Lab Soft News

Building the IT Infrastructure of RHIOs

Mr. Histalk picked up on an interesting nuance in a speech by David Brailer at the recent HIMSS meeting. Here is the link to the Brailer speech. Brailer is the national coordinator for Health Information Technology for the Department of Health and Human Services. Here is the link to the comment from HIStalk (scroll down) with a cut-and-paste from it below.

Does David Brailer really support the IT work of RHIOs? Some have speculated that he doesn't, preferring as a business-friendly guy to instead to turn the whole package over to a big-bucks vendor like IBM or Cisco and let RHIOs deal only with the politics and collaboration. An interesting quote that may reinforce that speculation, one I must have missed in his HIMSS speech: “I did not start out by believing the national solution will be a network of regional networks. Our goal with the national health information network is to allow those who do not want to participate in RHIOs to not have to do it." I'm not so sure that the work being done by various groups is operating under that same assumption.

If you believe that regional sharing of health data is important and that regional networks are an important next step, the question arises about how to build the network infrastructure. It seems to me that there are at least two possible approaches -- bottom up and top down. If you choose the former, it is highly likely that the regional components will never be sufficiently compatible to interoperate on a national basis. If you choose the latter (i.e., Cisco or IBM builds the infrastructure; regional RHIOs address the collaboration and politics), I still don't think that it would work. As I have stated in a previous note, the large health systems within a region are the primary stakeholders in any project to share clinical data on a regional or even national basis. I am hard pressed to identify any major incentives for them to share clinical data outside of the regional area from which they draw their patient populations.

Here a great article on the NHIN (national health information network) to be theoretically formed by the coalescence of the RHIOs. Below is the money quote; boldface emphasis mine.

The second element of the NHIN is interoperability, a concept that is cumbersome to fully capture and quantify but one that is synonymous with its organizational catalysts, RHIOs. Interoperability represents an organizational and technical infrastructure that allows for information to be readily and securely shared between a variety of health care stakeholders. The difficulty in assessing the benefits of interoperability is that the magnitude of improvement is dependent upon the level of participation by other stakeholders. The more organizations that participate, the greater value to all constituents. In addition, the types of benefits attributable to interoperability are less tangible and predictable, accruing variably to different types of stakeholders across a variety of health care processes. In the end, it looks like we would be getting a technical infrastructure that has the potential to do some good things, but it would be difficult to quantify those benefits today. The risk would primarily reside in the implementation of the technology and whether providers actually would use it.

The bottom line from this analysis is that the benefits of a RHIO/NHIN are highly dependent on the number of participants (the so-called network effect), that there are variable benefits by stakeholder, and that the estimated cost of such a project (total capital costs of $156 billion over five years and $48 billion in annual operating costs) will be staggering.

CDC Acquires Realtime Emergency Room Data Feeds

The CDC is setting up a system designed by Informatica and SAS Institute to gather clinical data from the emergency departments of 31 hospitals in 10 large cities and send it to the CDC for analysis as part of a surveillance program for a bioterrorist attack. Here is the link to the news story and below is an extract from that story. The bold face emphasis is mine.

The Centers for Disease Control and Prevention (CDC) has quietly begun working with 31 hospitals in 10 large cities to create a system that can send real-time data feeds from emergency rooms to the CDC. The program, which aims to help officials prepare for and respond to a pandemic of avian flu or a bioterrorist attack, will add 350 hospitals to the list this year....At this point, the CDC's systems are about a month away from being able to analyze incoming data from the initial 31 private hospitals during a catastrophic event,...As part of the IT upgrade, the CDC is building a real-time data warehouse that will support 20 TB to 30 TB of data over the next 12 to 18 months...The developers are using extract, transform, and load tools from Informatica Corp. to send data to the warehouse. They're also using business intelligence and data-mining tools from SAS Institute Inc. to analyze the information coming from the hospitals. The hospitals are sending the data -- including patient symptoms, diagnoses, and geographic information -- to the CDC over the Internet as Web services messages using the ebXML standard to guarantee reliability and secure the exchange of the messages. Beth Israel Deaconess Medical Center in Boston began sending feeds to the CDC on Dec. 21, about a month after the agency asked it to join the program.... The hospital now sends the data to the CDC as Web services every 15 minutes.

Here are some of my initial reactions to this information. You may have additional reactions which I urge you to submit as comments.

• Here is an example of what can be accomplished from a healthcare IT perspective when an arm of the government gets very interested in biosurveillance and bioterrorism. It also looks like the technology to support the program is readily available so there may be some lessons to be learned here for RHIO development.
• Note the reference to the fact that the project has proceeded quietly.  Pretty hard to keep a project involving so many hospitals under wraps for too long.
• The article is silent about the patient record confidentiality aspects of this project but it appears that only symptoms, diagnoses, and geographic information are being transmitted to the CDC. The individual patient data would need to be anonymized or at least deidentified to be HIPAA compliant. However, in the event of a bioterrorist attack, affected patients seen in the hospital emergency rooms would need to be quickly and specifically identified.
• Here is a good link to a CDC web site that discusses syndromic surveillance which is the process used for early detection of bioterrorist attacks. Obviously, lab data can and will play an important role in such early detection.
• I find it interesting that the CEO of Informatica, one of the two prime contractors for this bioterrorism surveillance project, is Sohaib Abbasi and that he established an Islamic Studies program at Stanford in 2003 with a gift of $2.5M. Here is the link from Campus Watch about Abbasi and below is a quote from the article.

Retired Oracle executive Sohaib Abbasi and his wife, Sara, hope to change that, Sohaib Abbasi said. The Atherton residents, who were born in Pakistan, donated $2.5 million to establish the Islamic studies program, Stanford announced Friday....Abbasi said he and his wife made the gift "to ensure this becomes the premier program of Islamic studies in North America."

Tension Between For-Profit and Non-Profit Cultures

Fauzia Khan makes a very interesting point in her response to my recent post about the strategies and approaches used by software sales representatives. The note discusses, among other things, the tension between the scientific/non-profit clinical lab culture as compared to the for-profit/commercial culture of vendors. Here is her comment:

Science/academia being godly and vendors/commercial world being evil goes very deep in almost all cultures including the American culture. Science/academia selflessly pursues 'ideals' that 'make life better' and vendors go after 'money'. However these stereotypes are changing.

I believe that her point is very important and one that I was remiss in not suggesting. Part of the antipathy on the part of lab professionals toward sales representatives is clearly that the later group represents the "crass and commercial." Note also her comment that such a feeling even pervades American culture which was founded on commerce and entrepreneurship and lacked a history of a landed gentry proud of its idleness. She is also correct in pointing out that these stereotypes are changing. Listed below are some salient examples of how and why these changes are coming about in the U.S.:

• Those of us working in hospitals are aware of how many decisions are spreadsheet-driven and how interest in the development of some healthcare delivery options (e.g., ramping up cardiology services) is stimulated by attractive "profit margins."
• Lab professionals are being increasingly marginalized and stigmatized for being a cost center in the inpatient arena in this DRG era despite the major contribution of the clinical laboratories to the diagnosis and monitoring of disease
• Prominent scientists and physicians are increasingly being tagged for ethical and financial lapses, suggesting that they are not totally immune from the temptations of financial gain. In short, the sanctity of the non-profit sector of healthcare is eroding.

So I guess that the take-home lesson, at least in healthcare, is that we are all working, or should be working, toward the common goal of improving the system and that everyone has an important role to play. I know that this sounds trite but it's where I ended up with this short essay.

Defining Technology. Will It Disappear?

A recent quote from Carly Fiorina, former CEO of HP, caught my attention. Here is the link. Below is a quote from the article. The boldface emphasis is mine.

... Fiorina told the audience that by 2030, technology will be so integral to daily living that we will take it for granted. "We have entered the main event of technology...We are talking about an era where technology is woven into the fabric of life and almost disappears. It means that technology is in everything and everywhere," Fiorina said.  ... "In the next five years, the biggest areas of technological change will be in cybersecurity and health care. That is why I have chosen to sit on some company boards," she said.

It occurred to me that I did not have a firm definition for technology in my mind. Following is a definition that I like with the link:

1. The complete set of knowledge about how to produce in an economy at a point in time, including techniques of production that are available but not economically viable. 2. The set of production functions available to an economy. 3. Referring to industries that are experiencing, or recently have experienced, technological progress.

I believe that the notion of a "point in time" is important for the definition of technology. One does not view a tool or device that was readily available in your early formative years as technology. For example, teenagers today, I am sure, do not view PCs, cell phones, or the Internet as technology. For them, the wonder of these one-time "technologies" has faded and they take on the trappings of utilities like electricity or heat. Therefore, I do not believe that technology per se will every totally blend into the background because new inventions will [hopefully] continue to emerge for our amusement and to support our production functions.

Technocrats from outside healthcare always seem to view the industry as a sponge that effortlessly absorbs and adapts to technology. Note Fiorina's decision to position herself in healthcare as an avenue for technological change. This attitude is one way to explain the recurring waves of enthusiasm for large healthcare computer networks such as RHIOs, discussed here in a previous note. However, I think "outsiders" fail to appreciate the deeply conservative nature of the healthcare delivery system, in part conditioned by the continual threat of malpractice actions and the consequences of rapid change. Healthcare tends to look high-tech to an outsider because the pharmaceutical, imaging, and IVD industries continuously develop new products and insert them into the healthcare delivery system, saying: "Why don't you try this? It's new."

Critical Assessment of HIMSS

HISTalk publishes this rag about HIMSS:

"What can we do to raise the quality of HIMSS? The sessions I attended were, for the most part, shallow....The exhibit halls are an amazing waste of money. Not that I am against exhibiting....One of the most common comments we heard in our booth was, 'You're showing us real, released software? Everyone else showed PowerPoints and mock ups.' .... Where is the dignity in our industry? I work in healthcare IT because I care about healthcare, and can touch more patients through IT than treating them one at a time. I think about how much money is spent not only on the booths, but the parties and receptions, bar tabs, cars to give away, and even transportation of all this stuff.... And can attendees please just boycott the paid actors who don't know anything about IT who are paid to look good and read a script? Is there a way to get the real business back into this otherwise valuable networking opportunity?"

I am really feeling very virtuous about Lab InfoTech Summit that starts next Wednesday. I am having trouble pushing the number of exhibitor hospitality suites past  numero uno. No booth babes. Every exhibitor has one "table" of the same size. All of the software really works. What is it in the genes of the lab crowd that demands this firm grasp on reality?

Inside the Mind of a Software Sales Representative (Part 2)

In a previous post, I linked to an interview by Mr. HISTalk who was able to corner a healthcare software sales representative and get him to spill the beans about some of the tricks of the trade. There was so much good information in this interview that I decided to parse it out across a series of notes on Lab Soft News. This is the second post in this series. Below is a set of Q and A's from the interview followed by my comments and analysis. Boldface emphasis is mine.

Q. How do you determine who are the decision makers and influencers?

A. One of my favorite tactics to flush out reality vs. pretenders is to threaten to pull out of the deal. For example, "Based on the information that I have about your project so far, I am not sure that my product is a good fit with your goals. This is a very difficult and expensive sales process and I don't want to waste your time or my company's resources on this.....blah, blah, blah."  Their response to that is very telling about how much influence they have.  With all that said ... sometimes you just have to go through the motions until you figure out who has the juice to make the deal happen. 

Q. How do you plan your sales strategy vs. the personalities involved?

That mostly comes from experience. One of the most important lessons I have learned in sales is you have to adapt to the style that works for an account. You can usually predict it. Academic medical centers have the "ivory tower" mentality. You have to be very deferential to them. Community hospitals are more like "good ole boy" networks. They want to know you and be your friend. IDNs and corporations are tougher to read but the locale often is telling. Stereotypes are usually somewhat true. Small town folks are more "homey" and want to hear about your kids. Big city folks want to get right down to business.

The most challenging situation is the dynamic of "we are the virtuous customer and you are the evil vendor." I was in a meeting last week where a senior executive of a large hospital opened the meeting by telling me how much he hates software sales people. He said "The only difference between a used car salesman and a software salesman is that the car guys know their product well enough to know when they are lying." I had never met this person before then. You can never win that guy over, but you have to at least neutralize him. Assassination is not beyond consideration!

The question about who has the "juice" to make a deal happen is very complicated with an LIS purchase so I think that the interpretation above is a bit of an over-simplification. In large healthcare systems, I envision four groups that can influence an LIS selection, each of which can contribute what I will call one token: (1) lab directors, medical technologists, the lab manager, and lab scientists, generally representing the needs and wants of their particular lab but sometimes with a broader perspective; (2) the surgical pathologists who have a vital interest in the functionality of the anatomic pathology module of the LIS; (3) hospital administrators and the CIO/central IT staff; and (4) purchasing and legal representatives. The latter group do not have the juice to reject a vendor but rather to sometimes slow down the process.  The first three groups have the essential juice but each group tends to have only one "token" that it can assign to a favored vendor in a collaborative process.  However, there are cases where a very strong individual in any one of these groups with a heavy hand can dominate the process.

I have no comment on the ivory tower mentality in an academic setting. The comment about IDNs and corporations is very interesting. Obviously, some IDNs and hospital chains have a very corporate mentality and take LIS purchase instructions from C-level executives. My guess is that this style will be made very obvious early in discussions -- the participation of lab professionals in the LIS selection process may be just for show. Some IDNs formed by the merger of two previously independent hospitals can provide an interesting case study. In some such cases, the two hospitals maintain their previous cultures. This can prove to be an LIS sales challenge in that the two hospitals in the so-called integrated system will seek to maintain different LISs primarily to impede the integration process.

The notion of a vendor being evil is widely prevalent in the clinical laboratory world and is a view that I have never understood or subscribed to. I believe that part of this phenomenon may be due to a culture clash between the personalities of pathologists and other lab professionals and a company's sales force. The former tend to be relatively quiet and introspective and the latter are ..... well, less quiet and introspective. For many lab people, their major goal is to get past the sales cycle relatively unscathed so that they can project themselves into the system installation cycle and interact with the technical people who are more able to understand their problems and respond to their needs.

Three Major Lab Data Domains

A recent press release from Hilgraeve that you can read here got me thinking about medical information domains in general and lab data domains in particular. Following is a cut-and-paste that caught my attention from the article (emphasis mine).

Physician practice management systems are the primary repository for patient data in this country. The Federal government has a 10 year goal to provide easy access to that data, which they project will reduce national healthcare costs by 10 percent,” explained Jeff Beamsley, president of Hilgraeve.

Because I tend to view clinical data from a lab-centric perspective and also because lab data constitutes such an important component of clinical data, I envision three separate data domains in the world of healthcare and medical research: (1) PMSs/office EMRs, as emphasized in the quote above; (2) hospitals and health systems; and (3) clinical research systems (e-research systems), primarily in located academic medical centers but also in other settings such as standalone clinics dedicated to support of clinical trials.

I am note sure what Jeff Beamsley meant by primary repository for patient data in the quote above. Does he mean primary in terms of the volume of data, criticality of the data, or the breadth of the delivery of healthcare services? For now, lets ignore this question and focus on the idea of the need to integrate data from the three clinical data domains describe above. I will assume that this integration is extremely important in terms of the quality of longitudinal care delivery and also the economics of care delivery.  For example, we could reduce the number of tests performed if we provided ready access to previous lab results, particularly when treating chronic ailments.

Lab data constitute a critical and common data element across all three of these separate data domains. As evidence for this assertion, the lab industry has spawned a new category of software, lab portals, to help bridge the gap between health systems and private office domains. Initially deployed by reference laboratories to facilitate test order entry/result reporting for private physician offices, this same software is now being deployed by hospital-based labs to support lab outreach programs. So called healthcare portals will soon be used throughout large medical centers.  See the previous note about the business model for Medicity.

I don't need to comment on the criticality of lab data in the clinical research data domain. Moreover, most significant clinical trials involve acutely ill patients who are being simultaneously being tested for their underlying illness and to reveal both benefits and harmful effects of any investigational new drug. There will be substantial overlap between hospital LISs and e-research systems.

Because labs and lab data span these three domains and because lab professionals continually strive to integrate the three for reasons of efficiency, quality, and patient safety, the labs in general are an ideal test bed for new IT product development with the goal of supporting the integration and interoperability of these domains.  As noted above in the example of the lab portal, integrative systems are already arising spontaneously, driven by the clinical laboratory business model.  If more grant money were made available for lab test integration, this process could be accelerated.

Inside the Mind of a Software Sales Representative

Mr. HISTalk was able to corner a healthcare software sales representative and get him to spill the beans about some of the tricks of the trade.  I thought that coverage of this item would be of value to the readers of Lab Soft News. Here is the link to the original interview. By the way, there is so much good material in this piece that I am going to present only a portion of it below and will provide additional material in subsequent posts. Emphasis below is mine.

Q. How do you determine who are the decision makers and influencers?

A. That is probably the single biggest challenge: determining who is really a decision maker and who isn't. ...in the end you have to rely on your guts to determine who really makes the call. All those other people who are sometimes wasting your time have their place as well. The key to big-ticket sales is having a good internal coach. Even if they don't have the pull to make things happen, they are a great sounding board for refining your message and identifying the real issues. The biggest people-challenges are the gatekeepers who want to limit vendor access to those decision makers. They are usually the ones who wrote the RFP (to which every vendor answered "yes" to every question) and insist that you "follow the established process." My biggest indicator of a person's influence in the deal is their level of honesty in the dialog. If they are only interested in setting up demos or "events" but can't articulate coherent business reasons for the project, then they probably don't carry much influence, regardless of their title. Maybe I'm a jaded sales rep, but if I can't get them to have a meaningful discussion about the project other than "send me some specs on XYZ" then I'm not going to win that deal and I'll focus on other accounts.

One could write an entire book on the basis of the wisdom contained in this short paragraph alone. Here are some of my reactions to some of the nuggets that are presented.

1. In any lab operation, there are those who revel in process and procedures, there are those who know and love the science and technology that forms the basis for any lab test, and there are those who understand the laboratory as a business. Those who understand the business are not necessarily those with MBAs. Sometimes the designated decision-makers referred to above come from one of these three groups and sometimes they have none of these talents. If this is the case, the lab has a problem.
2. The point about the "internal coach" is very important. A smart sales rep cultivates such key internal people in order to understand the organizational culture and priorities. If a member of the LIS selection team is too vocal a supporter of a particular LIS, it may work against the selection of that choice. A more subtle approach is for that person to function as an internal coach, thus providing an edge for the favored LIS.
3. Because many vendors will respond "yes" to most items in the RFP response as noted above, all "yes" answers need to be validated during site visits and scripted system demos.  It is also critical to append the vendor's response to the final contract so that the dishonest "yes" answers will return to haunt the vendor. Most honest vendors will reply "in development" when they do not offer a desired feature.
4. The RFP is only a means to an end. Do no fall in love with the document and do no infuse too much power or influence in authors of the RFP.
5. I love the reference to the need for lab personnel to articulate "coherent business reasons" for deploying a new LIS.  If you can't come up with such a list, you should terminate your shopping expedition and spend the capital on other lab improvements.

A Look at the World's Largest Central Laboratory Network

Q, Name the world's largest central laboratory network. 

A. That's easy! It's Covance.  Here is the company's description of itself from it's own web home page (emphasis mine):

We provide a portfolio of preclinical and clinical development and commercial service offerings — delivered through industry-leading nonclinical testing services, the world's largest central laboratory network, and a global team of clinical trial professionals and cardiac safety experts.

Covance is one of the world's largest CROs (contract research organization) and when a company manages clinical research, it is also provides laboratory testing.  As noted here, company sales for 2005 amounted to $1.25 B and its one years sales growth for that same year was 18.4%. This company link indicates that the company operates in 18 countries and has more than 7,000 employees worldwide.

Hoover's financial service describes the Covance mission in the following way:

Behind every great drug company stands a contract research organization (CRO). Covance develops and carries out preclinical and clinical trials of potential commercial drugs. Services include toxicology studies, biostatistical analysis, clinical data management, and study design and monitoring. Covance will also assist a company in moving a drug through the regulatory approval process.

I have commented previously on the topic of offshoring of laboratory services here and here. If I were asked to pick a candidate for the company most likely to succeed in managing such a business, I would select Covance. It is already in the business of performing and contracting for sophisticated offshore clinical lab testing, managing global biologic specimen logistics, and aggregating and analyzing lab test results from around the world. It also understands lab QC and credentialing on a global basis. Given their success as a CRO, however, they may not be looking for another business.

To illustrate their current skill in contracting for offshore lab testing, here is a description of the company's current laboratory reach into China:

In China, which is a key location for clinical trials — including those related to the treatment of diabetes — Covance utilizes the Hua Shan Hospital's Center of Laboratory Medicine as a regional laboratory service provider to meet biopharmaceutical sponsors' in-country testing needs. With this certification, sponsors are assured that their Covance clinical trial specimens are tested in a global network of NGSP Level 1 certified laboratories — in China, as well as in Sydney, Australia; Singapore; Geneva, Switzerland; and Indianapolis, Indiana in the United States.

A Peak at the Military Healthcare Information System

As an exercise in analyzing the challenges of deploying a large healthcare information system and also a view of the military mind, let's look at some recent events relating to the DODs CHCS II system. Here is a link to a recent press account.

The Defense Department’s $1.2 billion health care information system got a new name Monday....Assistant Defense Secretary William Winkenwerder said “this new health care information system called AHLTA” is a major accomplishment in military medicine. Later, in a press conference, Winkenwerder acknowledged that the new system was CHCS II, only renamed. In fact, CHCS II has been under development since 1998 and has yet to be deployed to about 40 percent of DOD health care facilities worldwide....Winkenwerder said he did not like the CHCS II name because it suggested that the system was not No. 1. AHLTA, which will hold the health records of 9.2 million military personnel, family members and retirees, will be one of the largest and most advanced health information technology systems.

Here are some of my personal reactions to this statement:

• Is it not somewhat premature to refer to an unfinished project as a major accomplishment in military medicine? However, an alternate explanation is that the project should be considered a major accomplishment even in its state of unfinishedness.
• Renaming such an IT project will probably not significantly change the way that is has been designed, rolled out, or accepted by users. Some would view this as merely a cosmetic change.
• A health IT system that is eight years old and not deployed by 40% of the target facilities may already be dead but not yet pushed over. It all depends on the pain threshold of current and potential users.
• Placing a "II" or "2" in a name of a project sometimes indicates that it is second in a series and not innately inferior. See this biography of the Ford family with references to Henry Ford II.

As you can see here, the Healthcare IT Guy is also a little dubious about the success of the project.