Healthcare informatics needs a new strategy to make it more ubiquitous and popular. Frankly, it has been more academic and rhetoric than substance till date. With only limited functionalities being implemented and that too mostly administrative and accounting instead of clinical ones that would have the maximum impact on patients and patient care, healthcare informatics is at a cross-road.

To most health informaticists it is perhaps both mind boggling and galling to note that even in the best of technically advanced places, clinical information systems (CIS) do not account for more than 10%. Although the very basic forms of EMR (level 1-4/7) has seen an increase, it is important to note that while this is definitely a beginning it is hardly expected to make the difference that CIS can actually make.

Patients are largely unaware and hence are unable to create the demand for such systems. The care providers are largely unaware or less aware or possess sketchy knowledge about the various functionalities and the impact that they can have on care and care delivery. In their defence however it must be said that working or demo systems that can provide sufficient knowledge hardly exist. With manufacturing companies making systems on-demand and with most installations being site-specific having been custom-developed for that person or institution, systems that can work for everyone are not readily available.

For healthcare information industry to grow, it needs to become a buyer’s market driven by demand. The supply side has already robust and any existing or fresh demands can easily be met. It is the demand side that needs full attention and that is not only the most challenging but most expensive as well. Marketing it will be the key, particularly marketing it right. It is vital to inform, educate (not coach or coax, that’s selling), demonstrate and encourage patients and care providers alike that adopting CIS does indeed make better care a reality and non-adoption is simply impractical and foolhardy in this day and age.

CIS will allow for faster information percolation, better access to past information, and improved care through quick and easy analysis. Patients will neither have to remember nor rue lost memory for missed medications, appointments, information, and faulty care resulting due to failure to inform vital facts on time like past conditions, etc. Care providers will not have any cause to feel bad about not receiving an information on time, missing vital clues like allergies or risky medication doses, making errors of judgement or incorrect diagnosis. Personalised treatment protocols can be designed for a particular patient rapidly and they can be closely monitored to ensure that the proper treatment is delivered. Home care can be the preferred option instead of institutionalised care (although institutions are not very keen on this for obvious reasons) and distance would not be a barrier to receiving best of available care from a care provider of one’s choice.

Lastly, getting something for nothing or cheap stuff never really works in any situation. While paying over the odds is not advisable either, a proper investment (time, money, human) is necessary for success. This is applicable for any industry including healthcare informatics.

Or in other words, what would the various stakeholders gain out of having these systems? Let us examine who the principal stakeholders are. Loosely, one can state that these are as follows.

1. Patients - without whom there is no healthcare
2. Payers - without whom there is no healthcare delivery
3. Care Providers - without whom there is no healthcare delivery services
4. Managers - without whom there is no efficient healthcare delivery services

Healthcare Information Infrastructure (HIE/HII)

Health Information Infrastructure is going to be core information infrastructure of over all e-Health Solution. IT will provide the capability to electronically move clinical information between different health care information systems while maintaining the meaning of the information being exchanged.

The goal of HII is to facilitate access to and retrieval of clinical data to provide safer, timely, efficient, effective, equitable, patient-centred care.

It is intended to provide an information drive healthcare to improve access and quality of care to patients. The objective is to ensure availability of the right information to the right person about the patient that will enable better clinical diagnosis and decision-making.

The Healthcare Information Infrastructure (HII) is the healthcare informatics infrastructure and will provide access to information about the patient at the point of care.

The HII will be designed to provide coverage to remote areas with existing infrastructure constraints to facilitate availability of care to these remote areas over solutions like tele-consultation and telemedicine can be a reality in near future.

This infrastructure will enable healthcare information exchange across the centres. Hence it can also be termed as Healthcare Information Exchange.

UHID (Universal Healthcare Identifier)/UID

The Universal health identifier (UHID) will be a number to uniquely identify a patient. It will be used for identification of patients when clinical care is rendered, with automated linkage of computer based records on the same patient, for creation of lifelong electronic health files, with a mechanism for protection of privileged clinical information.

The Health Information Infrastructure will require unique identification of the patient. The patient identifier will be handled at multiple levels and will be used to aggregate the patient clinical data from different sources in one single view. The patient base will be required to identify uniquely based on a set of patient demographic parameters.

These parameters can be used in the hospital system or in the HII for identification of the patient whenever the patient ID is not available.

Registry (Patient Registry)

It will be registry of Patient with entry of Patient Identity and it will provide the following basic services:

• The ability to create a single logical view of the various patient representations. This will be done through a linking technique and a probabilistic matching algorithm that will ensure the appropriate grouping of patient identities under the correct patient.
• The ability to query for a patient either through local patient identifiers or through local patient demographics. This will ensure that the existing tools available to the physician will be able to return a match without alteration to their view of the patient.

Clinical records would mean any type of electronic records that contains clinical data in granular or free-text form, structured or unstructured.

Text mining refers to the process of deriving high-quality information from text and involves the process of structuring the input text (usually by parsing, along with the addition of some derived linguistic features and the removal of others, and subsequent insertion into a database), deriving patterns within the structured data, and finally evaluation and interpretation of the output. Typical text mining tasks include text categorization, text clustering, concept/entity extraction, production of granular taxonomies, sentiment analysis, document summarization, and entity relation modeling (i.e., learning relations between named entities). [Wikipedia - http://en.wikipedia.org/wiki/Text_mining]

While using usual data mining techniques should have been fine, clinical data is frequently entered as free-text. The great complexity and user-unfriendliness of interfaces that force clinical care providers to enter data in structured form with adequate degree of granularity only is a great disincentive.

To mitigate and attain a degree of balance of sorts, as things currently stand allowing free-text entries is the most prudent option to adopt. The pay-off being the necessity of using text-mining to cull the required data from these free-text contents to allow proper data analysis.

While text-mining provides the appropriate tools to populate the data warehouses, searching for the right words makes the process inefficient. Using appropriate terms and their corresponding codes for clinical terms (e.g., SNOMED-CT) to cull the appropriate words from these fields definitely improves it. With more useful data being extracted faster, both quality and quantity of data for mining is significantly increased.

However, this use of codes introduces an intermediary stage where the clinical databases are text-mined using terms from the codes. Next, the corresponding codes are extracted and these are then made to go through the ETVL process to actually populate the data warehouses.

Once the data warehouses and marts, as appropriate, are populated, the rest of the process follows usual data mining and knowledge discovery in clinical databases.

The first conference that I am attending is IAMI 2009 [http://www.iami2009.com/] being held at Novotel Hyderabad Airport, from 13th - 15th November 2009. With its theme  ”IT enablement of healthcare services”, the Seventh International Conference on Medical Informatics (IAMI 2009) is expected to bring together researchers, educators and medical practitioners from across the world to share the results of their research work and practical experiences.

The goal of the conference is to provide a forum for professionals, researchers, educators and practitioners from both institutes and industry to meet and share cutting-edge advancements in the field of Medical Informatics.

A very important aspect of this conference is a major session on eGovernance in the healthcare sector. The session will devote discussion on the various Indian eHealth Initiatives, eGovernance Projects including the best bractices of the successful eHealth Projects, recent eHealth Trends and the use of mobile technology in Health Governance.

I will be making a presentation on WAHN - wide area health network - a conceptual network that allows seamless information flow amongst the various stakeholders to allow for quick and easy access to required and necessary healthcare information. The information may exist in paper or verbal or electronic form.

The second conference is Health Informatics Asia [http://www.healthcareinformaticsasia.com/] to be held in Singapore from 8 - 11 February, 2010.

This programme is expected to provide plenty of opportunities to pick the brains of some of the best in the field when it comes to daily best practice in implementing and managing healthcare informatics aspects:

• Plenary panels defining and guiding the future of healthcare informatics in
• Stream panels that provides the opportunity to quiz the expert son more niche areas of healthcare informatics
• Breakout sessions where persons are able to sit down with peers and have their say when it comes to challenges and opportunities for healthcare providers

• Aging population and how to best tackle the necessary implementation and strategy plans
• Solutions for hospitals with limited budgets and which strategies need to be in place
• Hear from NGOs such as UN and WHO how they think rural and lesser developed regions should be equipping themselves to provide better healthcare to the population
• Get latest technology solution updates and hear what speakers think about upgrades of legacy systems and how they think plug-and-play scenarios can be achieved
• Hear the physician’s perspective on informatics and how the changes are affecting various departments and what change management initiatives are necessary for a smoother transition
• Meet and network with peers from Asia Pacific and Beyond. Enjoy regional and international information exchange and benchmark with the best in class

I am scheduled to speak on the importance of dashboards in healthcare, with the focus being on

■ How healthcare management dashboards help in business efficiency

■ How Clinical dashboards help in clinical excellence

■ Using these dashboards as the most effective way to provide a bird’s-eye view of the results of intelligent analytics

I recently read a blog at http://hcrenewal.blogspot.com/2009/10/from-down-under-story-of-deployment-of.html

This is yet another example of “those who do not learn from history are condemned to repeat it”, I am afraid. People listen, hardly ever learn.

There is no one or easy answer to this conundrum. There are pros and cons for either. Let us examine them. As it is difficult to say which came first, product or services, let us take services first since it is more generic.

Services will let one to design and develop a solution de novo, ie from scratch. This allows things to be less cluttered as the design is made on a clean slate. As the services company will be expected to have already worked on a similar solution, they will have plenty of best practices and ready-made modules that can easily be incorporated. Additionally, by rigorously following all the well-established and standardized processes the solutions can usually be relied upon pretty well.

Being essentially one-off and completely dependent on the requirements provided by the client, solutions are not very mature as clients themselves would normally not have visualized all the requirements. Consequently, they would require extensive enhancements.

This is usually how services companies make money. They usually lose money when going through the bidding process and trying to get “the deal done”. Since all enhancements are done on a time and man (”T & M“) basis, when the clients realize that the solution needs upgrading, much time and effort and money are spent.

Products are ready-made solutions that require minimal to no cutomization, although they may require some configuration in terms of setting up like users, access privileges, etc. Although, it is not uncommon to find up to 40% customization, ie, extensive reworking and redesigning to be carried out.

Clients are however constrained to use whatever is available and their workflow may not be to their liking. There is no guarantee that the products conforms to any of the established and expected standards and since most manufacturers try to be as secretive as possible and cloak themselves in NDA’s, the clients is left none-the-wiser regarding whatever they are buying. Payment is usually upfront and most often includes periodically updates and upgrades to it.

Quite simply, in services, the clients call the shots and, in products, the manufacturers. There is a definite pay off in either cases.

The best way forward depends on the clients. If the clients are mature and know exactly what all they want, they should go in for services. On the other hand, if the clients cannot be bothered with waiting for solutions to be put in place, they should go in for products.

So, which one is best? Services for large institutions, products for smaller ones. The size is determined more by the maturity of clients than by volumes. Price is almost always the overall deciding factor. Degree of maturity of clients is not something that the clients will readily confess to, while the ability to pay is something that is relatively widely known.

The ROI is usually easier to calculate in products, while it is more tricky in services since the payments keep on happening over a period of time and almost unpredictable basis.

Enough said.

eHealth is not happening. Every other business area has been successfully penetrated by ICT that not only delivers on its promise but actually does more. Notwithstanding having this well-documented and well-proved fact with tangible evidence thereof visible everywhere, healthcare continues to remain one major area where there is hardly any ICT presence worthy of mention. Is it only apprehension, reluctance and ego that makes it so? Is the lack of proper solutions the crux of the matter?

As is usual, the problem lies somewhere between these extremes. While it would be easy and simplistic to say that there are faults on both sides, it is mostly due to ignorance, reluctance to change and incorrect focus in the solution development process that the things have come to such a pass.

A requirement is what a particular product or service should be or do. Consequently, a software requirement is what the software should be or do. Any user is only concerned with “what” the solution can or is able to do. The “how” hardly bothers him other than as an idle curiousity at the most. In fact, it should not be otherwise.

A “busy” interface is a poor interface. An interface with spelling errors is a put off and constitutes a P1S1 bug, while an interace requiring too many clicks (more than 3) to accomplish a task is a definite no-no. It is vital to remember that it is the user interace that the user gets to see and interact with. If it is not easy to work with in terms of presentability and usability, irrespective of the technology involved, it will never be widely used.

Equally important is to realise that an interace is not a painting meant for hanging on the walls of an art gallery. It must be able to help the user to get the job done in the shortest possible manner as easily as possible. However, it must be pleasing to the eye too. If the user needs to look at it for any significant length of time, it is better that he does not have to stare at stuffs that are ugly.

Anything beyond the UI is a “black box” to the user. What goes on there and how is of no concern to him nor should it be so. He may wonder but not ponder. As a passing fancy he is absolutely within his rights to delve into it, but only as such.

Technology is important, but only as far as system performance, stability and security are concerned. Uptimes, rapid request responses and denying unsanctioned access is the key. Not what language was used. Not whether the architecture conforms to SOA or client-server or n-tier or whatever. Not what platform dependence or independence it has.

While I would like to concentrate on UI in a different blog, it would not be out of place to mention over here that for any software, nay any, solution, the UI is the key. It is the face of the solution and unless it is ‘up to the mark’, the solution is without a hope or a prayer. Users, given time, can be made to learn and adapt to it, but an intuitive UI goes a long way in making the solution acceptable to one and all. No usability, no use.

Attended the conference. It was held at HICC, Hitec City, Hyderabad. Nice venue, but a bit far from the city centre. The rains did not help either. Gladly, had a very good friend to help me with travel.

Spoke on the use of dashboards in the session entitled “Technologies for clinical excellence and business efficiency”. The negative part was that I had to deliver a 20 minute (normally available) in 10 minutes since the previous speakers took more than their allotted time, which was a shame. Being the last speaker and that too immediately prior to lunch was an additional factor and came in the way of any Q&A session.

The attendance in the conference was larger than 2008 although the number of exhibitors were lesser. This was understandable. Delhi is the capital and has many nice places to visit on a one-day or two-day sightseeing trip, which is not really feasible in Hyderabad, charming though the city is.

The quality of speeches was also better. It was most heartening to note that things have begun to get a move on in the field of eHealth. A number of companies showcased their wares and many had stuffs that I always felt were important and needed. My points-of-view were vindicated to a large extent.

I got to renew old acquaintances and make new ones. It was great to meet and exchange ideas. The community, being not a large one in India yet mostly due to lack of serious demand as well as good supply, comprises of a bunch that is capable of doing aplenty. However, things are not yet flying.

This brings me nicely to the very important point of lack of ICT penetration in healthcare. This I will deal in a subsequent blog as I have some of my own opinion and observations regarding this. Suffice to say, the real reason is because of a fundamental error being committed by all sides. Concentrating on technology instead of usability.

What amazes me is that the importance of requirements in any software development, nay, any development can neither be over-emphasised nor is unrecognised. Yet, this is an aspect more ignored than respected. More on this later.

All in all, a successful conference I should say.

A few months ago I had the pleasure of informing that CDAC has come up wth two SDKs to allow healthcare solutions to be HL7 and/or DICOM-compliant with the minimal of pain and without having to maintain a separate group for this.

I take great delight in informing the Indian healthcare IT community that CDAC has launched both DICOM and HL7 SDKs v1.0 in April 2009 on their Foundation Day event and same is now fully available for evaluation and commercial use.

People can download fully functional time-limited evaluation of SDK in any platform of choice (.NET/Java on Windows/Linux/Mac) along with Samples etc. Complete online API documentation is also available at the site. They can communicate with the developers and other users through open forum at http://medinfo.cdac.in/forum, and generate online quote and license purchase through secure online shop at http://medinfo.cdac.in/shop.

The salient features are:

• Common API framework for JAVA and .NET allowing user to target platform of choice.
• Integrates with Rapid Application Development Tools so programmer can continue using the IDE of choice.
• Apart from standard deployable packages, custom packaging allows to target specific memory, storage and cost requirements.
• Cost effective complete implementation of the standard provides high Return-On-Investment.
• Suitable for both a Protocol Expert and a general object-oriented programmer.
• Start early with SDK using variety of samples, tutorials, test codes, documentation available with the toolkit.
• Designed to easily deliver and update revisions to standard.
• Comprehensive support and upgrade options for times when help in needed.

DICOM

• Implements NEMA’s DICOM PS3.0-2004 standard
• Complete object-oriented implementation of the standard
• Manipulate data using Dataset and/or IOD DataModel approaches
• Complete data dictionary with customization flexibility
• Support default or specific Transfer syntaxes for all SOPs
• Comprehensive Error / Warning Logging capability to assist debugging
• Allows customization or extension by implementing provided interfaces
• Validation modes provide ability to work with non-compliant or previous version datasets
• Comprehensive set of tutorials and API documentation along with the user guide and sample test codes
• Efficient handling of memory and native platform multicore / multiprocessing capabilities

HL7

• Implements ANSI approved HL7 v2.5 standard
• Complete object-oriented implementation of the standard
• Implements all standard defined data types, value tables, segments, messages, queries/events
• Provides network communication capability with advanced application-level support for security / compression
• Comprehensive Error / Warning Logging capability to assist debugging
• Allows customization or extension by implementing provided interfaces
• Validation modes provide ability to work with custom or previous version messages
• Comprehensive set of tutorials and API documentation along with the user guide and sample test codes
• Efficient handling of memory and native platform multicore / multiprocessing capabilities

Typically prices are around Rs. 60000 (Indian rupees sixty thousand only) max for each SDK. There is a nominal royalty to be paid (Rs. 500 - Indian rupees five hundred only) for each installation, although royalty-free options are also available. More information is available on their website.

All queries may be directed to the website http://medinfo.cdac.in where complete information regarding SDKs is available.

The SDKs are available only to Indian companies.