Science topic

Clinical Data Management - Science topic

Explore the latest questions and answers in Clinical Data Management, and find Clinical Data Management experts.
Questions related to Clinical Data Management
  • asked a question related to Clinical Data Management
Question
4 answers
Hi, 
I am looking to explore the association of  genes with clinical outcome, 
I tried a bit the Oncomine site, however I will be happy to know about alternatives.
Thanks, Ran
Relevant answer
Answer
Try www.kmplot.com. The Kaplan Meier plotter is capable to assess the correlation between the expression of all genes (mRNA, miRNA, protein) and survival in 30k+ samples from 21 tumor types including breast, ovarian, lung, & gastric cancer. Sources for the databases include GEO, EGA, and TCGA.
  • asked a question related to Clinical Data Management
Question
4 answers
Hello,
I am looking thesis topic in clinical research management. I am planning start an internship at CRO company. Areas of interest- Comparative study or retrospective study; GCP , Adverse events reporting or clinical data management etc. Any suggestions will be helpful
Thank you
Relevant answer
Answer
thanks for the information. I am right now digging into a few areas, need to narrow it down to one or two. I am too curious and have an interest in adverse events reporting. I will definitely look more into it. Danial Hassan
  • asked a question related to Clinical Data Management
Question
5 answers
Today, already two vaccines has shared promising early findings. The Pfizer’s vaccine showing more 90% of efficacy and Moderna vaccine showing 95% efficacy.
The current predictions continue to project the availability of a safe and effective vaccine by the next winter. However, given the role of Data and Safety Monitoring Board (DSMB) towards early trial termination either for safety, efficacy, or futility.
What would be the role of DSMB in the COVID-19 vaccine discovery towards stop for efficacy rule?
Recently, infecting trial participants to speed up COVID-19 clinical trials has been under discussion. Given the efficacy of the mentioned vaccines, would be ethical to infect participants from those trials given that the placebo group could be injected with saline or vehicle used for COVID-19 injection?
Relevant answer
Answer
Patrick Gad Iradukunda The closest to a controlled scientific study is the human challenge studies, of which the two vaccine trials are not. In the human challenge studies, volunteers are deliberately infected with a known strain of the SARS-CoV-2 and kept under closed surveillance environment. One group will receive the vaccine and the other will get the placebo and you monitor the results. Both vaccines did one step closer to this by testing on monkeys and got good results. I believe this is a strong indication that it would work on human as well although in some cases results from monkeys cannot be translated to human. see
  • asked a question related to Clinical Data Management
Question
4 answers
I am working on identifying biomarkers of the outcome of treatment in a disease. I have 20 potential biomarkers and more than 30 potential confounders. Actually, I know that some of those potential confounders affects treatment outcome. However, before determining the association between each biomarker and the therapeutic outcome, I want to know which of these confounders affects the expression level of each biomarker. I want to understand the impact of covariates on the expression of biomarkers to include them as confounders in multivariate analysis.
Relevant answer
Answer
some info in the link
or directly in the attachment.
Knowledge-based approaches seem more appropriate if your field is not completly new.
  • asked a question related to Clinical Data Management
Question
19 answers
I'm looking for a COVID-19 updated database containing patient information (age, sex, date disease was diagnosed, date of death/recover, etc). Is there any?
Relevant answer
Answer
Dear Daniel, there is a wonderful R package that includes data on COVID-19:
Best wishes
  • asked a question related to Clinical Data Management
Question
4 answers
Hi, I am new on SPSS, I hope you can provide some insights on the following. I have a data base of patients which contain multiple variables as yes=1, no=0. I want to work on this data based on multiple cases selection or subgroups, e.g. patients with variable 1 (1) which don't have variable 2 (0), but has variable 3 (1) and variable 4 (1). Variable 4 includes selected patients from the previous variables based on the output. In other words, let’s imagine we have a database from 10000 patients with crohn’s disease, I want to select ulcer location (loc-1, loc-2, loc3 and loc-4), for later comparison. However, the patients, based on ulcer location, should also be subclassifed as patients with hyperglycemia (1), which also have skin rash (1) and received corticosteroids (1). After I would later compare the same selected group with patients with hyperglycemia (1), which also have skin rash (1) and did not received corticosteroids (0). How can I do it using SPSS? I think you have to use the select cases tool, but I don’t know how to select cases (or variables) upon cases (or variables). Thank you very much in advance.
Relevant answer
Answer
Go to "Select cases", then select "If condition is satisfied" and then create an expression which includes all of your variables you want to include for a certain critierium. Use AND and OR operators.
For example ((VAR1 = 1) AND (VAR2 > 0)) OR (VAR3 = 0)
  • asked a question related to Clinical Data Management
Question
2 answers
Publicly available Database having a variable for Drug abuse and HIV as a prevailing medical condition. Thank you.
Relevant answer
Answer
Center for Disease Control: You will see HIV under the disease specific index.  Good luck.
  • asked a question related to Clinical Data Management
Question
3 answers
We are having a database having physiological data of people (sytolic pressure. diastolic pressure, heart rate, cholesterol, smoking habits etc.).
- What can Machine Learning do to such a database?
- How to apply Machine Learning to such a database?
- What tools can be used
- Do we have some out-of-the-box codes for applying it to such a database?
Relevant answer
Answer
Applying "machine learning" to a database just for the sake of applying it seems like a waste of time and resources. Scope creep can be problematic if one leaves the project too open-ended. There are probably hundreds of ways you could apply machine learning to your particular physiological database but what is the end goal? There are lots of factors to consider, which depending on what type of data you have and how many patients are in the database, can impact what you do. I would first define a research goal and hypothesis. In your case, the aim would hopefully have some clinical utility. Also before diving into that, one should know what type of sampling was used to generate the data in the database. Is it representative of the overall population or a particular sub-population? Is it biased towards some particular type of condition, etc? Once you have a clearly defined project aim and figured out the limitations of your data etc, you can refine your search of relevant machine learning approaches, apply that approach on your retrospective data-set and then validate it on a prospective data-set.
  • asked a question related to Clinical Data Management
Question
2 answers
I want to use i-stat to measure PCO2, PO2 and Ph on crab hemolymph at 20°C, does anyone know if the temperature interferes? 
Relevant answer
Answer
Thank you very much Julio.
  • asked a question related to Clinical Data Management
Question
6 answers
I have a dataset with multiple visits recorded for patients. I want to use only data from enrollment for an analysis. How can I code so I limit the analysis for the enrollment visit? 
Relevant answer
Answer
Dear Mehdi,
I assume, you are working with Stata, since your question is classified "stata programming"?
Second assumption: you have several records (observations) for each patient, one record per visit?
If both is yes, here is how you can do it:
1. step: you need an identifier for the enrollment visit. Is it the first (oldest=smallest) date? If yes, sorting by date makes the enrollment visit  the first observation of a patient. If this does not hold, look for a condition that is fulfilled only for enrollment visits, then generate an indicator variable with the command  "generate vi = cond(put here the condition, 1, 0) " This command will create vi having value 1 if the condition is true, and 0 otherwise).
2. step: select the observations that are renrollment visits:
a) In case you can use the date, the Stata command is:
bysort date: keep if [_n] == 1
b) in case you use the indicator variable
keep if vi == 1
After that your working file only contains the enrollment visits. Take care to save it under a different name than the original file..
I hope that helps,
Greetings, Klaudia
  • asked a question related to Clinical Data Management
Question
5 answers
Hi everyone, 
I am trying to predict some diseases using health check data.
The attributes are Age, BMI, Waist, Systolic, Diastolic blood pressure, HDL, LDL ... etc. I have to calculate distances (for kNN ...) or similarities (for Rough Sets ...) among instances. Certainly, we can use Euclidean distance and distance based similarity as well. In fuzzy approach it also possible to calculate similarity between instance "x" and "y" in attribute "a" by similarity_a(x,y) = 1 - |a(x) - a(y)|/(a_max - a_min) and combine with other attributes by a t-norm. However, those calculations are for arbitrary attributes which are mutually independent.
So for health check data, is there any considerable way to calculate distance/similarity?
Thanks in advances,
Relevant answer
Answer
Hi,
You could use the Heterogeneous Value Difference Metric (HVDM). In this way, you could work simmilarities for numerical and nominal attributes. In the case of medical attributes such as HDL, blood pressure I would try an experiment taking as a basis the "normal" values and calculate the deviation from them. Hope you find this useful!
  • asked a question related to Clinical Data Management
Question
3 answers
I need examples of a few publications using the Humedica database.
  • asked a question related to Clinical Data Management
Question
3 answers
This question is related to HIV drug class called fusion inhibitors. There is only one Fusion inhibitor called T20 (Enfuvirtide) approved by the FDA. I couldn't find the exact reason why the Roch and its partner company halted the development of T1249 in the middle of Phase II clinical trials despite of having good results!
Relevant answer
Answer
T20, Fuzeon, shows little acceptance by the patients, because it must be injected twice daily and causes local reactions. This problem is expected to remain for T1249.  Increasing the half-life of C-peptide entry inhibitors by conjugation to larger molecules has most likely so far has not been successful because these peptides loose efficacy in conjugated forms. That is at least what we have observed. 
  • asked a question related to Clinical Data Management
Question
7 answers
If in the trial data, there we have missing values in the treatment assignment, can we use multiple imputation to solve this? Because in trial the treatment assignment is independent of all other variables, so if we use other covariate to impute the missing treatment assignment, will it be effective?
Relevant answer
Answer
If you can reasonably assume that the process that lead to missingness of treatment is completely unrelated to the treatment itself then a multiple imputation method may help to overcome. If a larger proportion of treatment information is missing, or a clustering of missings (on time scale, covariables) is detectable, I would doubt on the independency assumption.
In any instance, I would recommend to analyze the complete case situation, and compare it with the analysis of imputations. Officially it is recommended to repeat the imputation process at least 5 times, but my experience is that there may still be a need for more repetitions of the process. Some of the datasets that I had analyzed showed that estimate stabilization may need 20 (or even more) repetitions.
In case of high rates of missingness (like >20%) it may be legitimate to introduce a new pseudoclass "missing" in your treatment variable, and analyze it with the model, featuring a "complete case" analysis. In case of completely-missing-at-random situation, the estimates of the "missing" category should be about mid of control and treatment.
Whether the process of imputation will be effective depends on rate and pattern of missingness, the sample size, the statistical method and model, the associations between variables, and the appropriateness of your assumptions on the process generation missingness.  You'll have to give it a try, and critically compare the different analysis variants.
There are several software products that may help you to solve the problems, in SAS (PROC MI and MIANALYZE; Macro IVEware ), R (MI, MICE anf ICE) and also specialized packages like SOLAS. For a (limited) review of these software solutions please read: http://www.jstatsoft.org/v45/i01/paper
  • asked a question related to Clinical Data Management
Question
3 answers
Finding free data from randomized clinical trials seems quite difficult. We would like to find a number of trials like these, to use their data with our students. I would appreciate any help.
Relevant answer
Answer
You may try clinicaltrials.gov
In advance search, limit the search just to the trials which have results.
  • asked a question related to Clinical Data Management
Question
2 answers
My project enrolls newborns through a pregnancy and birth surveillance program. Sick neonates are taken to a facility and blood is drawn. This specimen is transported to a laboratory where it is aliquoted and tested for pathogens. The data is then entered into the database. It's a complex study and I know for a fact that there have been very few studies like this. Moreover, most of those studies didn't publish their M&E methods. I am hoping to find articles on M&E activities for similar studies. If not similar, a project with multi-level monitoring and evaluation system may serve my purpose.
Relevant answer
Answer
Thanks Mr. Bilal
  • asked a question related to Clinical Data Management
Question
9 answers
Can anyone help me to find clinical data for angiography?
Relevant answer
Answer
The resolution in terms of the observer's ability to resolve features on the image would be improved by increasing signal to noise ratio (SNR), which is related to the 'graininess' of the image. SNR is crudely proportional to the square root of dose. So to double SNR you would need to quadruple dose. This is an over simplification as it doesn't take into account the quantum detective efficiency of the detector and other factors such as beam energy, anti-scatter grid usage and image processing.
Increasing dose is one way of improving image quality but it's not the only way. As others have pointed out, the issue needs to be looked at in the context of the application. Some examination require very high image quality while others do not.
  • asked a question related to Clinical Data Management
Question
1 answer
Sealing refurbishing and repair.
Relevant answer
Answer
Sealing , Repair are the conservative treatments most apropiatte
  • asked a question related to Clinical Data Management
Question
6 answers
I would like to have a software to manage large amounts of clinical data.
Relevant answer
Answer
Try OpenMRS, It is an open source and has a team of individuals who will help you set it up and guide you through its use. It is also being constantly improved by OpenMRS implementors. www.openmrs.org/download/. Hope it suits your use.
  • asked a question related to Clinical Data Management
Question
1 answer
I want to know how computer of IT experts may contribute in the field?
Relevant answer
Answer
IN CLINICAL RESEARCH the principal investigator of a research (usually in collaboration with a biostatistician) defines what data to collect in what form and timing using which method. After the data collection starts data are entered into computer databases. These databases may be very simple like a spreadsheet or complex like relational database with GUIs and server systems ensuring that all accesses and modifications to the data are controlled and logged according to regulatory requirements. An IT expert can contribute to designing an appropriate database or a secure data entry and storage system. Sometimes data are collected with electronic devices that need to be interfaced with the storage system to automate acquisition. Later he can help querying the data serving the needs of statistical analysis.
IN CLINICAL PRACTICE IT experts design and run databases that serve the diagnostic and treatment processes. An IT expert can take part in developing the specification of a system, coding, debugging, running, supporting it. Sometimes clinical databases are also used for research purposes.