IN THE CIRCUIT COURT OF THE FIFTEENTH JUDICIAL CIRCUIT OF FLORIDA, IN AND FOR PALM BEACH COUNTY. CASE NO.

: RICHARD COTROMANO, BETHANY COTROMANO, FRANK DECARLO PAULETTE DECARLO, GREGORY DUNSFORD, JENNIFER DUNSFORD, JOYCE FEATHERSTON, BILL FEATHERSTON, ROBERT T. NEWFIELD and TRACY NEWFIELD, all on behalf of themselves and all others similarly situated Plaintiffs
vs.

UNITED TECHNOLOGIES CORPORATION,Pratt & Whitney Group, A Connecticut Corporation, and PALM BEACH AGGREGATES,LLC,a Florida Corporation. Defendants. /

CLASS ACTION COMPLAINT Plaintiffs, RICHARD COTROMANO, BETHANY COTROMANO, FRANK

DeCARLO, PAULETTE DeCARLO, GREGORY DUNSFORD, JENNIFER DUNSFORD, JOYCE FEATHERSTON, ROBERT T. NEWFIELD and TRACY NEWFIELD, all on behalf of themselves and all others similarly situated, sue Defendants United Technologies Corporation, Pratt & Whitney Group, and Palm Beach Aggregates, Inc., a Florida Corporation, for damages as prayed for specifically below. I. This is an action for damages that exceed Fifteen Thousand Dollars ($15,000.00).

JURISDICTION and VENUE 2. RICHARD COTROMANO and BETHANY COTROMANO reside in Palm

Beach County, Florida. Together as husband and wife, they owned a residence on 76th Road North in the Acreage, an unincorporated residential community in Palm Beach County. They sold it through a short sale after the Florida Department of Health officially declared their daughter to be a member of a pediatric brain tumor cluster in 2010. 3. FRANK DeCARLO and PAULETTE DeCARLO reside in Palm Beach County,

Florida. Together as husband and wife, they own a residence on 80th Lane North in the Acreage, an unincorporated residential community in Palm Beach County. The Florida Department of Health officially declared their daughter to be a member of a pediatric brain tumor cluster in 2010. 4. JOYCE and BILL FEATHERSTON reside in Palm Beach County, Florida.

Together as husband and wife, they own a residence on 80th Lane North in the Acreage, an unincorporated residential community in Palm Beach County. The Florida Department of Health officially declared Joseph Michael Baratta, Joyce's deceased son, to be a member of a pediatric brain tumor cluster in 2010. The property is currently subject to a foreclosure action. 5. GREGORY DUNSFORD and JENNIFER DUNSFORD reside in Tennessee.

Together as husband and wife, they owned a residence on 85th Street North in the Acreage, an unincorporated residential community in Palm Beach County. They sold it through a short sale after the Florida Department of Health officially declared their son to be a member of a pediatric brain tumor cluster in 2010. 6. ROBERT T. NEWFIELD and TRACY NEWFIELD reside in Palm Beach

County, Florida. Together as husband and wife, they own a residence at 7309 Banyan Boulevard

in the Acreage, an unincorporated residential community in Palm Beach County. The Florida Department of Health officially declared their daughter to be member of a pediatric brain tumor cluster in 2010. 7. PRATT & WHITNEY is a division of UNITED TECHNOLOGIES

CORPORATION, a Connecticut corporation that possesses, maintains, and runs an engineering facility in northwest Palm Beach County adjacent to the Corbett Wildlife area. (Hereinafter "Pratt Whitney") 8. PALM BEACH AGGREGATES,LLC is a Florida Corporation operating mining

and associated interests and owning land in Palm Beach County. (Hereinafter "Aggregates"). Aggregates owned, and then for some time leased, the land that is currently used as an aquifer impoundment by the South Florida Water Management District - the L8 reservoir. THE CLASS REPRESENTATIVES 9. As stated in paragraphs 2-6 above, each of the named Plaintiffs is or has been an

Acreage property owner is a parent of a victim of the cancer cluster. Complaints addressing the injuries to those children have been separately filed. The remedies sought herein are limited to compensation for the diminution in value of property caused as a consequence of the stigma arising from the cancer cluster designation and do not include claims for physical injuries suffered by any member ofthe putative class. 10. The named Plaintiffs seeking to represent the putative class have already sustained

a separate significant injury that is common to every proposed class member. The class sought to be represented is comprised of every current or former Acreage area landowner who owns or owned an Acreage property at any time since August 24,2009.

11.

The named Plaintiffs are well aware of and are willing and able to accept the

responsibility to represent the members ofthe class. 12. Each ofthese representatives has owned property in the Acreage during the time of

the public disclosure of the Acreage cancer cluster and has sustained a loss as a consequence of the existence ofthe cancer cluster. CLASS REPRESENTATION ALLEGATIONS 13. 14. This action meets the requirements of Florida Rule of Civil Procedure 1.220. The Pleading requirements set forth in Rule 1.220 (c)(A) are as follows: the claim

is maintainable as a class claim pursuant to Rule 1.220(b) (1)(B): the prosecution of separate claims for diminution in value by individual members ofthe class would as a practical matter, be dispositive of the interests of other members of the class who are not parties to the adjudications, or substantially impair or impede the ability of other members of the class who are not parties to the adjudications to protect their interests. Alternatively, even if separate actions for diminution in value were maintainable, the claim is maintainable as a class claim pursuant to Rule 1.220 (b)(3): the questions of law or fact common to the claim or defense of the representative party and the claim or defense of each member of the class predominate over any question of law or fact affecting only individual members of the class, and class representation is superior to other available methods for the fair and efficient adjudication ofthe controversy. 15. The common issue Pleading requirement set forth in Rule 1.220 (c)(B) are met as

follows: The Questions of Law or Fact Common to the Claim ofthe Parties and the Class: 1. Did environmental contamination caused by the Defendants increase the incidence of brain tumors in the Acreage? 2. Did the contamination and resulting cancer cluster cause a diminution in property values within the Acreage?

3. What is the geographic limitation ofthe contamination? 4. What is the magnitude and temporal limitation ofthe contamination? 5. What is the magnitude ofthe contamination stigma within the cluster area? 6. What damages in addition to the diminution in property values are recoverable such as punitive damages and prejudgment interest? 7. Did the Defendants act with reckless disregard? 8. Whether the recovery included in section Chapter 376, Florida Statutes allows for recovery of stigma value damages under the circumstances alleged herein. 9. Whether Plaintiffs may seek the recovery of fees pursuant to F.S. 376.313 and, if so, the amount of the fee claim. 16. The typicality Pleading requirement set forth in Rule 1.220 (c)(C) is met as

follows: the claim or defense of the representative party is typical of the claim or defense of each member of the class.: the claim of the named Plaintiffs for the recovery of damages caused by a uniform stigma which was in turn caused by the Defendants' pollutive acts and the resulting cancer cluster is typical of the claims shared by all persons who have owned property in the Acreage at any time since the public disclosure of the increased incidence of cancer in the Acreage. 17. follows: i. The population in the Acreage is around 30,000 people. There are over 10,000 residential properties in the Acreage. ii. The members of the class include all past and current property owners of residential lots in the Acreage within the time frame of 2009 to the present. The designated cancer cluster area consists of 850 Census blocks, which is depicted on Exhibit A at page 23. The numerousity Pleading requirements set forth in Rule 1.220 (c)(D) are met as

iii.

The representative party can fairly and adequately protect and represent the interests of each member of the class. There are, in fact, no better

representatives than the families who suffered the primary damages of the cancer cluster, the affliction of a brain tumor. The counsel representing these representatives have prosecuted class actions in the past and have been investigating the cluster on behalf of the victims since 2009 before the DOH declaration.

18.

The Pleading requirements set forth in Rule .220(c)(E) are met as follows: The

claim itself is based on the fact that the entire Acreage community carries a uniform stigma caused by one thingthe Defendants' pollutive acts and the resulting cancer cluster. The facts and circumstances ofthe cancer cluster and the resulting diminution in value are described specifically in the allegations below. FACTUAL ALLEGATIONS The Brain Tumor Cancer Cluster 19. In 2009, the Department of Health declared that the level of pediatric brain tumors

diagnosed in the Acreage was significantly escalated for the years 2006-2007. In 2010, the CDC confirmed the escalation and the 2008 diagnoses were added to the cluster. 20. The report, attached as Exhibit A, compared the number of cancers diagnosed in

the Acreage (the observed incidence) with the population-weighted number of cancers diagnosed in the County and the State (the expected cancer incidence.) The August 2008 report stated that cancer was generally elevated throughout the Acreage and that pediatric brain tumors were significantly elevated top a level of more than three times the expected rate of occurrence.

21.

The DOH further reported that the brain cancer incidence among the entire

Acreage population as a whole was sharply rising, whereas that same trend was not occurring in the county or the state. 22. In 2010, the Center for Disease Control (CDC) confirmed the pediatric cancer

cluster was confirmed. The 2008 diagnoses among children were tallied and that year alone confirmed the existence of a cluster for the next three year period, 2008-2010. 23. The cancer cluster designation was based only on pediatric brain tumors. The

state stopped looking at diagnoses of brain tumors made in adults in 2007, although it appeared in the report that the entire population of the Acreage was experiencing a steady, sharp increase of cases. 24. According to self-reported data from the community, the 2008-2010 incidence of

brain tumors for the entire Acreage population is above 30. From 1997-2007, the highest number of cases that should be expected in any three year period based on both state and county comparisons is 5.1. Based on the self -reported data the pediatric cancer cluster declared by the State for the years 2004-2008 would be expanded to a general brain cancer cluster from 20042010. 25. 26. The fact is that brain cancers are significantly escalated in the Acreage. When the cancer cluster was declared in 2009, property values in the Acreage

declined in direct response to the declaration. 27. Property values will continue to decline as investigation into the cause of the

cluster continues to confirm the presence of industrial contamination in the Acreage and the causal link between that contamination and the significantly escalated incidence of cancer among Acreage residents.

RADIATION IN THE ACREAGE 28. It is generally accepted by nearly all branches of medicine and particularly by

epidemiologists, radio-oncologists, health physicists, and neurologists that exposure to ionizing radiation can cause a brain tumor and will significantly increase an individual's risk for developing a brain tumor. 29. Two types of ionizing radiation, including gamma and/or beta emitting

radionuclides, have been detected throughout the Acreage: naturally occurring and non-naturally occurring. Note, "naturally occurring" does not mean naturally presentonly that fission is not required to cause the isotope to be in existence. 30. Those radionuclides that have been detected in the Acreage at above natural

background rates but may be termed naturally occurring include Uranium-234, Uranium-235 Uranium-236, and elevated levels of the decay products of Thorium, particularly Lead-214 and Bismuth-214 and Actinium-228. (Naturally Occurring Radioactive Material or NORM.) The presence of technologically enhanced concentrations of NORM products(TE-NORMS)in water systems can be associated with the metal processing industry, particularly airplane engine and turbine manufacturing; however, high concentrations of these radionuclides in an environment are more readily associated with mining operations, including the type of shell rock mining conducted by the Aggregates. 31. In addition to TE-NORM, radionuclides have been detected in the Acreage that These include Cesium-137, Iridium-191, and Cadmium-109.

are not naturally occurring.

Contamination of Iridium-191 and Cadmium-109 is particularly associated with tge metal

processing industry generally and with airplane engine and turbine manufacturing and testing of the type conducted by Pratt Whitney at its Palm Beach Facility. 32. Both these types of radionuclides (TE-NORM and the industrial, non-naturally

occurring material) have been detected in the groundwater and in the surface water structures moving water through the canals that recharge the surficial aquifer from which the Acreage residents source their drinking, bathing, and irrigation water. 33. The aquifer underneath the Acreage moves at a southeast direction away from the

western edge ofthe Pratt Whitney campus and into the Acreage. 34. The Aquifer underneath the Acreage is recharged by the M Canal, which has been

significantly and periodically influenced by the Aggregates rock pits since at least 2003 and perhaps earlier. 35. In fact, there is little to no distinction between the surface and ground water in the

L8 Basin which includes both the Acreage and the Defendants' business operations. 36. The described carcinogenic contaminants are present within the L8 Basin as a

result ofthe operations of Pratt Whitney and the Palm Beach Aggregates. 37. Pursuant to provisions of Chapter 376, Defendants are strictly liable for the

leaching to of these contaminants into the waters of the state in violation of applicable laws, regulations, or permits. PRATT WHITNEY LIABILITY 38. Pratt Whitney has used the previously referenced radionuclides in its operations at

the Pratt Whitney campus, a campus with numerous canals and sheet-swamps that are adjacent to wetlands in the Corbett Wildlife Management Area(Corbett).

39.

Pratt Whitney used either the specific radionuclides or the parent radionuclide of

every naturally occurring radionuclide listed in paragraph 22 in its operations at the Pratt Whitney campus. 40. Pratt Whitney has failed to use and maintain these materials safely and has

emitted these materials into the surface and ground waters in Corbett and into the Acreage as part of routine misuse, dumping and neglect. 41. The wetlands in and around Pratt & Whitney drain into the Acreage surface

waters at varying rates each year but particularly in the times of extremely heavy rains such as have been documented to occur at least once every decade from the time Pratt Whitney began working with these materials (circa 1958). 42. Pratt Whitney has used some or all of these radionuclides in its operations at test

stands located on the western edge of the Pratt Whitney campus, a portion of the campus where the groundwater runs southeast at a velocity and trajectory that would carry and has carried contamination into the Acreage community. 43. In the 1980's, Pratt Whitney reported a portion of the contamination that had

built- up over time on its property to the Department ofEnvironmental Regulation. 44. The site should have been, and was at various times slated to be, an EPA

Superfund site, which would have included "a record of decision" and clean-up plan that was authored by, executed by, and performed by EPA. A Record of Decision (ROD) is a public document that explains which cleanup alternatives will be used to clean up a Superfund site. The ROD for sites listed on the NPL (NPL Site Listing Process) is created from information generated during an EPA conducted Remedial Investigation/Feasibilty Study (RI/FS).

10

45.

Instead, Pratt Whitney exercised influence through the state agency, the

Department of Environmental Regulation, to avoid the sort of government investigation that should have accompanied the preparation of an agency-authored "Record of Decision." 46. In order to avoid exposure for the full extent of its clean-up liability and to avoid

the disclosure of its use of several hazardous and radioactive products, Pratt Whitney exercised influence on the DER through then DER employee Rick Reis to have the clean-up, and more importantly investigation, performed by Pratt Whitney with limited EPA oversight. 47. Shortly after securing its ability to maintain control over the site, Pratt Whitney

hired Rick Reis away from DER to ensure that the company would have minimal exposure. 48. The purpose of these acts was to fraudulently conceal from the government and

the public the extent of Pratt Whitney's failure to follow environmental protection law. In other words, UTC/Pratt Whitney eschewed disclosing the full extent of contamination and liability for proper environmental clean-up in order to protect the company's government contracting abilities. PALM BEACH AGGREGATES LIABILITY

49.

Palm Beach Aggregates and/or its predecessor in interest, GKK Aggregates, have

conducted an aggregate mining business to the west ofthe Acreage since at least 1993. 50. The aggregate mined by Palm Beach Aggregates is of particular value because it

consists of limestone rock, the density of which is sufficient to meet the criteria for road and highway construction. Until 2008, the mining included a dewatering or dredging process through which Palm Beach Aggregates at various times deposited waters including high-concentrations of Naturally Occurring Radioactive Materials(NORM)into adjacent surface waters.

11

51.

At times prior to 2003, the dewatering violated permits either because the water

emitted was in excess of the volume permitted or the water was deposited into non-permitted recipient water bodies, such as the L8 Canal. 52. In 2003, Palm Beach Aggregates entered into a Consent Final Judgment with

South Florida Water Management District (SFWMD) through which many of the pits Palm Beach Aggregates was in the process of mining or had already mined were condemned to SFWMD for use as a reservoir. 53. For some time prior to this condemnation, Palm Beach Aggregates had been

negotiating with other prospective buyers. The deal struck with SFWMD was for over two hundred million dollars. 54. The first condemnation was of parcels 00404317000001030,

00404320000001040, which included many of the pits Palm Beach Aggregates had already mined. In February and August of 2007, Palm Beach Aggregates conveyed parcels These additional

00404329000005010 and 00404320000001010 to the District as well. conveyances were performed in accordance with the 2003 Consent Judgment. 55.

According to the 2003 Consent Judgment, Palm Beach Aggregates retained a

reservation of Occupancy, Possession, and Use, which allowed them continue to excavate or dredge pits. Palm Beach Aggregates also reclaimed pits for reservoir use, or constructed the reservoir. The Consent agreement also required Palm Beach Aggregates to certify the cells for water storage use before turning them over to SFWMD for use as a reservoir. 56. Pursuant to the Consent Judgment, Palm Beach Aggregates retained control ofthe

pits and specifically provided indemnity for any pollution caused by the pits for some time until

12

the process of handing over the pits was complete, which was not until sometime after May 2009. 57. Palm Beach Aggregates continued its mining activities at parcels it still owned

around the Reservoir including 00404329000005020 and 00404329000003020. 58. From 2003 until September 2008, the Aggregates continued the

dewatering/dredging process through which technologically enhanced NORM (TE-NORM) was increased in the designated reservoir pits. 59. Before 2003, the recognition that Palm Beach Aggregates operations had caused

or may have caused and would continue to cause contamination within the waters of the pits and the adjacent canals within the Acreage was known by Palm Beach Aggregates and its consultants. 60. The discovery of Palm Beach Aggregates' responsibility for contamination of the

Acreage has been hindered by inconsistent and inaccurate reporting to state regulatory agencies by Palm Beach Aggregates. 61. TE-NORM has from time to time at various times escaped the confinement of the

pits and has entered adjacent canals flowing into the Acreage, contaminating both surface and subsurface waters within the Acreage and materially contributing to the increased incident of brain cancers in the Acreage. Radiation at the Case Homes 62. The levels of radionuclides found at case homes in 2013 greatly exceed the

Maximum Contaminant Level (MCL) for alpha emitters set for drinking water by the Environmental Protection Agency of 15 picocuries per liter (pci/L.) The EPA set that standard because ingestion of alpha radiation over time can cause cancer.

13

63.

The radionuclides found within the Acreage include gamma and beta emitters,

which do not have to be ingested to cause cancer. The MCL for beta was 15 pci/L prior to the EPA recalibrating the level to a maximum for beta activity at 4 millirems per year. 64. The MCL for radium, an alpha producing NORM product, is 5pcil/L. The level

for radium is set lower than for alpha emitters generally because it will degrade into radon, lead214, and bismuth-214 (beta and low level gamma emitters) within a relatively short period. 65. Testing in 2013 detected elevated levels of lead-214 in water sourced from the

Newfield and DeCarlo well and the homes of other adult cancer victims. It was detected at 1088 123 pci/L at the Newfield home. It was detected at the DeCarlo home at 26634.7pci/L. 66. This trend continued in the testing done in the adult victims wells for Lead-214

and for the other thorium products including Actinium. The trend decreases as the testing proceeds further east. Such a trend can be explained by the influence of Palm Beach Aggregates on the M canal and/or the influence of Pratt Whitney on Corbett. 67. In 2013, Iridium-192 was detected in the Newfield well at 189 22.2 pci/L. It was

detected at the DeCarlo home at 41.216.49 pci/L. In other words, as the groundwater moved southeast away from Corbett, the concentration of this non-naturally occurring radionuclide decreased at the time oftesting. 68. The referenced contaminants were detected in the soil where Newfield,

Featherston and DeCarlo systems back-flush. The half-life of Iridium is 72 days. 69. The referenced contaminants were found in the water and back-flush soils at the

homes of other adult brain tumor victims as well. 70. The referenced contaminants were detected in the surface water adjacent to the

former Cotromano residence 0511.016 pci/L and in the ITID impoundment at 6.11512.32. In

14

other words as the surface waters move away from Corbett, the surface water concentration of this non-naturally occurring radionuclide decreased at the time oftesting. 71. The only surface water that did not have radionuclides was the water coming east

into the Acreage from outside the L8 basin. 72. Considering that the half-life of Iridium is 72 days, the quantity of Iridium causing

the present contamination would of necessity be dramatically higher the longer the span of time between the origination of the contamination and the time of testing when the contamination is detected. 73. Such an exposure greatly increases the risk of developing a brain tumor and was a

material and substantially contributing cause ofthe brain tumor cluster. 74. The industrial radionuclide Cadmium-109 was also detected in the Newfield

backflush soil. 75. In addition to the 2013 detection of these industrial radionuclides, the DEP 2010

investigation detected Cesium-137 in one of the five soil samples it tested in its 2010 investigation of the Newfield home, in five of the five samples tested at the DeCarlo home, in one of the five samples tested at the Newfield home, in three of the five samples tested at the Cotromano home, in one of the five samples tested at the Dunnsford home, and in four of the five sample taken at the Featherston home. 76. The DOH study detected Cesium-137 in the soil of every case home it tested

except one. It tested twelve case homes. 77. Cesium-137 has a half-life of 30.17 years and decays into Barium-137, which has a

half-life of 153 seconds before becoming stable barium. Stable barium has been detected in the soil as well.

15

78.

Cadmium, barium, and lead are neurotoxins which can pass the blood-brain barrier

and cause brain tumors. 79. The accumulated exposure to naturally occurring and non-naturally occurring

radioactive products leached into the environment by these Defendants was a material and substantially contributing cause ofthe brain tumor cluster. 80. The cancer cluster caused by these Defendants was a material and substantially

contributing cause of the diminution in value damage sustained by these representatives and the community as a whole. COUNT ONE Diminution in Value via Negligence 81. 82. Plaintiffs re-allege those facts and matters set forth in paragraphs 1 through 80. Defendants owed a duty of care to the surrounding neighborhoods to exercise

caution in their use and exploitation of resources including radionuclides and to keep such materials within the confines oftheir property. 83. Defendants failed to exercise reasonable care by allowing the referenced

contaminants to escape the confines oftheir property. 84. At times, Defendants have realized a portion of the damages caused by their

neglect but instead of remediating the problem, they allowed their properties to continue such emissions by failing to remediate contamination safely. 85. Defendants failures caused the materials to be leached or emitted into the ground

water and surface waters adjacent to their properties and eventually into the Acreage community. 86. These breaches caused a significantly increased incidence of brain tumors to occur

in the Acreage, which in turn caused a diminution in the value of the Acreage properties.

16

87.

The Class Representatives and all others similarly situated as included in the

definition ofthe Class have sustained the damage of that diminution in value. WHEREFORE, Plaintiffs demand judgment for damages and an award for attorney fees and costs pursuant to F.S. 376.313 and further demands trial by jury. Plaintiffs reserve the right to assert a claim for punitive damages upon satisfaction of statutory prerequisites. COUNT TWO Diminution in Value via Strict Liability 88. 89. Plaintiffs re-allege those facts and matters set forth in paragraphs 1 through 80. As alleged above, both Palm Beach Aggregates and Pratt Whitney emitted

contaminants into the surface and groundwater in excess of the limits allowed by governing agencies, statutes and, when applicable, their permits. 90. Pursuant to F.S. 376.313, Pratt Whitney and Palm Beach Aggregates are strictly

and jointly liable for any and all damages emanating from the effluences as previously described. 91. The emissions caused the Representatives and the entire class damages including

an ongoing diminution in value. 92. Pursuant to that F.S. 376.313, Plaintiffs may be awarded costs of litigation

(including reasonable attorney's and expert fees) if the court determines such an award is in the public interest. WHEREFORE,Plaintiff demands judgment for damages and an award for attorney fees and costs pursuant to F.S. 376.313 and further demands trial by jury. Plaintiff reserves the right to assert a claim for punitive damages upon satisfaction of statutory prerequisites.

17

DATED this

l Vt "--fri

day of AUGUST,2013.

ack Scarola/Florida Bar No: 16 440 Mara R. P. Hatfield/Florida Bar, o.: 37053 Primary E-mail: mrh@[Link] Secondary E-maih-r dtm@[Link] Searcy Denney Scarola Barnhart & Shipley, P.A. 2139 Palm Beach Lakes Boulevard West Palm Beach, Florida 33409 Phone:(561)686-6300 Fax: (561)383-9539 Attorney for Plaintiffs

18

ACREAGE CANCER REVIEW PALM BEACH COUNTY

August 2009

Florida Department of Health Division of Environmental Health Bureau of Environmental Public Health Medicine

DIVISION OF

Environmental Health

Charlie Crist Governor

Ana M. Viamonte Ros, M.D., M.P.H. State Surgeon General

EXHIBIT "A"

TABLE OF CONTENTS Executive Summary Level 1: Preliminary Cancer Cluster Assessment Background Florida Cancer Data System (FCDS) Review Other Facts about FCDS and Cancer Rates Brain Cancer Epidemiology Determination of the Geographic Area The Acreage Community Population Data FCDS Data Search Acreage Analysis Preliminary Assessment of Environmental Agents and Contamination Concerns in the Acreage Results (Malignant Only) Type of Brain Cancers Results with Benign and Borderline Brain Cancers Additional Analysis Limitations Conclusions Recommendations References Appendix A. Attachment I 3 5 7 7 8 9 10 10 10 11 11 12

13 15 15 16 16 19 19 21 22 32

Executive Summary Florida Department of Health (DOH) concluded a Level ll cancer inquiry in the Acreage community of Palm Beach County in response to resident concerns. Providing accurate public health information using sound scientific investigation principles is the best way in which DOH can help residents of the Acreage make informed and appropriate decisions at this time. This inquiry was designed to determine whether DOH could find evidence of elevated cancer rates in the Acreage community. In other words, whether the number of observed cancer cases in the community is higher than the number of expected cases based on cancer rates found within Palm Beach County or within the state as a whole. A Level I inquiry uses existing data sources. In this case the Florida Cancer Data System (FCDS) was the source for cancer data and the 2000 Census provided the most recently available published population data. All forms submitted by residents were verified against FCDS records. The time period examined was 1995 to 2007, with more analysis focused on the later years (2000-2007). The study included all types of malignant brain cancers as well as central nervous system (CNS) tumors. Benign and borderline brain cancers and CNS tumors were reviewed beginning in 2004, when data were first available. Based on review of the literature it is important to note that brain cancer is not a single uniform disease but different diseases with different risk factors. A search of FCDS records identified 1369 total cancer cases among all age groups in the Acreage. Twenty three cases of malignant brain cancers and other CNS tumors were found among residents of any age in the period 1995-2007, 18 of these occurred in the period 2000-2007. Six of these 18 cases were pediatric brain cancers and CNS tumors. In addition to these cases, 9 additional cases of borderline or benign brain and other CNS cancers were found for 2004-2007, none among children. Looking over a ten year period, five different types of brain cancer were identified among children; eight different types of cancer were identified in adults. In order to explore whether a particular type of brain cancer was elevated, state registry data for a 10 year period was also reviewed. No one particular type of brain cancer or CNS tumor appeared to be more common than expected when compared to statewide data. The analyses showed increased rates of total cancers particularly for the years 2000-2007. Females, particularly children, experienced an increased rate of brain cancers during this period. Highest rates of brain cancers were seen in 2005-2007. The most recent data by age-groups available for the Acreage population are from the 2000 census. The population in the Acreage has grown in recent years. Some of the increases in rates of total cancers and in brain cancers in the Acreage may simply be because there are more people there now than in 2000. Conclusions: 1369 cancer cases were found among Acreage residents of any age during 1995-2007. Six pediatric brain cancer cases of five different cancer types were found among children in the Acreage during the period 1997-2007. Fifteen brain cancers and other CNS tumors of eight different types were found among adults during this period, 1997-2007. Brain cancer rates in girls were higher than expected over the same time period. Total cancer rates were also elevated in the Acreage. Rates have been increasing from 2000-2007, although based on changes in population this may not represent a true increase.

 The increases in total cancer and in brain cancers may be a reflection of population numbers that are not current. A Level II cancer cluster inquiry will follow and the activities will be: To recalculate cancer rates using more recent population estimates. To verify if any particular type of brain cancer is more common than expected. To describe the residence history and other risk factors of pediatric brain cancer cases through interviews. To continue coordinated efforts with Florida Department of Environmental Protection.

Excerpt from Florida Department of Health Cancer Inquiry Protocol:

Level 1: Preliminary Cancer Cluster Assessment

The Level I phase is intended to primarily involve collection of information surrounding the cases, the geographic area in question, the cancer types involved, and any suspected environmental exposures and contaminants perceived to be associated. Cases submitted via the Patient Listing Form will be verified against cancer incidence records collected by the state cancer registry, FCDS. Cases not verified by the FCDS database or a medically reliable source (e.g. physician's record, medical facility record, or death record) are excluded from data analysis. Once verification is complete, a descriptive analysis of cases is conducted. This may include but is not limited to data summarization by cancer type, sex, age at diagnosis, race, and/or other basic factors. A literature search should be undertaken for inquires that are perceived to be linked to an environmental or occupational agent. The literature search will be conducted to identify information related to environmental agents and cancer. State environmental resources will also be searched to determine if the site is proximal to a listed National Priority List site or DEP identified-contaminated site and to gather all available documents on risk assessments done at this site. The Level I phase of the inquiry is viewed as the data gathering step; cancer cases are identified and verified, environmental contaminants are identified and verified where applicable, and the situation surrounding the inquiry is more thoroughly described. At the end of this phase, the inquiry may be considered closed in certain situations. Inquiries may be closed when: s very few or none of the cancer cases are verified or; s when only a wide variety of disparate cancer types are identified; or s when only other factors come to light that do not plausibly support the possibility of an increase in cancer. Completion of the following step(s) is encouraged before proceeding to Level II: 1. Case verification is complete. 2. Literature search is conducted on the suspected environmental agent and cancer. For inquires that are perceived to be linked to an environmental or occupational agent, the investigation and confirmation of the following criteria are encouraged before proceeding to Level II: 1. There has been a minimum time period from the start of suspected contamination to present that allows for a cancer latency period to have occurred. The range for cancer latency is generally considered 8 to 20 years or more depending upon the type of cancer. Latency for pediatric cancers is thought to be much shorter, and is generally considered 1 to 8 years. 2. The suspected contaminants are identified and have carcinogenic potential at the levels found at the site based on published scientific literature and/or consultations with experts at the state and federal level, such as within the Agency for Toxic Substance and Disease Registry (ATSDR) or within the EPA. 3. A completed pathway for the contaminant(s) that allows for human exposure.

4. The suspected duration of exposure for the population is at least one year unless a one time highly concentrated acute exposure is suspected. Duration of exposure parameters may be viewed by the State Toxicologist.

Background: On May 13, 2009 an email came into the Florida Department of Health, Bureau of Epidemiology,(State Epidemiologist, Dr. Richard Hopkins) requesting assistance in investigating a possible increase in childhood brain cancers in an unincorporated area west of Palm Beach called the Acreage. The individual was a concerned resident whose child had been recently diagnosed and treated for brain cancer. The resident stated that in the course of treatment for her child she had learned about numerous other families in the area with similar cancers and felt that there were an alarming number of brain cancers, particularly among children in her area. The Florida Department of Health (FDOH), Division of Environmental Health (EH), began an investigation using the Florida Cancer Inquiry protocol and accompanying attachments and this inquiry is numbered CI2009001PB. On May 14, 2009 an initial contact letter was sent to the citizen providing an explanation of the general steps involved in the Cancer Inquiry process and providing some additional information and links. A dozen copies of the Patient Listing Form (PLF) were also included and the initiator was asked to complete, sign and return the form to FDOH. The resident had indicated that she would be willing and comfortable asking other members of the community to complete this form. These fellow residents had the option of completing the form and sending it in to FDOH on their own or allowing the citizen initiator to send it in with other forms as a package. The letter was mailed and emailed to the resident and the PLF was provided as a template attachment. Information provided on the PLF is compared to data in the state cancer registry to verify address, time of diagnosis and type of cancer. The related PLFs are used to get an initial overview of the types of cancers seen in the community as well as the size and location of the geographic area in question. Attachment 1 includes a copy of the initial contact letter and template PLF. For this inquiry, FDOH began a Level I inquiry (the first level of investigation) and utilized existing published data to accomplish the review for the period 1995-2007. The Florida Cancer Data System (FCDS)and 2000 census data (both available electronically to the Department) were the primary data sources for this Level I review. The purpose of the Level I review was to determine whether there is an increase in brain cancers or pediatric brain cancers in the Acreage area over what may be expected based on FCDS data during the time period 1995-2007. This type of assessment cannot be used to link an exposure to a health outcome such as cancer. In other words, causality cannot be established. Studies or reviews such as this Level I review presented here for the Acreage are not able to show a link between specific cancers and specific causes such as environmental contaminants. Florida Cancer Data System (FCDS) Review: The FCDS was established in 1981 and contains information on cancers diagnosed since then through 2007(at the time of this report the 2007 data was available but still considered provisional). Hospitals, free standing ambulatory care centers, radiation therapy facilities, pathology laboratories, physician offices and dermatopathologists offices are required to report incidence data to FCDS. The FCDS is an award winning cancer registry, receiving the highest level of national certifications possible which reflects receiving or exceeding national standards for data accuracy, completeness and

timeliness of reporting. The FCDS serves as the repository of cancer data as reported for state residents. Over the years, numerous steps have been taken to improve the completeness of the data set, the timeliness of the reporting and to expand the number of facilities required to report as well as the types of cancers reported. Hospital discharge data, ambulatory surgical center data and death certificate data are also linked with FCDS on an annual basis in order to identify any possible missing cases. FCDS annually also links with data provided by the Florida Association of Pediatric Tumor Programs(FAPTP) in order to assure all pediatric cancers known to FAPTP are represented in FCDS. Of particular importance for this investigation are changes to reporting requirements effective in January 2004. All patients with a newly diagnosed, benign or borderline brain or central nervous system (CNS)tumors, diagnosed on or after January 01, 2004, whether being treated or not were made reportable to FCDS. There are a few exceptions to the broad reporting requirements for cancer in Florida; a few of these may have some relevance in the Acreage investigation and are listed below: a. Data on non-melanoma skin cancers are generally not reported to FCDS. This is not a cancer of concern in the Acreage area but as some analyses have been examined for 'all cancers' this exception is noted. b. Reporting of some early stage in-situ cancers, such as early stage cervical cancer, is also not thought to be complete and thus early stage cervical cancers are not included in this analysis. Early stage cervical cancer is not a cancer of concern in the Acreage area but as some analyses have been examined for "all cancers" this exception is noted. c. Veteran Administration hospitals (VA's) are not required to report to FCDS. FDOH routinely receives records from military facilities (bases). Acreage residents who received health care solely though a VA facility (not military base care) may not be represented in FCDS records. Other Facts about FCDS and Cancer Rates: FCDS contains data on Florida residents only and thus former Florida residents who move out of state and are then diagnosed with cancer after they are considered residents of other states may not be present in the registry. Some states do reciprocate and provide data to FCDS on Florida residents who are diagnosed with and receiving care in other states. Persons with multiple primary cancer diagnoses contribute multiple records to the data base, meaning there is an individual record for each instance of primary cancer diagnosis and one individual may contribute more than one record. Some cancers are known to metastasize to the brain, in these situations, the cancers are recorded by FCDS based on the primary site of cancer and not counted as a case of brain cancer. All records list the date of the initial diagnosis for the primary cancer and all records contain the address at the time of diagnosis as well as the current or last known address. Residence history (time lived in a community or at a particular address) is not found in FCDS or in any other vital statistics record.

In addition, records are unduplicated so that there exists only one record for each individual primary cancer, regardless of the number of sources (such as laboratories, physician offices, hospitals) that report the case. Brain Cancer Epidemiology: Central nervous system tumors are the most common solid tumors occurring among children, with glioma (consisting of astrocytomas, ependymomas and other glial tumors) as the most common type of malignant CNS tumors among children. The incidence of malignant pediatric brain tumors appears to be on the rise based on National Cancer Institute data and various factors such as improved diagnostics and possible increased exposures to environmental triggers have been hypothesized to play a role, although this increase is still being studied. Fortunately, survival rates have increased in recent years. In general, white children and males are at greatest risk for pediatric brain cancers. There are few known risk factors for brain tumors in children. Ionizing radiation and certain specific genetic conditions are documented, and a family history of brain cancer does appear to also increase the risk. Numerous recent studies have explored a variety of risk factors for brain cancers: consumption of nitrates, prenatal or postnatal exposure to radiation, parental occupation involving certain chemical exposures, electromagnetic field exposure, home treated with pesticides, insecticide exposure, exposure to farm animals, exposure to sick pets, having a tonsillectomy, neonatal distress, neonatal infections and other maternal and childhood exposures have all shown weak associations with increased risk of brain cancers in one study or another but have not consistently and repeatedly been seen as a risk factor. One issue that limits studies on adult or childhood brain cancer is the rarity of brain cancers and the variety of types of brain cancers that comprise this diagnosis. The majority of brain cancers remain of unknown cause (Ross and Spector, 2006; Preston-Martin, Munir and Chakrabarti, 2006; Hemminki, Kyyronen and Vaittinen, 1999; Preston-Martin, Yu, Benton and Henderson, 1982; Linet et al., 1996; Cordier, et al, 1994; Gold, et al, 1979; Singer and Byrne, 2001; Singer and Byrne, 2002). Epidemiologic studies examining pesticide, herbicide and insecticide exposure with brain cancer incidence illustrate the difficulties in summarizing associations between brain cancers and exposures. Several case-control studies have shown an association between household pest exterminations and an increase in childhood brain tumors but others have not shown such an association. According to a review article published in 1999, most studies examining maternal and childhood pesticide exposure found an association with increased risk of childhood brain tumors, but some did not. In addition, some studies examined all childhood brain tumors while others explored only specific types such as neuroblastomas, making comparisons between studies difficult. Any assessment of cancer concerns in a community, especially where the concern is related to childhood cancers, involves heightened citizen concern and awareness of all adverse health outcomes. It is important to realize that cancer is not a rare disease when viewed over an individuals' lifetime. The American Cancer Society estimates that 1 in 3 individuals will be diagnosed with cancer in their lifetime. In 2006, over 100,303 new cases of cancer were reported in Florida with Palm Beach County reporting 8,064 cases. In 2006, there were 1,346 cases of malignant brain cancer and other CNS cancer cases reported with 117 cases in Palm Beach County. There were 122 cases of malignant pediatric brain cancer in 2006 and 9 cases were in Palm Beach County. The

2006 age-adjusted incidence rate of all cancer for the state of Florida was 433.13 cases per 100,000 individuals; Palm Beach County had an age-adjusted rate for 2006 of 426.58 cases per 100,000. Overall, the four most common types of cancer in the US are lung and bronchus, prostate, breast and colorectal cancers. Florida shares this same pattern. In general, whites have a higher incidence of most cancer types. Cancer rates, stage of cancer at diagnosis, survival rates, and types of cancer vary by age, by sex and by race and all of these components have changed over the last decades with changes in medical practices and in treatments. Determination of the Geographic Area: For the FDOH review of the cancer data for the Acreage area a determination of the geographical areas and boundaries was first made on the parcel designation for the area obtained from the Florida Department of Revenue (zoning map). The parcel designations are most often used for zoning, building and permitting purposes but first served to define the area. The parcel designation does not have population estimates associated with it. The corresponding zip codes (33411, 33412, and 33470) were explored and determined to encompass an area much larger than the actual Acreage. The census blocks for the Acreage area (2000 census data) were also explored and found to exactly correspond to and exactly overlay the parcel boundaries for the Acreage (Map 1-2). This correspondence between census blocks and parcel designations allowed for population estimates specific to the Acreage area. Census blocks were used for this Level I analysis and 850 census blocks were used to define the area of concern for the Acreage inquiry. Once cancer cases were selected from FCDS based on general area (zip codes), addresses were further geocoded using Arc GIS in order to be placed within or outside of the Acreage area for analysis. Only those cases geocoded to be within the Acreage area census blocks were included in the analysis. The Acreage Community: The Acreage is located in an unincorporated area roughly 15 miles northwest of Palm Beach and is located in the areas near Royal Palm Beach and Loxahatchee Groves. Most of the residents live on one acre or more lots designed to provide a private rural setting and much of the building began in the 1980's. The area is served by private drinking water wells and septic tanks. The area is crisscrossed by mostly dirt roads, canals and storm water drainage channels and some residents have farm plots, pets, horses and other types of rural land usage. The area previously had citrus groves, agricultural acreage and nurseries active on-site and some of those companies are still in operation in the area. Population Data: No recent population data were available for the Acreage area that provided estimates by age group and by gender. Population by age group, race and by gender was extracted for the 2000 census blocks from the US Census Bureau fact finder website ([Link] Population data were not available by census block for the 1990 census years, and were only available for a much larger geographical area

10

(census tracts) for 1990. For the purpose of this Level I inquiry, it was decided to use the 2000 population data and assume a stable population for the period under study. For the Acreage community in 2000, the census data counted 29,036 individuals. Table 1 provides a breakdown of the population by gender and age group. By comparison, the 2000 census data lists 69,608 individuals in the three zip codes that serve the Acreage area, slightly more than two times the Acreage community. The limitations of using the 2000 census data (and assuming no increase) as the population source for the time period are discussed in a later section of this report. For all areas of the analysis population data was multiplied by the number of years examined. For instance, the population data for the three year estimates were calculated by multiplying the 2000 Acreage population numbers by three: 3 x 29,036 = 87,108. Additionally, to be consistent with Acreage population estimates, the population data for Florida and Palm Beach County also were obtained from the 2000 census and population data for analysis were calculated as stated above.

FCDS Data Search: FCDS was searched for cancer cases related to the Acreage area using the three relevant zip codes as broad catchments areas for the period 1995-2008. 2008 cases were captured although 2008 data are not considered complete enough by FCDS standards for use in the analysis. All cases of cancer having any patient address within the three zip codes broadly defining the Acreage area were pulled and then further geocoded to the area. Only cases geocoded to the Acreage area were included in this study. Of note, residents whose cancers were diagnosed after leaving the Acreage area, who were diagnosed and treated in a VA facility, and those who did not seek and receive confirmed diagnoses would not be present in the data set. Acreage Analysis: The primary focus of this report involves malignant cancers of the brain for the period 1995-2007 and benign and borderline tumors of the brain for the period 2004-2007. As another analysis tool, a category all cancer' is also examined for the general population in order to better understand the overall cancer burden of the community. Where possible, results are shown for adults (> 20 years of agel versus pediatric (0-19 years of age) cancers, and by gender. As the population of the Acreage is predominately white, results by race are not presented. All rates were age adjusted. For purposes of comparison, rates for Palm Beach County and for the state of Florida were used to produce expected counts of cancer. In assessing whether the Acreage is experiencing an increase in cancer, Standard Incidence Ratios (SIRS) were used. This involves comparing the observed number of cancer cases to a number that would be expected if the community were experiencing the same rate of cancer as a larger comparison area (in this case Palm Beach County or the state of Florida). Specifically, this is done by calculating rates for the comparison area minus the Acreage. These rates were multiplied by the population in the Acreage based on 2000 census data. The final number was the expected number of cases in the Acreage. The observed number of cases in the Acreage was then divided by the expected number of cases in the Acreage. This ratio of observed over expected is

11

called an SIR. All rates (and the SIRS) in this study were age-adjusted. The SIR is considered elevated when it is over 1.0 and it is considered statistically significant if the 95% interval between the lower and higher confidence limits does not include 1.0. As an example, if 100 cases are observed and 25 cases would be expected based on comparison rates, the SIR would be 4. This elevation would be considered statistically significant if the lower confidence intervals surrounding this were at least 1.1 (does not include 1.0). For the period from May until August 11, 2009, the Acreage community forwarded 140 PLFs to the FDOH. The information on these forms was checked against FCDS records, and 90 PLFs contained information that could be verified as a cancer case listed within FCDS. One PLF listed an Acreage address that could not be located with an Acreage address in FCDS; further research indicated a miscoding in the registry address field and this record was recoded and included in this study (this record is included in the 90 verified PLFs). This was the only cancer case that use of FCDS data alone (without PLFs) would have missed and this case was in the FCDS data base but with an incorrect zip code. Sixteen forms concerned cancers diagnosed in 2009 and were too recent to be located in the FCDS data base. Eight PLFs contained information on a noncancerous diagnosis (epilepsy, non-melanoma skin cancer, etc.) and therefore were not found in FCDS. The remaining 26 PLFs that were not found in the FCDS records were related to addresses outside of the Acreage area, had incomplete information that did not allow verification against FCDS data, or the information provided was complete but were not able to be verified within FCDS. Of the 26 PLFs not able to be verified, only 12 listed an Acreagerelated zip code. Of the 90 PFLs that could be verified in FCDS,60 were geocoded to the Acreage, 12 were outside of the Acreage but in one of the three relevant zip codes and 18 were not in a relevant zip code. Additionally, 3 of the 90 PFLs were found in FCDS but were for cancers diagnosed prior to 1995. In summary,60 PLFs contained information on Acreage residents that could be verified and used in this analysis. Please note that FDOH did include all cases of cancer found in FCDS that could be geocoded to the Acreage area in our analysis, not just those cancer cases submitted through a PLF. Preliminary Assessment of Environmental Agents and Contamination Concerns in the Acreage: Most cancer cases are not able to be linked to an environmental contaminant and the etiology of brains cancer is largely unknown. The contaminant and accompanying risk factor most linked to cancer is cigarette smoking. This affects cancer rates of many different anatomical sites. For the Acreage area, no evidence of widespread contamination is known. The community is an unincorporated area that currently contains at least one active citrus grove, has some light agricultural use, and has numerous small farms including stables. Some of the homes may have been built upon old citrus groves or farm land. The area is adjacent to current citrus groves, farm land, sugar cane fields and is served by private wells. The area is also crisscrossed by canals and storm water drainage channels which can act as pathways for contaminants given the right conditions. There are no known National Priority Listing (NPL)sites proximal to the Acreage. The nearest active site currently undergoing remediation is the Pratt Whitney site which is 46 miles away from the Acreage. Contaminants, including 1-4 Dioxane, have been noted on-site but there is no evidence that any contamination has migrated near the Acreage.

12

Results (Malignant Only): Overview 1995-2007 An FCDS search for cancers in the Acreage area found 1,369 cases of any type of cancer (benign and borderline included in total numbers only from 2004 onward) in the period 1995-2007 among all age groups. Comparison to Palm Beach County and to state rates for the same time period yielded an expected number of total cancers of 1,056 based on state values (and 1,052 based on county values). Computation of an observed number of cancer cases over expected number of cancer cases yields an elevated Standardized Incidence Ratio (SIR)for this time period. This overall SIR is 1.3 and represents a statistically significant increase (the confidence interval does not include 1.0) whether compared to state or to county levels (Table 2). Analysis by gender also displayed elevated SIR's for all cancer for males and females (statistically significant). For the same time period (1995-2007), a total of 23 brain cancers and other central nervous system (CNS)tumors (22 brain cancers, 1 CNS cancer) were found among all age groups. Comparison with an expected number of brain and other CNS tumors yielded a slightly elevated SIR (1.2) but results were not statistically significant for the state or for the county ratio. Analysis by gender also yielded non-significant results, although the ratio for females was elevated particularly for malignant brain cancers and approached statistical significance (Table 2). Overview for 2000-2007 Based on resident comments, the increase in brain cancers was thought to be a more recent trend. The above analysis was also done for the period 2000-2007 in order to explore the concept of a more recent increase. For this shortened time period, a total of 1,026 cases of all cancer were found, producing an elevated SIR of 1.6 which was statistically significant and represented an increased SIR over the 1995-2007 analysis. Analysis by gender also yielded elevated SIRs for all cancer which were statistically significant (Table 3). In other words, the category "all cancers" was elevated (statistically significant) for the Acreage area when compared to what would be expected based on county or on state rates for the same time period and this pattern held for males and for females. For the same time period, a total of 18 brain cancers and other CNS tumors(17 cases of malignant brain cancer) were found among all age groups. Comparison with the expected number yielded slightly elevated SIRs' but these again did not reach statistical significance. Analysis by gender yielded primarily non-significant results, although the ratio for malignant brain cancers among females when compared to Florida state rates was significantly elevated (Table 3). The analysis for the period 2000-2007 was also done for adults (age 20 or greater) and for children (age 0-19). Among adults, rates for all cancers were elevated (statistically significant) as were rates for males and females (data not shown). These results are similar to those seen for the period 1995-2007 and again for 2000-2007. For children, the FCDS search identified 15 total cancer cases and 6 cases of brain cancers (malignant) in this age group. All cancer rates for pediatric patients were not elevated in this time period. Pediatric brain cancer rates were somewhat elevated with an SIR of 2.4 (comparison to state). Analysis by gender revealed an SIR of 4.4 among female children for brain cancers only (5 observed and 1.1 expected in the state comparison).

13

The SIR for brain and other CNS tumors was also elevated (4.0) for female children. The results for male children were non-significant but there was only one case among males in this time period (Table 4). As can be seen in the provided tables, the total number of pediatric cases, particularly brain cancer cases, is small. Brain cancers are rare among children, and although 6 cases appear to represent an excess over expected, these very small numbers make analysis and interpretation difficult. In some years, no cases of brain cancers were reported to FCDS from the Acreage area. For this particular analysis presented in Tables 2-4 for the period 1995-2007 and for 2000-2007 only malignant brain cancers and other CNS tumors were addressed since benign and borderline brain cancer cases were not reportable until 2004. Analyses including borderline and benign brain cancers are presented later in the results section. Three Year Rates: Given the small number of cases of brain cancers (n=18) it is not possible to explore each year separately. These 18 cases represent malignant brain cancers among all age groups for the years 2000-2007. Discrete three year periods were explored in order to better define issues surrounding the increased rates of all cancers and of brain cancers (1996-1998, 1999-2001, 2002-2004, 2005-2007). Further exploration of the rates was done for all cancers and for brain cancers considered in discrete three year increments to obtain a more refined statement of whether there may be increased rates of these cancers in the Acreage. Beginning in the period 1999-2001, and continuing through to 2005-2007, counts for all cancer were elevated in the Acreage when compared to expected counts based on county and on state values (Table 5). Beginning in the period 1999-2001, counts for brain cancers and other CNS tumors were somewhat elevated over expected but these small elevations were not statistically significant (Table 6). Male residents of the Acreage appeared to have increased counts of all cancers beginning in the period 1999-2001 although these were statistically significant for the years 2002-2004 and 2005-2007 only (Table 5). This was true for both the county and state level comparison (county comparison not shown). Males did not show an increase in brain and other CNS cancer cases over expected in any of these three year time increments in either the state or county comparison (Table 6). Female residents of the Acreage appeared to have an increased count in all cancers beginning in 1999-2001 continuing to 2005-2007 with all periods statistically significant in both county and state comparisons (Table 5). Female brain cancers and CNS tumor rates were statistically elevated for the period 2005-2007 in both comparisons and in 1999-20001 for the county level comparison (Table 6). Pediatric Cases Pediatric cancer cases were also explored for discrete three year time periods within the years 1996-2007. There were 18 cases of any cancer among children age 0-19 in the time periods 1996-2007 and this did not appear to represent a significantly elevated rate among children in the Acreage area for any of the discrete three year periods. This is in contrast to what was seen among adults where initial analysis did indicate an increase for all cancers among adult Acreage residents in years beginning 1999-2001 (adult data not shown for three year increments). However, brain cancers and CNS tumors counts

14

were elevated among children (4 observed and 1 expected)for the period 2005-2007 and this result was significant. Children in the Acreage did not show an increase in all cancers in these time periods but did show a statistically significant elevation for brain and CNS tumors for the period 2005-2007 with an SIR of 3.9 (Table 7). Of note, although the 2008 data are not complete, 4 cases of brain and CNS tumors among Acreage residents are currently known to the registry, 3 of which are among children (one male and two female). Pediatric cases for brain cancer occurred throughout the time period 1995-2008 with most cases appearing in 2004/2005 and in 2008. Type of Brain Cancers: Brain cancers are a diverse group of tumors with numerous tumor types (McCarthy, et al, 2002). Different types of cancers have different risk factors. In other words, it is unlikely different types of brain cancers share the same cause. Table 8 presents brain tumor types among pediatric and adult cases found in the Acreage and the corresponding percentages these types of brain cancers represent for the state as a whole (over the last 10 years). As would be expected, the most common type of brain cancer among Acreage children was Pilocytic Astrocytoma; this is also the most common type of brain cancer among Florida children representing approximately 30% of all cases of brain cancer among children in the Acreage (and 25% of all cases among Florida children). The other types of brain cancer among children found in the Acreage (Ependymoma anaplastic, Glioblastoma, Glioma, Mixed Glioma, etc) are represented by one case each during the time period 1997-2007. The distribution of adult types of brain cancer found in the Acreage also follows an expected pattern based on the state values for the last ten years. Analysis by type of brain cancer would be helpful in furthering understanding this issue in the community, given that most types of brain cancer had only one case each and the most frequent type (Pilocytic Astrocytoma) had 2 cases making sub-analysis by type impossible. Results with Benign and Borderline Brain Cancers: In 2004, Florida statute required the additional reporting of all borderline and benign cases of brain cancer to the FCDS. For the period 2004-2007, this added 9 additional cases of brain cancer and other CNS tumors to the Acreage area count. Comparisons of these observed cancer counts to expected counts for benign and borderline cases based on state and county rates showed no elevated rates, nor did analysis by gender show any elevated counts (confidence intervals surrounding SIRs included 1.0). No cases of borderline or benign brain cancer occurred among children in this time frame (data not shown). Combining borderline, benign and malignant brain cancers for the period 2004-2007 resulted in a total of 23 cases of brain cancers and other CNS tumors among Acreage residents in this time period. When borderline, benign and malignant brain cancers only (CNS tumors excluded) are examined, elevated SIR'S of 2.0 (statistically significant) are found. Similar results and SIR were found for this same period for malignant brain cancers (data not shown). Analysis for adults did not show any statistically significant increases. Analysis for females also produced a statistically significant SIR (2.9), an increase not seen among males (Table 9). In summary, analyses for females show a statistically significant increase in both malignant brain and other CNS tumors (20052007)and benign, borderline and malignant brain and other CNS tumors (2004-2007).

15

Combining borderline, benign, and malignant brain cancers for the period 2004-2007 for pediatric cases also produced evidence of significantly increased counts for brain cancers and for brain cancers and other CNS tumors. This increase was most strongly seen when the Acreage was compared to state values. Analysis for pediatric cases by gender again revealed the pattern of increased counts of brain cancer for females with an elevated SIR of 5.5 (Table 10) and increased counts for brain and other CNS tumors with an elevated SIR of 4.7. As no cases of borderline and benign brain cancer were noted among children in the period 2004-2007, these results are best viewed as associated with the malignant brain cancer cases seen in 2004/2005. The SIR related to pediatric cases of malignant only brain cancers among females for the period 2004-2007 were 5.3 (2.1, 13.5 confidence interval) for the comparison to Palm Beach County and 6.8 (2.7, 17.3 confidence intervals) for the state comparison (data not shown). These elevated SIRs are not surprising as the majority of cases fell in the years 2004 and 2005 and inclusion of preceding or succeeding years would produce lower SIRs (Table 4). The very few cases (n=4) among females limits the interpretation and further analysis. Of note, there were 5 cases of benign and borderline brain cancer reported in 2008 recognized to date in the FCDS. One of these cases was in a child.

Additional Analysis: In order to further explore these rates in a larger geographic area, the above analyses were conducted using data from the three zip codes that include and surround the Acreage (33411, 33412 and 33470)as well as these three zip codes minus the actual Acreage. The analysis combining zip codes also showed an increase in the rate of all cancers for the period 1999 to current and an increase in the rate of brain cancers for 2005-2007 only. A similar pattern held with increased rates of all cancers among adults but no evidence of an increased rate of brain cancer among adults. For the combined zip codes an increased rate of pediatric cancers and pediatric brain cancers was identified primarily in the period 2005-2007. These same zip codes were reanalyzed removing the Acreage area in order to assess whether the increase in the zip codes as a whole was most likely due to an increase solely among the Acreage area. These analyses indicated the same patterns as the Acreage and the zip code analysis (all cancer elevated, brain cancer elevated in children for the period 2005-2007 and no adult brain cancer increase). The SIR's were reduced however, indicating that a larger portion of the total zip codes increase was due to the increase in cancer rates in the Acreage area. Limitations: This analysis was conducted as a Level I (first review) investigation of data associated with a possible increase in cancer, specifically brain cancer, in the Acreage area. One of the limitations in this situation is the relative rarity of brain cancers and of other CNS tumors. There are 27 cases of brain and other CNS tumors (malignant, borderline and benign combined)among all age groups for this period 2000-2007 making statistical analysis and interpretation difficult. The small sample size makes estimates unstable and confidence intervals wide. The single largest limitation in this analysis was the lack of accurate current population data. We assumed the 2000 census data for the Acreage area as a stable population

16

denominator. The population in the Acreage was likely to be less than that in the period 1995-1999 and has grown, perhaps substantially, in the period 2001-2007. Current population estimates (unofficial) for the Acreage area range from 40,000-50,000 individuals, an increase of 38 72%. Counts of birth for this area have shown an increase of 48% from 499 births in 1995 to 738 births recorded in 2007. The Acreage is an unincorporated area in Palm Beach County and data from the Palm Beach County Profile does not provide specific information on the Acreage. However, the Palm Beach County Profile listed no residents in Loxahatchee Groves in 2000 with an estimated 3,232 residents by 2008, Royal Palm Beach had 21,523 residents in 2000 and 31,567 by 2008, and Wellington had 38,216 residents in 2000 and 55,076 in 2008. Estimates for these adjacent areas clearly indicate an area growing in population. The total unincorporated areas within Palm Beach County had 521,447 residents in 2000 and 562,807 projected by 2008 and some of this growth would be reflected in the Acreage itself. Data received through demographers with the Planning Department of the Palm Beach County School District also show an increase in school age children (ages 5-14) in specific Study Area Codes relevant to the Acreage, 4,904 children based on the 2000 census and 5,019 children by 2008 including private and home schooled. By all of these supplemental sources, it is clear the Acreage area has been growing in the time period 2000-2007. For the purposes of this report, population numbers by age, available in census data, were necessary to provide expected counts of cancer. We decided to use the established census numbers for the Level 1 review rather than estimate population growth for each age group for the years 2001-2007 from the 2000 census data. By using the established census population figures (acknowledged to be low) we are most likely underestimating the expected counts of disease. The expected population cancer counts are used as the denominator to calculate the SIR. As a result, the SIR may have been inflated and possibly overestimate the magnitude of excess cancer seen in this analysis for the Acreage, particularly for later years. In other words, it is possible that the increased rates of cancers seen in the analyses could be attributed to underestimation of the population (underestimation of the population for recent years would produce observed increases over time in SIRS). Exploration of better estimates for the population in the Acreage by age group may be considered should further studies of this area result. Brain cancers are a group of diseases with many specific types diagnosed in the Acreage. Lumping of these different types for analysis purposes may not make sense from a medical/ biological perspective and it is likely that the risk factors for one type are not the same as risk factors for another. There is little known about risk factors for pediatric brain cancers. Family history and genetics does play a role in some types, and ionizing radiation is the primary known environmental risk factor. Many other behaviors and risk factors have been studied but none have been able to be shown as risks over repeated studies. Residence history is also not ascertained at this level of investigation which is limited to review of existing data bases. Therefore, long term residents who have moved away and are then diagnosed with cancer will be missed from this type of analysis. Residents new to the Acreage who are then diagnosed with cancerwill be counted as resident cases, regardless of the time that they have been living in the Acreage and that any relevant exposures were most likely from outside of the Acreage. Cancer is a disease that takes a number of years to develop between initial exposure time or initial trigger and the diagnosis, including signs and symptoms. This latency

17

period may be affected by the type of cancer in question, the age, sex and race of the individual and other contributing factors, such as smoking status. Many cancers have a latency period of 15-20 years and in some cases latency may be 30 years. Pediatric cancers have a much shorter latency. Personal interviews, surveys, questionnaires and any other method of gathering exposure or other risk factor information specific to each case or to Acreage residents in general were not part of this Level I review and are not a routine part of a Level I review. Because there was no collection of personal information, no assessment of risk factors in this population was made for this study. Limiting the study area to the census block areas that comprise the Acreage may be an artificial boundary for this concern. Analysis of the larger zip code area and the zip code area minus the Acreage residents indicated that the majority of any excess was related to the Acreage area. However, there is no biological reason why any excess would be limited to within these artificial boundaries, particularly as no environmental issues have surfaced at this time. Although all rates and SIRs were age adjusted, adjustment by race was not done in the analysis for this review. Adjustment by or stratification by race may provide slightly different SIRs than were produced here and most likely would have led to lower SIRs. In conclusion, the underestimation of the denominator discussed at length earlier may be driving the increase in all cancer rates in this community seen in this analysis and may also be contributing to the detected increase in pediatric brain cancers. The increase in all cancer among Acreage residents was also explored by cancer type for the discrete three year periods and increases in lung and bronchus, prostate, breast, colorectal, nonHodgkin's lymphoma, liver, leukemia, kidney and bladder cancers were seen, particularly in the later years. This overall increase in most cancers, considering the multi-factorial risk factors associated with such a wide range of cancer types, also points to the underestimation of the Acreage population denominator, as the driving factor in the increase among all cancer rates.

18

Conclusions: The results of this Level 1 analysis suggest: FCDS provided the data needed to address the cancer concerns in the Acreage for this Level I review and did correctly capture nearly all cases used in this review. The Acreage community experienced increased "all cancer rates for the period 1995-2007 among all ages and genders, and particularly for the period 2000-2007 and 2005-2007. It is possible the increased rates are in large part due to an underestimation in the Acreage population counts. The Acreage community experienced increased rates of malignant brain cancer, as well as malignant, benign and borderline brain cancer during the time period 2000-2007 among females, particularly children. This increase in brain cancers were most pronounced in the period 2005-2007. It is possible the increased rates are in large part due to an underestimation in the Acreage population counts. The overall number of cases is quite small in the time period 2000-2007 with 18 cases of malignant brain cancers and other CNS tumors among all age groups(6 among children). No specific type of brain cancer appears elevated. In many years only one case of any type of brain cancer was noted and in some years no cases of brain cancers were noted. Inclusion of borderline and benign cases of brain cancer (for 2004-2007) increased the case count by 9 to a total of 27 cases (6 among children, no additional cases). Combining malignant, benign and borderline brain cancers in the analysis also produced an increased rate for the period 2004-2007 but only when CNS tumors were not considered. For children, combining borderline, benign and malignant cancers revealed the same pattern as seen throughout the analysis with increased rates for female children. It is possible that the increased rates are in large part due to an underestimation in the Acreage population counts. The use of more precise denominator data that provide a more current population estimate by age group may reduce or eliminate significant findings. Recommendations: A Level II inquiry is recommended for consideration to further understand the increases noted. Given the small number of brain cancer cases found and the various types of brain cancers found with no one type found in excess, proceeding to a Level II inquiry should involve discussion of the merits and feasibility of the various steps involved, The following methodological issues should be addressed in a Level II Investigation:

19

 Denominator data should be addressed and more current population estimates by age group determined. Brain cancer types should be further described. Alternate study designs for a Level II study should be explored and discussed and may include a case control study design with interviews of both cases and controls or a more detailed study design which address denominators, race adjusted rates, geographic boundaries and cancer subtypes.

20

REFERENCES Breslow N.E. and Day N.E. 1993. Statistical methods in cancer research, volume II: the analysis of cohort studies. New York: Oxford University Press Census 2000. Bureau of the Census. 2000 census of population and housing. Summary file 1. Washington: US Department of Commerce, 2005. Available at [Link] Accessed July 2009 Cordier S., Iglesias M., et al. 1994 Incidence and risk factors for childhood brain tumors in the Ile de France. Int J Cancer 59 (6) 776-782. Florida Cancer Data System (FCDS). [Link] Fritz A., Percy C., et al. 2000. International Classification of Diseases for Oncology Third Edition. Geneva: World Health Organization Gold E., Gordis L., et al. 1979. Risk Factors for Brain Tumors in Children. Am J Epi 109 (3): 309-319. Hemminki K., Kyyronen P., Vaittinen P. Parental Age as a Risk Factor of Childhood Leukemia and Brain Cancer in Offspring. 1999 Epidemiology 10:271-275. McCarthy B., Surawicz T., et al. Consensus Conference on Brain Tumor Definition for Registration. 2002 Neuro-Oncology April:134-145. Preston-Martin S., Yu M., et al. N-Nitroso Compounds and Childhood Brain Tumors: A Case-Control Study. 1982 Can Res 42:5240-5245. Preston-Martin S., Munir R., Chakrabarti I. Chapter 62. Nervous System, In: Schottenfeld D., Fraumeni J.F., 2006. Cancer Epidemiology and Prevention, Third Edition. New York: Oxford University Press, pp. 1173-1195. Ross J., Spector L. Chapter 65. Cancers in Children, In: Schottenfeld D., Fraumeni J.F., 2006. Cancer Epidemiology and Prevention, Third Edition. New York: Oxford University Press, pp. 1251-1268. Singer M., Byrne J. 2001 Epidemiology of Brain Tumors in Children. Part 1: Descriptive Epidemiology. Childhood Brain Tumor Foundation, Germantown, MD. Singer M., Byrne J. 2002 Epidemiology of Brain Tumors in Children. Part 2: Risk Factors. Childhood Brain Tumor Foundation, Germantown, MD.

21

APPENDIX A. MAPS AND TABLES

22

Map 1. Geographical depiction of the Acreage community located in Palm Beach County as defined by parcels

EMI 11101111111111111111111
" MIMI

Iso ..-:
COW y Hwy 809
IIIIIIIIIIII .IIIIIIIII

ammo

IIMMINIIIIIIIIIIIIII

=MOM 11111111i
Milli

MIMI OniMINNIMMr

MOM tam iiimmViAMMii IIIIIMsainini wIIIIIIIIIIIIIPIIIIIIE

. 1111.111111 =

3= a \ ,,,.... /
mom i -

In Mall

MINN NM

111111111,11.

J
MN

li I 1 I ill. rt

oak
NS MN

-, e 1 .4% ffii Ag
g

IM

II

.e.

r1
CO ., I ,i

tili .,.
al .

I , Ebi

MI r 1 I
RI . .. . . ----I ---Ljl

J /

IN US H ElKaimur: "11.1 .11111111 1 .111.1111 1 11" 1111i NV OS

ALIFerdrtfr iiN 62 NM \ "r.. q

Streets Highways Acreage Parcels ProjectionCoordinate System: GCS Norlh_American_1983 Data Sources: Florida Department of Revenue 2009 Map Date: July 2009 LIlAhira 1,=-

Disclaimer This product is for reference purposes only and Is not to be construed as a legal document. Any reliance on the information contained herein Is at the user's own risk, The Florida Department of Health and its agents assume no responsibility for any use of the information contained herein or any loss rest:thine there ham.

Parcels are geographical units used to define land ownership - population data and location of residents can not be determined using this geographical unit therefore another geographical unit was required.

23

Map 2. Geographical depiction of the Acreage community as defined by the 2000 US census blocks

County Hwy 809

WIM
,[Link]

111111M1.111.1.1.111111.

%mom ME \

MEI ME lei s. mommin 111111=11111101111111 1 I }

----

k
.

Vim ei ilaM ml

I pli li c_ i _ . 1111 11 m

F-

____/

OR '''

.., ..._ _
-1 ( j 1 1 ri t____

_, _..... _ -,,,-

. ..1 ....TagA
a
a

i 41
CD , .._, ,

US H

:: at

'8

Lit c,,i9, -1 ,r*: ir e --r,

in I-1 .-1-I 1 ! l T i t gM er. c

'r--

m.
-"' -

10 -1

ass
s

Streets Highways 2000 Census Blocks

Projection Coordinate SyStern: GCS florthAmerican_1953 Data sources' Florida Department Revenue 2009 Map Date: July 2009 LI,IAHr.1

or

Disclaimer: This product Is for reference purposes only and is not to be cznstrued as a legal [Link] reliance on the information contained herein is at the user's own fist, The Florida Department of Health and its agents assume no responsibility for any use of the information contained herein or any loss resulting there from.

%-...

iL

The geographic unit that corresponded with the parcels used to define the Acreage community was the 2000 US census block. Census block geography was used to obtain population information and to determine if a case resided in the Acreage community.

24

Table 1. Population for Acreage Community as defined by US census blocks by 5 year a e group and vender(2000 US Census Male Under 5 years 5 to 9 years 10 to 14 years 15 to 19 years 20 to 24 years 25 to 29 years 30 to 34 years 35 to 39 years 40 to 44 years 45 to 49 years 50 to 54 years 55 to 59 years 60 to 64 years 65 to 69 years 70 to 74 years 75 to 79 years 80 to 84 years 85 years and over Total population 1213 1471 1541 1191 452 581 1180 1872 1736 1263 816 538 322 237 149 102 29 19 14,712 Female 1099 1389 1429 999 439 784 1394 1929 1590 1058 732 493 307 248 194 108 74 58 14,324 Total 2312 2860 2970 2190 891 1365 2574 3801 3326 2321 1548 1031 629 485 343 210 103 77 29,036

25

Table 2. SIRs for the Acreage Community by gender [Link] to the state and Palm Beach County (1995-2007) Compared to Florida Compared to Palm Beach County Observed 95% Confidence 95% Confidence Expected # Expected # SIR SIR # Cases Interval Interval Cases Cases (1.2, 1.4) 1.4) 1,055.8 1.3 (1.2, 1,051.9 1.3 1,369 All Cancer 1.1 (0.7, 1.7) 20.3 1.2 (0.7, 1.8) All Cases Brain & other CNS cancers 23 19.5 (0.7, 1.8) 19.0 1.2 (0.7, 1.8) 18.4 1.2 22 Brain cancers only ----,----. (1.1, 1.3) 1.2 553.1 1.3 (1.2, 1.4) 684 531.9 All Cancer 0.8 (0.3, 1.4) 11.8 0.8 (0.4, 1.5) 9 11.5 Male Brain & other CNS cancers 0.7 (0.3, 1.4) 11.1 0.7 11.0 (0.3. 1.4) 8 Brain cancers only
-,,,,,,,, ,,,,,,N, , N, ,, . , ti, , ,,,,, , S , e 'C.t<.X,,,,,,,,,,,,,. ... , ,,at, , , ,,, , ,,, , .,.3 , a

All Cancer Female Brain & other CNS cancers Brain cancers only

684 14 14

531.3 8.1 7.6

1.3 1.7 1.8

(1.2, 1.4) (0.9, 2.9) (1, 3.1)

515.2 8.6 7.9

1.3 1.6 1.8

(1.2, 1.4) (0.9, 2.7) (1, 3)

ender compared to the state and Palm Beach County (2000-2007 reage Communityb.. gender Compared to Florida Compared to Palm Beach County Observed 95% Confidence Expected # 95% Confidence Expected # SIR SIR # Cases Interval Cases Interval Cases (1.4, 1.6) 675.3 1.5 (1.5, 1.7) 1.6 658.0 1,026 All Cancer (0.9, 2.2) 1.4 13.0 (0.9, 2.2) 1.4 18 12.7 All Cases Brain & other CNS cancers 1.4 (0.9, 2.3) 12.1 1.4 (0.9, 2.2) 12.2 17 Brain cancers only .. ._,_ (1.4, 1.6) 350.3 1.5 1.6 (1.5, 1.7) 327.1 516 All Cancer (0.6, 2.1) 1.1 2.2) 7.5 (0.6, 7.3 1.1 Male 8 Brain & other CNS cancers (0.5, 2) 7.1 1.0 (0.5, 2.1) 7.0 1.0 7 Brain cancers only ! (1.4, 1.7) 1.6 324.9 (1.4, 1.7) 329.8 1.5 509 All Cancer (1, 3.3) 1.8 5.5 (1, 3.4) 1.8 10 5.5 Female Brain & other CNS cancers (1.1, 3.6) 2.0 5.1 (1, 3.5) 1.9 5.3 10 Brain cancers only

26

Table 4. SIRs for the pediatric (0-19 ears residents in the Acreage by gender (2000-2007 Compared to Florida Compared to Palm Beach County Observed 95% Confidence Expected # 95% Confidence Expected # SIR SIR # Cases Interval Cases Interval Cases 14.8 1.0 (0.6, 1.7) 15 15.9 0.9 (0.6, 1.6) All Cancer 2.2 (1, 4.7) All Cases Brain & other CNS cancers 2.8 2.1 (1, 4.7) 2.8 6 2.5 2.4 (1.1, 5.2) 2.7 2.2 (1.1, 4.9) 6 Brain cancers only Male All Cancer Brain & other CNS cancers Brain cancers only All Cancer Brain & other CNS cancers Brain cancers only 5 1 1 10 5 5 9.6 1.5 1.5 6.4 1.3 1.2 0.5 0.7 0.7 1.6 3.8 4.2 (0.2, 1.2) (0.2, 3.8) (0.2, 3.8) (0.9, 2.9) (1.7, 8.9) (1.8, 9.8) 8.2 1.5 1.4 6.5 1.3 1.1 0.6 0.7 0.7 1.5 4.0 4.4 (0.3, 1.4) (0.2, 3.7) (0.2, 4) (0.8, 2.8) (1.7, 9.3) (1.9, 10.2)

Female

Table 5. Three year SIRs for all cancers diagnosed among Acreage residents by ender (1996-2007 Compared to Florida Compared to Palm Beach County Observed 95% Confidence 95% Confidence Expected # Expected # SIR SIR # Cases Interval Interval Cases Cases (0.8, 1) 228.4 0.9 (0.7, 1) 234.7 0.9 200 1996 - 1998 (1.1, 1.3) 1.2 (1, 1.3) 247.9 249.3 1.2 292 1999 - 2001 All Cases (1.3, 1.5) 254.0 1.4 (1.3, 1.6) 252.4 1.4 353 2002 - 2004 (1.7, 2) 1.8 (1.8, 2.1) 253.5 237.4 2.0 462 2005 - 2007 1996 - 1998 1999 - 2001 2002 - 2004 2005 - 2007 90 140 169 249 110 151 184 119.2 127.2 125.2 115.3 115.5 122.0 126.8 0.8 1.1 1.4 2.2 1.0 1.2 1.5 (0.6, 0.9) (0.9, 1.3) (1.2, 1.6) (1.9, 2.5) (0.8, 1.2) (1.1, 1.5) (1.3, 1.7) 118.4 129.1 132.4 130.6 109.9 118.8 121.7 0.8 1.1 1.3 1.9 1.0 1.3 1.5 (0.6, 0.9) (0.9, 1.3) (1.1, 1.5) (1.7, 2.2) (0.8, 1.2) (1.1, 1.5) (1.3, 1.8)

Male

..,

1996 - 1998 1999 - 2001 Female 2002 - 2004

2005 - 2007

213

121.5

1.8

(1.5, 2)

122.8

1.7

(1.5, 2)

27

Table 6. Three year SIRs for brain and other CNS cancers diagnosed among Acreage residents b ender"I , Compared to Palm Beach County Compared to Florida Observed Expected # 95% Confidence Expected # 95% Confidence # Cases SIR SIR Cases Interval Cases Interval 1996 - 1998 2 4.5 0.4 (0.1, 1.6) 4.6 0.4 (0.1, 1.6) 1999 - 2001 5 3.9 1.3 (0.6, 3) 4.6 1.1 (0.5, 2.6) All Cases 2002 - 2004 6 5.1 1.2 (0.6, 2.6) 4.9 1.2 (0.6, 2.7) 2005 - 2007 9 4.8 1.9 (1, 3.6) 4.9 1.8 (1, 3.5) 1996 - 1998 1999 - 2001 2002 - 2004 2005 - 2007 1 1 4 3 1 4 2 6 3.0 2.4 2.9 2.7 1.5 1.5 2.3 2.2 0.3 0.4 1.4 1.1 0.7 2.6 0.9 2.7 (0.1, 1.9) (0.1, 2.3) (0.6, 3.6) (0.4, 3.3) (0.2, 3.6) (1.1, 6.7) (0.3, 3.2) (1.3, 6) 2.7 2.6 2.9 2.8 1.9 2.0 2.1 2.1 0.4 0.4 1.4 1.1 0.5 2.1 1.0 2.9 (0.1, 2.1) (0.1, 2.1) (0.6, 3.6) (0.4, 3.1) (0.1, 2.9) (0.8, 5.3) (0.3, 3.5) (1.3, 6.2)

Male

1996 - 1998 1999 - 2001 Female 2002 - 2004 2005 - 2007

Table 7. Three year SIRS for pediatric (0-19 years) residents in the Acrea I e Communit 1996-2007) Compared to Palm Beach County Compared to Florida Observed Expected # 95% Confidence Expected # 95% Confidence # Cases SIR SIR Cases Interval Cases Interval 1996 - 1998 3 4.2 0.7 (0.3, 2.1) 4.7 0.6 (0.2, 1.9) All 1999 - 2001 2 4.3 0.5 (0.1, 1.7) 5.1 0.4 (0.1, 1.4) Cancers 2002 - 2004 5 6.5 0.8 (0.3, 1.8) 5.5 0.9 (0.4, 2.1) 2005 - 2007 8 6.2 1.3 (0.7, 2.5) 5.7 1.4 (0.7, 2.8) ...., 1996 - 1998 Brain and 1999 - 2001 Other 2002 - 2004 CNS 2005 - 2007 1 1 1 4 0.6 0.5 1.2 1.2 1.6 2.0 0.8 3.3 (0.4, 8.9) (0.5, 11.4) (0.2, 4.5) (1.3, 8.5) 0.9 1.0 1.1 1.0 1.1 1.0 0.9 3.9 (0.3, 6.3) (0.2, 5.6) (0.2, 5.1) (1.6, 10)

28

Table 8. Morphology for brain and other CNS tumors among Acreage residents and Floridians with correspondingpercentages (1997-2007 Brain Cancer Morphology Astrocytoma, NOS Astrocytoma, anaplastic Chordoma, NOS Ependymoma, NOS Ependymoma, anaplastic Glioblastoma, NOS Glioma, malignant Malignant Rhabdoid Tumor Pediatric Cases Medulloblastoma,NOS (0-19 years) Meningioma, malignant Mixed Glioma Neoplasm Neuroblastoma Oligodendroglioma, NOS Pilocytic astrocytoma Other Morphologies Total Astrocytoma, NOS Astrocytoma, anaplastic Chordoma, NOS Ependymoma, NOS Ependymoma, anaplastic Glioblastoma, NOS Glioma, malignant Malignant Rhabdoid Tumor Adult Cases(20 Medulloblastoma,NOS years or older) Meningioma, malignant Mixed Glioma Neoplasm Neuroblastoma Oligodendroglioma, NOS Pilocytic astrocytoma Other Morphologies Total Acreage Community Count Percent 0 0.0% 0 0.0% 0 0.0% 0 0.0% 1 16.7% 1 16.7% 1 16.7% 0 0.0% 0 0.0% 0 0.0% 1 16.7% Florida Count Percent 138 9.4% 40 2.7% 2 0.1% 3.9% 57 31 2./% 4.8% 71 218 14.8% 5 0.3% 172 11.7% 1 0.1% 19 1.3% 41 2.8% 28 1.9% 47 3.2% 366 24.8% 238 16./% 1474 100.0% 1092 772 41 346 13 6368 758 0 70 287 272 1329 16 480 149 825 12,818 8.5% 6.0% 0.3% 2.7% 0./% 49.7% 5.9% 0.0% 0.5% 2.2% 2./% 10.4% 0.1% 3.7% 1.2% 6.4% 100.0%

0
0

0.0%
0.0%

o
2 0 6 2 2

0.0%
33.3% 0.0% /00.0% 13.3% 13.3%

o
1 0 6 1 0 1

0.0%
6.7% 0.0% 40.0% 6.7% 0.0% 6.7%

0
0 1 0 1

0.0%
0.0% 6.7% 0.0% 6.7%

0
0 15

0.0%
0.0% 100.0%

29

Table 9. SIRs for the Acreage residents diagnosed with benign, borderline, and malignant brain and other CNS tumors by gender (20042007 Compared to Florida Compared to Palm Beach County Observed 95% Confidence 95% Confidence Expected # Expected # SIR SIR # Cases Interval Interval Cases Cases 15.8 1.5 (1, 2.2) (1, 2,2) 23 15.6 1.5 Brain & other CNS cancers All Cases (1.2, 3.3) (1.2, 3.4) 7.4 2.0 7.4 2.0 15 Brain cancers only Males Females Brain & other CNS cancers Brain cancers only Brai & other CNS cancers n Brain cancers only 9 6 14 9 7.3 4.3 8.3 3.1 1.2 1.4 1.7 2.9 (0.7, 2.4) (0.7, 3) (1, 2.8) (1.5, 5.4) 7.0 4.3 8.6 3.2 1.3 1.4 1.6 2.8 (0.7, 2.4) (0.7, 3.1) (1, 2.7) (1.5, 5.4)

Table 10. SIRs for the Acreage pediatric (0-19 years) residents diagnosed with benign, borderline, and malignant brain and other CNS tumors by sender(2004-2007 Compared to Florida Compared to Palm Beach County Observed 95% Confidence Expected # 95% Confidence Expected # SIR SIR # Cases Interval Interval Cases Cases (1.2, 6.3) 1.9 2.7 (1, 5.2) 5 2.2 2.3 Brain & other CNS cancers All Cases 3.2 (1.4, 7.4) 1.6 2.6 (1.1, 6) 2.0 5 Brain cancers only Males Females Brain & other CNS cancers Brain cancers only Brain & other CNS cancers Brain cancers only @Md 1 4 4 1.2 1.1 1.1 0.8 0.9 0.9 3.7 4.7 (0.2, 4.8) (0.2, 5) (1.5, 9.6) (1.9, 12.1) 1.0 0.9 0.8 0.7 1.0 1.2 4.7 5.5 (0.2, 5.5) (0.3, 6.5) (1.9, 12.1) (2.2, 14.2)

30

31

Attachment I

32

FLORIDA DEPARTMENT OF

Charlie Crist Govornor

Ana tvl. Viaimrito Ros. \4P_11. State Sur ;eon Genera'

14

May 2009

. 11 1 .11111111111111*
Thank you for your recent email concerning a possible excess of brain cancer cases in your community. The preliminary data for the years 2000 to 2007 show similar incidence and mortality rates of all brain cancers in Palm Beach County as compared to the state. A cancer inquiry will be initiated as requested. You are at the first step in the cancer inquiry process. Please distribute the enclosed Cancer Patient Information Forms to those individuals with cancer in your community of concern. Explain to them why you have initiated this cancer inquiry and that this information is necessary to assist the Florida Department of Health (FDOH) in investigating your concern. Information in this letter can assist you in this process. Due to confidentiality requirements: each individual needs to complete and sign their own form. These forms may be returned individually or as a packet to Sharon Watkins. Division of Environmental Health, 4052 Bald Cypress Way Bin A-08, Tallahassee Florida 32399.(The address is printed on the form.) It is important that these forms be returned, so ask each person to provide as much information as possible. Please ask each individual to return the enclosed forms within 6 weeks. It is important that you follow-up with them to find out if the forms have been returned. If we do not receive any forms within six weeks, our office will assume that no further action is necessary and will close the inquiry. If more than six weeks is needed to complete and mail the forms, please contact me at 850-245-4444 x3939 or email Sharon_watkinsdoh.[Link]. After the Cancer Patient Information Forms are returned please allow us six weeks to complete the next step of the cancer inquiry process. This phase is primarily for collection of information surrounding the cases, the geographic area in question, the cancer types involved, and any suspected environmental exposures and contaminates. As you are working on this concern, realize that although cancer is very serious and frightening, it is a surprisingly common disease. Current information shows that approximately one out of three Americans will develop cancer in their lifetime, and cancer will affect three out of four families. Also. the risk of developing cancer increases with age, so as the population ages, more cases of cancer in our communities are expected. The term cancer actually covers not one but many diseases that share the common feature of abnormal cell growth. It can occur in almost any part of the body. Each cancer type develops differently and has different risk factors. For example, the main risk factor for lung cancer is cigarette smoking, but for skin cancer it is sun exposure. On the other hand, it really isn't known what causes some cancers like breast cancer.

Division of Environmental Health 4052 Bald Cypreii:s Way, in Ai S Tallahris-sce, El.. 32399-1720

4411111111111M6 Page 2 of 3 14 May 2009

Many people believe that something in the environment causes the cancer they see in their community, but behavior accounts for most of the known cancer risks. Factors such as smoking, poor diet, heavy alcohol use, and sexual and reproductive history can all contribute to developing cancer. It is estimated that less than 10% of cancers are caused by environmental exposures. In contrast; cigarette smoking alone causes about 30% of cancer. In addition. family history is important and contributes to some types of cancer. Most cancers take a long time to develop. It is usually decades from the time someone is exposed to something that might cause cancer to the time that cancer is discovered. This is one of the reasons that cancer is more common in older adults. In addition. the few chemicals that are linked to cancer must have fairly long and/or concentrated exposures before they will actually cause cancer. An apparent increase of cancer in one community or group is called a cancer cluster. If a suspected cluster includes cancers of different types, it is probably not a 'true" cancer cluster_ For example, if someone called and reported that there were many people with cancer in their community, but the kinds of cancer included lung, breast, leukemia; and prostate, it is usually determined not to be a "true" cancer cluster. A confirmed cancer cluster is a relatively rare occurrence and few documented clusters have been able to be linked to an environmental agent. The state of Florida does have a procedure to examine potential cancer clusters. A "true" cancer cluster usually has one or more of the following characteristics: only one type of cancer is involved: the cancer is occurring in an unexpected age group; there is a very rare type of cancer involved; there is a scientifically established causal relationship between the type of cancer and the suspected exposure. In addition, in order to asses a cancer-contaminant link the contamination would need to be found in a vehicle that could reasonably lead to human exposure such as being documented in ground water used for drinking, or in soil samples, etc.(we call this having a completed biological exposure pathway); the level of contamination found off-site would need to be in amounts that could potentially cause harm based on known toxicological information; and the type of cancer that is of concern should be known to be linked to the contaminants in question. The fact that cancer is common does not mean that you, your family, or FDOH should not be concerned. Many cancers are preventable, and early detection saves lives as well as associated medical costs. There are national and state goals designed to lower the number of people with cancer. In order to accomplish this. awareness about cancer prevention and screening needs to increase. Thank you for your interest in this issue. Enclosed are several items to help you understand more about cancer. Additional information can also be obtained at these Web sites: [Link] http:[Link]/

. . 11111 1111.11111r Page 3 of 3 14 May 2009 If you have any questions, please contact me at 850-245-4444 x3939 or sharonwatkins@[Link]. The Florida Department of Health is committed to protecting the health of all Floridians.

Sincely,

, -'"Sti aron WatkirfiS, PhD Bureau of Environmental Public Health Medicine Division of Environmental Health Florida Department of Health

cc: Tara Hylton, Cancer Epidemiologist, Bureau of Epidemiology, Division of Disease Control, Florida Department of Health Barbara Johnson, Senior Community Health Nursing Supervisor, Division of Epidemiology and Disease Control, Palm Beach County Health Department Richard Hopkins, Acting State Epidemiologist, Division of Disease Control, Florida Department of Health

PATIENT LISTING FORM

Please provide the requested information below to aid the Florida Department ofHealth in responding to your cancer concern. Please note that each individual needs to complete their ownform. Forms can be returned individually or as packet to thefollowing address: <Insert contact person and mailing address>

First Name: Birth date: Type of Cancer (diagnosis): Smoking History (Y/N):

Middle: Gender:

Last Name: Race:

Date of diagnosis:

Length of residency (Years/Months):

Occupation:

l authorize the Florida Department of Health to use the patient identifiable information l have provided above to search the statewide cancer registry database and conduct appropriate analyses to respond to my cancer concern.

Signature

Date

l hereby authorize, <Insert name of lead contact>, to submit my patient identifiable information jointly with my local community residents.

Signature

Date