14 February 2005. Good research strategy entails verifying the information you find to ensure it is accurate, complete and up to date. While not every question demands research, professionals often depend on online search tools to start the process.

A lawyer who wants to find an appellate-level case on point, for instance, or words and phrases related to, or synonymous with, key terms that describe a certain legal scenario, might first conduct searches via LexisNexis’ free LexisONE service. It's a frugal, quick, and many times productive, way to initiate case law research. But woe be to the professional who relies on it -- or any other search tool -- as the sole means for finding information for making critical business or legal decisions.

An issue brought to light last week serves as a case in point. A member of the electronic discussion group, LAW-LIB, submitted a question about LexisONE that led to concerns about the keyword searching capabilities of the service. You have to enter a keyword to launch a search -- unless you know the exact case citation. But keywords do not retrieve several important U.S. Supreme Court decisions.

As law librarians David McFadden and Kent Olson discovered, you cannot retrieve cases decided prior to 1908 or between the years, 1945 and 1975, with a keyword query.

LexisONE's coverage of U.S. Supreme Court opinions runs from 1790 to present. Enter a Supreme Court case citation, and you will find it, including some earlier decisions reported in U.S. Reports. But, as law librarians David McFadden and Kent Olson discovered, you cannot retrieve cases decided prior to 1908 or between the years, 1945 and 1975, with a keyword query.

The question presented to the group pertained to strategy for finding cases involving a particular judge or attorney. The requirement to include a keyword seemed to complicate the task, especially since the search form allows for entering a litigant's, judge's or attorney's name.

Olson suggested formulating the query with propriety Lexis syntax. For example, searching opinionby(name) as a keyword finds decisions written by the judge whose name you enter. You might have to run several queries with different date restrictions because LexisONE limits retrieval to 100 cases per search.

Another technique involves entering a keyword that will always appear in a case; for example, court or opinion. Then use the search limits (litigant, judge, attorney) to find decisions involving a particular judge or lawyer. However, because the LexisONE qualifier for "judges" retrieves cases heard by a particular judge -- whether or not the judge wrote the opinion -- it actually yields broader results than the more precise query Olson recommended.

But when McFadden tried these strategies, he discovered he couldn’t find key U.S. Supreme Court cases, such as Plessy v. Ferguson (1896) or Brown v. Board of Education of Topeka (1954). Additional keyword queries, such as public education, right to remain silent, or the single term, discrimination, found gaping holes in the search results. Some queries limited to the years, 1945 to 1974, found no cases at all.

A spokesperson for LexisNexis said the company began investigating the problem as soon as it learned of the LAW-LIB discussion. "We have discovered that the underlying structure of our Supreme Court databases changed recently, due to the size of the databases. This change did not have an impact on our for-fee customers, but it was not properly translated into the LexisONE display. We are in the process of making the appropriate corrections so our LexisONE users will have access to the full date range of the Supreme Court opinions. We expect this correction to be made over the weekend."

This morning, however, the trouble affecting the database remains uncorrected.

Technical glitches could plague any online service, whether free or not. But when they corrupt a portion of a database, or cause a single feature to malfunction, they may go undetected for a while. Had McFadden not puzzled over the missing key cases he knew existed, the problem might not have been brought to light quickly. And if it hadn’t, professionals conducting business critical research would have had one option for avoiding potential disaster -- verification through some other means.

At the risk of belaboring this point, I ask readers to recall the consequences of unverified research performed a few years ago by a Johns Hopkins doctor. His task was to discover whether there were any known potentially harmful affects in using the chemical, Hexamethonium, as an inhaler for asthma patients. His research found none, and later, an asthma study patient died. According to Baltimore Sun news accounts (search the Baltimore Sun archives for hexamethonium), though, evidence of a link between the drug and lung damage existed in medical articles published during the 1950s.

Don't let bad research habits hide critical information. Verification might not always save a life, but it might save your case.