An Integrated Future for Clinical Trials

Abstract

The purpose of this paper is to provide a perspective of how the future of clinical trials will be focused on integrated web based technologies. There are two aspects of clinical trials explored, that of Data Capture, or the data that is captured for conducting clinical trials, and of Patient Recruitment for clinical trials.

Data Capture

  Today, if you researched how the use of technologies will affect the workflow and the capturing of data for clinical trials  in the future, you are likely to find a scenario where the EMR / EHR plays a central role. For example the following scenario is described in the textbook “Biomedical Informatics: Computer Applications in Health and Biomedicine” (Shortline, & Cimino, 2006), in reference to use of EHRs supporting clinical trials:

With the introduction of computer-based patient record (CPR) systems, the collection of research data for clinical trials can become a by-product of the routine care of the patients. Research data may be analyzed directly from the clinical data repository, or a second research database may be created by downloading information from the online patient records... the interaction of the physician with the medical record permits two way communication, which can greatly improve the quality and efficiency of the clinical trial.  (p. 10)

  The above solves a lot of issues that currently exist when conducting a clinical trial; it helps saves time by removing double entry from source documentation to the case report forms (CRF); it reduces errors in the data being collected, it integrates the clinical trial protocol within the EMR to help with protocol compliance, and in general appears to greatly reduce the cost and time to conduct a clinical trial. However, conducting a clinical trial solely through an EMR / EHR has some roadblocks that put this kind of solution into question. Examples of such roadblocks are ownership of the clinical trial data and real time data access for clinical trial sponsors via internet technologies (generally a clinical trial sponsor is usually a pharmaceutical company, a medical device company or the government), the ability to configure a EMR to address study protocol decision rules and data elements, ability to perform the required clinical trial monitoring and querying, and the cost and efficiently of such a solution.

  There is a strong argument to be made that the future of clinical trial technologies is not with a sole EMR solution, but rather through a solution that integrates current clinical trial software, such as electronic data capture and general clinical trial management systems, over the internet with the other health IT systems, one of which is the EMR.  This is highlighted in the article “EHR and EDC Integration in Reality” (Gudrun Zahlmann et, al., 2009), which supports an argument that to solve the issue of making clinical trials more efficient, the solution must be multiple health IT systems that share data, allow for remote access, and not are not focused on the EMR as the sole solution. Further support for this argument is found in the fact that no EMR only clinical trial solutions, or pilot projects or viable future roadmaps, could be found when conducting Google searches. In fact the only solutions that have shown any rate of success are a few integrated pilot solutions such as the solution created by Technical University of Munich and Siemens Medical Solutions (www.siemens.com/medical) (Gudrun Zahlmann et, al, 2009). In the Munich project, multiple systems, including an EMR, PACS, and Lab system were successfully integrated with a clinical trial electronic data capture system (EDC) all communicating with the use of multiple standards such as HL 7 and CDISC. The EDC system utilized web based technologies which made it accessible by the sponsor, thus allowing for data ownership and remote access.

  One important aspect of this integration is that it leveraged the strengths of the multiple systems for their function. For example the EDC system was used to ensure CFR Part 11 compliance, validate data, allow for clinical trial monitoring and maintain a separation of ownership of data from that of the University of Munich and the sponsor. In relation to this, the PACS system was used to store any uploaded images and had the ability to link into the EMR and EDC system for image retrieval.

  Clinical Trials have highly specialized documentation / data capture needs that are specific to the requirement of individual trials. Health IT systems on their own are not capable to address all of these requirements on their own (Martin Dugas et, al, 2008). The integration and leveraging of systems, such as an EDC system, which is designed to specifically address clinical trial requirements is a key concept in future solutions for clinical trials.

Patient Recruitment

  Identifying patients for clinical trials is one of most pressing issues when conducting clinical trials. According to a study conducted by the University of Münster of 100 clinical trials, less than a third of the trials reached their original recruitment targets (Martin Dugas et, al, 2008).  There are multiple reasons for this low rate of patient recruitment, complex study inclusion and exclusion criteria (Martin Dugas et, al, 2008), patient populations and general lack of the ability to identify patients that are good matches for clinical trials. The ability to provide a solution in which a general health IT system is utilized to identify and match patients to clinical trials would have a very positive effect on success of the trial. Again, this calls for a integrated solution between a clinical trial management system (CTMS) and other IT systems such as the EMR, billing and Lab system.  However, most of the health IT systems today are not integrated to solve this problem (Martin Dugas et, al, 2008). This problem is really two problems, first there is the problem of identifying patients that are potential candidates for clinical trials, and second is the problem of notifying physicians that a patient is a potent candidate for a clinical trial. In order to solve these problems, integrated solutions need to be built that can match complex inclusion / exclusion criteria of a study to patient information stored in the health information databases. To solve the second problem, these solutions will need the ability to notify physicians, that the patient is a potential study candidate. A third, separate, but connected aspect of this is passing the patient information on to where the clinical trial is being conducted. Most likely the amount of patient information will include the details of the electronic record.

  If the location of the clinical trial is not located within the physicians health IT system the patient information will need to transferred across health IT systems to the unaffiliated locations. In order for this process to take place, a secure, web base solution will need to be built, ideally this would use the patient controlled personal health record (PHR) such as Microsoft’s Health Vault, or Google Health. By using the PHR the patient could control the possible bidirectional flow of information from one health IT system to the other.

  There is a lot of promise for the future of patient recruitment for clinical studies; much more it appears then the solutions for capturing clinical trial data. One current solution that is showing a signs of success in this area, is a solution developed by the Mayo Clinic in collaboration with Centerphase Solutions Inc (http://www.centerphasesolutions.com/). The solution searches Mayo Clinic’s EMRs of seven million patients, selecting appropriate patients for each study. This process is helping to speed clinical trials and eliminate potential errors that result in the low rate of enrolment in studies (Lee, T., February 4, 2010).

Conclusion

  It is apparent that the future of clinical trials is in the integration of health IT systems with current and future clinical trial software. This integration will need to be robust enough to handle the complex needs of individual clinical trials, while taking advantage of the general care IT systems such as the EMR. It will need to address the needs of the individuals capturing data for conducting clinical trials, recruiting patients, and for the organizations sponsoring the trials. In addition the role of the patient controlling their own clinical trial data (and data flow) will be an important factor in tying everything together.

References

Dugas, M., Lange, M., Berdel, W. E., & Müller-tidow, C. (2008). Workflow to improve patient recruitment for clinical trials within hospital information systems – a case-study. Retrieved December 04, 2010 from Trials: http://www.trialsjournal.com/​content/​9/1/​2.

Lee, T. (2010). Mayo Clinic looks to tap EMR database to craft clinical trials. Retrieved December 04, 2010 from Mass Device: http://www.massdevice.com/​news/​mayo-clinic-looks-tap-emr-database-craft-clinical-trials.

Shortiffe, E. H., & Cimino, J. J. (Eds.). (2006). Biomedical Informatics: Computer Applications in Health Care and Biomedicine 3rd edition . United States of America: Springer Science+Business Media, LLC. Smith, M. (2001). Writing a successful paper. The Trey Research Monthly, 53, 149-150.

Zahlmann, G. et al. (2009). EHR and EDC Integration in Reality. Retrieved December 04, 2010 from Applied Clinical Trials: http://appliedclinicaltrialsonline.findpharma.com/​appliedclinicaltrials/​online+extras/​ehr-and-edc-integration-in-reality/​articlestandard/​article/​detail/​641682.

Second Life: Doctor - Patient Interactions?

Over the past week I have been looking at Second Life (www.secondlife.com) with my avatar “devrjd windstorm”. This is my first experience with Second Life (SL) and my goal was to determine how the software could affect patient-doctor relationships; specifically in patient-doctor interactions. I feel that I met this goal and came to the conclusion that there are some good, bad and questionable aspects of SL and interactions.

Good: General Doctor-Patient Interactions

It makes sense to me that the most obvious uses of Second Life for doctor – patient interactions is when interactions are remote. This would be useful in cases where the patients are not able to physically see the doctor due to distance, a physical limitation, or a mental health issue. If the doctor had a SL “office” where the patient could go for a consultation, then this would provide a functional use of telemedicine.  The usefulness of this could be greatly expanded if SL would ensure a secure transmission of data and perhaps work as a “data portal” of PHR, EHR, and EMR data. By “data portal” I was thinking that SL would not actually store any patient / doctor data but allow the visibility of this data to be accessed through the SL interface. For example, a patient who is recovering from knee surgery is scheduled for the 1^st post follow up visit (meaning that they are at home, but not very mobile). The day before the scheduled visit, the patient has her husband take multiple pictures of her knee; she then uploads these to her PHR. The day of her visit she logs into SL and teleports to her doctor’s office. The doctor has permission to sync the images of her knee with the EMR and can thus see how the knee is doing and ask any needed questions, adjust medications and so on. The patient didn’t have to leave her house, and probably received a very high level of care from her physician.

Bad: Is this better than other forms of telemedicine?

In my scenario above I have what I consider an example of a basic potential use case of telemedicine. In trying to apply the above to SL, I kept asking myself if this was better than using some other forms of software, such as video conferencing, to achieve the same result. I think Second Life may provide a more physical feel of seeing a doctor, but I just can’t help the feeling that in use of telemedicine there will be more adoption of video than virtual reality. This is especially true in the fact that use of video/ video conferencing is being built more and more into the devices we use, such as laptop, netbooks and smart phones (iPads, not yet, but someday soon), and that the speed and reliability of these technologies seems to be reaching new levels where we can depend on them .  

By the way, where I do feel very confident that SL is a huge benefit to medicine is in training, education, and medical response simulation. Maybe even in general hospital administration, such as creating simulations for bed space – and seeing if there are ways to improve usage and care.

Questionable: Security

I tried to find some clear information on Second Life’s security model, and specifically what problems there might be with protected health information. I say tried, because SL does not make it very easy to figure out how their data is stored and if there is any way to allow for a secure level of data transmission in the case of a patient – doctor interaction (and access to PHR, EMR, EHR data). If SL is not able or unwilling to provide / ensure an extremely high level of data security than I do not see how it could be used in a meaningful way for patient doctors to interact. I would be interested to learn more about this – if anyone has info on this, please comment and let me know.

Notes:

 I should point out that had a hard time exploring health “places” as using the built in search engine for some of the applicable words did not return results as I would have expected. This seems to be consistent with the survey “A Survey of Health-Related Activities on Second Life” (Leslie Beard, BDes; Kumanan Wilson, MD, MSc; Dante Morra, MD, MBA; Jennifer Keelan, PhD - http://www.jmir.org/2009/2/e17/ ); that said, I did some Google searching and found:

·         A “place” call HealthInfo Island (much which is under construction)

·         “3D Virtual Worlds for Health and Healthcare” (Vanessa G. González, Macquarie University, Australia) http://journals.tdl.org/jvwr/article/view/664/504 .

·         http://doctorandpatient.blogspot.com/2010/06/worlds-first-virtual-ivf-clinic-on.htm

 In case you were wondering about my avatar name, no real meaning, just an abbreviation for developer [dev] my initials [rjd] and a random last name J

Health Insurance Exchanges

One of the biggest issues that came up during last year’s raging debate on the need to overhaul health care was the idea of government run health care insurance exchanges. Since the passage of the health care bill, this topic has seemed to lose focus in the media. However, I noticed in an article in The New York Times titled: “Health Care Overhaul Depends on States’ Insurance Exchanges”  (by Robert Pear; Published: October 23, 2010) that reminded me of how much of an impact these exchanges could/ will have on the success of the health care overhaul. Also, from an IT perspective, I was (am) interested in the technical aspects of how exactly these exchanges will work. I should also note that under national health reform, states must have a health exchange in place by January 1, 2014 – so this is something that states are either planning and/ or have a solution in place.

What is a State Insurance Exchange?

I had a hard time finding an official definition of a “State Insurance Exchange”; but here’s a definition that seemed to make the most sense to me – it’s from a 2008 study by the Minnesota Department of Health titled, accurately enough “Health Insurance Exchange Study”: 

“The basic idea of a health insurance exchange is similar to the concept of a stock exchange or farmers market – an exchange serves as a market clearinghouse, but not as a regulator or purchaser.3 It functions as a single place where people can go to find out about their health insurance options, and improves market competition among health plans by providing more complete and understandable access to information about the products and pricing available in the market.”

Why does an Exchange matter?

In my mind the most simplistic way to understand why this matters to the success of the health care overhaul, is to think about the cause and effect of a law stating that everyone needs to have health insurance. That is to say, that in order to realistically require some level of health insurance for everyone there has to be a organized (maybe controlled?) way for people to obtain that insurance, at a cost they can afford (or be subsided by the government) – and there has to be a way to provide enough data about the options so people can make an informed choice. And, of course for any of this to work you have to create a “market place”, as stated in the definition above, in which insurance companies not only compete but also activity participate in, maybe even be required to participate in.

How they work... and the IT side of things

There is a great graphic in The New Your Times article sited above that shows generally how Health Insurance Exchanges work, however like many other articles it does not describe any of the specific technical details of how an exchange works. That said, here are my general comments on what I did see or assumed:

·         First, the obvious - health exchanges are web based solutions. I could not find one example of a state exchange that was not web based.

·         For a visual of what the user interface of an exchange might look like I looked at the U.S fed. Employee’s site: http://www.opm.gov/insure/health/search/plansearch.aspx and the Massachusetts state’s “HealthConnector” https://www.mahealthconnector.org/portal/site/connector/

·         Both of the above are your basic web portal, with tools, information. I had tried to find someone working on more of a next generation web app style exchange (think iPhone app) but didn’t find anything

·         I couldn’t find any specific info on how the insurance companies communicate with the exchange – meaning are they using a specific standard (HL7?) and are they using the normal web service model / XML to exchange data. I’m assuming that this model is in use, but I didn’t see this detailed out anywhere.

·         I also tried to answer a general question I had as to if there was some link into a user’s patient health record system (PHR) and if there was any way for a tool, through the PHR to do routine searches of the exchange to determine if there was possibly better options for the patient – both cost and general insurance options. My thought was that it would be good if there was a way, say every 6 months to have an app update a patient’s PHR info with any changes in the exchange. So when the patient logged in they could determine if they wanted to make a change. Anyway, this does not exist (or else I couldn’t find it).

Please tell me what you know…

Please comment on any additional info you may have to the above; I would love to find more technical information on how the exchanges will work – especially how other entities communicate with them, what standards are used, and if these exchanges are just between insurance companies and the state or if the exchanges will also allow other tools/ apps to exist and be used by patients – or be tied into a patient’s PHR.

First Post

Hello!

This is the first post in my blog; as you can probably tell by there not being any other posts!

A little about me..

I have worked in Health IT for over 10 years, specifically in the area of clinical trials. I work for a small software company that produces web based applications that address the needs of both the business and clinical data capture aspects of the clinical trail environment. My role within the company is that of the software architect (who still writes a lot of code) and VP of the development department - meaning that I manage the developers, product managers and QA / testers. 

Purpose / Goal of this blog..

The purpose / goal of this blog is to provide a public space for me to distribute and receive information related to the health IT world in which I belong.