| Index | Last Week | Contact DQ News | Latest Issue |
| Featured Articles | Data & Information | Science & Technology | Education |
| Economics | Business | Law | Medicine |
Scanning Problems May Cost Retail Consumers Billions
An investigative article in the September, 1998 issue of Good Housekeeping magazine found that errors by checkout-line scanners are much more common than consumers realize. Retail overcharges may be costing consumers billions of dollars each year.
According to a 1996 Federal Trade Commission study nearly 5% of all retail transactions in the United States are scanned with an incorrect price. And other investigations by states attorneys general and consumer protection agencies have disclosed that the percentage is significantly higher for department stores, drug stores, hardware stores, and discounters. Overcharges may be two or three times as likely as undercharges.
According to Good Housekeeping, bar code scanning equipment is not to blame. Rather, the cause is bad store management, underpaid, undertrained, and overworked retail employees, and a careless and often negligent attitude by retail managers and executives. Unlike the grocery industry - which has spent millions of dollars hiring price verification managers to ensure accuracy - discount department stores and other retail outlets continue to allow erroneous prices to be generated in several ways. The first way errors occur is when incorrect prices are sent to stores from corporate headquarters. Another way errors occur is when there are sales and seasonal increases in business. Retail employees are hard pressed to post and change sale prices, or even post prices at all on some seasonal merchandise. A third way erroneous prices occur is when competitors lower the prices on the same or similar merchandise. Again, the posted price may not be the same as the scanned price.
The U.S. Commerce Department has set an industry goal of a 2% error rate - with the error equally divided between overpriced and underpriced items. Presently, the pricing error rate varies from about 5% to over 25%. Enforcement of laws against overpricing and retail price errors vary from state to state, with Michigan and California taking the lead. In these two states a combination of bad publicity, tougher regulations, and stiff fines has not only reduced the pricing error rate but underpricing mistakes have begun to exceed overpricing mistakes. The article was written by Bob Trebilcock and appears on page 110.
Historians Question Numerical Assessments
An article recently published in Lingua Franca - a magazine for college and university academics - questions the accuracy of historical data about long-dead societies, as well as estimates of prehistoric activity. While the field of "demographic historiography" has grown significantly as a specialty, only a few cultures in world history ever had the means, or the inclination, to make accurate numerical assessments of themselves.
Although China, Egypt, and Western Europe are well known for their large body of statistical records reaching back millenia, the same cannot be said for sixteenth-century West Africa or the societies of pre-Columbian America. These societies did not keep close track of their populations, and in many cases there were no written records. Therefore, much historical data about these societies is besed upon "educated guesses" at best. Much of the "data" about pre-Columbian America comes from early Spanish military and missionary documents, and incomplete censuses. Unfortunately, historians have used conjecture as much as poor-quality data to estimate population in pre-Columbian America. Data often drive historical theories. And getting historical theories accepted as "facts," drives the careers of historians.
Even when there are written historical records, statistical controversies can be heated. Estimates of the total number of Africans transported across the Atlantic between 1500 and 1900 have ranged wildly between 3.5 million and 100 million. Attempts by historians in recent years to estimate the number of Africans brought to the New World have hovered between 10 and 15 million. These estimates were derived from historical records and "guesswork."
According to Lingua Franca, most demographic historiographers believe that historic data should be viewed critically and with skepticism. If the last few American decennial censuses failed to count a significant number of Americans who belong to minority groups, how precisely can one estimate the population of societies that ceased to exist 400 years ago from the scant information available about them? The article was written by Lawrence Osborn and appears on page 49.
SEC Chairman Attacks Data Manipulation
In a speech delivered at New York University on September 28th, U.S. Securities and Exchange Chairman Arthur Levitt attacked the manipulation of earnings data to artificially and temporarily support stock prices, otherwise known as "earnings management." Chairman Levitt criticized companies that use certain aggressive accounting techniques to inflate their quarterly earnings and said the SEC would soon issue new accounting rules and guidelines intended to halt the abuses. He also called for a review of how the nation's public accounting firms use certain aggressive accounting techniques.
Mr. Levitt said that the Commission's enforcement division would focus on corporations that use certain accounting methods to "manage earnings" so profits can be increased or decreased at will in such a way that the bottom line does not reflect actual operations. He said that the commission was frustrated with companies that use factory closing or work force reductions as opportunities to take millions of dollars of one-time charges for "restructuring." The commission has also been critical of companies that acquire other companies and then write off much of the purchase price by calling it "research and development." The SEC recently blocked America Online from reporting its fiscal fourth quarter earnings for nearly two months because of disagreements over how much AOL should write off in its recent acquisitions.The commission has found that some companies take R & D write-offs after acquiring companies that don't do much R & D.
Mr. Levitt also called on the Financial Standards Accounting Board to pass new accounting rules quickly. He also asked both the American Institute of Certified Public Accountants and the Public Oversight Board to review whether auditors should change the procedures in performing an annual audit. Commissioner Levitt listed a total of nine actions he wants taken by accountants, the SEC, corporations, Wall Street, and investors. He wants investors and stock analysts to punish the stocks of corporations exposed telling half-truths.
Articles about Chairman Levitt's speech appeared in The New York Times (staff reporter Melody Petersen), The Wall Street Journal (staff reporter Elizabeth MacDonald) and USA Today (columnist David Henry) on September 29th.
Will Software Clean-Up Medical Data Mess?
A front-page article in the October 2nd issue of The Wall Street Journal examines efforts by software firms to solve the problems that have resulted from the torrent of medical information physicians must deal with every day. Physicians complain that they are drowning in paper. One software company featured in the article, Healtheon Corporation, has embarked on an ambitious project to fix the chaos of medical paperwork.
Medical information problems begin with the fragmented nature of medical data. There is no master databank for patients, and privacy advocates don't want one. Instead, limited information is scattered among doctors, insurers, labs, pharmacies, and employers. Data quality is erratic, and many parties don't trust the others enough to share information freely. Even if they wanted to, many parties can't share data and information because computer systems are incompatible. Moreover, every aspect of medical record-keeping is hard to standardize, and thus hard to automate. And physicians don't like to be ordered around by computer programmers. For all these reasons, high-tech attempts to resolve the medical data mess have been disappointing so far.
According to the Journal article, Healtheon's first major headache was caused by poor data quality. When Healtheon implemented a clerical system for a large group of San Francisco physicians, the system's designers discovered that obtaining reliable data about patients and their families was a serious problem. Healtheon had to resort to "inspect/repair" to obtain an acceptable quality level. Healtheon also ran into serious problems with insurance company approvals for specialist referrals. But the long development time needed to rewrite and debug software did not build confidence with the physicians who used the system.
According to the Journal, the opinion of those involved in designing and implementing medical data software systems is that development of such systems is, and will continue to be, a long drawn-out process. The article was written by Journal staff reporter George Anders.
Disputed Data, Lethal Consequences
A front-page article in the September 30th issue of The Wall Street Journal explored the consequences when warning labels for drugs that have serious side effects are inadequate. One drug, the analgesic Duract, was recently taken off the market because warnings of possible liver damage were not strong enough to prevent the drug from being mis-prescribed.
Of 830 patients who had taken Duract in clinical trials, 23 had unusually high levels of certain liver enzymes that indicated their liver cells were being destroyed too quickly. One FDA physician concluded there was a significant potential for serious liver damage resulting from long term use if the FDA's strongest warnings weren't included in prescription information provided to doctors, patients, and pharmacists. The pharmaceutical company that produced Duract, Wyeth-Ayerst, argued for less-stringent warnings, including a warning that doctors monitor patients for liver problems after 30 days. Wyeth-Ayerst's labeling proposal prevailed.
Unfortunately for patients, some physicians didn't read the prescription information. They not only didn't order liver tests for patients who took Duract for more than 30 days, they gave their patients prescriptions for hundreds of Duract pills. Patients began to become ill as their livers started to malfunction. Eventually the liver damage in several patients became so severe that they required liver transplants. The FDA ordered Wyeth-Ayerst to put its most stringent warnings on Duract labels. Although physicians started limiting Duract prescriptions to 10 days, patients continued to suffer life-threatening liver problems. After four patients died, the drug was pulled from the market in June. (Apparently, not every physician, pharmacist, and patient received the updated warnings about the drug.) According to the Journal, several other drugs have recently been taken off the market because of unforseen side-effects. The article was written by Journal staff reporter Rochelle Sharpe.
Go to: Data Quality Home Page
Comments: dqemail@aol.com (10/01/98)