|
Home |
Software Providers |
Consultants |
Articles |
Columns |
Reviews |
Headlines
|
|
|
- |
| - |
|
- |
Entire contents Copyright © 1999 Business Insurance "In search of reliable data; Electronic claim processing can
lessen chance for errors"
|
|
INTEGRITY, OR LACK thereof, was a common theme during the recent election campaign. But integrity is not only important in political candidates. It is also appropriate in risk management information systems. More specifically, how good is the data used by that risk management information system? Indeed, this topic is more important than any discussion of technological capabilities of evolving systems and vendors or how to design and select the system. Integrity is critical because most people -- especially managers -- tend to believe that anything that comes from a computer must be right. There are several places where data inaccuracy becomes a problem and, more importantly, what the risk management professional can do about it. When we think of data integrity, we must first think of the initial creation of the data: claim record, location number, employee, Social Security number, claim date. In short, creation begins at the employer. Ultimately, the data is transmitted through several sources to a record. (Throughout this article, I will use the example of a claim, because that is the most common data element evaluated and used by most risk management information systems.) Obviously, the real first exposure is to the claims organization or the employer, which may get the wrong information on the claim in the first place. Such errors frequently happen in workers compensation claims: The risk manager, seeking accurate payroll information to calculate the average weekly wage, must estimate the initial amount used to calculate the compensation rate and then submit it to the claims organization. That figure must later be revised to reflect the accurately produced payroll record. This involves rekeying of information or editing existing data. But in general, the first great risk of inaccurate data is in the reporting and recording of that claim record. The old process involved filling out accident reports or first reports of injury at the employer and sending the reports to the risk management department, which in turn sends the reports to the insurance broker or third-party administrator. The risk of data inaccuracy increases as data moves onward and is rekeyed. Self-insured and self-administered employers maintain the entire risk of data accuracy because they generate the raw data, input it into an information system, evaluate it and then handle reporting. That can be good if well-designed editing and control points are built into the claims/risk management information system and self-administered claims department. Unfortunately, many that I have seen do not have such controls. For example, one large self-insured and self-administered organization that had no built-in editing system to evaluate data accuracy and appropriateness was found to be $ 20 million dollars underfunded after audits. The secondary data sources --insurers, third party administrators, brokers, RMIS vendors -- are the next line of defense against data inaccuracy. These organizations are expected to catch errors made by the employer or, at least, catch mistakes made in transmitting the data from one of these secondary sources to another. That may be a major, naive assumption. Consider, for example, how many mistakes can be made in an average loss run where a manual entry of claimant, claim number, date of loss and location number or code are required entry fields. Many risk management information system vendors tell me that location code mistakes are among the most frequent problems. That may have been a mistake made at the employer level, but manual data entry at the secondary data source can also cause problems. This can be proven by the number of duplicate claims, erroneously opened and closed claims, and even liability claims which become property claims or vice versa. Where are these problems coming from? Typically, the answer lies in the conversion process. As we know, employers frequently change their service providers. And, because most service providers' claims information systems are different, the data from the older source must be accurately converted to the newer source. Who performs this conversion routine? Typically, it is the same type of individual who is entering claims information at the service provider -- one with little or no understanding of the claims or insurance process. Data entry personnel may not understand how critical it is to accurately translate the data from the old source to the new system without making any changes or arbitrarily inputting data to fill open fields. Disturbing as this seems, it probably happens more than the general public knows. In some cases, the original data may be abysmally incomplete. That is typically the case when one moves from an insurer-based program to a self-insured situation where the risk manager is looking for much more data to analyze. Historically, insurance companies have only collected information that was useful to them; they weren't very interested (although that is now changing) in capturing a lot of sub-data on a particular claim which is very useful to the risk manager today. This conversion process involves specific and important steps: mapping, file formatting and trial input. But the point here is that the conversion is a critical function and is many times overlooked by those interested in moving to a different RMIS vendor or other service provider. A conversion process that identifies errors in the old data will provide some comfort to the risk manager. Data can be transmitted from one source to another in three major ways:
One might say that the problem with data integrity will be reduced as we move from manual to electronic transmission. This may be true, but the problems of accuracy and integrity remain in terms of verifying data accuracy in the exchange and making sure that the conversion software routine does not mistakenly alter the data during the conversion process. Copyright© 1992 Business Insurance |