We’ve all heard the phrase “garbage in-garbage out”. It is usually quoted in reference to computer generated information in an unfavorable context. Derogatory comments are also heard often about human data entry errors. Regardless of its source, erroneous data distort summarized information drawn by computer programs on which a business depends to render mission critical decisions such as whether to send a $300.00 unpaid medical bill for collection or how much the business is worth to a potential buyer.

Transferring data between computer systems is a permanent part of the business community. Sharing data between dissimilar systems such as a patient management system (PMS) and an electronic medical records (EMR) system is essential. Upgrading from an older PMS or EMR to a newer, more fully featured one is inevitable. These are variations of a data conversion, both must be well managed to be successful and neither can be accomplished without cost.

A data conversion can be done manually or electronically. An informed decision on which method to choose requires a factual analysis of the options. The best decision will be the one which favors the business without regard to clerks who want overtime pay for data entry or the programmer who enthusiastically wants to automate everything in the office.

Manual data entry produces information that has been edited by the data entry person. Usually this approach delivers good results but not always. Using this method, data is entered from reports produced by the old system. Carefully prepared instructions guide the data entry team in cleaning up and standardizing the data. Old patients are easily discarded by simply ignoring them. However, this option is seldom used because the typical office lacks the skills and/or motivation to organize and see a project of this magnitude through to completion.

Starting a new system without pre-loading existing patient data is the most often used form of a manual conversion and is erroneously assumed to be the “no cost option”. In fact it is the most costly option. The old system is used for collections until all monies are collected or written off. The new system is used for all new business. All patients, old and new, are considered new on their first visit after the new system is implemented. They are asked to complete new patient forms from which the staff enters their data into the new system. This actually increases the cost of the data conversion. Benefits of this approach include eliminating old patients and spreading the cost over a longer period of time. The drawbacks include more admin time for every patient; the loss of valuable patient demographic data, gathered at significant cost over a long period of time; some patients will be annoyed; all patients will be inconvenienced; and most often these costs are not visible to the doctor.

An electronic data conversion will not introduce errors, it will reduce real costs, and it can make the transition to a new system far less painful. Many errors in the old system can be cleaned up without additional cost. Conversion timing can be coordinated with the implementation of the new system to nearly eliminate duplicate data entry. Old patients can be dropped using a combination of date of last visit and a balance of zero. Training on the new system is accelerated and patients experience a “seamless” transition when the staff has immediate access to the old system’s data.

Hiring the doctor’s nephew is probably not the best way to a successful conversion.

A successful electronic data conversion requires three key ingredients:

1. Hire a skilled database professional who understands medical information and its terminology.

2. Assign the most knowledgeable user to the project.

3. Select patient records that can be used to validate the data conversion and develop a plan to validate the conversion.

 Finding the data to convert is far more difficult than converting the data into the new system. The database professional skilled in medical system databases is a far better choice for this work than a typical programmer.

However, it is not necessary for the conversion analyst to have experience with the client’s old PMS. Converting data electronically requires more analytical skills than programming. I.E. finding the data and identifying links required by the new system. It is easy to say “the data isn’t in your files” but that usually is a weak excuse for not having the skills or not being willing to take the time to find the required data and/or the related linking information. If patient data can be seen in the old system, it is in the old database or is what we call a “calculated” value. Calculated values exist almost exclusively in financial data, which can be converted but often are not. On a rare occasion a skilled conversion analyst will be unable to find a piece of data. In such cases you should be provided with a logical explanation.

The second most important part of a successful electronic data conversion is a willing user who has expert knowledge of patient data in the old system. The user will validate the conversion by checking patient information from the new system against known patient information in the old system. It is necessary to allocate sufficient time for the client to thoroughly test the converted data files.

Third, select 10-20 patient files whose information varies enough to cover known, usual and unusual conditions found in the patient files. Examples of conditions to include in the test file: new patient, zero balance, old balance, guarantor, no guarantor, no insurance, etc. The converted data will have been checked for accuracy and completeness but the conversion team is limited in its ability to fully validate the data. The client must take the selected list of patients and validate the converted data and validate it against the old system data. This should be done immediately after receiving the converted files while corrections can still be made by the conversion team.

Finally, none of the options will produce a perfect data conversion, however, an electronic conversion is by far the most cost effective option in terms of real dollars. It provides a higher level of accuracy. Timing is predictable. And, in terms of large patient files, historical information, images and chart data, the electronic conversion is indispensable.

KW Norris

Vice President

Technology Consultants, Inc.

4125 SW 185th Avenue

Beaverton, OR 97006

503-356-4105 ext 11

503-939-9223 cell phone

503-356-4109 fax

kw@tech-consultants.com

www.tech-consultants.com

Mr. Norris is not an employee of InvestMed.  He co-founded Technology Consultants in 1995. Prior experience includes VP of Sales and Marketing for an IT recruiting firm, MIS Manager, IT Consultant and Software Developer. He earned a BS in Computer Science from Brigham Young University. He serves as Alumni Chairman in Portland and is active in fund raising for scholarships.