Printer Friendly

Data bank incomplete and future cloudy.

The National Practitioner Data Bank was established by Part B of the Health Care Quality Improvement Act of 1986, also known as Title IV of Public Law 99-660. The other part of this law, Part A-Promotion of Professional Peer Review, offers immunity to peer review bodies of hospitals and to persons serving on or assisting such bodies from private damages in a civil suit under federal or state law. It appears this part has benefitted peer review, although a U.S. Supreme Court decision in 1991 established that federal jurisdiction on antitrust in peer review could have a chilling affect. Part A was self-implementing, so the U.S. Department of Health and Human Services (HHS) did not issue regulations. This article focuses on the Data Bank and does not cover peer review.

Regulations for the Data Bank were published in final form on October 17, 1989 (45 CFR Part 60) and are included as an appendix to the NPDB Guidebook, which contains deitaled guidance for reporting, querying, and understanding the Data Bank. The Guidebook is available for $7.50 from the Government Printing Office. Although useful for someone who has little knowledge about the Data Bank, the Guidebook is now out of date. It was published in the Spring of 1990 and the intent was to update it regularly, but this has not been done. HHS has stated official answers to many additional questions since publication and has changed and added policies and procedures, but they have not been distributed uniformly.

When the National Practitioner Data Bank opened in September 1990, the reporting and querying of malpractice payments and adverse actions primarily against physicians were the only reportable actions. Section 5 of the Medicare and Medicaid Patient and Program Protection Act of 1987 (Public Law 100-93) authorizes the expansion of the Datab Bank to include reporting of licensing adverse actions against all health care practitioners. Although four years have passed since passage of this record law, proposed regulations have not yet been published by HHS, and there currently are not any known plans to publish them. Twenty months passed between proposed and final regulations for the first phase of the Data Bank, Title IV. There does not appear to be pressure from the nursing and allied health professions to urge HHS to implement Section 5 of the Data Bank. The National Council of State Boards of Nursing has changed its mind and now plans to keep its disciplinary data bank in operation. Although Title IV now allows hospitals to submit clinical privileges adverse actions to the NPDB on nurses, HHS said that as of January 3, 1992, only 35 of 1,039 hospital clinical privileges reports were submitted for practitioners other than physicians and dentists.

Title IV and Section 5 are the two main laws authorizing the Data Bank, but there are several amendments to them, including OBRA 1990-- the budget and catch-all law that affects many areas of health care. HHS decided to implement the Data Bank using an outside contractor and, after evaluating competitive proposals, awarded a contract to the Unisys Corporation at the end of 1988. The contract amount was $15.9 million for five years to develop Title IV, Section 5, and the research service. The research service is the reporting of information from the Data Bank to any person in a form that does not permit the identification of any particular physician, other health care practitioner, or patient.

Contractor and Cost Problems

Both Title IV and Section 5 were both to have been operational by September 1, 1989, according to the contract, but only Title IV was implemented and it was one year late. The research service was scheduled to open in September 1991, but it is not clear when it will begin. Despite this, spending to contractors has apparently been running much higher than the planned $3 million or so a year. Because of huge backlogs that resulted in delays of months to receive responses to queries, Unisys hired two subcontractors to keypunch report and query forms. And because of a critical General Accounting Office report in the summer of 1990, HHS hired another computer contractor to oversee Unisys and advise HHS on technical aspects of development and operation of the Data Bank.

Despite Data Bank costs that are four times higher than in the Unisys contract (see below), an executive of one of the largest malpractice insurance companies said that in 1992 there continue to be "rather frequent" data entry or keying errors in the many malpractice reports they submit to the Data Bank. He said that reports that are miskeyed and then corrected by the insurer sometimes are found to have other errors after the initial errors have been corrected. Physician executives should advise their doctors to study malpractice report verifications very closely, as the data on the reports may not be what the insurer submitted.

These much higher than expected costs resulted in a quick increase in the query fee from $2 to $6 in spring 1991. Because of the mass query of the entire medical staff of a hospital required by law every two years, this caused concern at hospitals, as well as at state licensing boards. Hospitals are required to query, so they could not do much about it. But state medical and dental boards are not required to query, only to report. As a result, most of these boards have not queried the Data Bank. As do some medical malpractice insurance companies, some boards now ask the applicant to submit a self-query, for which there is no charge by the NPDB, or just use the data Bank of the Federation of State Medical Boards (FSMB), which does not charge state medical boards for full access to its disciplinary reports. FSMB reports now number almost 30,000 since it began its systematic data bank in the 1950s. FSMB data do not include malpractice payments. Some medical boards are irritated that they have to tract payments of a few hundred dollars from the federal Data Bank, which are nuisance payoffs.

Soaring Data Bank costs also are the reason given by HHS for not implementing the research service or Section 5. The Congress appropriated $6 million for the Data Bank in the 1992 fiscal year (including $5 million in user fees), about double the $3 million per year in the Data Bank contract with Unisys. Although the costs have doubled, less than half of the functions are or will be operational in 1992, and the volumes are lower than projected. Therefore, the federal government is paying Unisys about four times the amount in the contract for work that some users state is not of satisfactory quality. It is puzzling why costs in 1992 have soared, because the current software has already been paid for and future phases of the Data Bank are apparently not being developed this year. There is a rumor that HHS asked the AMA and FSMB in 1991 to take over the Data Bank, but they refused.

Even with these expanded funds, work in 1992 on the research service and/or Sections 5 for 1993 implemention seems unlikely. And if query volume decreases to the Data Bank. HHS may decide to increase fees again just to fund continuing Title IV operations. Several Data Bank experts recently confirmed that development of Section 5 in 1992 is deferred. HHS wants to go ahead with the development of the research service in 1992 but has no opening date, so it still could be a late as 1993 before it is operational.

HHS in January reviewed a draft small claims study to determine whether to stop collecting reports on malpractice payments of less than $50,000. Some persons at HHS and insurers favor a floor under which payments would not be reported. The AMA has advocated this since 1986, when the statute was created, so there may be a real possibiblity for a change by Congress. A public report to Congress on the Data Bank is due in March, and the results of the small claims study may be mentioned.

Research Service

The Public Health Service Amendments of 1987 (Public Law 100-177) amended the Data Bank in several respects, including the authorization to establish the research service by adding the following to the end of Section 425 of Title IV: "Information reported under this part that is in a form that does not permit the identification of any particular health care entity, physician, other health care practitioner, or patient shall not be considered confidential. The Secretary (or the agency designated under section 424<b>), on application by any person, shall prepare such information in such form and shall disclose such information in such form."

HHS and Unisys have not worked out procedures to respond to research requests. Although the Data Bank is programmed in standard IBM Cobol and CICS computer language, Unisys told the Data Bank Executive Committee in July 1991 that the way the software is set up, requests for data under the above law cannot be easily made. (The Executive Committee is a group of 18 health care leaders from the AMA, AHA, and other associations who meet two or three times year to be briefed by HHS and Unisys.) Unisys stated that data could not be provided without extensive programming. This is puzzling, because all that is necessary to comply with the above four-year old law is to blank out certain identifier data elements, such as practitioner name, license number, etc., and the computer hardware and software is all standard IBM format. If HHS supplies a computer tape of Data Bank reports without identifiers, outside researchers could have it analyzed in a few weeks. Unisys would not even provide the Data Bank's Executive Committee in July with requested basic data, such as the number of queries, so that they could be compared to FSMB data.

Why are these research data important? As suggested above, they could verify whether or not the Data Bank is obtaining complete data by comparing NPDB data to FSMB data. Also, they could be a resource of malpractice information for physicians and their organizations. Malpractice insurance companies are not interested in the data because they already have extensive data. But physician executives could learn much from this data. As described later, the Data Bank appears to be forcing many more claims into litigation rather than the heretofore much more common settlement outside the court. The research service data would be most helpful for physicians, and both plaintiff and defense attorneys, in pricing claims on almost a real-time basis, because a report has to be filed with the Data Bank within 30 days of payment. The great majority of reports in the Data Bank are malpractice payment reports, 84 percent (21,090 of 25,034) as of January 3, 1992. The other two types of reports are state medical board reports and hospital clinical privileges reports, which numbered 2,810 and 1,093 for the first 16 months of the Data Bank. There were also 41 adverse actions from professional societies.

The research service is also important to certain consumer groups who advocate that all data in the Data Bank, with practitioner identifiers, should be open to the public. (No group wants data revealed with patient identifiers.) Ralph Nader's Public Citizen Health Research Group advocates complete disclosure, as does the People's Medical Society of Emmaus, Pa., which publishes a great deal on consumer health care issues. Even the congressional sponsor of the Data Bank, Rep. Ron Wyden (D-Ore.), may want complete release of data, with practitioner identifiers, from the Data Bank. But the votes to pass this are not currently available in Congress, so it will not happen in 1992.

Security, Confidentiality, and

Reliability of Data

While there is an argument for complete openness of data from the Data Bank, there is also a good argument for the data confidentiality that currently is supposed to exist. Most of the data in the Data Bank are malpractice payment data for payments of any size, even one dollar. Dental groups have sued HHS over the requirement to report very small payments. Clearly some and perhaps even many of the payments reported to the Data Bank are nuisance payments by insurance companies to avoid much larger legal and other expenses to pursue a defense of a claim. Title IV requires a small-claim study to determine if there should be a $10,000 to $30,000 floor, under which payments do not have to be reported, as described above.

The American Medical Association continues to be concerned that the security, confidentiality, and veracity of data are not satisfactory. Its concerns may be well founded, because anyone can obtain a Data Bank identification number and be registered on the Data Bank merely by signing a registration form. There is no separate verification that the medical group practice, hospital, HMO, or state agency is a legitimate, operating health care entity under Title IV. The AMA continues to urge that responses from the Data Bank be sent by certified mail with return receipt, and that the Data Bank verify that the entity conducts formal peer review as required by Title IV. But HHS will not accept either suggestion.

Data Bank forms may be ordered by anyone, without providing a Data Bank ID. Photocopies and facsimiles of blank forms are allowed so that anyone in a hospital or other entity could obtain one, along with the entity's ID, and submit a false report that would be accepted or send a query and intercept the response at a new address he or she submitted for the entity or from the mail room or mail slot of the correct address. The response from Unisys has been that the fines under the law will prevent these breaches of security. Fines may discourage fraud, but they certainly do not provide the security to prevent it, because there continues to be absolutely no verification of the existence or validity of many of the 14,000 or so entities in the Data Bank. The data for most of these entities are more than two years old and have not been verified since 1989 with independent data.

HHS has paid Unisys and others almost $10 million or more so far. The FY 1992 budget is $6 million, and there are not even elementary safeguards, such as checking with hospital associations at least yearly to learn which hospitals have merged, gone out of business, or are brand new. A caretaker or laid-off manager in a rural Texas hospital or one of the 100 or so other hospitals that go bankrupt or merge each year could be feeding reports and/or making illegal queries of the Data Bank. Because there has been no effort to check any of the entities with outside sources, such as licensing agencies, associations, etc, to ensure they are legitimate, currently operating entities, there are inactive entities on the Data Bank file that are improperly classified as active. Who's to say that a few of these are not being used improperly? And if one is caught, who is going to prosecute and fine the person committing the fraud? Unless the fraud is repeated and significant, it is unlikely the HHS Inspector General and the Department of Justice would prosecute.

A computer-generated verification copy of an adverse action or malpractice payment report submitted to the Data Bank is routinely sent to the accused physician. But if the address for the doctor is an old one or a false one, the copy may never reach the physician because Unisys does little or nothing to correct addresses for verifications returned by the Postal Service. Many physician addresses on malpractice reports are almost five years old and out of date. Unisys prohibits staff from calling insurers for current addresses. Also, as described above, after almost two years of operation, there may be extensive data entry errors.

There also have been computer system bugs that have often prevented delivery of verifications, such as the fact that no foreign country is printed on the address of the verification, not even Canada. This occurs even though the name of the country is provided by the submitter on the original report.

As a result of Unisys' lack of follow-up work to correct addresses and fix computer bugs, hundreds and possibly thousands of verifications have not reached physicians. They have been returned by the Postal Service to Unisys, which has not reprocessed them. Therefore, it is possible that a false or incorrect report could be accepted by the Data Bank and never be disputed because the practitioner never received the verification.

The reliability, accuracy, and completeness of data in the Data Bank have yet to be proven. In late 1990 and early 1991, there were a number of cases of errors, but these could be excused as start-up problems. By now the system should be running flawlessly, but is it? An informal survey in January 1992 revealed that many believe the Data Bank is very expensive, considering the little new data it provides and the administrative burden to hospitals and insurers. The number of malpractice reports filed with the Data Bank in its first year of operation is far fewer than any projections. At some point it is imperative that an analysis be done to verify that there is full compliance by reporting entities and complete, proper processing by Unisys. This activity was part of the Unisys contract, but, like other activities, it has fallen by the wayside.

HHS recently affirmed that it has no plans to require more than self-certification when an "entity" submits an entity registration form or a Data Bank report or query form. HHS said it is impossible to determine if an entity has formal peer review without a big staff to check it. The Public Health Service of HHS, which oversees the Data Bank, has no plans to study entity compliance, although it has noticed cases where state boards know of adverse actions that hospitals are not reporting within 15 days. The HHS Office of Inspector General published Data Bank regulations for civil money penalties in May last year. The Office stated in 1990 that the first year of Data Bank operations would be a grace period, so perhaps there will not be any investigations until 1993. HHS has no plans to study the affect of the Data Bank on the medical malpractice businesses, but it may change its mind in the future if some of the early clues prove true.


As of January 3, 1992, there were 21,090 medical malpractice reports submitted to the Data Bank. This is for the first 16 months of operation and is less than half the number expected. Also, 1,055,117 queries have been made, also less than the expected number. Only 11,132 matches were made. Some experts believe the number is low because fewer physicians are willing to settle because that would result in a report to the Data Bank. For malpractice insurers with consent clauses in their insurance contracts, some physicians are withholding consent to pay off small frivolous claims, forcing them to go to juries that will reject the claims. Another reason for the low number may be that many claims may have been expedited and prepaid prior to the opening of the Bank. Also, there may be a compliance problem, although virtually all medical malpractice payers were identified prior to the opening of the Data Bank. They were sent complete materials, invited to training, etc. It may also be possible that Unisys has not processed all reports that it should have.

Some in the medical malpractice insurance business in 1988 and 1989 opposed the large number of data elements being collected. Early in 1991, HHS proposed to expand the elements but, after receiving 66 mostly negative responses decided not to proceed. HHS is working with insurers to obtain malpractice research data separate from the Data Bank.

Disputes Concerning Data

Bank Reports

Both the Title IV statute and regulations, policies, and procedures establish the right of a physician to dispute the facts of a Data Bank report. HHS, hospitals, and medical malpractice insurers all urge that the respective parties review and discuss the report before it is submitted and informally resolve disputes before submitting a formal dispute. Despite this, more than 2,000 disputes have been submitted so far, and, as of mid-January 1992, HHS has resolved or is close to resolving 30. HHS has approximately 70 disputes pending for which it will decide in favor of the practitioner, typically a physician, or the entity, usually a medical malpractice insurer or a hospital. Therefore, despite a large number of disputes, HHS is requested to decide a small number, about 100 so far. Because of protracted legal review at HHS, the 100 disputes reaching the Secretary level take a long time to decide.

More and more reports in the Data Bank are being flagged as disputed, and recipients of disputed reports may not be able to easily use the data. A reviewer of a disputed report may check with the Data Bank at HHS at 301/443-2300 to find out if the practitioner has formally asked the Secretary of HHS to review the report. In most cases, the answer will be no, and the dispute flag in the report will be dropped so that the report is accepted. The AMA is recommending that HHS place an explanatory statement in the disputed computer report until the dispute is resolved, but HHS has refused so far to do this.


The time has come for Congress and the Administration to reassess the incomplete, poorly operating, and very expensive Data Bank. A number of critical articles have been written in recent months, and the report to Congress on the Data Bank is due this month. No group involved with the Data Bank is satisfied with its quality, cost, or performance.

Jim Lapinski is an independent health care consultant in the Washington, D.C., area. He spent much of the past four years working on the National Practitioner Data Bank as an employee of Unisys Corporation.
COPYRIGHT 1992 American College of Physician Executives
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:National Practitioner Data Bank
Author:Lapinski, Jim
Publication:Physician Executive
Date:Mar 1, 1992
Previous Article:Efforts to implement national health reform in the United States.
Next Article:Analyzing vendor and supplier contracts.

Related Articles
Supersensitive supersecrecy.
Meeting focuses on future AMA leadership role.
Club Med; how the medical fraternity keeps the government from exposing the skeletons in doctors' closets.
The National Practitioner Data Bank: the first 18 months.
Changes to National Practitioner Data Bank reporting - effective or not?
Constitutional rights versus malpractice insurance settlements.
Doctors with lax manners tend to get sued most, Vanderbilt studies show.
Researchers question reliability of National Practitioner Data Bank.
The medical industry: the errors of its ways.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters