPV, timeliness, validity) as well as qualitative data like acceptance of the healthcare personnel and simplicity of the system in performer view. Either primary data from interviewing healthcare personnel or secondary data such as EMR (which transformed into CRF) were collected. The problems were classified depend on data collection and source. For EMR data, we encounter uncompleted, invalid data that could be interfere the main result. We also experienced unclear, difficult to understand question during interviewed, due to the limited of time, we were not testing the questionnaire as well.
-
In my opinion, informatics workforce in Thailand still open for everybody, there are many problems that need health informatician to explore, design solution, and implementation. The main challenge is how to implement interoperable data sharing between healthcare facilities. If we succeeded, seamlessly data sharing will improve data accuracy, validity, and immediately decision to response to situation. It also improve data security, privacy and confidentiality that enhance data creditability too.
-
Data sharing is actually benefit at all. Some of most common diseases in Thailand might be very rare disease in other part of the world. They could use these data for research, education, or references. However, data sharing policy must be made for protect, provide confidentiality to data source itself, such as identifier concealment, encryption, specify data utilization policy, or selective data for sharing.
-
From my little knowledge, American people have to pay larger amount for healthcare than other country because they do not have health coverage for all people. Only some group of people like employees, elderly with many health conditions, or veteran soldiers have health insurance from central government. Other people must seek their own health insurance. This system originates from individualism and liberalism, they believe health care is goods or services which individual freely choose to access and individually made. Another reason is universal health coverage is the symbol of socialism, it strongly against liberalism. Even thought they have health insurance; some group of people might have to pay some service fees to hospital or pharmacy because of co-payment system. However, universal health coverage is not fit for all country. There are many factors to consider for establishing health coverage or health system. It take time for advantages and disadvantages studies and shaping the health system.
-
In my opinion, problems of net pracharat are inadequate access point. Somebody did not want to leave the house or properties to access the internet. In theoretical, 30 Mbps. is fast enough for loading songs from internet within few seconds, streaming at 720p, even video call and conference, but still have complain about speed of internet. I think this issue might occur from low quality hardware. It affected stability of internet connection and interrupt performance of internet. Another concern of net pracharat are data security and confidentiality. If someone want to attack this network, the vulnerable people such as children, elder with low digital literacy will become easy victims. Internet traffic also easily tracking by bad person, it easily know what they search, query, or using social network.
-
Using EMR make benefit for both provider and patient. In example, for provider, we could service same routine but faster, more decisive, and safer to patient. Data in EMR has more security, privacy, and confidentially if we use the right technique to apply in database. For patient, they will receive faster service and safety treatment. If possible, they could know their health status, and receive health suggestion for some issue like hypertension or DM lifestyle modifications. In the downside of applying EMR, bad things are how to start using EMR with minimal resistance, we could cover this problem by good change management. Secondly, the implementation phase. How to implement EMR in the hospital, where to start first, and how to transform into new system seamlessly.
-
Suggestion for dealing with challenges:
1. Missing data – to dealing with missing data, we have to explore at upstream of data, The data sources. Try to define what is the main problem of data missing such as some variable is out-of-date, some variables are hard to entry or understanding, some variables is not to be added at first. If we try to understand these problems, we could reduce missing data at downstream. However, missing data is inevitable. We could use statistical methods to deal with them like imputation or complete-cases analysis but be caution, we are introducing information bias into the result.
2. Selection bias – selection bias usually occurs when sample size is too small and specific in some of populations. To dealing with this issue, we could identify and validate data in our data and data in general population in which representative or not. We also increase sample size and recruit more eligible samples to reduce bias.
3. Data analysis and training – In general, big data could not use old-fashioned method to manipulate and analyzing. So, we have to provide infrastructure for data storage, manipulation, analyzing, and interpretation insight from data.
4. Interpretation and applicability of result – we need to use result to build an application or optimization of process in healthcare, or visualize result, and insight of our data.
5. Privacy – data encryption is the one of standard method to protect healthcare data, need to apply in all level of operation. Another issue is authentication and authorization management by using methods like 2FA, Access control, or logging issue -
1.Disseminate information – every change start within person, giving the knowledge and information to changing attitude and behavior on corruption and also increase addressing skill on corruption.
2.Improve detection and enforcement – after all personnel has skill to address corruption within organization, this issue is about how to build confident of personnel to report every corrupted actions. The enforcement and punishment of corrupted act must be appropriate and equity.
3.Increase transparency and accountability – transparency and accountability are how to manage organization clearly and could tracing every action and decision without any influence from everything.
4.Reduce monopolies – monopoly destroy transparency in organization. It easy to make corrupted action or decision within influence of some benefits. Reducing monopolies not only reduce corrupted act but also improve quality of product to appoint in organization, and generate benefit to customers. -
As we frequently seen in hospital setting whatever provincial or rural hospital. Today trend is established electronic medical record system. With this improvement, we can provide more service to the patient with more efficiency and safety. EMR could change how we store medical records into privacy and secure platform. EMR also guide healthcare personnel to perform service harmoniously. However, change always create barriers. Possible barriers could occur are investment, data migration and change management. In some setting, investment might be the most important issue even you plan to set on-prem or cloud platform. You have to convince the executive board to comply with your project. Data migration is later consideration, how you migrate paper-based data to electronic-based; manual entry or another. Lastly, change management is how to convince all level of healthcare personnel to believe this system is boosting routine efficiency and willing to implement this system in routine workflow.
-
My organization is the huge scale data lake of Thailand health data center. Health-related data was generated largely every day. It is also sensitive to long downtime because of large number of end-user. So, in my opinion, it prefers to use data replication server to use as disaster recovery because it could serve as backup system faster than data backup and recovery, easily scalable by investing server and storage or using cloud providers. The downside of data replication technology are they are harder to implemented and vastly investment of infrastructure.
-
If we implemented high availability technology for HIS, there will be benefit in both patient and hospital. For patient, they will get faster and reliable service from hospital. these benefits come from minimizing downtime that could interrupt patient service in the hospital interface, also RAID and load balancing technology that improve speed of connection within hospital. The hospital will get benefit that they can work seamlessly even some of system is fail. They could assure that the crucial data are not destroyed, and their server are not overload until physically damage by using load balancer.
-
According to my experience, it was not about information system directly, but I prefer to share some story. I ‘m not sure what happened exactly but I believed that my Facebook and Line account was hacked by someone. My confidentiality was broken. It was sending some message to somebody that I did not write it. It affected me about the credibility of my account and my personal. So, I decided to suspend both Facebook and Line account and make the new one. At this time, I use new set of passwords contained around 15 character and log out every time I did not use my personal PC or mobile phone.
-
In the real world, if other organizations request for the raw data of the system like surveillance data, research question, or survey data. They should submit the proposal that contain such as the objective of the study, what and how much the data will used, and how to protect the data. This proposal will be going to the ethical committee board for consideration to authorize the data access. In this situation, I will follow these protocols in order to ensure the data privacy and security. The data will be use in the beneficial way, to generate new knowledge, enhance the system, and lead to disease elimination in the country.
-
In this situation, I choose not to tell my friend, because I will respect to patient autonomy and confidentiality. In addition, it strongly against the Personal Data Protecting Act, it is illegal. In this case, only attending physician have right to disclose the data to my friend’s husband. After all, their attending physician will follow the HIV counselling guideline and giving advice to my friend respectively
-
From the previous, I think my organization did not rely on ADKAR as much. They communicated in the board that we are going to use the new imaging system. They did not address what is the new system, why we use the new one, or risk of not use this system. However, the operation team from software vendor provide training, and coaching the technicians from radiology department including nurses and physicians. Finally, all healthcare personnel were used this system seamless with the routine work. Technicians got the support from software operation team occasionally to evaluate the work, maintenance, and remote support.
-
In couples years ago, my hospital implemented the new radio-imaging system. To convert from film to the computer image file. It made clearer, higher quality, and flexible image. In my opinion, the success of this system was result of many factors. The data of this system was quite simple, and demand by users. System design was usable, met the objective and secure enough for confidentiality. The operation section was stable, also easy to learn too. All healthcare personnel were happy to use, not increase the workload and interrupted the routine workflow.
-
DSS in my experience from the past usually comes in the reminder and alert. The reminders are the pop-up message from the previous visit after the patient’s interface was selected. The alert comes in many forms such as critical lab result alert, drug allergy alert, and drug interactions. These features aim to reduce the clinical error and improve quality of care. This system is user-friendly and did not interrupt the current workflow. In my opinion, they are influencing factors to implement DSS.
-
ICD system serves as the standard that all healthcare facilities communicate the same language. If there is no ICD system, healthcare facilities were not sharing the data for each other. Also, the transfer the data to the higher level. So, we cannot monitor the situation of some disease because we use ICD for query the data. It will hard to refer the patient between the hospital, it will cause misunderstanding in the diagnosis between hospitals, affect both patient safety and quality of care.
-
I used to work at district hospital that use fully EMR before. I encounter some problems like program did not allow me to prescribe some drug but unclear reason, program crashed sometimes, and patients complain that physician did not pay attention to them. During FETP training, I explored the EMR systems in the hospital. I found the most issue from user was the redundancy. They had to entry the data too many times in the different systems. Another problem is how to communicate or give feedback the problems to the developer and executive level. To minimize these problems, we need to start from the designing phase, getting all requirement from all stakeholders as much as possible. The good information system will cause benefits more than chaos.
-
-
I think the MOPH’s HDC consider as a big data because it fits for 5Vs. The data generate from all health care provider such as hospitals, health promoting hospitals, it is a mass volume of data. The data generate every day and sending to data center each month as batch processing. It can be trustful and origination. It has own value and many purposes like monitoring, analyzing pattern of health status, supporting policy decision, and input for machine learning. Moreover, it lack of variety because it contain only structured data, no JSON, image, video is collected.
-
I have seen and be a user of “HDC” or big data MOPH that provide raw data of patient visit in Thailand, also aggregate data like KPI, dashboard. In raw data part, I used for my project and some data analysis.
This information system could provide overall health data in Thailand which use to monitor health status, provide knowledge after analyzing, also predict some future problem via machine learning. This system has difficulty in data collection and data cleaning. For example, each hospital sending the data, but they do not know the quality of data that they sent, sometimes they sent the duplicate data unintentionally. It makes data cleaning process to be more difficult and sometimes damages the data structure. -
In this case, I will propose the board to use cloud server because of some reasons. The main reason is manpower, this hospital has only an IT officer which will be struggle if we decide to use on-prem server, it hard to established, maintenance, and troubleshooting. Cloud server has less expensive to set up, no need IT officer to maintenance and easily scalable when needed. Cloud server also has high level of security protocol, in both physical and logical if we use hybrid or private cloud providers. I will propose the board to use cloud service as software as a service (SaaS) because of the manpower too. Only one IT officer cannot develop, test, deploy, set up infrastructure, monitor and maintain the service. So, we have to use the package software of cloud provider for making everything runs. It could be some problem in flexibility issue in the software but, there is the better way to initiate the cloud-based application. In the future if we have a larger, expert IT team. We can develop our own application and migrate to cloud service as platform as a service (PaaS) or infrastructure as a service (IaaS) to save cost and more flexibility.
-
Enterprise architecture (EA) could be divided into 4 parts. First, business architecture should involve all stakeholder of the project to defined goals and problem which the project needs to solve, planning steps, design workflow and organization of the enterprise. Then, information architecture which involve database specialist, data engineer, data analyst or data scientist. These guys should define what is the data and information need to collect, how to storage, process, and utilize the information. Third part is application architecture might be consisting of developer, programmer in order to create the software or application to support the information team and create for solving the problem. Lastly, infrastructure architecture, which refer to tech-leader, IT specialist, infrastructure specialist to establish and request for the hardware, software, network in this enterprise. All of four parts should be on the same way to develop solution which solved the existing problems.
-
In this case, the absence of the enterprise architecture (EA) cause the major point, lacking interoperability. Every hospital has their own UI, database but no sharing for each other. In the smallest unit, it could affect the continuity of patient care if the patient shopping doctors around. In larger scale, officers in each hospital need to work redundantly for sharing the data to the national data center for monitoring disease situation and surveillance. It could cause intolerance if we change or remodeling the architecture of data center, surveillance program or the policy. EA comes to relive this problem by creating the unified blueprint that stable enough for working but flexible for changing. This blueprint will be the goal or final shape of the system which all hospital and organization will be develop into together.
-
I worked as the epidemiologist in residency training program called “FETP” in past 2 years. I used to play with some secondary data especially in Hypertension, explored the information system of MoPH and faced some challenges too. So, the public health informatics is the consist of public health knowledge such as disease and prevention, information technology (computer science and tech-stack), and soft skill range from managing resources, leadership, communication skills and lifelong learning.
-
Yes, of course. The main reason are the security, cost and maintenance and support issues. Cloud provides better security than the on-prem database like strong authentication and authorization. It cost lower and do not need maintenance by IT support, help desk in the hospital. To establish on-prem data server, we must invest a lot of money for server, data storage device, networking etc. if you want to scale up on-prem database you must invest your money again. But, cloud use cloud engineer to establish the infrastructure and database and programmer for deploy application. Cloud also easily for scaling up if you want the bigger database.
-
Another kind of chart for this week.
-
