|
|
|
Index - Major Sections
Site Map
Product and Services _______________ Index - Same Level Subject
Index - Child Subjects |
Computer Applications One of the best uses of the computer is the data acquisition and data analysis of data. Data transmission can just as well be done with a fax machine and/or telephone. In order for data to be analyzed, it must first be collected in a structured format and "coded." Design Principles Principles of design have been extensively research by the likes of Microsoft and other application vendors. Their designs doesn't make an application more correct than another design...but it does go a long way! We do not need to re-invent evidence based process. The workflow of healthcare professionals (HCP) today was developed based on "Paper" documentation. The paper based procedures and requirements are now of the past. It is archaic and obsolete. The statement that the system must be designed to match the workflow of the physician is a very good reason NOT to design the system that way.Today's systems must be reengineered to take advance of the power of the computer. This requires examining every aspect of the Healthcare Processes including the workflow, coding system, to the financing of healthcare. Healthcare is based on several very simple processes.
These processes may cycle over and over from the more general to the more specific along a "discovered path" as more data becomes available. The inclusion of the Prediction/Prevention is a often missed process....even though it is probably the most important. The task of the Healthcare Management Information System (HMIS) is to support the healthcare professional in obtaining the necessary data and providing decision-support. The models (rules, concepts) we develop are just that..."Models." They must be constantly analyzed and refined for the "individual" under care. There is no "one way" to do something and we never let the computer make the final decision for us. The more the data process collection methods can be structured, the better decision support systems (alerts, automation, suggestions, etc) will be. Navigation Navigation of the system can be performed in many ways. There should be a number of methods to do any one task. There are "side-bar", "tool-bars", and "Menus" all of which makes it possible to navigation to your destination. Naturalness While doing what is "natural" is easy, it may not be correct or the most efficient and effective process. Wither something is "familiar" is entirely dependant on what has been learned and performed previously. Designing a work-flow based on what is being done (so you will not need to spend a lot of time training) is a sure way to reduce the value of a Healthcare Information System (HIS). Almost all Computer applications that fail fail because they try to duplicate what is being done and not what needs to be done. This is a special problem in healthcare and probably the main reason for healthcare professionals not accepting a system. Workflow has been a subject that has been long studied in other businesses. Minimizing Cognitive Load The HIS is all about minimizing cognitive load...at any one time. Presenting "ALL" data on the same screen in order that the user do not have to change screens is a recipe for errors. Studies have show that people can only accommodate a certain amount of data at any one time. Data should be organized by importance and detail....from the more general to the more specific. Data should be "easy to access", not that it should all be shown at one time. Many business managers use the concept of "Management by exception"....if it is not an abnormal/exception then I do not want to see it. A system that makes it too "easy" may do exactly what you do not want it to do....it makes the user stop thinking. Anything that is "routine" may put the user to sleep! There should be a careful balance. Efficiency Efficiency, as a test metric, is NOT "the speed with which a user can successfully accomplish the task at hand." Efficiency in healthcare is the comparison of what is actually produced (outcomes of patient care) with what can be achieved with the same consumption of resources (money, time, labor, etc). It is the skillfulness in avoiding wasted time and effort. In Healthcare the task is the care of the patient. It involves many individual tasks from many different people. Just as building a "house" requires the corporation of many different individuals all doing their jobs. It is the same in healthcare. While it make take more time for one individual in his role to do "their" work in order for others to do theirs, then that is still efficiency. Doing something "fast" and having it cost more money is not being efficient. Doing something "fast" and making more work for others is not being efficient. Efficient from a healthcare point of view is a "team effort" not how fast an individual employee can enter "his data" into the application. While one individual may have data that is "difficult and time consuming to enter" , what is more important, is wither that makes it easier for another team member to do his job. Therefore, the InHCc definition of Efficiency. "Efficiency is the corporation of all healthcare team members in such a way as the best outcome is delivered to the patient with the lest amount of money, time and labor. " Data Meaning Data fields are designed to completely descript that concept or entity. These data fields are called "attributes" of the concept. These attributes are include in a structure such that they are easy to complete and meaningful to anyone that uses the system. While narrative data fields are included, it is highly discouraged. There is very few times that these text fields will be required. New Data fields are easy to add or delete if no longer required. All data can be viewed in the language of choice (if the translations have been created) Principles of Evaluation While most, if not all, papers discuss the factors for "clinician acceptance and system success" as "Efficiency of Use" and Minimizing likelihood of user error"...this is only a very small part of what can be accomplished with a HIS. InHCc thinks that the most important is the ability to analysis data. As said in the first paragraph, the computer can be used to analyze large data sets collected over a large number of individuals, over time with many different "problems," treated with many different therapies. This type of analysis is often referred to as Data Mining (Data Mining). In general, Data Mining does NOT need a "hypothesis" to work, but rather the outcomes of Data mining may, indeed, "suggest" a hypothesis for additional research. There are some processes that are so complicated that a computer may be able to analyze and "visualize" the process better than a human. Such processes examples may those of biological signals such as a ECG, Electroencephalograms, or body chemical changes. Abnormal signals can be recognized immediate and an alarm can be sounded without waiting. Many imaging now has software applications that can automatically read the results. Based on the data that is entered into the computer, "Expert Systems" can suggest the cause of the problem, probability of the answer given being correct, suggest other alternatives, and suggest the therapy based on the latest research evidence medicine. Injections of medicine can automatically be programmed into an automatic drug dispensary (pain medicine, insulin, heart medicine, fluid balances, etc) based on biological signals...and now they are testing "robot surgery systems"...Wow! Healthcare Specific Negative findings It is required by many Certification organizations that negative findings be recorded. An example is shown, in it complete form: History of present Illness: 42 year old woman reports
wheezing. She denied dyspnea or a cough. She denied a nasal discharge Now is this a joke or what? Does any busy healthcare professional have time to report every negative finding and what about the next Healthcare Professional trying to read this....??? And if the poor healthcare professional forgets to say that something is negative....does that mean that he could be subject to a nasty law suit for mal-practice This type of summary examination reporting is
The problem with this type of reporting is that it is very hard for the Healthcare Professional to separate what is happening and what is not happening...The "Nos" are hard to read and they are next to the "yeses." Summary Notes One of the best example of a "Historical" process is that of the HCP having to write "Summary Notes". Healthcare Professionals (HCP) have up to now spend a large proportion of their productive work day writing notes. Summary notes should be a thing of the past. There are many problems with "writing notes" besides taking a lot of time to write and read. Some of the problems with text notes that could impact of the efficiency, effectiveness and safety of healthcare are the following:
Healthcare Professionals have a lot of material that they must digest. In order to be able to take in as much as possible, professionals tend to "skim" over articles and information...they do not read every word of the sentence. This skimming of information can cause misreading/misunderstanding of the notes by other Healthcare Professionals. This is dangerous and can problems in the care of the Client. Summaries can easy be developed from the detail by the computer and presented in any format There is no lost of detail since each click on the summary can present the complete detail of the procedure. There is no "forgetting" to include something in the summary that would make a difference in the treatment of the client. And it would be easier to read by the next HCP that requires the information. Caregivers see a combination of "All the Notes" at one glance instead of having to pick through all the support personal's notes. In summary, the computer can do it better, faster and it would save the HCP significant amounts of time. Summary records do not have to be created as long as the detail records are not deleted (a must requirement of any longitudinal healthcare system). With all detail records archived, it is a simple matter to combine these records in any format required. It is suggested that we let the computer produce the summary notes automatically in name:value format. Protocols Protocols can easily be developed based on Evidence Based Medicine. Have you check the These protocols can be presented for the HCP's selection at the appropriate time for his selected. There is no need for the wide variations and expensive non-valued procedures that exist in today's healthcare. Protocols also enhances safety and quality of care by "reminding" the HCP of those processes that are required. Coding Some of the problems of "Manual Coding" are
(See the section on coding) Clinical Drawings With the ability today to use "cam recorders" to photo objects, there is really no need for "clinical drawings." A photo can not only take the actual picture but it can also record color... However, until applications can be developed to "analyze" the imaging, descriptive data must still be recorded for analysis. Ophthalmology is a good example where photographic "images" have almost complete replaced clinical drawings. It may be soon that the entire examination process is video recorded and read with special application programs. X-ray imagines already can be read with these applications...and they are getting better. Research Research is another area where changes can be made. While the standard golden rule of the "Real" procedure and the "Pseudo" procedure and all the implications of not providing the real procedure to the test group may be required for the "HCP's published article", data mining today, can determine more valid relationships faster and cheaper with no ethical dilemma. Design should be based on a careful review of how processes should be preformed effectively and efficiently using IT technology and the data that is needed in each step, This workflow analysis has been performed for years in business but seems to have completely overlooked in healthcare. The system cannot be designed by looking at how healthcare professionals think about and carry out processes. There are no two healthcare professionals that think the same way or had the same education and training or live in the same area. Physicians are never trained in exactly the same way. They do not go to the same school, they do not live in the same locations, and they do not have the same beliefs about health care....Maybe for once, we really should have a look at best practices! As seen from the definition in the Stakeholder section, there are many shareholders...the most important being the individual. However, is seems that from reading the literature (in regards to accepting a Information System), the physician is the principle stakeholder. Much is made of the fact that it is hard to get the healthcare professional to "buy into" information systems. That it takes too much time for them to enter the data, or that they have to enter too much data, or that is not the way they do things. Physicians also say that information systems (Computers in this case) are too obtrusively and it interferes with their face to face meeting with their clients. The physician is only one small part of the overall system...admittedly the most important...but NOT THE ONLY part. Now think about this...if the information that the physician collects is used for data analysis to improve the overall health of the population and the procedures that he is required to follow are evident based medicine that lead to more cost effective medicine...Is it really important that it takes him a little longer to enter the data, or that it is not what "he is use to doing." Think about the other shareholders that his processes do benefit. Since Healthcare is one of the largest industries in any country, how can we not consider the other players? InHCc believes that the system should be designed by professionals. A good example, is Microsoft's Design Guidelines. Patient's information should be immediately available. Summary Hierarchal data should first be presented, and "drill down" detail information, that is required and only what is required, can be viewed by clicking on a summary record. No information is deleted. All procedures are different, screens should be designed that present all concepts that is required and the data sets that are required by each concept. Form layout With today's technology there is not a lot of problems with designing forms for any and all stakeholders from the "same data" There is no need to worry wither the "user may not feel if there view is represented" (ASTM E1384-07). With xml and style sheets, it is not only possible to produce any screen format but also it can be produced for almost any "device". With the Microsoft© Report writer, it is possible for the user to design screens "ad hoc" by easily moving data around much like the design of the Excel spread sheets. Once this data screen is designed, it is possible to output the data in a multitude of formats from word documents, pdf documents, text, and excel spread sheets. "Power Users" can not only design their own forms "on the fly" but also select only the data that they need. Any type of format that the "Standard Organizations" produce is obsolete! This discussion will continue as our project moves forward. Ease of Learning
It is true that the more a user can apply prior experience to a new system the lower the learning curve. However, it is also true that "bad habits are hard to forget." Sometimes what has been done before is a real handicap in learning new improved work flows and processes. As has been said in another section of this report, the greatest failure of computer applications in producing good returns on their investment is the attempt to "duplication" what already exists. An electronic system is "different." It is capable of doing more and better things in different ways. The constant attempt by developers to make physicians "comfortable" with what they are doing by duplicating existing methods, is probably the greatest determine to the successful implementation of healthcare systems. Training requires "repeatable" reinforcement. Plan for it! In one job, InHCc actually "planned into the plan" the "failure" of the initial training. It was assumed that additional training would be needed after the staff had time to experiment with the system. The training went something like this:
At InHCc we have successfully trained doctors to be input data in a single afternoon. However, in order to teach them why and how to use the data they were creating took several weeks. It is the last item (how to use the data) that is the hardest and the most valuable. As a given, any system should include "on line help", context sensitive information, and examples of data use. User Satisfaction The least important in selecting a new system, but causes the greatest problems. In order to rate anything, you must have more than one object to rate...and the measures that will be used in the rate. "Satisfaction" is completely objective and most of the "users" have never seen another healthcare application much less worked with one. While it may have been rather dramatic, InHCc had one manager (a very very good one) tell his staff...."you either use the application and use it correctly or find another job!" It seems that the staff had a 100% increase in their satisfaction rating overnight. Links
|
|
|