Opinion
No. 2003 CA 0079.
April 2, 2004.
ON APPEAL FROM THE 19th JUDICIAL DISTRICT COURT PARISH OF EAST BATON ROUGE, LOUISIANA DOCKET NO. 470,997, DIVISION "J" HONORABLE CURTIS A. CALLOWAY, JUDGE PRESIDING.
LYNN ERIC WILLIAMS, Jr. KEVIN R. KRAUSE WILLIAMS KRAUSE, L.L.C. Metairie, LA and LYNN E. WILLIAMS WILLIAMS LAW FIRM Baton Rouge, LA, Attorneys for Plaintiffs-Appellants Williams Law Firm and Carey Brown.
LLOYD J. LUNCEFORD TAYLOR, PORTER, BROOKS PHILLIPS, L.L.P. Baton Rouge, LA, Attorney for Defendant-Appellee Board of Supervisors of Louisiana State University and A M College.
Before: CARTER, C.J., PARRO, and GUIDRY, JJ.
This is a mandamus action seeking certain records from the Louisiana Tumor Registry (the LTR), a statewide cancer registry program administered by the Louisiana State University Health Sciences Center under the supervision of the Board of Supervisors of Louisiana State University and Agricultural and Mechanical College (LSU). The Williams Law Firm and Carey D. Brown, Sr., appeal a judgment denying their motion to compel production of certain records and for clarification of an earlier judgment of the trial court. We affirm in part, reverse in part, and render.
FACTUAL AND PROCEDURAL BACKGROUND
On March 30, 2000, the Williams Law Firm, a Professional Law Corporation (the Firm), petitioned the Nineteenth Judicial District Court, Parish of East Baton Rouge, for a writ of mandamus directed to the custodian of records of the LTR. The petition alleged that the Firm had requested certain information from the LTR under the provisions of the Public Records Act, LSA-R.S. 44:1, et seq., and that the LTR had refused to produce the information, claiming it was confidential and not subject to disclosure. According to the petition, the Firm reiterated its request, explaining that no confidential information was sought, but only raw numerical data. This second letter clarified that the information sought was the following:
An amending petition was later filed, seeking to add the River Region Crime Commission (the Commission) as a plaintiff. However, the amending petition was not accompanied by a motion seeking leave of court to file it and was not served until after an answer had been filed. At a hearing on June 18, 2001, the trial court ruled that no answer was required to the amended petition and allowed an oral amendment to add Carey D. Brown, Sr., the first vice president of the Commission, as a plaintiff. Brown is also of counsel to the Firm. This appeal was taken by the Firm and Brown. In this opinion, we will refer to the plaintiffs, collectively, as the Firm. We note also that Brown's first name was shown as "Kerry" in one portion of the record, but as "Carey" in the petition, judgment, and briefs to this court.
The letters requesting information and the LTR's letters of response were attached to the petition.
1. The raw numerical data of totals for each and every cancer listed in the current Louisiana Tumor Registry publication by zip code and by Parish for each and every Parish in the State of Louisiana by year for each year of the years of 1985 through 1999;
2. The raw data of the Pediatric Cancers for each and every zip code and by Parish for each and every Parish in the State of Louisiana by year for each year of the years of 1985 through 1999. The pediatric cancers should be provided as listed in the 1989 LOUISIANA TUMOR REGISTRY (LTR) according to the IARC pediatric cancer classification system (page 18 of the 1989 LTR publication). These cancers include, but are not limited to: leukemias[,] lymphomas, neuroblastomas, rhabdomyosarcomas, retinoblastomas, etc.; and,
3. The respective population totals for each parish for each year of the years of 1985 through 1999. These populations were used in calculations published in the respective LTR publications since 1985 to present.
The LTR responded to this second request by again refusing to produce any of the information sought in the first two paragraphs, indicating that the calculations sought in those paragraphs had not been made, that there were no extant documents meeting those descriptions, and that it was not legally required to generate new computer programs or create lists or documents to satisfy a request for information under the Public Records Act. The LTR further advised that if zip-code-specific and parish-specific responses were created for each type of cancer, most of the totals would be zero or one, which would meet the definition of case-specific data exempt from disclosure for confidentiality reasons. The LTR did offer to compile an answer to the third request, based on information available from the U.S. Census Bureau. The LTR's refusal to provide any of the information that was available only from its records prompted the filing of this mandamus action, to which the LTR filed an answer on May 17, 2000.
On May 22, 2000, a hearing was held, during which the testimony revealed there may be some information that the LTR could reasonably provide. After the hearing, the court withheld ruling on the mandamus request to allow the parties to meet and determine whether some information could be furnished to satisfy the Firm "and moot this controversy." The Firm scaled down its request, and the LTR provided some information during the following year. However, this production was also unsatisfactory to the Firm, and a second hearing was held on June 18, 2001. After hearing the testimony and reviewing the documentary evidence, the court ruled on June 22, 2001, that the LTR was to make available to the Firm for inspection or copying those records in its custody and in the form in which they exist, but was not required to write new computer programs to create the records. The LTR moved for a new trial for reargument only, complaining that the judgment announced a principle of law, but did not apply that principle to the Firm's request and did not specify which records, if any, the LTR was required to produce. As a result, the parties were each interpreting the judgment "in mutually exclusive ways." The court granted the motion and heard additional arguments, after which the judgment was amended to add that all case-specific data in LTR's custody was exempt from disclosure. The amended final judgment was signed December 14, 2001.
Claiming production was required by the amended final judgment, the Firm sent letters to two of the LTR's regional registries, seeking annualized information concerning 21 cancers in each of seven parishes. These requests were again denied by the LTR on the grounds that the responses would reveal case-specific information and would require the writing of new computer programs to reformat data to produce the information. The Firm then filed a motion to compel production in accordance with the amended final judgment and to clarify that judgment. After a hearing on the Firm's motion, the court denied the motion to compel production and denied the motion to clarify the December 14, 2001 amended final judgment. The Firm appealed.
MOTION TO STRIKE
After the appeal was lodged, the Firm filed a motion with the trial court to supplement the record, seeking to include an exhibit called "2000 Louisiana Health Report Card," which was discussed at various hearings and was purportedly considered by the trial court in reaching its conclusion. The trial court ordered the record supplemented. The LTR moved this court to strike the exhibit on the grounds that it was never filed into the record at the trial court.
Having reviewed the record, including the transcripts of the hearings held in connection with this matter, we conclude that the exhibit was not filed in any of those proceedings, and therefore, did not become part of the trial court record. It was inappropriate to order the trial record supplemented with a document that was never offered, introduced, or admitted into evidence. See Greenfield v. Lykes Bros. S.S. Co., 02-1377 (La.App. 1st Cir. 5/9/03), 848 So.2d 30, 33. This court cannot consider evidence that was not part of the record made in the trial court in this suit. White v. West Carroll Hosp., Inc., 613 So.2d 150, 154 (La. 1992). Accordingly, the motion to strike must be granted. However, to the extent this and similar annual reports were discussed by the witnesses in the trial court proceedings, the information provided in their testimony is properly part of the record before this court.
APPLICABLE LAW
Article XII, Section 3 of the Louisiana Constitution provides that no person shall be denied the right to examine public documents, except in cases established by law. This constitutional provision has been codified in the Public Records Act, LSA-R.S. 44:1, et seq. (the Act), which includes in its definition of "public records" all documentary materials, including information contained in electronic data processing equipment, "having been used, being in use, or prepared, possessed, or retained for use" in the performance of any function under the authority of the constitution or laws of this state. LSA-R.S. 44:1(A)(2)(a). Louisiana Revised Statute 44:31 states:
A. Providing access to public records is a responsibility and duty of the appointive or elective office of a custodian and his employees.
B. (1) Except as otherwise provided in this Chapter or as otherwise specifically provided by law, and in accordance with the provisions of this Chapter any person of the age of majority may inspect, copy or reproduce, or obtain a reproduction of any public record.
(2) The burden of proving that a public record is not subject to inspection, copying, or reproduction shall rest with the custodian.
The right of access to public records is fundamental; therefore, access may be denied only when the law specifically and unequivocally denies access. Hilliard v. Litchfield, 01-1987 (La.App. 1st Cir. 6/21/02), 822 So.2d 743, 746. The legislature, by the public records statutes, sought to guarantee, in the most expansive and unrestricted way possible, the right of the public to inspect and reproduce those records which the laws deem to be public. There was no intent on the part of the legislature to qualify, in any way, the right of access. As with the constitutional provision, the statute should be construed liberally, and any doubt must be resolved in favor of the right of access. Landis v. Moreau, 00-1157 (La. 2/21/01), 779 So.2d 691, 694-95; Title Research Corp. v. Rausch, 450 So.2d 933, 937 (La. 1984).
Louisiana Revised Statute 44:32(B) recognizes that portions of the requested material may be nonpublic and allows the custodian the option to separate that portion from the requested material. Simply because material requested may contain nonpublic records is not a reason for restricting access. Elliott v. District Attorney of Baton Rouge, 94-1804 (La.App. 1st Cir. 9/14/95), 664 So.2d 122, 125-26, writ denied, 95-2509 (La. 12/15/95), 664 So.2d 440. However, the request for production of information cannot be so burdensome as to interfere with the operation of the custodian's constitutional and legal duties. Any restriction or limitation imposed by the custodian places the burden on the custodian to justify the restriction or limitation.Id. at 126; Local 100, Serv. Employees' Int'l Union v. Forrest, 95-1954 (La.App. 1st Cir. 5/10/96), 675 So.2d 1153, 1156, writ denied, 96-1499 (La. 9/20/96), 679 So.2d 441. The custodian need only produce or make available for copying, reproduction, or inspection the existing records containing the requested information, and is not required to create new documents in the format requested. Nungesser v. Brown, 95-3005 (La. 2/16/96), 667 So.2d 1036, 1037.
Louisiana Revised Statute 44:4 enumerates certain specific exceptions to the Act, among them, "any records of information defined as 'confidential data' as provided in R.S. 40:3.1." LSA-R.S. 44:4(19); see In re Philip Morris, Inc., 97-2708 (La.App. 4th Cir. 1/28/98), 706 So.2d 665, 669. Louisiana Revised Statute 40:3.1(A) defines as confidential data:
All records of interviews, questionnaires, reports, statements, notes, and memoranda procured by and prepared by employees or agents of the office of public health or by any other person, agency, or organization acting jointly with that office, including public or private colleges and universities, in connection with special morbidity and mortality studies and research investigations to determine any cause of condition of health, . . . [which] shall be used solely for statistical, scientific, and medical research purposes relating to the cause or condition of health. . . .
Pursuant to LSA-R.S. 40:1299.81, the LTR exists under the auspices of the president of the LSU System as a statewide registry program for reporting cancer cases for the purpose of gathering statistical data to aid in the assessment of cancer incidence, survival rates, possible causes of specific cancers, and other related aspects of cancer in Louisiana. It is charged with collection and dissemination of cancer incidence data on a statewide level. The confidentiality of the information reported to the LTR is addressed in LSA-R.S. 40:1299.85, which states, in pertinent part:
We have used the current version of the relevant statutes in this opinion, because the amendments that have occurred since this suit was filed do not impact the issues in this case.
D. All information reported pursuant to this Part shall be confidential and privileged. The [LSU System] president shall take strict measures to ensure that all identifying information is kept confidential.
E. All information regarding case specific data, as distinguished from group, tabular, or aggregate data concerning patients . . . contained in records of interviews, written reports, and statements . . . shall be confidential and privileged and shall be used solely for the purposes of the study. Nothing in this Section shall prevent the [LSU System] president from publishing compilations relating to morbidity and mortality studies which do not identify case specific data or sources of information.
The LTR is authorized to exchange information with other cancer registries in order to obtain complete reports of Louisiana residents diagnosed or treated in other states, as long as those registries in other states agree in writing to keep non-aggregate, case-specific information confidential and privileged. LSA-R.S. 40:1299.87(B). Another sub-paragraph of LSA-R.S. 40:1299.87 provides:
F. No case specific data shall be available for subpoena nor shall it be disclosed, discoverable, or compelled to be produced in any civil, criminal, administrative, or other proceeding, . . . Nothing in this Section shall supersede the provisions of R.S. 40:3.1(A) through (H).
The Act specifically recognizes and continues in effect the exceptions, exemptions, and limitations on disclosure contained in LSA-R.S. 40:3.1, 40:1299.85, and 40:1299.87. LSA-R.S. 44:4.1(B)(24).
Under LSA-R.S. 44:35(A), any person who has been denied the right to inspect or copy a record under the provisions of the Act may institute proceedings for the issuance of a writ of mandamus, injunctive or declaratory relief, together with attorney fees, costs, and damages. If the person seeking the right to inspect or for a copy of a public record prevails in such suit, he shall be awarded reasonable attorney fees and other costs of litigation. If such a person prevails in part, the court may in its discretion award reasonable attorney fees or an appropriate portion thereof. LSA-R.S. 44:35(D). If the party requesting enforcement of the Act is required to appeal the action of the trial court, that party is entitled to an award of additional attorney fees and costs for prosecuting the appeal. Thornton v. Dep't of Public Safety, 536 So.2d 595, 597-98 (La.App. 1st Cir. 1988).
ANALYSIS
The Firm assigns as error the trial court's denial of its motion to compel production of records and to clarify the amended final judgment signed on December 14, 2001. In its second assignment of error, the Firm contends production of the requested information would not violate confidentiality statutes. The Firm also argues the trial court erred in failing to award attorney fees and costs for the writ of mandamus that it was required to bring after the LTR refused to comply with its public records request. The LTR contends this court has no jurisdiction to examine the December 14, 2001 amended final judgment, because it was not appealed; only the judgment of July 1, 2002, denying the motion to compel and to clarify the earlier judgment, is before this court. The LTR also re-asserts its position that to produce the information sought by the Firm, it would have to perform substantial additional work creating new records and would have to provide case-specific information.
At the outset, we reject the contention that this court cannot look beyond the July 1, 2002 judgment to address the substance of the trial court's amended final judgment of December 14, 2001. In the judgment on appeal to this court, the trial court denied production under its amended final judgment and denied clarification of that judgment. Inherent in the judgment before us are the legal conclusions of the trial court that its earlier judgment needed no clarification and required no further production from the LTR. Because those conclusions are before this court by way of the appeal of the July 1, 2002 judgment, it is appropriate and necessary for this court to examine those conclusions to determine whether they are legally correct. Motion to Clarify Judgment
The judgment of July 1, 2002, denied the Firm's motion to clarify the trial court's earlier amended final judgment, thus implying that no clarification was needed. In construing a judgment, the entire context must be considered, and in the event of doubt or ambiguity it is proper to consider the pleadings, subject matter of the suit, reasons for judgment, and other matters of record in order to arrive at an interpretation consistent with a proper decree on the facts and law presented.State, Dep't of Transp. Dev. v. Sugarland Ventures, Inc., 476 So.2d 970, 974 (La.App. 1st Cir.), writ denied, 478 So.2d 909 (La. 1985). It is apparent from this record that the implicit conclusion of the trial court — that its amended final judgment of December 14, 2001, needed no clarification — is insupportable. As noted by the LTR in its own earlier attempt to have the trial court clarify its first judgment, the amended final judgment merely states principles of law and does not apply them to the facts of the case to determine precisely which documents or records, if any, were to be produced in response to the Firm's request under the Act. The judgments of the trial court in this case are the equivalent of entering a judgment, after the trial of a personal injury suit, stating only that "[e]very act whatever of man that causes damage to another obliges him by whose fault it happened to repair it," without specifying in the judgment what damage was caused, who caused it, and how much to award for repair. The trial court's denial of the motion to clarify was reversible error. Generally, we would remand for the trial court to correct this unfortunate situation. However, the trial court has already had two opportunities to clarify its judgment and has failed to do so. The record in this case contains sufficient evidence for this court to clarify the trial court judgment. Therefore, this court will examine the evidence and render a judgment specifying which documents and records, if any, the LTR must produce to the Firm in response to its public records request under the Act.
See LSA-C.C. art. 2315.
Summary of Evidence
At the first hearing on May 22, 2000, the trial court heard testimony from Vivien Wai-Mei Chen, Ph.D., the director of the LTR. She explained that before any information could be provided in response to the Firm's request for "raw numerical data," the data would have to be cross-checked and verified to eliminate errors and internal inconsistencies, a process that would take six months to a year. Then it would take at least three weeks for one person working full-time to perform the calculations to extrapolate the data in the form requested, and she would have to personally review everything, which would take an additional week. She estimated that to comply with the request for just the adult cancers by zip code, the LTR would have to tabulate 2.7 million items of information in the form of raw data. Chen further explained that the LTR does not report its data by parish or zip code for each kind of cancer, because most of the results would be so few in any one category that fluctuations from one year to the next, while high when shown as percentage changes, would be statistically insignificant. To avoid the misleading nature of such fluctuations, the LTR grouped its data by reporting annually only the top five types of cancer for each parish or by combining data from several parishes in regional reports covering five-year time periods. Chen also said the LTR was planning a publication of parish-specific data for a ten-year aggregate time period, but the preparatory work to verify the data had not begun for that project. Chen defined any data pertaining to one case as "case-specific" data, which the LTR could not disclose under Louisiana law, and said data in which the response was zero or one for a particular type of cancer in a zip code or parish met that definition. She said that although parish and zip codes were on the collection forms and were entered in the computer at the data collection centers, that data was not validated and could contain inaccurate or misleading information. This hearing adjourned without a ruling from the court.
During the following year, the LTR provided the Firm with some of its existing publications and monographs containing summarized data on cancer in Louisiana. The LTR again explained that its compilations were only in five-year increments for multi-parish regions based on historical health districts and for the seven-parish "industrial corridor" along the Mississippi River. Therefore, satisfaction of the Firm's request for information by parish and zip code would still require new tabulations and reports, the preparation of which would disrupt the work of the LTR. The LTR also re-emphasized that many of the results would show only one incidence of the disease in a particular parish or zip code, which would be case-specific data exempt from disclosure. Other than its generally available publications, the LTR did not give the Firm any information concerning the annual incidence of adult cancers.
The correspondence between the parties and all of the information produced by the LTR following the first hearing were later introduced into the record.
The Firm significantly scaled down its request, reducing it from data concerning over 90 types of cancer in children and adults to less than 20 types of childhood cancer, broken down annually by parish. In response to this request, the LTR compiled a table showing statewide total counts of each of sixteen categories of childhood cancers for the years 1988 through 1996, broken down by year of diagnosis. The LTR eventually produced a series of tables showing annual parish-specific information for sixteen categories of childhood cancers by year of diagnosis, but omitting the results for cells that would have shown a zero or a one. The LTR contends this was a voluntary production not required by the Act. Also, although the LTR had to perform considerable verification and other work to produce this information, its brief to this court indicates the work was already in progress for the LTR's 2001 comprehensive monograph,Childhood Cancer in Louisiana (1988-1996), a copy of which was provided to the Firm.
A "cell" on the table is any blank spot in which a number would appear. For example, on the table for Acadia Parish, one cell shows that the number of leukemias diagnosed in 1995 was three. In Orleans parish, eighteen cases were diagnosed in 1992. The table identified eight of these as brain cancers and four as leukemias; the remaining six were not identified by type, apparently because the cells in which the identifying numbers would have appeared would have shown a one. As a result of the data omissions on the tables and the relative rarity of childhood cancers (only about 125 cases diagnosed each year in Louisiana), most of the cells for most of the 64 parishes contained only the symbol showing that a zero or a one had been suppressed. For a number of parishes, all of the cells contained this symbol, even when the totals for those parishes showed that some cancers had been diagnosed during a particular year. Of course, the suppression of the ones and zeros completely obscures the incidences of rarer types of cancer.
Because the Firm was not satisfied with these responses, it reiterated its original request for information concerning all the cancers reported to the LTR, by parish and zip code for the specified years. At the second hearing on June 18, 2001, Catherine Correa, Ph.D., the LTR's assistant director and data manager, testified concerning the collection and receipt of data, its entry into the format prescribed by the North American Association of Central Cancer Registries (NAACCR), its verification, and its management to produce reports meeting NAACCR standards. She said that although the LTR has gathered information for all 64 parishes since 1988, that information is reported only for multi-parish regions corresponding to historic health districts. Correa said information was collected in electronic and paper form at regional registries, which then transmitted the information to LTR's central office. The central office also received information from out-of-state treatment facilities. The data compilation at the regional registries was done using a software system called "C/Net," on which data was entered in various fields. Correa described C/Net as a regional registry data collection tool, not an analytic tool, and said C/Net software was not used at the central office of the LTR. Preliminary validation checks were performed at the regional registries before the data was transmitted to the central office; significant additional verification and coding was done at the central office. Zip code information was never verified, because this was not a variable used by the national reporting systems. Therefore, the LTR also did not use zip codes as a reportable variable.
Correa stated that to prepare the childhood cancer report for the 64 parishes, the LTR had received all the data from the regional registries and entered it into "SEER*Stat," which is an analytic program, not a database. The data also had to be coded to identify the childhood cancers. The central office then used SEER*Stat to analyze the data and generate the tables. Because C/Net is not used at the central office, Correa said she could not extract information simply by querying for a specific data field, such as a street, but said this could probably be done by someone at each regional registry. She explained that the database at the central office is not static; after the data is processed and verified, it is extracted to an analytic file. The records for each cancer patient stay in the database and are continuously modified as new information is received on that patient. A single individual might have multiple records coming from various hospitals where that patient has been examined and/or treated, might have multiple reportable incidences as a single cancer recurs after a remission, or might have multiple diagnoses reflected in the data collected by the regional registries. Correa explained that only at the central office can all the records concerning an individual patient be consolidated, verified, and analyzed.
Correa testified concerning a diagram she had drawn representing the constituent parts of the central registry database, which was shown as a large rectangle. Within that rectangle were 23 "SAS" data files, 21 of which were shown as diamond shapes, representing data sets compiled and collected by the seven regional registries, and 2 of which were shown as triangles, representing data sets from other states. The top row of seven diamond shapes represented the data as it was first received from the regions; known as the "trash can," this data did not necessarily fit the criteria needed for further processing. As the data flowed through the processing and verification steps, it moved to the second level of diamonds and ultimately, to the third level. The two triangles within the large rectangle showed out-of-state data concerning Louisiana residents; this data was processed separately and remained segregated to satisfy agreements with the registries in other states. Outside of the large rectangle were seven cylinders, representing the C/Net data collected by the regional registries. Correa confirmed that this is the data that is moved into the central office database when it is received. Correa's diagram also showed a pentagon and a hexagon outside the large rectangle; one of these represented the data that had been re-coded and extrapolated into a separate childhood analytic file for the childhood cancer monograph; the other represented the latest volume of "Cancer in Louisiana." Once completed for a monograph, this data became static and was no longer changed or manipulated in any way.
Correa said she had to apply SEER*Stat and write a program using menu-driven commands to extract the statewide childhood cancer data from the SAS files in the central registry database. To produce the parish-specific data on the sixteen childhood cancers that was provided to the Firm, she had to write a similar program to cull this information from the separate childhood analytic file. Correa admitted that she could write programs and menu-driven commands that would produce the information concerning all 90 cancers affecting the adult population by parish. However, she could not produce this information by zip code, because zip codes were not a separate variable in the data.
Chen testified again at the second hearing. She explained that NAACCR is the only professional organization for central cancer registries in all of North America. One of its functions is to annually evaluate each state's cancer registry and certify it based on standards of quality and completeness. The LTR enjoys the highest certification level awarded by NAACCR. Chen has held various elected positions in NAACCR, including serving as president from 2001 to 2003. The National Cancer Institute conducts a cancer surveillance program called SEER, which is an acronym for Surveillance Epidemiology and End Results. SEER participation is not open to all the cancer registries in the fifty states and Canadian provinces, but is limited to those of the highest quality. Among the seventy cancer registries, only the registries in nine states have qualified for the SEER program; the LTR is one of those. Chen said the statewide tabulation and parish-specific compilation tables for sixteen childhood cancers that were prepared by the LTR for the Firm were not reports that the LTR would generally have available, but were prepared by writing some original computer programs to respond to the information request. To satisfy the confidentiality requirements of Louisiana law, those tables suppressed all data for cells that would have contained a zero or a one. Chen also confirmed that annual parish-specific data was compiled and reported to the Office of Public Health for the top five cancers appearing during a given year in a given parish.
The Firm produced the testimony of Vince Nguyen, an electrical engineer with considerable expertise in computer data retrieval. He testified that he had reviewed documentation from the LTR, including information concerning all the software programs used in compiling and analyzing its data. Having examined the data collection software, C/Net, he said the data compiled in that database could be exported to one of many programs, including the statistical analysis program, "SAS," used by the central office of the LTR. He said that to retrieve the information sought by the Firm, he would export the C/Net data to a program such as Microsoft's standard "Excel" or "Access." Then a simple query could be written using three variables, asking the computer to find all persons in a particular parish having a particular type of cancer in a given year. Depending on the number of records the computer had to search, the results would be available within seconds or minutes. Nguyen acknowledged that the statistical analysis program used at the central office could not be queried for information, explaining that, "the SAS is not a database system in itself, so you can't really query from the SAS system." However, he said that once the data was in SAS, it could be exported to Excel or Access using just "four clicks of a mouse." Once in those formats, simple queries would produce the information sought by the Firm. This process would not involve writing new programs.
The hearing transcript shows this expert's surname first as Nguyen and later as Nyugen; we have used the first spelling throughout this opinion.
On cross-examination, Nguyen further explained that if there were a list of 100 persons, showing their names, addresses, and annual incomes, that information would comprise a data set and would be stored in a file. That file as a whole could be pulled into an SAS system and could also be exported as a whole to a more common database program. Once in a database program, a query could sort for all people making under $100,000 per year living in a particular area. Such a query would not work while the data was in SAS. However, Nguyen said that, as he understood the process used by the LTR, the regional registries used C/Net to collect data and then exported the data to Correa at the central office, where she put it into the SAS system. Based on that, he concluded that she would have in her possession both the C/Net raw data and the data that had been further processed by SAS. Any of this data could be exported to one of the database programs he had discussed, and could then be queried to provide the information sought by the Firm. The exception would be for the zip code information, which he understood was not entered as a separate field and, therefore, was not available from the C/Net data.
Patricia McGovern Williams, Ph.D., the president of the Commission, also testified on behalf of the Firm. She described the Commission as a private, non-profit corporation, the purpose of which is to foster honesty and integrity in government, political campaigns, and environmental protection and regulation. To accomplish its goals, the Commission facilitates identification, investigation, and gathering of evidence of political and environmental crimes, then reports this evidence to law enforcement and regulatory officials at the local, state, and federal levels. Her area of expertise is in environmental science and regulation. Williams said she is employed in the Department of Medicine at the LSU Health Science Center in Shreveport, where she is an associate professor of medicine and director of the occupational toxicology outreach program. She uses epidemiology as a tool in population studies. Her work has included medical surveillance studies of communities exposed to toxins to determine whether there were clinical warning signs of developing diseases in the population. She said when conducting such studies, she observes state and federal confidentiality requirements, and within those guidelines, has been able to report the results without deleting ones and zeros. Williams opined that reporting ones and zeros for certain diseases within a locale is not "case-specific," stating:
[F]or the way the data has been requested, you would have to have only one child in every parish under the age of 15 for that to be case specific . . . or you would have to have only one adult 15 years and older in every parish or any parish of the State of Louisiana and that doesn't exist. And that would be the only way that it would become case specific.Clarification of the Judgment
Based on this evidence, one conclusion can be reached easily, that being that it is not necessary for the LTR to provide by zip code the raw numerical data of totals for each and every cancer listed in its publications for each of the specified years. Both Chen and Correa testified the LTR does not use zip-code-specific data for any of its reports; the zip code is not entered as a separate variable in the data base or verified for accuracy. Therefore, even though the zip code forms a part of the address when the data is compiled, this information cannot be retrieved in the form requested by the Firm simply by querying for all cancers within a particular zip code. Nguyen's testimony confirmed this conclusion. Therefore, under the jurisprudence applying the Act, the LTR is not required to create new programs to access and report the information sought by the Firm by zip code. See Nungesser, 667 So.2d at 1037.
The request for adult cancer information for each parish is more problematic, because the evidence indicates that the parish is a separate variable used in the database when information is entered and is also used by the LTR in generating some of its reports. In particular, Chen stated the LTR has parish-specific data that it uses to prepare an annual report to the Office of Public Health showing the top five cancers appearing during a given year in a given parish. Correa also said since 1988, the LTR has received data from all the parishes, and admitted she could use menu-driven commands to produce parish-specific information concerning all 90 cancers affecting the adult population for the specified years. Correa and Nguyen testified that the regional registries use a database collection software from which this information could more easily be generated, but acknowledged that the verified data at the central office could be exported from the analytic files and queried for the parish-specific information.
Two problems emerge from the evidence. First, according to Correa and Nguyen, the software programs used at the central registry are not database programs that can respond to a simple query for parish-specific information. Therefore, before this information could be generated in the form requested, the data would have to be exported to such a program and then queried using menu-driven commands. Second, although the regional registries use a simpler database collection software that could be queried for parish-specific information, the raw data accumulated at this level is not necessarily accurate, because it may include reports from multiple sources for a single cancer occurring in a single patient. Correa indicated there could be as many as ten records per patient in the data received from the regional registries. Therefore, until the data is thoroughly vetted at the central registry level, the raw numbers collected by the regional registries would be likely to over-report the incidences of cancer in a given year for a given parish.
Despite our recognition of these problems in producing annualized parish-specific data, this court is not persuaded that the LTR has satisfied its burden of proof that production of such information is not required under the Act. The main reason we reach this conclusion is that, at some point each year, the LTR must be satisfied that it has accurate parish-specific information concerning incidences of cancer occurring during that year. If it did not have that information, it could not use it to compile the annual report for the Office of Public Health showing the top five cancers in each parish, which both Chen and Correa described. We agree with the Firm's argument that in order to determine which of the types of cancers constituted the highest incidences in each parish for a particular year, the LTR must know how often each of those types of cancers occurred within each parish for that year. At this point, just before this information is summarized for the Office of Public Health, the verified, parish-specific, raw numerical data must already have been extracted to make it available within the LTR for the annual report. Since the parish-specific information on adult cancers is available and, more importantly, is used by the LTR to produce its annual compilation for the Office of Public Health, it is at that point a public record and is subject to production from the central office under the Act. Accordingly, for those years for which such a report has been prepared for the Office of Public Health, the parish-specific data must be produced. Production of such verified data from the central office would obviate the duplications and inconsistencies that would be inherent in production from the regional registries.
In briefs and oral arguments, the LTR contends this report does not use annual data, but summarizes the five most common cancers in each parish over the preceding five years. However, the Firm contends, and we agree, that even if this is how the data is summarized for the report, the annual information for each parish must be available within the LTR in order for the latest year's data to be added to the summary.
The initial requests sought information from 1985 through 1999. It is not clear from the record that the LTR used parish-specific information for its compilations of annual reports throughout this period.
Because the data at the regional registries is inaccurate, it is not used by the LTR in that form to produce any of its reports. We note that for this reason, the data at the regional registries does not form a part of the public record.
Additionally, we are not convinced that it is necessary for the LTR to "write new programs" to produce the data for each parish in the format requested by the Firm. To construct a simplistic analogy, it seems that when cancer incidence data is received and entered into a database program using separate fields or variables, one of which is the parish where the incidence occurs, this data entry is comparable to inserting the information into separate file folders, each of which is labeled for a specific parish. The verification and re-coding done by the analytic programs is comparable to manually checking the information in each file folder to eliminate duplication and inconsistencies, and then checking each folder against all the other folders to further verify its accuracy. The process of moving the verified data from an analytic program to a database program where it can be queried is comparable to removing all the completed file folders from the filing cabinet and putting them on a table where they can be accessed. The process of querying the database program using the parish as a variable is comparable to sorting through the file folders for each parish and counting the number of incidences that appear in each folder. The mere fact that these activities occur electronically, rather than manually, does not constitute "writing new programs to create records." The same steps would have to be taken by the custodian of records to retrieve the information from a file cabinet.
We conclude that once the LTR receives and verifies the information and uses the parish as one of its variables in producing its reports, it cannot then refuse to produce information on a parish-specific basis on the grounds that the process of accessing that data constitutes the writing of new computer programs to create new records. Therefore, we further clarify the amended final judgment of the trial court to state that production by the LTR of parish-specific, verified data showing the raw numbers of cancers occurring annually does not constitute the writing of new computer programs to create the records.
Production of "Case-Specific" Information
The "ones" and the "zeros" quandary is the final hurdle to be cleared by this court in clarifying the trial court's amended final judgment. The LTR has not argued that confidentiality provisions apply to all of the raw numerical data sought by the Firm. However, it insists that when the data shows that either one or zero incidences of a type of cancer occurred within a parish within a certain year, that result is case specific and must be suppressed to comply with LSA-R.S. 40:3.1(A) and (F), 40:1299.85(E), and 40:1299.87(F). The LTR defines case specific as any data pertaining to one specific case. Therefore, if it provided full tabular results, some of the cells would show one incidence of a particular type of cancer within a parish for a particular year. Thus, one specific case would be referenced by that number. The Firm argues that the purpose of exempting production of case-specific data is protection of the privacy of the individuals whose information becomes part of the LTR data base. It contends that revealing the ones and zeros on a table would only be case specific if there were only one person in the parish, in which case that person's incidence of a particular type of cancer would be revealed. As this obviously is never the case, the Firm maintains that the inclusion of ones and zeros on a table does not reveal case-specific information.
Unfortunately, the statutes that shield case-specific data do not define the term. Nevertheless, there are clues in the statutory language and jurisprudence. In LSA-R.S. 40:1299.85(E), case-specific data is "distinguished from group, tabular, or aggregate data concerning patients. . . ." The information sought in this case is the simple expression in tabular form of either a zero or a one. As previously noted, the suppression of these numerals results in reports that do not reveal any of the occurrences of the rarer forms of cancer over time in a particular parish. While these small numbers might be statistically insignificant, as Chen asserted, the incidences of rare cancer that would be revealed by these numbers may implicate crucial public health concerns. Therefore, these omissions are substantively significant.
The language of the statutes and the jurisprudence suggest that the underlying purpose of protecting case-specific data is to safeguard the privacy of the individuals and to ensure their cooperation in providing information by maintaining the confidentiality of the data. The LTR contends the addition of these words to the statutes in 1990 was a statewide, legislative expansion of the decision in Marine Shale Processors, Inc. v. State of Louisiana, through Dep't of Health and Hospitals, 572 So.2d 280 (La.App. 1st Cir. 1990), in which this court shielded from disclosure the responses on questionnaires of patients ticipating in a study of childhood neuroblastomas. However, in that case, the fact that there were five cases of this rarely-occurring childhood cancer in a particular locale was already known, as were the identities of the five patients suffering from the disease. The information that was withheld was other data, such as the date of diagnosis, age at diagnosis, sex, race, and other such identifiable characteristics that would "tend to reveal to which of the five cases it applied. . . ." Id. at 284. Moreover, the 1990 amendments did not protect case-specific data; these words were not incorporated into the statutes governing the LTR until 1995. Our research disclosed no cases in which this term was interpreted or defined in the manner suggested by the LTR. Indeed, the issue appears to be res nova.
LSA-R.S. 40:3.1 was added by 1990 La. Acts, No. 59, § 1, effective June 26, 1990. However, it does not use the term "case specific" to describe data that is confidential.
The words "case specific" appear only in LSA-R.S. 40:1299.85 and 1299.87; they were added by 1995 La. Acts, No. 1197, § 1, effective June 29, 1995.
Because the simple expression of a zero or a one on a table, without any other identifiable characteristics, does not tend to reveal the identities of the patients or any other private information concerning them, and because the suppression of these numerals may obliterate some of the most important information from the data collected and reported by the LTR, we conclude that the LTR's definition of case specific is overly broad. The portion of the trial court's amended final judgment stating that "all case specific data in the custody of defendant are exempt from disclosure" is clarified to the extent that the expression of zeros and ones in tabular form does not constitute the disclosure of case-specific data and does not violate the confidentiality provisions of the relevant statutes or confidentiality agreements with other states.
Motion to Produce
Having thus clarified the trial court's amended final judgment, it is obvious that the trial court's denial of production of additional records pursuant to that judgment was also erroneous and must be reversed in part. The Firm's requests for parish-specific numerical data showing the annual incidences of adult cancers is subject to the Act and must be produced by the LTR from the central office where the verification process has been accomplished, but only for those years during which an annual report using that information was prepared by the LTR. In addition, the parish-specific numerical data concerning sixteen childhood cancers that was provided by the LTR should be supplemented by completion of the tables to include zeros and ones in the appropriate cells. To the extent the trial court's denial of production is applied to the production of zip-code-specific information or to information from the regional registries, we find no error in that portion of the judgment.
Attorney Fees and Costs
In its third assignment of error, the Firm contends the trial court erred in failing to award attorney fees and costs, pursuant to LSA-R.S. 44:35(A). However, although the cited statute does authorize such awards, they must first be requested. The statute states that a person who has been denied access to public records under the provisions of the Act "may institute proceedings for the issuance of a writ of mandamus, injunctive or declaratory relief, together with attorney's fees, costs and damages." In this case, although the Firm instituted proceedings for a writ of mandamus, it did not ask for attorney fees, costs, and damages in its petition. Nor did it submit any evidence to the trial court upon which an award of "reasonable attorney's fees" for the successful litigant could be based. Therefore, this court is unable to make such an award.
CONCLUSION
The judgment of the trial court denying the Firm's motion to compel production of documents from the LTR is affirmed as to the Firm's request for information by zip code and from the regional registries. In all other respects, the judgment denying the motion to compel production and the motion to clarify the amended final judgment of December 14, 2001, is reversed. Judgment is hereby rendered as follows:
(1) The zip-code-specific information sought by the Firm is not maintained by the LTR in the form requested and cannot be accessed without writing new programs and creating new records; therefore, the LTR is not required to provide the requested information by zip code;
(2) The parish-specific information sought by the Firm is maintained by the LTR at its central office in electronic files that can be sorted for this variable and is used by the LTR in preparing an annual report to the Office of Public Health showing the five most common cancers occurring within each parish; therefore, the LTR is required to provide the requested information by parish from its central office for those years for which it used this data in its annual report;
(3) The inclusion of ones and zeros on tables showing raw numerical data only does not constitute case-specific data and does not violate confidentiality statutes or agreements; therefore, the LTR is required to include these numbers in the information produced for the Firm in response to its current or future requests.
All costs of this appeal, in the amount of $538.59, are assessed to the LTR.