Forum Replies Created
-
Posts
-
-
-
-
New technologies might have more benefits and usefulness compare to existing technologies. However, the change can be step-by-step, start with small pilot setting and then scale up to the whole organization when we see the good results and good practices in real life. This method can allow us to compare results and performance both new and old technologies as the same time and can gain more acceptance when people see the obvious difference.
-
I think social and peer influence may play an important role on an individuals’ perceived ease of use or perceived usefulness of a new technology. I also think ecological models which recognize multiple levels of influence on human behaviors, including: Intrapersonal/individual factors, which influence behavior such as knowledge, attitudes, beliefs, and personality can explain what can affect perceived usefulness of a new technology both external and internal factors as a holistic way.
-
-
I really like the discussion of common misinterpretations of power on page 345. On the point 24, “If you accept the null hypothesis because the null P value exceeds 0.05 and the power of your test is 90%, the chance you are in error (the chance that your findings a false negative) is 10%.”
This is very normal misinterpretation even I took many courses of Statistics but still miss sometimes. It is something that we really need to give in mind that we need to state “If the null hypothesis is true”, the chance you are in error is 10%. However, if the null hypothesis is false and we accept it, we are 100% in error.
Thank you for a nice wrap up paper!
-
-
I am a Public Health officer working for the Division of Mental Health Promotion and Development, the Ministry of Public Health,Thailand. I have been applied statistics to my work work from time to time during the process of data analysis to support policy planing and development. Also I have been involve to the research project of my department. So I have to use a lot of statistics and I use STATA program to analyze data.
-
Age-specific mortality rate
Definition
An age-specific mortality rate is a mortality rate limited to a particular age group.Calculation
The numerator is the number of deaths in that age group; the denominator is the number of persons in that age group in the population. For example, in the United States in 2003, a total of 130,761 deaths occurred among persons aged 25–44 years, or an age-specific mortality rate of 153.0 per 100,000 25–44 year olds.Usefulness
Age-adjusted rates allow you to compare health statistics (like death rates) between population groups, even though the size of the groups or the age of group members might be very different. -
I have a question while doing first two assignments and reading some articles about decision tree classification. Actually, I do want to try the data mining techniques that we have learned on a large dataset of mental health and some socioeconomic factors that I have. Base on my best knowledge, if I just want to see a pattern and the association of the clusters that I found, I can use clustering techniques and some statistics and if I want to classify and create a model to predict the mental health issues, I can use a decision tree technique (you guys can help to correct my understanding 555). Decision tree has its way to test the performance of a tree model but when I read article about this method they always compare the performance of their models with other methods such as eXtreme Gradient Boosting, Elastic Net, Quantile Ordinal Regression – LASSO, Linear Regression, Ridge Regression. My question is do I need to compare my tree model with some of these methods?
-
-
Ebola also known as Ebola hemorrhagic fever, is a rare and deadly disease caused by an infection of one of the Ebola virus strains. Ebola can cause diseases in humans and nonhuman primates (monkeys, gorillas, and chimpanzees). EVD is mainly transmitted through direct contact with infected bodily fluids and contaminated materials. At the early stage, the disease is characterized by initial-flu symptoms, high fever, severe headache followed by pharyngitis and abdominal pain, whereas the late stage of the disease is marked by vomiting, diarrhea, rash, and internal and external bleeding. The disease was first identified in Sudan and Zaire in 1976 where it infected over 284 people with a mortality rate of 53%, and subsequently, few months later after the first outbreak, another outbreak emerged in Yambuku, Zaire, affecting 315 and killing 254 people.
Model structure:
The SEIR model, which have 4 compartment in the model;
1. the susceptible compartment S,
2. the exposed compartment E,
3. the infected compartment I
4. the removed compartment R (which includes all individuals who either recover from the disease or die).
N is the total system population.
𝑁(𝑡) = 𝑆(𝑡) + 𝐸(𝑡) + 𝐼(𝑡) + 𝑅(𝑡)Constant parameters and their numerical values appearing in the epidemic model
λ Rate of increase (recruitment rate) of susceptible humans 0.6321
μ Natural mortality rate of susceptible humans 0.9704
β1 Rate of infection from susceptible to exposed humans 0.2877
β2 Rate of infection from exposed to infected humans 0.7613
β3 Rate of infection from infected to recovered humans 0.4389 [
β4 Rate of infection due to wild animals from susceptible to exposed humans 0.1234
β5 Rate of infection due to wild animals from susceptible to infected humans 0.2431
β6 Rate of infection due to domestic animals from susceptible to exposed humans 0.4000
β7 Rate of infection due to domestic animals from susceptible to infected humans 0.3000
μ1 Natural mortality rate of exposed humans 0.0432
μ2 Disease induced mortality rate of exposed humans 0.2006
μ3 Natural mortality rate of infected humans 0.0656
μ4 Disease induced mortality rate of infected humans 0.9764
μ5 Natural mortality rate of recovered humans 0.6704R-code
library(deSolve)SEIR.dyn <- function(t,var,par) {
S <- var[1]
E <- var[2]
I <- var[3]
R <- var[4]
N <- S+E+I+Rbeta1 <- par[1]
beta2 <- par[2]
beta3 <- par[3]
beta4 <- par[4]
beta5 <- par[5]
beta6 <- par[6]
beta7 <- par[6]
gamma <- par[7]
sigma1 <- par[8]
sigma2 <- par[9]
sigma3 <- par[10]
sigma4 <- par[11]
sigma5 <- par[12]
sigma <- par[13]#Differential equations
dSdt <- gamma-sigma*S-(beta1+beta4+beta6)*S*E-(beta5+beta7)*S*I
dEdt <- (beta1+beta4+beta6)*S*E- beta2*E*I-(sigma1+sigma2)*E
dIdt <- beta2*E*I+(beta5+beta7)*S*I-(beta3+sigma3+sigma4)*I
dRdt <- beta3*I-sigma5*R
list(c(dSdt, dEdt, dIdt, dRdt))
}beta1 <- 0.2877
beta2 <- 0.7613
beta3 <- 0.4389
beta4 <- 0.1234
beta5 <- 0.2431
beta6 <- 0.4000
beta7 <- 0.3000
gamma <- 0.6321
sigma1 <- 0.0432
sigma2 <- 0.2006
sigma3 <- 0.0656
sigma4 <- 0.9764
sigma5 <- 0.6704
sigma <- 0.9704SEIR.par <- c(beta1,beta2,beta3,beta4,beta5,beta6,beta7,gamma,sigma,sigma1,sigma2,sigma3,sigma4,sigma5)
SEIR.init <- c(0.1,0.1,0.1,0.1)
SEIR.t <- seq(0,90,by=1)SEIR.sol <- lsoda(SEIR.init,SEIR.t,SEIR.dyn,SEIR.par)
TIME <- SEIR.sol[,1]
S <- SEIR.sol[,2]
E <- SEIR.sol[,3]
I <- SEIR.sol[,4]
R <- SEIR.sol[,5]
N <- S + E + I + Rplot(0,main= “EBOLA SEIR Simulation”, xlab= “Time”, ylab= “Number of population”, col= “Blue”, xlim = c(0,90), ylim=c(0,1000))
lines(SEIR.sol[,2],type= “l”, col= “Blue”)
lines(SEIR.sol[,3],type= “l”, col= “Red”)
lines(SEIR.sol[,4],type= “l”, col= “Green”)
lines(SEIR.sol[,5],type= “l”, col= “Yellow”)legend(locator(1),legend=c(Susceptible,Exposed,Infectious,Recovery),col=c(‘Blue’,’Red’,’Green’,’Yellow’),lty=rep(1,4))
Reference: Rafiq, M., Ahmad, W., Abbas, M. et al. A reliable and competitive mathematical analysis of Ebola epidemic model. Adv Differ Equ 2020, 540 (2020). https://doi.org/10.1186/s13662-020-02994-2
-
What intervention(s) you are considering in your modelling?
I chose Ebola as my study case. I would like to study the early response or intervention for Ebola. In August 2014 the WHO published the Roadmap for response, outlining three phases of response initiatives to combat the outbreak. In October 2014, during the first phase, the UN Mission for Ebola Emergency Response (UNMEER) was launched. UNMEER had several aims: that 70 percent of cases would be isolated and that 70 percent of the burials would be conducted in a safe manner. Approximately 2 months after the UNMEER initiated interventions were implemented, the national weekly case counts decreasedHow it will be added to the model structure?
I would like to explore the potential benefits of the intervention by looking at the time of implementation by moving the time of interventions in the transmission model.What are the characteristics of the intervention(s)?
The characteristics of the intervention include the transmission parameter and also the hospitalization rate and the case fatality rates for those hospitalized. -
Topic: Suicide
Rationale:
Suicide and suicide attempts constitute major public and mental health problems in many countries. The risk factors of suicide include not only psychological and other individual features but also the characteristics of the community in which the people live. Therefore, in order to better understand the potential impacts of community characteristics on suicide, the regional level effects of suicide need to be thoroughly examined.Research question:
How does reginal risk factors and discern spatial patterns in suicide risks affect suicide rates in each region in Thailand? and Can these risk factors predict reginal suicide rates?Reference:
Phillips, J. (2013). Factors Associated With Temporal and Spatial Patterns in Suicide Rates Across U.S. States, 1976-2000. Demography, 50(2), 591-614. Retrieved September 7, 2021, from http://www.jstor.org/stable/42920539Congdon P. The spatial pattern of suicide in the US in relation to deprivation, fragmentation and rurality. Urban Stud. 2011;48(10):2101-22. doi: 10.1177/0042098010380961. PMID: 22069804.
-
It was a very informative presentation about blockchain technology for health. Which brings so many points that we need to think ahead before adopting in Thailand such as training, acceptability, private and security. It will be a big change that will scare stakeholders but its potentials that you present are the opportunity to overcome many health system challenges!
-
https://tinyurl.com/24b32ch8 here is my dashboard!
I selected only important key indicators that I think it is in public interest to explain the COVID-19 situation. And I used simple graphs to explain the data for general public.
Feel free to comment my dashboard. -
Here is my dashboard https://tinyurl.com/2n9d2ab2
The following are benefits of each visualization for Policy decision making;
– visualize the pattern of the outbreak in each continents overtime;
– Comparing the total number of confirmed cases, recovered, and deaths;
– I used Ribbon chart to see the different of total cases between 2020 and 2021;
– Scatter plot with the play fiction is the excellent to see correlation between 2 variables. -
See my baby step of creating a dashboard by using PowerBI here >>https://tinyurl.com/ryp22874
Benefits for decision making;
– International resource allocation such as vaccines, masks, PPE to low-and middle income countries;
– International Public health measures such as traveling rules. -
Link: https://coronavirus.jhu.edu/data/hospitalization-7-day-trend/inpatient-capacity
This dashboard is called Coronavirus Resource Center, created by Johns Hopkins University.
What I like?
– It provides well-analyzed data and information. I think it is a good example of how to visualize data, interpreted data, and disseminate with explanation texts properly,
– it track how the novel coronavirus is spreading around the globe with up-to-date visuals that give context to the data collected on Johns Hopkins University’s COVID-19 map,
– Data in Motion is brilliant! with a short video of data makes their data even more interesting!
– They have topics names for all data that they presented which is easy to catch my eyes for what I want to search or read on their dashboard like an online magazine.
What I don’t like?
– some graph is too fancy, I need to spend sometime trying to understand what it mean. However, with text explanation can solve this problem,
– seem like their analysis is for academic groups, some contents is hard to understand by general people which is sad because it all good contents! -
Here is my summary! https://ibb.co/N7BsdF1
-
-
Here is my wrap-up of week 2, check it out! https://ibb.co/2vbHnXd
-
Here is my exotic summary!
http://prnt.sc/12yb1x5 -
– Related to research objectives
– Avoid directly identify birth date of participants
– No redundant questions
– Even you have some open-ended questions but I think it’s okay to have there
– You did a good grouping related variables
– Good design for multiple answers Q
– Identify correctly all unitswell done!
-
-
-
-
All of these processes- audit trial/time stamp, access control level, edit check and logical check, data backup and recovery plan, are crucial for data quality and integrity. Unfortunately, I never have any experience on clinical research or any research that use these processes. However, I do have some experience on a systematic review which is my dissertation when I was study my first master degree. I adopted the PRISMA guideline to conduct my systematic review which had a data management flow like we are studying.
For example;
– On the step of searching for relevant studies, I needed to stated the date of my search along with my search strategy;
– I had my peer to cross check my abstract and title screening and extracting data.
Even I don’t have an experience on clinical research that use these full steps, but I do have an experience of using data from Web-application for mental health screening which has a strict protocol and policy on data privacy and security. So I need to get a permission from a committee of my department to get a password before accessing to people’s data. I do believe that we have data backup plan but I just don’t know how my department storage the data. I need to ask them next time 🙂 and I just found that we really cannot do editing check and logical check when dealing with a large data set, right? -
I only have experience on a small research which was not clinical trials and did not involve colleting any biomedical data.
Many steps of data management workflow I have done as the following;
– Protocol discussion
– Data design
– Data Acquisition (Paper based only)
– Data entry and processing
– Data Manipulation and analysis
– Study report
As Ajan Saranath mention, database access control and data security were tasks that we needed to performed and it was a requirement from our ethic committee.
If I have a chance to go back to improve my project, I would like to:
– Develop a data management plan
– Do QC and QA
– database access control
– database lock and security -
I do have experience on data collection for research proposes. I have been involved a implementation research of a community-based mental health practice. We used mix-method research design to explore both implementation and effectiveness outcomes. We decided to collect primary data both qualitative and quantitative method. For qualitative data, we used focus groups to collect data and we used closed-questionnaires to survey for quantitative data. There were many challenges during data collection process as the following;
– Data collector training which we have to ensure that all collectors have enough knowledge and skills before collecting data.
– Instruments- there are so many process to test and ensure quality of your instruments especially, a questionnaire.
– Data missing -
A IT team used to be just a team who only take care health professionals’ computers. Currently, with the advanced in technology and data science, IT teams in many organizations are responsible for data management and system development with health professionals, public health officers, epidemiologists, and statisticians. However, IT professionals and health professionals often viewed as separate. This is a big challenge when we need some one who have all knowledge of IT, Public Health, and data science. A leader with the vision is also one of the most important challenge to build capacity of health information workforce. Which contribute to the opportunity of their workers to be trained and work in the right job.
-
I have hold a data set of mental health problem screening during the COVID-19 outbreak. Since the web-based application has launched in April last year, there are more than 1 million people across the country reach it. We have been trying to produce value added data and publicize it nationally on our dashboard aiming to rise awareness of mental health problems, and also use it as a data base or evidence base to advocate on mental health policy action. Importantly, we would like to share this data to local policy makers. So they can understand the mental health circumstance they are in and take an action to address their problems.
-
I think the problem is not about the money, if higher spending lead to better outcomes. However, that’s not the case in the United States. Despite significantly higher healthcare spending, America’s health outcomes are not any better than those in other developed countries or developing countries. The United States actually performs worse in some common health metrics like life expectancy, infant mortality, and unmanaged diabetes. And yes, like Kridsada mentioned, capitalism health system might lead worse outcomes than other designed health systems.
-
This really happened in my department! We have a mental health large dataset like 2 million responses. Some guy took around 50k responses to analyze associations between variables.They found associations with a small p-value. They interpreted the results and used it without validating with other information and did’t aware that with that large number of sample, every thing can be statistical significance.
-
-
-
-
-
-
-
-
-
-
-
-
-
Such a good question! Personal data on this application though out processes have been protected following Personal Data Protection Act, 2019. Moreover, We set Personal Data Protection Policy of the Department of Mental Health and inform every one who access to this application that they have right to refuse or give us a consent form to use their data with our data protection protocol.
-
-
You have risen a very good example of data sharing! Surveillance data, especially the data of communication diseases is crucial to control the outbreaks and to prevent the pandemics. I think many countries have concerned about the issues of data sharing that you mentioned. They have been trying to set a standard on data collection and promote it to other countries. For example, in Europe, disease surveillance networks have been in operation since the 1980s and are currently integrated and managed under The European Surveillance System (TESSy), a web-based technical platform which collects and disseminates national surveillance data from all European Union (EU) and European Economic Area (EEA) countries using standardized formats.
-
According to your discussion about inadequate access point, Just thinking that before this kind of the national projects kick off and implement, they should study feasibility of the project and do a pilot to see what happen in the real world setting. Which might be able to prevent some basic problems like quality of hardware, access point, and maintenance plan.
-
