README.TXT for National Practitioner Data Bank, adapted by IRE/NICAR Reports Received from September 1, 1990 through March 31, 2009 updated June 2009 ------------------- DATA INTRODUCTION ------------------- The National Practitioner Data Bank Public Use Data File contains selected variables from disclosable malpractice payment and adverse licensure, clinical privileges, professional society membership, and DEA reports (adverse actions) received by the National Practitioner Data Bank concerning physicians, dentists, and other licensed health care professionals. Information provided to the National Practitioner Data Bank is confidential and can be provided only to authorized users, such as hospitals, HMOs, and state licensing agencies for puroses including professional credentialing,licensing and peer review. However under section 426 of the Act (42 USC 11135), as implemented by regulations at 56 CFR 60.11(a)(7), data may be released to "a person or entity who requests information in a form which does not permit the identification of any particular health care entity, physician, dentist, or other health care practitioner." This file is released in accordance with that provision to facilitate research use of National Practitioner Data Bank information by persons interested in medical malpractice, licensing, discipline, and quality assurance issues. To maintain confidentiality, Variables which identify or would allow identification of individual entities or practitioners are excluded from the public use file. In order to assure confidentiality to all types of practitioners, the smallest geographic unit identified in each record in this file is a State. Personal details about the practioners, incident dates, patient details, or settlement amounts that could identify a specific case are described using standardized categories or ranges. It also includes reports of Medicare and Medicaid exclusion actions taken by the Department of Health and Human Services Office of Inspector General. Malpractice payers, state licensing agencies, hospitals, other entities, and professional societies are required to report this data to the National Practitioner Data Bank under the provisions of Title IV of P.L. 99-660, the Health Care Quality Improvement Act of 1986, as amended. Please note that reports are required for malpractice payments involving all types of licensed health care practitioners. Reports concerning adverse actions are required only for physicians and dentists. Adverse actions taken against other types of licensed practitioners may be reported voluntarily and are included in the data file. Therefore, the NPDB's parent agency strongly cautions users that the adverse action reports in this data set for practitioners (other than physicians and dentists) represent neither the universe of all actions taken nor a random sample of such actions. Exclusions are included for all types of individual (as opposed to institutional) practitioners. For specific detailed information on the National Practitioner Data Bank and the types of data reported to it, you may review the National Practitioner Data Bank Guidebook. The guidebook is included with this dataset as a .pdf document, or it can be downloaded at http://npdb-hipdb.com/npdbguidebook.html. Other information concerning the NPDB is available at http://www.npdb-hipdb.com. The National Practitioner Data Bank Public Use Data File contains one record for each disclosable report in the National Practitioner Data Bank as of the date specified at the beginning of this read.me file. The record format is the same regardless of whether the record is a malpractice payment, an adverse action, or an exclusion report. Malpractice payment records contain blanks for adverse action and exclusion-related variables; adverse action records contain blanks for malpractice payment and exclusion-related variables. Exclusion records contain blanks for malpractice payment and adverse action-related variables. Every year IRE/NICAR downloads a new version of the entire dataset. Record counts may not match from year to year because reports may be voided or corrected. ------------------------------------ NOTE TO PREVIOUS USERS OF THIS FILE ------------------------------------ BEGINNING WITH FILES DATED DECEMBER 31, 2005 AND LATER, four new field of license codes (codes 148, 165, 175, and 470) have been added for the LICNFELD variable and the description of LICNFELD in this documentation was updated to reflect the additions. Users should be cautious in interpreting results involving the new codes. Although these codes were first available to reporters on October 17, 2005, for this data file any previously filed reports which included a written in �other, specify� response that fit one of the newly available codes was coded to the new code in this file. However, reports which may actually involve a practitioner in a field with newly available code but which were actually reported using an old code could not be recoded to the new code. For example, a Certified Nurse Aide (new code 148) that was previously reported as a Nurse Aide (code 150) was not recoded to 148 since we have no way to tell from the record that the practitioner was actually a certified nurse aide. But if the individual was previously reported with "Certified Nurse Aide" written in the "other, specify" field, then the report was recoded to 148. :::For users of this file before March 31, 2004::: The March 31, 2004 and later versions of the Public Use File are substantially different from previous versions in content and format. New variables have been added concerning malpractice payments. Some old variables have been dropped and old values converted to values for the new variables. Some variables have been renamed. These changes reflect changes to the NPDB�s malpractice payment reporting requirements effective January 31, 2004. Specifically, AGEGROUP has been re-named PRACTAGE; ALGNNATR, ALEGATN1 and ALEGATN2 have replaced MALCODE1 and MALCODE2; OUTCOME has been added; TOTALPMT has been added to represent all past and expected future payments for the reported practitioner for this particular incident. PAYMENT is retained. It represents, as it has in the past, the amount of the reported payment, which in mostcases is the total payment. In addition, payment amount groupings have been changed for larger payments for the PAYMENT variable. These new groupings also apply to the new TOTALPMT variable. PYRRLTNS, which explains the relationship of the reporting entity to the reported practitioner, PTAGE, the age of the allegedly injured patient in 10 year increments, PTGENDER, and PTTYPE, inpatient, outpatient, or both, have been added. Only records reported on or after January 31, 2004 in the new reporting format include values for the new variables except ALGNNATR, ALEGATN1 and ALEGATN2. For older records these three variables have values translated from the previous MALCODE1 and MALCODE2 variables. The EXCLTYPE, EXCLSTAT, and EXCLYEAR variables have been deleted because all legacy format Exclusion records havebeen converted to new format Adverse Action Records. The information previously found in these three variables is now found in the analogous variable in exclusion Adverse Action records. The RECTYPE variable was also changed to accommodate the new malpractice payment report type. BEGINNING WITH FILES DATED DECEMBER 31, 2003 AND LATER, this documentation file changed. The code list for the REPTYPE variable was updated. BEGINNING WITH FILES DATED JUNE 30, 2003 AND LATER, this documentation file changed. The code lists for the following variables were updated: LICNFELD, MALCODE1, AACLASS1, BASISCD1, and TYPE. New date of first use and date of last use columns also have been added to the description of the values for some variables. The date of first use and date of last use columns indicate the dates that the values were offered as selection criteria for report submission. When no date of first use is provided, the value has always been available. When no date of last use is provided, the value is currently available. The format for the data file was not changed. BEGINNING WITH FILES DATED SEPTEMBER 30, 2002 AND LATER, THE FORMAT OF THIS FILE HAS CHANGED. THE ADVERSE ACTION CLASSIFICATION VARIABLE NAME (AACLASS) HAS BEEN CAHNGED TO "AACLASS1" AND FOUR ADDITIONAL ADVERSE ACTION CLASSIFICATION VARIABLES (AACLASS2, AACLASS3, AACLASS4, AND AACLASS5) HAVE BEEN INCLUDED IN THE FILE. THEREFORE EACH RECORD MAY HAVE UP TO FIVE ADVERSE ACTION CLASSIFICATION CODES. BEGINNING WITH FILES DATED APRIL 30, 2002 AND LATER, THE FORMAT OF THIS FILE HAS CHANGED. THE BASIS FOR ACTION VARIABLE NAME (BASISCD) HAS BEEN CHANGED TO "BASISCD1" AND FOUR ADDITIONAL BASIS FOR ACTION VARIABLES (BASISCD2, BASISCD3, BASISCD4, AND BASISCD5) HAVE BEEN INCLUDED IN THE FILE. THEREFORE EACH RECORD MAY HAVE UP TO FIVE BASIS FOR ACTION CODES. Beginning with files dated Dec. 31, 1999 and later, the format of this file has changed. Records of exclusions from participation in Medicare and Medicaid (individuals only, not organizations) have been added to the file and variables concerning exclusions have been added. BEGINNING WITH FILES DATED APRIL 30, 2002 AND LATER, THE FORMAT OF THIS FILE HAS CHANGED. THE BASIS FOR ACTION VARIABLE NAME (BASISCD) HAS BEEN CAHNGED TO "BASISCD1" AND FOUR ADDITIONAL BASIS FOR ACTION VARIABLES (BASISCD2, BASISCD3, BASISCD4, AND BASISCD5) HAVE BEEN INCLUDED IN THE FILE. THEREFORE EACH RECORD MAY HAVE UP TO FIVE BASIS FOR ACTION CODES. Beginning with the file of Dec. 31, 1999, the PAYMENT AMOUNT variable also has been changed. Ranges of payment amounts have been established and all payments within a range are coded to the midpoint of the range. In addition, beginning with the file of Dec. 31, 1999, the variables included for adverse action records have been changed. Files dated Dec. 31, 1999 and later also contain revised variables for counts of the number of reports of various types for each practitioner. See the Layout.txt for details concerning the new or changed variables. :::For users who used this dataset earlier than Sept. 30, 1999:::: Beginning with the Sept. 30, 1999, version of this file, the former "PROCYR" (year current version of the record was processed into the NPDB) variables has been changed to "ORIGYEAR" (year original version of the report was processed into the NPDB). This variable does not change if a report is corrected modified in a later year. The "TYPE" variable also replaces "ENTYTPDB" in the Sept. 30, 1999, and later files. This reflects a change in ENTITY TYPE CODES introduced during the summer of 1999. As Reporting ENTITIES RE-REGISTER with the NPDB, the type codes shown in this file will increasingly reflect use of the new codes; however since some reporting included in this file were reported by entities which no longer exist of no longer registered, some reports will permanently use the old entity type codes. Beginning with the file dated April 30, 1999, each record contains the variable "FUNDPYMT" which identifies "malpractice payments made by state funds." In most cases these payments are in addition to payments made for the same practitioner for the same incident by a primary insurer. --------------------------------------------- INVENTORY OF FILES INCLUDED WITH THIS DATASET --------------------------------------------- NPDB.DBF - Main data table LAYOUT.txt - Record layout for NPDB.dbf BROCHURE.pdf - Data Bank information brochure FACTSHEET.pdf - provides background and information for the National Practitioner Databank ANNIVERSARY.pdf - information on health care changes made from 1986 to 2000 README.pdf - Readme file provided by the Department of Health and Human Services, includes much of the information found within this file and the layout with some additional contact and background information. REGULATIONS.pdf - Definitions and laws provided by the Department of Health and Human Services. GUIDEBOOK.pdf - Compliance requirements for NPDB. LEGAL.TXT - important legal information. ----------------- WHAT YOU CAN DO ----------------- Several stories available through the IRE have been produced using the NPDB. In a 2005 series, "Special Treatment: Disciplining Doctors," a Washington Post investigative team found that hospitals were underreporting problematic doctors to NPDB. In the story notes submitted to the IRE Resource Center, the team reported �54 percent of all hospitals have never reported a disciplinary action to the data bank and some hospitals limit punishment to less than 30 days because longer sanctions must be reported.� (Story #22640) Like several other stories listed here, the Post series provides a good example of how the NPDB can be used in conjunction with other data sources to identify possible cases of unaddressed physician misconduct. In his 2003 series, "The Cost of Courage," Pittsburgh Post-Gazette reporter Steve Twedt identified several cases where physicians who identified colleague misconduct were themselves listed in the NPDB for improper conduct.(Story #21089) The Hartford Courant examined several doctors who continued to practice medicine despite being sued for malpractice. By setting standards for the doctors that the paper would focus on, such as having had a minimum number of suits filed and paying a minimum dollar amount in malpractice payments, The Courant was able to focus on several of the most dangerous doctors. The reporters working on the project sent certified letters to the doctors they reported on before their series was published. In the letters they explained to the doctors what they were doing and what they had found out about them and asked the doctors if what they had found was accurate.(Story #16564) NICAR NOTE: this story preceded major changes in the level of detail disclosed in the publicly available NPDB. The current version of the database has been further adapted to prevent its use in identifying individuals. The New York Daily News revealed 15 of New York's most-sued doctors. This investigation found that these 15 doctors alone had cumulated 412 malpractice suits and their insurance firms spent more than $26 million in compensation for the doctor's patients. (Story #16585) The Florida Sun-Sentinel focused on plastic surgery and found gaps in training, criminal activity, malpractice claims and uninsured practioners. (Story #16073) A medical dictionary is recommended to help with accuracy and a better understanding of the subject. The Division of Quality Assurance recommends that because of the size of this file, analysis be performed using statistical software such as SPSS, SAS, etc. The file may be too large to import into most spreadsheet programs, such as EXCEL, QUATTRO, or LOTUS 1-2-3. These programs are not designed for statistical analysis. Database programs, such as ACCESS or dBASE, are more likely to be able to handle a file of this size, but are not designed for statistical analysis. Although they may be used to create tables and count records with specific characteristics, the use of spreadsheet or database programs with this file is likely to be a slow and cumbersome process. Word processing programs cannot be used to analyze this file. If you have any questions about the processing of the data on this CD, please call NICAR at (573) 884-7711. Questions specifically about the data or methodology in gathering it should be directed to: Practioner Data Banks Branch Office of Workforce Analysis and Quality Assurance Bureau of Health Professions Health Resources and Services Administration U.S. Department of Health and Human Services 5600 Fishers Lane, Room 8-103 Rockville, MD 20857 Phone: (301)443-2300 Fax: (301)443-0238 Contact: Jiaying Hua Contact phone: (301)443-1898 Contact email: jhua@hrsa.gov -------- STORIES -------- To order these stories, please contact the IRE Resource Center at (573) 882-3364 and give them the STORY NUMBER. STORY NUMBER 21742 Frank Mullen of the (Reno) Gazette Journal used databases and documents to track Nevada malpractice lawsuits in 2004. Mullen reported that Nevada is experiencing a "medical malpractice crisis" in which doctors are leaving the state in droves because their malpractice premiums have skyrocketed. STORY NUMBER 16564 This series, done by the Hartford Courant in April 2000, investigates several doctors who, despite multiple malpractice lawsuits and disciplinary action taken against them, continue to practice medicine. Jack Dolan and Mike McIntire examine the arguments for and against making the National Practitioner Data Bank, a list of disciplinary actions against doctors, open to the public. The reporters write about several patients who have been permanently damaged or killed in the care of doctor whose names are in that database. STORY NUMBER 16585 This 2000 investigative report reveals 15 of the most sued doctors in New York. Russ Buettner and William Sherman of The New York Daily News analyzed the National Practitioner's Data Bank and 10 years worth of records and suits, and found that these 15 doctors cumulated 412 malpractice suites and their insurance firms spent more than $26 million in compensation for the doctor's patients. ---------- TIPSHEETS ---------- The IRE Resource Center has about 30 tipsheets about doctors or the health care system, most of which are available on the IRE Web site. To access the tipsheets available, type the word guest -- note all lowercase -- for both username and password. If the information is not available on the Web, please call the Resource Center to place an order. There are a few tip sheets available specifically concerning the National Practitioners Data Bank. Among them are Jack Dolan's tipsheets with information about how the Hartford Courant matched this data to other public data sources to determine the identities of many of the doctors. The numbers for two of these handouts are 1440 and 1378, which are also available to download on the IRE Resource Center Web site.