谁了解PRODIGY

n1703005

有谁比较了解PRODIGY的？你对PRODIGY应用于中国的CIS有什么想法吗？

n1703005

PRODIGY is a source of clinical knowledge, based on the best available evidence, about the common conditions and symptoms managed in primary care. The team at the Sowerby Centre for Health Informatics at Newcastle has developed, and keeps up-to-date, around 170 guidance topics for both acute and chronic illnesses, including most of the conditions that can be managed by extended formulary nurse prescribers. The guidance is structured to support both decision making in the consultation and learning outside of the consultation - so that knowledge can be accessed in an appropriate format where and when it is needed.
<br>
<br>RODIGY guidance has been available since the end of 1998 through RFA compliant primary care computer systems as a decision support system (Release 1), and from the website. In 2002 the guidance was also published as a book.
<br>
<br>On the website PRODIGY guidance is available as full text documents, quick reference guides and Patient Information Leaflets (PILs). Additionally the guidance can be viewed through a web browser.
<br>
<br>A much improved version of the decision support system (Release 2) has been developed and is awaiting a strategy for national implementation. This new system offers pathways, alerts, prompts, and reminders - aiming to match different professionals&#039; requirements and workflow.
<br>
<br>这里，我们主要是想PRODIGY R2对中国现有的CIS会有什么帮助或合作。

张琨

感觉Prodigy是经过大量临床实践后的一个经过dataming后的结果，可以帮助建立临床路径、治疗预警等服务功能。
<br>与国内的结合有机会但也有明显的局限。从临床科研角度来说可以帮助国内医生在治疗技术上有些参考，工作方法上对于CIS公司开发产品有些借鉴。但具体内容由于中英医疗服务模式和知识的差别，感觉受限制比较大。
<br>
<br>楼主是哪里的阿？能否介绍一下？还有你自己对这件事的想法呢？交流要有进有出
<br>

n1703005

<!--quote-张琨+2005-08-19 12:14--><div class='quotetop'>引用张琨 &#064; 2005-08-19 12:14)</div><div class='quotemain'><!--quote1-->交流要有进有出<!--quote2--></div><!--quote3-->
<br>说的完全正确。来这个论坛，就是来向大家学习交流的。
<br>
<br>我来自SCHIN（Sowerby Centre for Health Informatics at Newcastle），即PRODIGY的开发研究机构。^_^
<br>
<br>正如你所说的，PRODIGY的背后，确实有大量资深的医生、药物学家、护士及相关学者在做authoring。PRODIGY R1是基于Protege的ontology开发的。现在已经成为英国GP必读的一份Guidance。我们在9月将会参加在上海举办的由Shanghai Ministry of Science & Technology，British Consulate General和Shanghai Technology Network联合举办的一个交流会。旨在寻找潜在的合作伙伴共同寻找国内的商业机会。
<br>
<br>我们也意识到中国现有的传统的医疗流程和英国存在一定的差别。但首先，PRODIGY knowledge是一个knowledge的平台，而并不是一个软件成品，换句话说，是一个需要载体来释放的knowledge；其次，上海的community care环节的酝酿，可能会导致一个重大的医疗流程改革。
<br>
<br>个人以为，由于医疗信息技术特有的气质，相对于技术上的不同实现，其对信息这个概念背后的knowledge的含金量要求要更为偏重与依赖。
<br>
<br>由于对国内的医疗系统现状，还需要更进一步的了解和研究，目前的不少想法可能会有误区，希望论坛上的各位专业人士指正。

n1703005

谢谢你的回复。但我对于你提到的<!--quote-linforest+2005-08-21 07:05--><div class='quotetop'>引用linforest &#064; 2005-08-21 07:05)</div><div class='quotemain'><!--quote1-->国内信息标准<!--quote2--></div><!--quote3-->和<!--quote-linforest+2005-08-21 07:05--><div class='quotetop'>引用linforest &#064; 2005-08-21 07:05)</div><div class='quotemain'><!--quote1-->机构和系统之间的语言都不通<!--quote2--></div><!--quote3-->不是很明白。能否给出更进一步的解释或举例？

KrZ

My 2p worth....
<br>
<br>ROIGY Guidance decision support covers areas from patient condition management mainly via scenarios, therapy group, therapy details and prescribing. The underpinning decision support module ontological development uses our criterion model to express the precondition of scenarios and prescriptions. This approach gives much needed flexibility in comparison to some of the scripted constrain language. Execution of the model is done via the execution engine that interprets client record by the use of virtual medical record pipeline engine (VMR mediates the terms/drug terms used by clinical system to PRODIGY local terms for processing against the decision support ontology).
<br>The workflow of the recent Torex implementation goes something like this:
<br>GP loads up patient medical record from within Torex system. This starts up the VMR server which then interrogates the currently loaded patient record (mediating terms used in torex to Prodigy terms). The mediated terms then gets passed to the trigger module which uses our trigger knowledgebase to determine the validity of alert and reminder (say reminding GP to take blood pressure if it is absent etc, this type of data entry is also termed quick data entry within Prodigy). At this point, GP are presented with a set of options which includes a set of actions to perform, repeat prescription etc as well as the entry point to Prodigy guidance browser which will walk through the management of conditions through actions (such as print out PILs etc) as well as prescriptions.
<br>Finally, all data write back are done via the VMR server again.
<br>In terms of coding standards say Arden Syntax, it is in large irrelevant in this case because the integrating point is at patient medical record level, rather then at the execution level. Hope this helps
<br>

jing.wang

KrZ, your explanation with the workflow helped. Thank you. Here are some follow-up questions I was hoping that you could also help clarify. Linforest touched on some of the points, and I&#039;d like to tease out more details.
<br>
<br>1. If I understood you right, the VMR server is a defacto translator or language broker between the front-end clinical application and Prodigy. The role of VMR is to allow proprietary systems at both ends to pass coded information back and forth. If this is correct, then is it fair to state that the design is a proprietary integration, an alternative to the standard-based integration by Arden Syntax?
<br>
<br>2. Would you elaborate on the development effort to integrate a currently non-compatible CIS with the VMR? In terms of resource, time, and also with regard to upgrade and/or change?
<br>
<br>3. With the CDS module, what is the governance structure that controls the content? How does it reconcile among different interest parties such as medical, government regulatory and maybe private payers? Are there cost/benefit formulas and approval criteria?
<br>
<br>4. Lastly, the CDS module and its application so far described seemed to be hard-coded, if I may. Is this a true statement? What&#039;s your vision of the use of AI in this application?
<br>
<br>Thanks in advance.
<br>

KrZ

hi, thanks for the reply.
<br>VMR server does serve as a translator and or language broker of TERMS between clinical system and prodigy decision support. I misunderstood the earlier question on the use of Arden Syntax, thought was referring to the terms passed across the system rather the structure of the packet. As far as I know, the packet structure conforms to HL7 standard (which in a way i guess related to Arden Syntax). Two key persons were responsible for the Prodigy specific data structure, Ian P and Pete J (HL7 UK technical committee). So it is hardly surprising that it would be HL7 compliant. The access to the clinical system were done via the use of their api, and the prerequisite is that their data transferred (not necessary stored) conforms to HL7.
<br>The collaborative work between prodigy and torex system involves someone works their ass off to get compatible output from clinical systems (namely pete), pardon me French lol. In house wise, a lot of head scratching, as you can possibly imagine, software ppl and doctors trying to communicate with each other…emmm… ain’t pretty.
<br>

jing.wang

KrZ, thanks again.
<br>
<br>I think you hit one of the major challenges right on - how technical talents and clinical experts can communicate and understand each other well. Good or bad, your organization is not in this sticky-n-thick alone, it&#039;s everyone and everywhere, well, almost. Personally, I think the key is inter-discipline skill sets.
<br>
<br>Aside from that, other critical-path blockers such as medical terminology differences between UK and China, lack of one single standardized medical vocabulary in China or almost anywhere for that matter, differences with protocol formulation due to demographical and socioeconomical discrepancies between the two countries, political and financial ramifications inherent in China... to name a few, in my humble opinion, are far more significant and daunting.
<br>
<br>CDS is a noble idea and the dream of many. To get there, we&#039;ll have to take baby steps.