Progress 08/15/16 to 08/14/17
Outputs
Target Audience:Veterinarians in private food-animal practice in the USA, including but not restricted to members of the American Association of Bovine Practitioners. Veterinarians and other professionals involved in the regulation of prescription drugs for food-animals in the USA. Veterinarians and other professionals working within the pharmaceutical industry for food-animals. Veterinary students. Similar groups of professionals outside the USA. Livestock producers. The general public. Changes/Problems:Reviewers were correct to cite lack of detail in personnel for software development as a weakness of the work plan in the Phase I proposal. Good software developers are hard to find, expensive, and usually have enough employment options that they will not commit to an unfunded project. Immediately on receiving notification of probable funding, we began work on resolving this weakness along a three track plan. A job description was posted to an online agency for software developers. One of the respondents was then hired on a per-hour basis to develop one of the essential software components, a user management system according to specifications. An RFQ was developed for an application comprising a minimal subset (about half) of the features described in the work plan and submitted to an agency that is a leader in the kind of software we had selected for this project. After some further refinement of software specifications we were offered a quote of $30,000 with satisfaction guaranteed, and about 4 months projected development time (including a substantial delay until coding would begin). We ended up not accepting this offer. The PI began building the necessary knowledge, and writing code himself. In this, $500 invested in the track 1 developer was a great assistance, but there were significant problems with the code provided by that developer, not reflecting lack of skill or knowledge of the developer, but the inherent difficulty in communicating complex ideas. It became clear that Track 3, with occasional use of outside developers per Track 1, was the safest and most economical plan. There is overhead in learning new skills, but also great inefficiencies in communication between experts in different fields. In this case, having the coding done by a person who understands the entire project and also has a deep understanding of the needs of users offers efficiencies that offset the training costs (mostly time). There will also be a continuing advantage, when other software developers inevitably need to be added, in having a project director with deep knowledge of software internals and coding tools. A no-cost extension was asked and granted, partly because software development took longer than scheduled, but also to develop beyond the Phase I objectives and present the best possible Phase II proposal. What opportunities for training and professional development has the project provided?The PI has invested considerable time in the months before official initiation of the project and throughout in learning new software design and coding tools and upgrading previous skills. These include the IntelliJ IDE, Google Dev Tools, Git, Scala, Play Framework, Slick, Javascript, ReactJS, MySQL, HTML, CSS, Amazon Web Services, Google Web Services, Docker, Kubernetes, and other bits and pieces of computer technology along with the basic information science of protocol design. How have the results been disseminated to communities of interest?Protocol management is currently an area of great interest to practicing veterinarians, so it has not been difficult to get attention. The complexities involved in writing and managing individually customized protocols have often been underestimated, and moving from interest to implementation remains a very large undertaking that will take time to develop. Early dissemination has focused on the AABP where Dr. Ehrlich has high visibility as a past Director and one of the founders and managers of the very successful email discussion group "AABP-L". This position of authority must not be abused, so the dissemination strategy has been to make a few short postings to AABP-L announcing milestone developments in the project, as well as joining the general discussion when that is topical. More detailed tutorials are posted to http://forum.VCPR.ORG, which is currently more like a blog than a discussion forum as active discussion on protocol management are more likely to be within AABP-L. AABP-L reaches an audience of well over 2,000 veterinarians. The application now also includes the capability to send email to all registered users, and this route is used to announce important upgrades. There are currently 142veterinarians who have registered to use the application, while documentation including YouTube videos and tutorial excercises are being assembled on the forum.At this stage, veterinarians serving swine, poultry, and other food animals have received little attention, but similar methods can be used with those groups as the project progresses. VCPR.ORG operated an educational exhibit at the annual AABP conference in 2016 in Charlotte, NC, and will do so again in 2017in Omaha, NE. As an organizer of the preceding 17th ADSA Discover Conference on Dairy Analytics, Dr. Ehrlich was asked to give introductory remarks at the 31st Discover Conference on Big Data in Dairying. That conference provided an opportunity to discuss technology and possibilities with a wide variety of dairy scientists from around the world. Dr. Ehrlich writes occasional articles, including on protocol management and VCPR issues,in "Hoard's Dairymen", the leading lay publication for dairy farmers. What do you plan to do during the next reporting period to accomplish the goals?This project is currently in a no-cost extension period during which the protocol management application will remain available to registered users, also with continued user support and development of the application itself. Other support activities, including an educational exhibit at the 2017 AABP conference will also continue. Efforts will continue to develop relationships with potential Phase III partners, and a Phase II research proposal will be prepared and submitted.
Impacts
What was accomplished under these goals?
The ultimate objective of this project is a change in the entire culture of the management anduse of pharmaceuticals in food animals. This can best be accomplished by leveraging the professional knowledge and ethical standards of veterinary practitioners, represented by only about 5,000 doctors in private food-animal practice serving nearly 100 million beef and dairy cattle, 70 million swine, and 9 billion poultry. This will require development of efficient tools enabling veterinarians to prescribe then track usage on each farm that they serve. Feasibility of this approach is demonstrated by the pilot application which has attained all the specific objectives detailed in the work plan and which continues to be developed with additional features and refinements,and also by very strong participation and interest in the project by practicing veterinarians. The FDA holds responsibility for writing and enforcing regulations on use of pharmaceuticals in animals. The USDA holds responsibility for ensuring the safety of food, including that it is free of drug residues, and for enforcing standards of animal welfare which include adequate medical care. Private-practice veterinarians stand in the middle, and take responsibility that their clients adhere to regulations, treat their animals humanely, maintain food safety, and survive in a competitive economic environment. There is a desperate need for better tools to enable practitioners to meet all these demands. In 2009 Congress passed the Health Information Technology for Economic and Clinical Health Act (HITECH) providing up to $27 billion to promote adoption and appropriate use of Electronic Medical Records (EMR) for people. Much has been learned, some of it applicable to food animals, but of course there are vast differences between human medicine and agriculture. This project is informed by developments of the EMR, as well as the software developed for social applications on the Internet, but designed around the basic needs of farms and veterinarians from the ground up. Goals stated in the project initiation form are rather general for this reason, and the process of development has been and will continue to be one of experimentation and adaption as we gradually add functionality and prune complexity, homing in on what works. The project can be though of as involving a single iterative activity, the conception, implementation in software code, testing, and refinement of features needed within an Internet application for design and management of veterinary treatment protocols. Veterinarians were enlisted to assist in testing as soon as the most basic functions were in place. Features have been added at the request of users, and the user interface simplified where users had difficulty completing specific tasks. Feedback from users is received via multiple routes including nonintrusive monitoring of their work, email, and a feedback utility built into the application. In the listing below, each feature constitutes a change in knowledge, in that both investigators and users better understand the requirements of an efficient process for managing treatment protocols. Each feature that is implemented in software also constitutes a change in condition, because software components as they are developed and tested can be used and recombined in various ways in this and potentially in other applications. As the overall objective at this stage of development is proving feasibility, expecting changes in action would be premature, though widespread interest from many directions, especially from practitioners, indicates general appreciation that changes in actions are inevitable for the way that pharmaceuticals are managed for food animals. Application features implemented to date: Privacy and security are basic necessities. We established an Internet server for the VCPR Protocols Manager(TM) at HTTPS://protocols.VCPR.ORG with user management services allowing a veterinarians to fill a form requesting an account, for the system administrator to approve the account (after checking license information), and for basic login/logout and password management functions. TLS encryption (https) is available, but not required at this point. A data model for treatment protocols has been designed and implemented within a database, and also a REST API which allows exchange of protocols within the system and also with external software systems that may develop (such as dairy records software). Each Protocol consists of any number of treatments, an Instructions section containing text and images organized into semantic subsections, and metadata such as protocol identifier, author, and timestamps. A drug database, assembled from FDA-approved drug labels andlicensed from a commercial service, is incorporated into the software components used to design treatments within a protocol, such that species, condition (indication), product (drug), dose, and route of administration can be selected from the database via a search utility and selection lists. This database automatically adds FDA label warnings and withholding information for on-label treatments. a rapid lookup utility can display full FDA label information for any drug in the database on demand. Sophisticated logic based on AMDUCA1 rules is built into the user interface, to classify treatments as they are assembled and categorized choices presented to the user with respect to the AMDUCA algorithm. Users can easily add rich text and images to treatment notes or any of the Instructions sections in a protocol. Images are automatically uploaded and stored securely in the Internet cloud. A feedback utility makes it convenient for users to send comments or suggestions while using the application. Users can add accounts for their clients, then prescribe a protocol creating a timestamped immutable snapshot as a PDF document, and metadata stored to the database. The PDF can be emailed, printed, and recalled from the server at any time (requiring the password of the owner). Utilities for monitoring and managing prescribed protocols are rudimentary at present, and being developed further. The author of a protocol can choose to share it with individuals or all veterinarians using the system. When a protocol is shared, it can be cloned then further edited by the user or users to whom it has been shared. The original author can add further editing at any time to the master template, or undo the sharing. The system is deployed entirely within commercial Internet cloud services, and can be scaled up rapidly and easily. Deploying a new version is quick, automated, and the updated application and data is immediately available to all users without any action on their part. Version upgrades are sometimes deployed several times in a week, allowing rapid response to issues as they arise. 1.AMDUCA: AnimalMedicinalDrug Use Clarification Act (1994) 21 CFR 530
Publications
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2016
Citation:
Presentation at Cornells Summer Dairy Institute, http://www.vet.cornell.edu/summerdairyinstitute/
7/12/2016
- Type:
Websites
Status:
Accepted
Year Published:
2016
Citation:
VCPR.ORG forum website
http://forum.VCPR.ORG
- Type:
Other
Status:
Submitted
Year Published:
2016
Citation:
Article for "Hoards Dairyman"
"SOP or Protocol - What is the difference?"
|