Redland Soccer Club, Inc. v. Department of Army of United States

55 F.3d 827, 1995 WL 289681

• Court: Court of Appeals for the Third Circuit
• Decided: May 15, 1995
• Docket: 93-7829
• Status: Unknown

Opinion

OPINION OF THE COURT

HUTCHINSON, Circuit Judge.

Appellants are three groups of plaintiffs whose separate actions were consolidated by the district court because they all claimed they were harmed by exposure to toxic wastes appellee, the United States Army (“Army”), had deposited in lands once a part of the New Cumberland Army Depot (“Depot”). In all three eases, plaintiffs sought class certification, but the district court denied class certification. 1 One group of plaintiffs consists of workers for the township (“Township Workers”) who converted the area the Army used as a landfill into a soccer field after the Army had transferred it to the township that adjoins the Depot. The second group consists primarily of residents living near the landfill (the “Neighbor Plaintiffs”). The third group are persons, primarily children, who played soccer (the “Soccer Plaintiffs”) on the field created on the site of the Army’s landfill. This third group includes two children of the Elliott family, Todd and Tracey (the “Elliotts”). Tracey suffers from leukemia and Todd from enlarged lymph nodes.

Except for the Elliotts, the primary relief all parties seek is medical monitoring. They appeal the district court’s final order for the Army on all their claims, which was entered following orders granting the Army’s motions for summary judgment. Their appeals raise several important issues. Ultimately, we will affirm the orders of the district court with respect to all appellants except the Elliotts, the only plaintiffs who have been able to produce evidence of actual harm by medical evidence showing the Elliott children are suffering from conditions that require medical attention beyond the medical services everyone in the general population should have. Our reasons, which differ somewhat from those of the district court, follow. 2

I. Factual History

A. The History of Marsh Run Park

1. NCAJD’s Use of the Land as a Landfill

The New Cumberland Army Depot is located just east of the Harrisburg Airport on about 974 acres of land, between the Pennsylvania Turnpike to the south and some railroad tracks and the Susquehanna River to the north. From 1917 until the mid-1950’s, the Depot used a fourteen-acre tract of land in its extreme southeastern part as a landfill to dispose of various wastes. The former landfill is bordered by a railroad embankment and the Susquehanna River to the north and by Marsh Run Creek (“Creek”) and an access road on the south. A drainage pipe carrying storm water from the Depot once emptied out onto the western part of the tract. When the landfill was closed in the mid-1950’s, Depot employees covered the debris with eighteen inches of dirt taken from the bank of the Creek and then spread coal ashes over the landfill’s surface, adding another six to seven inches of cover. The Depot’s perimeter fence was moved westward so that the land, then known as Marsh Run Field, was no longer within the fence. The Depot remains adjacent to the west end of Marsh Run Field.

2. Transfer of the Land to Fairview Township and Conversion Into a Soccer Field

In 1970, an executive order directed the Army to identify and dispose of its excess acreage. The Depot identified its former landfill as excess and engaged in negotiations with adjoining Fairview Township (“Township”) officials to transfer the former landfill to the Township for use as a public recreational area, which would include soccer fields. The land was formally transferred to the Township in 1976. According to both Army officials and Township employee, E. Robert McCollum (“McCollum”), it was common knowledge that the area was once used as a landfill by the Depot. Indeed, the tract was commonly referred to as “Pineapple Junction” because of old canned goods that were known to have been disposed of there. No one from the Depot or the Army informed the Township that the landfill contained potentially hazardous or toxic substances. Whether the Army knew the landfill was contaminated before the transfer is a point of contention. 3

In 1981, the Township began excavating and leveling the site, now known as Marsh Run Park (“Park”), for use as a soccer field. The soccer field was completed in 1982 and was used by the Redland Soccer Club from 1982 until the Park was closed on August 28, 1987.

3. The Park Closure and Tests for Contamination

In the 1960’s and 1970’s environmental concerns intensified in this nation. New laws and regulations reflected this growing concern, and the Army began investigating how wastes were disposed of at its facilities. The Depot was included. In 1972, the United States Army Environmental Hygiene Agency did a study on the Depot’s wastewater discharges’ effect on local streams and waters, including the Susquehanna River and the Creek, a stream which flows through the whole southern end of the Depot. This study concluded that the Depot’s discharges had no apparent deleterious impact on the Susquehanna River, but that they did have “a significant, adverse impact” on the plant and animal life in the part of the Creek lying within the Depot’s boundaries. In June of 1978, the Army recommended that the waters of the Creek within the Depot be closed to recreational use because low levels of polychlori-nated biphenyls (“PCBs”) had been detected in them.

Most of the environmental studies done at the Depot thereafter focused on lands within the Depot and excluded the closed landfill, which was no longer Depot property. In September of 1977, Depot officials discovered a document indicating a one-pound container of potassium cyanide, a toxic substance, had been buried in the landfill. Depot officials contacted the Township and asked for permission to dig it up and remove it. Depot officials were unable to locate any other documents detailing the contents of the landfill, and to date all its contents have not been identified. 4

a. Woodward-Clyde Soil Testing Report Dated July, 1987

In 1986, the United States Army Corps of Engineers (“Corps”) hired Woodward-Clyde Consultants (“Woodward-Clyde”) to perform soil testing at the former landfill to see whether any contamination existed there. This was done pursuant to the Defense Environmental Restoration Account (“DERA”), a program established under 10 U.S.C.A. § 2701 et seq. (West 1983), to investigate and remedy environmental contamination at former Department of Defense sites. The testing was done in March of 1987 when the field was still being used by the Redland Soccer Club.

The parameters of the study were determined by the Corps. Woodward-Clyde installed three monitoring wells surrounding, but not on, the landfill. It dug five test pits: one in the cut area on the northern edge of the Park to obtain background soils, one in the area of the cyanide canister burial and the remaining three on the field itself. The test pits in the field were four to five feet in depth. Two samples were removed from each pit, one near the surface and one at mid-depth. Surface soil samples were also taken, but not from the soccer field area. Groundwater was sampled in the monitoring wells, surface water was sampled at two locations along the site’s boundaries and samples were obtained from in or near the Creek.

The testing demonstrated a “significant presence of contaminants in some areas” of the Park and contamination in most of the soil and sediment samples. Test pit soil samples contained organic contaminants and all surface soils contained elevated levels of petroleum hydrocarbons. Groundwater samples contained elevated concentrations of metals. Woodward-Clyde recommended further testing, including testing of the surface soils from the playing fields and surrounding areas “where fill is visible at the surface.” Appellants’ Appendix (“App.”) at 950a. Following receipt of the Woodward-Clyde report, the Army and the Township, by mutual agreement, closed the Park to further public use and the Army repossessed the land in order to conduct additional testing.

b. Corps’ Public Health Evaluation Dated June 1988

On May 25, 1988, nine months after the Park was closed to public use, the Corps sampled surface soils from seven areas on the soccer fields and one off site in an effort to determine whether the surface soils of the former landfill presented possible human health hazards. The samples were analyzed for the presence of volatile organic compounds (“VOCs”), semi-volatile organics, PCBs, metals and cyanide. In four of the eight locations low levels of polycyclic aromatic hydrocarbons (“PAHs”) were detected. Lead was the only metal detected at levels significantly above those found in the off-site sample. The report considered exposure pathways of dust inhalation, skin contact and inadvertent ingestion of soil by hand-to-mouth contact. The Corps concluded that the sampling results showed “no apparent increase in health risk to the children playing at Marsh Run Park” because the concentrations of contaminants were within the acceptable limits proposed by the United States Environmental Protection Agency (“EPA”).

c. EPA Soil and Groundwater Sampling Report Dated July 29, 1988

On June 11, 1988, the Army determined that the former landfill was an appropriate site for a remedial investigation study, which was also to be performed by the Corps as part of the Defense Environmental Restoration Program. Meanwhile, EPA officials also decided to conduct soil and groundwater sampling at the Park and make a detailed magnetic survey. Surface samples were taken by EPA on June 22 and 23, 1988 at ten on-site locations, stratified soil samples at three locations and groundwater samples at three on-site monitoring wells. EPA concluded that the Park’s surface soils were contaminated with lead and PAHs, its subsurface soils with lead and VOCs and its groundwater with VOCs. The magnetic survey showed three possible drum burial sites in the former landfill.

d. EA Engineering Remedial Investigation Report Dated January 1990

In conducting its remedial investigation, the Corps contracted with EA Engineering, Science and Technology, Inc. (“EA Engineering”). EA Engineering agreed to identify potential sources of contamination, define the nature and extent of site contamination and any immediate offsite impact to ground water, surface water and air, and to assess downstream ground water and surface water for human health and environmental risks. EA Engineering sampled waters from four nearby residential wells on August 31, 1989. It sampled ground water and soils from monitoring wells in two separate phases in February and August of 1989. The record does not contain the entire EA Engineering Remedial Investigation Report, and it does not indicate when EA Engineering took the Creek surface water and sediment samples it analyzes in the Report. EA Engineering did not sample the Park’s surface soils but instead used the results of the Corps’ and EPA’s surface soils testing in May of 1988. EA Engineering did not sample the waste fill itself but analyzed the surface soil and groundwater samples results to determine what contaminants might be flowing out from the waste fill.

EA Engineering concluded that the site was contaminated with PAHs but that the PAHs were not unique to the site. It concluded the site was also contaminated with trace metals including barium, lead, copper and silver. It also determined the fill was a potential source of YOCs. It found VOCs in the bedrock aquifer beneath the site and determined the source of this contamination was probably the fill. It found low-level VOCs discharging to the Creek upstream from and adjacent to the Park, but could not confirm whether this contamination was resulting from ground water flowing from the fill or from some other source. EA Engineering also concluded the trace metals were emanating partially from the fill and some other source. It found no contamination in any of the residential wells, all of which are hydraulically upgradient from the Park.

EA Engineering concluded that any contaminated ground water from the site would migrate north towards the Susquehanna River or flow into the adjacent Creek and that no residences are in the predicted migration path. It also concluded that the Creek’s surface waters upstream from the Park contained low-level volatile contaminants trichlo-roethane and 1,1,2,2-tetraehloroethane, and that the upstream sediments contained low-level PAHs. The Creek’s waters adjacent to and downstream from the Park contained low-level volatiles of trichloroethane and 1,2-dichlorothene. EA Engineering concluded that the source of these compounds was somewhere upstream, unrelated to the Park and that dust from the surface soils was not a significant exposure pathway because the field, when tested, was covered by dense grass.

As for human health risk, EA Engineering concluded that “the past use of Marsh Run Field as a soccer field ... resulted in very little risk to the children using the field.” App. at 1251a. It concluded that there would be potential health risk from any ingestion of on-site ground water, but that such ingestion would be highly improbable because it was unlikely any residential use would be made of the site in the future. Finally, EA Engineering concluded there was no risk to any of the nearby residents because their wells were not contaminated and would not become contaminated in the future. As for residents who may have eaten fish from the Creek, it concluded there was no carcinogenic risk.

B. The Township Worker Plaintiffs

The Township worker plaintiffs consist of seven individuals who either performed the excavation and levelling work while the former landfill was being converted into a soccer field or who mowed the grass and performed maintenance work at the Park after the field was constructed. Their deposition testimony and affidavits can be summarized as follows.

Sometime in 1980 or 1981, Fairview Township Engineer, Robert G. Hartman (“Hartman”), was assigned the task of landscaping the former landfill for recreational use as a soccer field. He conducted a topographical and perimeter survey with a field crew. The Township Workers then excavated and le-velled the land over a four month period during the summer of 1981.

When the Township Workers first arrived at the site, it was covered by brush and trees and was swampy in some areas. The Township Workers cleared the brush and trees and also removed what little topsoil was present at the site, which they put aside for later use. As they began to grade the site and move earth from a higher section of land to the north, near the railroad tracks, to a lower section to the south in order to level the ground, they began to unearth “junk,” including several barrels of white' powder, drums, canisters, broken glass, old coffee mugs, utensils, cans, wood, a railroad rail and gas masks. One of the Township Workers recalled someone digging test holes into the junk area and taking samples from them. No one recalled any Army personnel being on site at any time during the work. As excavation continued, the junk was covered up and used as part of the fill for the lower area. The Township Workers estimated that a cap of an average of three feet of dirt from the higher area near the railroad embankment was placed over the junk. After the fill was levelled, the Township Workers placed on-site topsoil and topsoil from off-site over the fill to a depth of two to six inches. The field was then seeded. The Township Workers testified they inhaled dust, waded through dirt and debris, sat in the dirt and ate their lunch there over the four-month period. One Township Worker recalled that his eyes burned or stung while he was running the grader and tearing up new soil and that the burning did not stop until the newly torn up soil was reburied.

During the excavation and leveling work, some of these same Township Workers constructed a drainage swale to redirect water flowing out of a pipe carrying storm drainage waters out of the Depot. During excavation for the swale, the Township Workers exposed groundwater. Some of the Township Workers performed repairs on a bridge over the Creek and removed debris from the Creek.

A Township Worker named David A. Kupp (“Kupp”) was responsible for mowing the field after it was constructed. He mowed once a week for approximately five hours. Kupp recalled the field as dry and only sparsely covered with grass and remembered the mower kicking up dirt and dust from the surface of the field about 20% of the time. He also spent several hours painting the bridge over the Creek while he stood on the banks and rocks and in the waters of the Creek.

None of the Township Workers are currently suffering from any physical ailment that they claim is the result of their exposure. There is no evidence in the record that any of them have been examined for health problems related to their exposure at the Park or that any doctor has personally informed them that they have an increased health risk because of exposure to toxic substances while working at the Park.

C. The Neighbor Plaintiffs

The Neighbor Plaintiffs are twelve residents living in the immediate vicinity of the Park and the Creek, plus some relatives who regularly visited them. Some ■ testified in depositions or affidavits that they waded in the Creek, fished in it and ate fish they caught there. Several testified they hunted in the former landfill area and ate pheasant, rabbit, squirrel or turtle they caught or shot there. Most of the Neighbors used the Park for walking or other forms of exercise.

On May 8, 1988, the Department of Environmental Resources (“DER”) tested residential wells for trace metals and VOCs and found none of the wells were contaminated. One Neighbor, however, testified that her well water was tested in 1990 by her employer and found to contain high concentrations of several chemicals, including lead. Another Neighbor testified that he was told his well water did not pose a health risk but contained traces of contaminants. He and his family stopped drinking from it. None of the Neighbors are currently suffering from any illness as a result of their exposure, nor have any been personally advised by a doctor that they have an increased health risk as a result of such exposure.

D.The Soccer Plaintiffs

The remaining 128 plaintiffs are members of the Redland Soccer Association (“Red-land”), adults and children who used the Park on a regular basis from 1982 to 1987 for soccer activities, and members of their immediate families who were with them during activities at the Park. Some of the Soccer Plaintiffs testified they helped build the soccer field in 1982, picking rocks and moving dirt around the field and then raking and seeding it, setting goalposts and lining the field in order to get it ready for play in 1983. This took five or six weekends with about six hours work per day.

One of the Soccer Plaintiffs coached a team for Redland. He testified that his team, boys sixteen years old and under, practiced two to three hours, three times each week at the Park and played one game each weekend, half at the Park. He also coached a team for boys fourteen years of age and under. That team also practiced three times each week and played a game once every weekend during the season. Half of this team’s games and practices were played at the Park.

Soccer’s spring season started in April of each year and ended in early June. Its fall season began in mid-August and ended in mid-November. Practices were canceled if it rained, but games were played no matter what the weather conditions were. If water was on the field, someone would shovel the water away or throw sawdust on it. The soccer coach recalled several players falling into the drainage ditch that ran the length of the field, which was sometimes dry and sometimes wet. Each practice started with calisthenics.

The record contains excerpts from the depositions of two of the soccer players. One of them, a goalie, testified he often fell in the dirt around the goal area, sometimes face down in puddles one to two inches deep. He also testified he sometimes got dirt in his mouth that he had to spit out. Both players testified at times they went into the Creek to retrieve balls. None of the players except Todd Elliott and his sister, Tracy Elliott are alleging that they are currently suffering from any ailment as a result of their exposure or that they-have been personally advised by a doctor that they have an increased health risk due to their exposure.

E.The Elliott Plaintiffs

The Elliott plaintiffs include soccer player Todd Elliott and his younger sister, Tracey Elliott, as well as their parents. Todd played soccer at the Park and Tracey, while attending practices and games at the Park, skipped stones in a stagnant creek near the field, sat on the grass and ate food, crawled and ran on the field and walked through a mud-filled gully near the parking area. The Elliotts allege that as a result of Todd’s and Tracey’s exposure to contaminants at the Park, Tracey is suffering from acute lympho-cytic leukemia and Todd suffers from enlarged lymph nodes and an increased risk of cancer. 5

F.The Parties’ Expert Reports 6

1. Plaintiffs’ Risk Assessment

The plaintiffs rely primarily on a report by Richard S. Greeley, Ph.D. (“Dr. Greeley”) of R.E. Wright Associates, Inc. entitled “Public Health Risk Assessment of a Soccer Field Near the New Cumberland Army Depot, Fairview Township,” dated January 24, 1992 (“Plaintiffs’ Risk Assessment”). The Plaintiffs’ Risk Assessment is limited to a study of the health risks for children and adults making use of the former landfill as a soccer field. It does not address any health risks to the Township Workers from their excavation work or to the Neighbors from their recreational use of the Creek and their ingestion of fish and animals from the Creek and Park or water from residential wells.

The Plaintiffs’ Risk Assessment contains the following summary:

Surface soil samples and soil samples from excavation of test pits on the soccer field have shown that volatile and semi-volatile organic compounds and inorganic chemical compounds are present in the soil. Some of these compounds are carcinogenic and others can cause adverse non-carcinogenic health effects. The assessment considered health risks arising from four primary pathways of exposure of the soccer players, referees and coaches to contaminant chemicals in the soil: (1) ingestion of contaminated soil; (2) ingestion of contaminated water on or near the field; (3) inhalation of contaminated dust; and (4) dermal contact with the contaminated soil or water.

The risk assessment was conducted based on U.S. Environmental Protection Agency risk assessment guidance documents.

The results of the calculations indicate that participation in games or practices at the soccer field for periods of time greater than 33 hours results in significant health risks for both children and adults.

The primary chemicals contributing to these risks are the inorganic chemicals arsenic and lead, and the base neutral polycyclic aromatic hydrocarbon ben-zo(a)pyrene. Other inorganic chemicals, polycyclic aromatic hydrocarbons, and semi-volatile organic compounds contribute lesser percentages to the risks.

App. at 2909a. The Report also states that increased risk may arise from absorption of chemicals through a cut, abrasion or perspiration, increased amounts of volatiles and dusts in the air during play and ingestion of contaminated soil or water during rough play when a player’s face comes in contact with the ground.

In arriving at his conclusion that adults and children using the field for soccer play or practice for more than thirty-three hours were exposed to a significant health risk, Dr. Greeley relied on EPA’s risk assessment procedure which consists of four steps: (1) data collection, evaluation and identification of chemicals of concern; (2) exposure assessment; (3) toxicity assessment and (4) risk characterization. Risk characterization involves, among other things, the calculation of carcinogenic risks, which are stated in terms of “risk per million,” and is arrived at by multiplying the calculated “increased risk of cancer” by 1,000,000. 7 For each pathway of exposure, Dr. Greeley added together the cancer risks for each of the carcinogenic chemicals found at the site to derive an increased risk of cancer for each pathway. He then totalled the risks for each pathway to arrive at a total “increased risk of cancer,” which he defined as an increased risk of cancer due to exposure at the site against everyone’s everyday risk of getting cancer.

Using the EPA guideline that treats an increased cancer risk which is greater than one in a million as “significant” and a similar guideline for non-carcinogenic health risks, Dr. Greeley concluded that children or adults playing or practicing soccer at the Park for thirty-three hours or more had an increased risk of cancer of one in a million and an increased risk of non-carcinogenic health problems of three in a million (children) and one in a million (adults). Children who played or practiced soccer at the Park for the maximum calculated exposure time of 1,350 hours had an increased risk of cancer of sixty-five in a million and of non-carcinogenic health risks of thirty-eight in a million. Adults with the maximum calculated exposure time had an increased risk of cancer of forty-six in a million and a non-carcinogenic increased health risk of eleven in a million.

In preparing his report, Dr. Greeley relied on the soil and groundwater sampling performed by Woodward-Clyde in March of 1988. Dr. Greeley also considered the soil and groundwater sampling results from the Corps’ study performed in May of 1988 and the EPA study performed in June of 1988, but he decided to rely solely on the Woodward-Clyde results because of “the non-homogeneity of the landfill/soccer field soil, as well as the difficulty in attempting to correlate samples taken at different depths at different times by different sampling personnel.” App. at 2921a. He reasoned that the Woodward-Clyde study was the most representative of the three, and its sampling was performed while the soccer field was still in use. He acknowledged, however, that Woodward-Clyde only dug three test pits within the immediate area of the former landfill, now the soccer field, and that the samples were not from the surface but were “ ‘near-surface’” samples and composite samples over the four-foot depth of the test pits. Therefore, he concluded “[t]he actual concentrations of the chemicals of concern in the soil to which the soccer players and adults were exposed may vary more or less from the values selected for this risk assessment.” App. at 2960a. He also considered and rejected additional exposure routes via ingestion or contact with the sediments in the marsh area adjacent to the soccer field and the surface water of the Creek because the concentrations of chemicals there were either below the detection limits of the testing method used or no larger than the concentrations in the test pit soils.

2. Plaintiffs’ Medical Monitoring Report

The plaintiffs also rely on a report by Susan M. Daum, M.D. (“Dr. Daum”) entitled “Medical Surveillance for Individuals Exposed to Hazardous Waste on Land Known as ‘Marsh Run Park’ in Fairview, Pennsylvania near the ‘New Cumberland Army Depot’” ( Medical Monitoring Report”) dated May 2, 1993. App. at 3006a. There, Dr. Daum states she relies on Dr. Greeley’s Risk Assessment and agrees with Dr. Greeley that risk levels above one times the background rate of one case per million is medically significant. She refers to “the exposures which occurred from the ... Depot and waste disposal site, whether through well water, or recreational activities on/in contaminated soil,” as having a risk estimate above one in a million but does not state where she obtained the risk estimate for well water exposure, in light of the fact that Dr. Greeley did not address well-water exposure in his Risk Assessment. App. at 3010a. Ultimately, Dr. Daum concludes that

the examinations [she recommends] ... are not out of the ordinary, but consist of the usual adult medical examinations recommended for all adults with the adult risk of cancer in our society from those carcinogen exposures which are already prevalent. It is because of the increased risk of the exposures at the Marsh Run area, however, that such examinations become more urgent, and access to such examinations should not be left to vicissitudes of employment, health insurance contract, or other individual economic difficulties so prevalent in current health care delivery.

App. at 3008a. Therefore, Dr. Daum did not recommend any specialized tests for any of the plaintiffs but did recommend routine physical examinations and preventative programs.

3. Plaintiffs’ Contributing Contaminants Report

Finally, the plaintiffs rely on a report by Richard C. Cronce, Ph.D. (“Dr. Cronce”) of R.E. Wright Associates, Inc. entitled “Evaluation Contributions of Contaminants to the Fairview Township Soccer Field” (“Contributing Contaminants Report”) dated May 19, 1993. App. at 3044. The purpose of the report is “to determine the possible pathways of migration of regulated compounds to the surface of the soccer field, thus exposing persons on the field to potential adverse health effects from these chemicals.” Id. at 3044a. In preparing his report, Dr. Cronce reviewed the various Army reports on the site and also performed a “site walkover” to observe present on-site conditions. Id. at 3044. The report concludes that periodic additions of contaminants to the surface of the soccer field are likely as a result of flooding, overland flow of discharge waters from a drainage pipe adjacent to the field, erosion of top soil which is revealing an underlying layer of coal ash and movement of the contaminants inside the landfill up to the surface soil either by VOCs moving up through pore space in the soil or semi-volatile organic compounds moving upwards as a result of pedoturbation, or physical soil mixing, which occurs as animals or insects dig or burrow in the ground. Finally, Dr. Cronce concluded that because some of the topsoil used on the field originated from a point along the Creek, it was probably contaminated.

Dr. Cronce did not perform any soil testing to confirm his hypotheses. He believed, however, that “[t]he presence of these contaminants on the existing surface has been documented and, therefore, the contribution of these contaminants from these various processes is highly likely.” Id. at 3049a.

4. Elliott Plaintiffs’ Expert Report

Finally, the record contains the affidavit of Peter W. Wright, M.D. (“Dr. Wright”), dated April 23, 1992 regarding plaintiff Tracy Elliott’s acute lymphocytic leukemia and plaintiff Todd Elliott’s enlarged lymph nodes. While preparing his report, Dr. Wright reviewed Tracy and Todd Elliotts’ medical records as well as extensive scientific and medical literature regarding the causes of cancer and acute leukemias in particular. He concluded, “based on a reasonable degree of medical certainty, [that] the chemicals ... found at Marsh Run Park are known to cause cancer, and some have been specifically implicated with acute leukemias, such as that which has affected Tracy Elliott. App. at 2233a-34a.” Dr. Wright further opined that “the acute lymphocytic leukemia of Tracey Elliott is related to her exposure to the [certain] ehemicals[]” [and that] “Todd Elliott, ... due to his exposure to the [these] chemicals, is himself at increased risk of cancer.” Id.

5. Defendants’ Expert Reports

The defendants present a number of expert reports refuting plaintiffs’ experts’ conclusions and assumptions. Jessica Herzstein, M.D., M.P.H. (“Dr. Herzstein”), a physician specializing in occupational and environmental health, reviewed the Plaintiffs’ Risk Assessment and the three soil sample analyses performed in 1987 and 1988. She concluded that no medical monitoring was necessary because the plaintiffs’ excess risk for cancer was extremely low and the risks of such tests outweighed the benefits. Dr. Herzstein also produced an affidavit addressing Dr. Daum’s Medical Monitoring Report and refuting its conclusion that medical monitoring was necessary for the same reasons given in her initial report. Defendants also produced an expert report by Martyn T. Smith, Ph.D. (“Dr. Smith”), a toxicology specialist, who also critiqued Plaintiffs’ Risk Assessment and concluded that the actual excess cancer risk posed to the Soccer Plaintiffs was zero. Dr. Smith also concluded that Plaintiffs’ Risk Assessment was flawed in the following respects: (1) it utilized unrealistic weather conditions; (2) it assumed an exceptionally high intake of surface water and soil; (3) it used test results from soil samples taken three to five feet below the surface; (4) it failed to take into account normal background levels of contaminants and (5) it used rodent studies for carcinogenic potency values, which are upper bound estimates of human potencies. Dr. Smith also produced an affidavit negating Dr. Greeley’s response to his critique of the Plaintiffs’ Risk Assessment.

James H. Jandl, M.D. (“Dr. Jandl”), a specialist in blood and blood disorders, reviewed Tracey Elliott’s medical records as well as the existing literature and research in the fields of hematology and oncology and concluded that there is no medically recognized evidence linking acute lymphatic (lymphob-lastic) leukemia to any chemical substances. He stated that the only known cause of this type of leukemia is exposure to ionizing radiation. He also reviewed Todd Elliott’s medical records and concluded Todd has no medical problem with respect to his enlarged lymph nodes.

Roger Minear (“Minear”), Director of the Institute for Environmental Studies at the University of Illinois and Professor of Civil Engineering, conducted a detailed review of the available documents concerning the Army’s use of the land as a landfill as well as the various soil studies and remedial investigations reports undertaken by EPA and the Army and pertinent literature. He concluded that the landfill has not caused surface contamination at the soccer field and that plaintiffs’ use of the subsurface soil test results to represent the surface conditions on the soccer field was not realistic or scientifically defensible. He also prepared a report critiquing Dr. Cronce’s Contributing Contaminants Report, concluding that Dr. Cronce’s hypothesized transportations of contaminants to the field has not been confirmed by any of the soil samples.

Finally, defendants rely on a report by Marilyn A Hewitt, P.G. (“Hewitt”), a certified professional geologist and former Pennsylvania DER hydrogeologist. Hewitt reviewed Plaintiffs’ Risk Assessment, as well as the soil test reports and other environmental investigation reports at the Park, maps, photographs, depositions and correspondence. She concluded that the exposure assumptions made in the Plaintiffs’ Risk Assessment were not consistent with standard EPA protocols for evaluating human exposure to contaminants when the use of the contaminated property is recreational, such as a soccer field. She found the Plaintiffs’ Risk Assessment was erroneous primarily because it utilized test results from soils as deep as three feet below the surface, whereas the standard EPA protocol called for use of surface soil samples no more than one foot deep. She also found the Risk Assessment failed to average the concentrations of contaminants in the soil samples and calculated the health risks using the maximum concentrations of contaminants, also contrary to standard EPA protocol. Therefore, she concluded that the Plaintiffs’ Risk Assessment contained an inflated estimate of the health risks associated with the soccer field. She also examined Dr. Cronce’s Contributing Contaminants Report and, using the available soil testing results, refuted Dr. Cronee’s assumptions regarding contaminants being contributed from other contaminated areas of the Creek or Depot. She refuted Dr. Cronce’s conclusion that the surface soils were contaminated by upward transport by volitization of contaminants within the landfill based on the “insignificant” concentrations of such chemicals in the soils at the ParkApp. at 3223a. Finally, Hewitt refuted Dr. Cronce’s assumption that pedoturbation had caused mixing of the surface soils with the contaminated subsurface soils based on the fact that the surface soils were tested after the field had been closed for use as a soccer field.

Both Dr. Greeley and Dr. Cronee submitted affidavits responding to defendants’ experts’ critiques of their reports.

II. Procedural History

On June 7, 1990, five of the plaintiffs filed a class action complaint (“Redland complaint”) seeking (1) injunctive relief and money damages under the Federal Tort Claims Act (“FTCA”), 28 U.S.C.A. § 2671 et seq. (West 1994), for remedial action, medical monitoring and emotional distress (“Redland FTCA Plaintiffs”); (2) injunctive relief and response costs under the Comprehensive Environmental Response, Compensation, and Liability Act (“CERCLA”), 42 U.S.C.A. § 9601 et seq. (West Supp.1994), in the form of remedial action, medical monitoring and reimbursement of plaintiffs’ litigation costs; (3) injunctive relief under the Pennsylvania Hazardous Sites Cleanup Act (“HSCA”), 35 Pa.Cons.Stat. § 6020.101 et seq. (West 1993), in the form of remedial action, monetary damages and reimbursement of plaintiffs’ litigation costs and (4) monetary damages for diminution of the Neighbors’ property values based on trespass. On February 26, 1991, the Elliotts filed a complaint under the FTCA alleging negligence and seeking monetary damages for past and future medical expenses, pain and suffering, medical monitoring and litigation costs. The district court consolidated these eases on September 30, 1993.

The district court denied class certification for the Soccer Plaintiffs, Neighbors and Township Workers on March 4, 1991. After the plaintiffs’ motion for reconsideration was denied on May 3, 1991, they filed an amended complaint joining an additional one hundred forty-five plaintiffs (collectively “Red-land Plaintiffs”) on August 26, 1991.

On December 12, 1991, the United States moved to dismiss plaintiffs’ claims for medical monitoring and injunctive relief pursuant to Federal Rule of Civil Procedure 12(b)(6). On February 12, 1992, the district court granted the motion in part, and dismissed the Redland Plaintiffs’ request for a medical monitoring fund under CERCLA. The court also dismissed all of the Redland Plaintiffs’ FTCA claims requesting injunctive relief.

On March 27, 1992, the United States moved for dismissal of the Elliotts’ complaint and for summary judgment. On June 23, 1992, the district court granted the motion in part and entered summary judgment in favor of the United States on all claims except medical monitoring, which it then left for trial.

On June 4, 1992, the United States moved for summary judgment or, in the alternative, partial summary judgment on the amended Redland complaint. On September 15, 1992, the district court granted the motion in part and dismissed the Redland plaintiffs’ citizen suits under CERCLA and HSCA claims for lack of subject matter jurisdiction. The court also granted summary judgment to the United States and dismissed the Redland Plaintiffs’ claims for attorneys’ fees and experts’ fees under CERCLA and for attorneys’ fees under the HSCA. The court did not address the Redland Plaintiffs’ entitlement to expert fees under HSCA. The court denied the United States’ request for partial summary judgment based on sovereign immunity, and denied the motion in all other respects.

On January 25, 1993, the United States moved to dismiss the Redland Plaintiffs’ FTCA claims for lack of subject matter jurisdiction and, in a separate motion, asked for summary judgment on all of the plaintiffs’ claims for medical monitoring and emotional distress, including the Elliotts’ medical monitoring claim. On June 1, 1993, all plaintiffs moved for partial summary judgment. On October 19, 1993, the district court denied the United States’ motion to dismiss but granted its motion for summary judgment on all plaintiffs’ medical monitoring and emotional distress claims and entered judgment against plaintiffs on those claims. It therefore denied plaintiffs’ motion for partial summary judgment.

During the proceedings in the district court, the parties became embroiled in a number of discovery disputes which the district court resolved in the United States’ favor and which plaintiffs now challenge on appeal. These include (1) an order dated January 14,1991 denying plaintiffs’ motion to compel discovery and sustaining the United States’ objections to several interrogatories; (2) an order dated August 13, 1992 denying plaintiffs’ motion to compel the production of one hundred thirty-nine documents and sustaining the United States’ assertion of the Deliberative Process Privilege; (3) an order dated January 29, 1993 denying plaintiffs’ motion to compel the production of five documents and sustaining the United States’ assertion of the Deliberative Process Privilege; (4) an order dated March 4, 1993 granting the United States’ motion for a protective order concerning plaintiffs’ notices of deposition and (5) an order dated November 16, 1993 denying plaintiffs’ motion for emergency relief concerning defense counsel’s contact with former Army employees who were potential witnesses for the plaintiffs.

On November 29, 1993, the court entered final judgment in favor of the United States and against the plaintiffs. On December 23, 1993 plaintiffs filed a timely notice of appeal.

III. Jurisdiction and Standard of Review

The district court had subject matter jurisdiction pursuant to 28 U.S.C.A. §§ 1331,1346 (West 1993) and 28 U.S.C.A. § 2671 (West 1994), as well as CERCLA, 42 U.S.C.A. § 9613(b) (West 1983). It had supplemental jurisdiction over the plaintiffs’ state law claims pursuant to 28 U.S.C.A. § 1367 (West 1993). We have appellate jurisdiction pursuant to 28 U.S.C.A. § 1291 (West 1993).

In reviewing an order granting summary judgment, we exercise plenary review. Viewing the facts in the light most favorable to the nonmoving party, we look to see if there was a genuine issue of material fact; and, if not, whether the moving party was entitled to judgment as a matter of law. See Fed.R.Civ.P. 56; Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249-50, 106 S.Ct. 2505, 2510-11, 91 L.Ed.2d 202 (1986); In re Paoli R.R. Yard PCB Litigation, 916 F.2d 829, 860 (3d Cir.1990) (“Paoli I”).

We review de novo the district court’s determination that the Redland Plaintiffs were not entitled to response costs under CERCLA and HSCA, including a health risk assessment, expert fees, attorney fees and other costs. See United States v. Hardage, 982 F.2d 1436, 1446 (10th Cir.1992). Finally, we generally review the court’s discovery rulings for abuse of discretion. See Marroquin-Manriquez v. I.N.S., 699 F.2d 129, 134 (3d Cir.1983), cert. denied, 467 U.S. 1259, 104 S.Ct. 3553, 82 L.Ed.2d 855 (1984). Nevertheless, we exercise de novo review over the standards the district court used in exercising its discretion. We find an abuse of discretion only if there is either an interference with a substantial right or a gross abuse that could result in fundamental unfairness at trial. Id. (citations omitted).

IV. Plaintiffs’ Claims Under FTCA .

A. Medical Monitoring

We will first address the Redland FTCA Plaintiffs’ argument that they presented sufficient evidence to the district court to withstand a motion for summary judgment on their FTCA claim that the Army’s negligence entitles them to medical monitoring. In Pao-li I, supra, we concluded that the Pennsylvania Supreme Court would recognize a cause of action for medical monitoring for plaintiffs who have been exposed to various toxic substances. See Paoli I, 916 F.2d at 852. In so concluding, we set forth four factors a plaintiff must prove in order to recover:

1. Plaintiff was significantly exposed to a proven hazardous substance through the negligent actions of the defendant.

2. As a proximate result of exposure, plaintiff suffers a significantly increased risk of contracting a serious latent disease.

3. That increased risk makes periodic examinations reasonably necessary.

4.Monitoring and testing procedures exist which make the early detection and treatment of the disease possible and beneficial.

Id. We stated that these factors must be proven by competent expert testimony. Id. (citation omitted). We did not, however, define the term “significantly exposed” or state what details an expert must testify to in order to establish it.

We recently revisited this issue in In re Paoli Railroad Yard PCB Litigation, 35 F.3d 717 (3d Cir.1994), (Paoli II), cert. denied, General Electric Co. v. Ingram, — U.S. -, 115 S.Ct. 1253, 131 L.Ed.2d 134 (1995). There, we noted a New Jersey Supreme Court decision which held that plaintiffs seeking recovery for medical monitoring must “ ‘have ... experienced direct and hence discrete exposure to a toxic substance[.]’ ” Id. at 787 (quoting Theer v. Philip Carey Co., 133 N.J. 610, 628 A.2d.724, 733 (1993)). Theer held a plaintiff who was exposed to asbestos while laundering her husband’s clothes could not bring a medical monitoring claim because, in part, “it was too difficult to quantify her exposure.” Id. We noted our uncertainty regarding the impact of Theer, but believed Pennsylvania courts would not adopt such a narrow view of the cause of action. We reasoned that “[sjome-one indirectly exposed to one chemical might have as much risk of disease as someone directly exposed to another chemieal[.]” Id. at 787-88. Accordingly, we declined to adopt a “per se rule requiring direct exposure, actual injury, and testimony about an individual’s particular level of exposure.” Id. at 788.

Nevertheless, we predicted that the Pennsylvania Supreme Court would set some limits on a medical monitoring claim, and we therefore adopted the Utah Supreme Court’s holding in Hansen v. Mountain Fuel Supply Co., 858 P.2d 970, 980 (Utah 1993). We stated that:

In order for a plaintiff to show significant exposure that causes a significantly increased risk to plaintiff of contracting a serious disease that makes periodic testing reasonably necessary, we think that a plaintiff must: prove that by reason of the exposure to the toxic substance caused by the defendant’s negligence, a reasonable physician would prescribe for her or him a monitoring regime different from the one that would have been prescribed in the absence of that particular exposure. This is because under this cause of action, a plaintiff may recover only if the defendant’s wrongful acts increased the plaintiffs incremental risk of incurring the harm produced by the toxic substance enough to warrant a change in the medical monitoring that otherwise would be prescribed for that plaintiff.

Hansen v. Mountain Fuel Supply Co., 858 P.2d 970, 980 (Utah 1993). The court continued:

[I]f a reasonable physician would not prescribe it for a particular plaintiff because the benefits of the monitoring would be outweighed by the costs, which may include, among other things, the burdensome frequency of the monitoring procedure, its excessive price, or its risk of harm to the patient, then recovery would not be allowed.

Id. at 788. “Significant exposure,” therefore, refers to an exposure which, either by duration or harm, is sufficient to cause a significantly increased risk, which in turn is sufficient to require a monitoring regime different from that normally required in the absence of such an exposure.

Here, in order to establish the first Paoli I factor of significant exposure, the district court reasoned that plaintiffs must show by clear evidence that they were actually exposed to toxins. The district court noted that “[t]here is, of necessity, a degree of speculation in a medical monitoring case. However, ... the allowable conjecture should be in regard to the amount or future effect of the exposure, not whether there was exposure at all.” Redland Soccer Club Inc. v. Dep’t of Army, 835 F.Supp. 803, 810 (M.D.Pa.1993). After reviewing plaintiffs’ expert reports, the district court concluded that nowhere did plaintiffs’ experts unequivocally state that plaintiffs had actually been exposed to any of the toxins alleged to be in the Park’s soils or that any of the toxins had actually entered any of the plaintiffs’ bodies. The district court’s conclusion was based on plaintiffs’ failure to present any evidence that the surface soils of the Park contained a level of contamination harmful to human beings.

The district court’s analysis focuses perceptively on an issue we believe is central in all toxic tort cases; namely, the requirement that the alleged wrong create some significant risk of harm to the plaintiff. Thus, a plaintiff must not only show exposure, but must prove that he was exposed beyond what would normally be encountered by a person in everyday life, so that the plaintiffs risk of being injured from the exposure is greater, in some way, than the normal risks all of us encounter in our everyday lives. In Paoli II, we chose not to delve into the issues of how much exposure there had to be to equal “significant exposure”, nor how “direct” the exposure must be. Instead, we simply required the plaintiff to prove indirectly the nature of the exposure by requiring him to show an “injury” (e.g., a need for medical monitoring greater than that what is required by all persons). 8

We do not believe, however, that the Redland FTCA Plaintiffs’ failure to produce evidence, in the form of blood or tissue tests, showing directly that they absorbed toxins from the field into their bodies is fatal to their claims. Defendants’ own expert stated generally that there are no medical tests which could have detected the presence of the toxins found at the Park, and even if a test existed which could have detected a particular toxin, it would have been useful only if it were conducted within one hundred twenty days of the plaintiffs’ exposure. App. at 82a. Requiring a plaintiff to produce this kind of evidence to support a finding of exposure to a toxic hazard would place an impossible burden on persons subjected to serious medical risk from toxic substances polluters have left to contaminate the environment and afflict the people who live near the wrongdoer’s waste deposits. Thus, even without this direct evidence, we believe plaintiffs may still satisfy the first Paoli I factor through expert testimony showing they were exposed to the toxins at issue at levels significantly above their normal background presence so as to require special tests or more frequent medical monitoring than medicine recommends for the general population. With this in mind, we turn to the Redland FTCA Plaintiffs’ expert reports to see whether any of them have produced evidence sufficient to survive summary judgment on their medical monitoring claims.

Surprisingly, we discover the record has no expert opinion on whether either the Township Workers or the Neighbors have been exposed to toxins to such an extent that they suffer such an increased risk of contracting a serious disease that supplemental medical testing is reasonably required. Of all the Redland Plaintiffs, the Township Workers who excavated and levelled the contaminated landfill for four months would appear to have the highest potential for significant exposure to toxins. Yet plaintiffs’ experts virtually ignore them, as well as the Neighbors, and focus their expert opinion almost exclusively on the soccer players, who would seem to have suffered a more limited risk of significant exposure, given the paucity of evidence in this record showing there were harmful quantities of contaminants in the surface soils of the playing field. Our review of the expert reports presented by the parties, as well as the soil testing analyses, indicates that the subsurface of the landfill was contaminated; but neither the EPA’s nor the Corps’ surface soil test results indicate contamination in the surface soil above background levels. It is undisputed that the soccer players’ primary exposure for any substantial length of time was only to the surface soils on the soccer field. Without evidence showing that the surface was contaminated, the Soccer Plaintiffs cannot show significant exposure.

Dr. Greeley’s report relies solely on the Woodward-Clyde soil samples which were taken at some depth below the surface of the Park, and he acknowledges that the exposure values selected for use in the risk assessment may vary from the actual concentrations in the soil. Nevertheless, we believe that Dr. Greeley’s report, as well as Dr. Cronce’s report in which he concluded that contaminants from the subsurface could migrate upwards into the surface and that flooding of the Creek could spread contaminants on the surface, permit a reasonable factfinder to infer that adults and children using the soccer field for thirty-three hours or more were exposed to carcinogens and noncarcinogens that increased their risk of illness beyond the one-in-a-million benchmark the EPA uses to measure significant risk. 9

Turning to the Redland FTCA Plaintiffs’ evidence concerning special medical monitoring, the only expert report concerning the need for and extent of medical moni toring is that of Dr. Daum. She expressly states that she does not recommend special testing for any of the plaintiffs: 10

I emphasize ... that the examinations suggested [below] are not out of the ordinary, but consist of the usual adult medical examinations recommended for all adults with the adult risk of cancer in our society. ... It is because of the increased risk of the exposures at the Marsh Run area, however, that such examinations become more urgent, and access to such examinations should not be limited to vicissitudes of employment, health insurance contract, or other individual economic difficulties ....

She acknowledges the considerable limitations of the currently known examinations and tests for the early detection of cancer, as well as the fact that some, such as lung cancer screening, create risks that outweigh the potential benefits. She also declines to recommend any specific surveillance tests for any other non-cancerous chronic diseases, for the same reasons. (App. 3011).

Thus, because all the Redland FTCA Plaintiffs (the Township Workers, the Neighbors and the Soccer Players) failed to introduce evidence that their exposure required a different medical monitoring regimen than that which would normally be recommended for them absent exposure, under Paoli II, we will affirm the district court’s order granting summary judgment to the United States on the Redland FTCA Plaintiffs’ medical monitoring claims.

B. Emotional Distress

The Redland FTCA Plaintiffs also seek to recover damages for negligent infliction of emotional distress caused by their exposure to chemicals at the Park. The record shows that none of the Redland FTCA Plaintiffs currently suffer a physical injury or a medically-identifiable effect from any exposure to chemicals at the Park. Therefore, we conclude the district court did not err in granting summary judgment to the United States on this aspect of plaintiffs’ claim. Absent some physical injury or impact, Pennsylvania’s governing law does not provide recovery for negligent infliction of emotional distress. See Wisniewski v. Johns-Manville Corp., 759 F.2d 271, 274 (3d Cir.1985) (citing Cathcart v. Keene Indus. Insulation, 324 Pa.Super. 128, 471 A.2d 493, 508 (1984)); see also Bubash v. Philadelphia Elec. Co., 717 F.Supp. 297, 300 (M.D.Pa.1989) (mere exposure not equivalent to physical injury).

We also believe the district court correctly distinguished Merry v. Westinghouse Elec. Corp., 684 F.Supp. 847 (M.D.Pa.1988). There, the court held that plaintiffs’ emotional distress claims survived Westinghouse’s summary judgment motion because the plaintiffs’ experts testified that plaintiffs suffered a present “physical effect as a result” of exposure to contaminated well water and that some plaintiffs “demonstrated acute physical symptoms of exposure to the chemicals.” Id. at 852. We have no such expert evidence here. Therefore, summary judgment against the Redland FTCA Plaintiffs on their emotional distress claims was appropriate. 11

V. The Redland Plaintiffs’ Claims Under CERCLA

We turn next to the Redland Plaintiffs’ argument that the district court erred in dismissing their claims for response costs, including attorneys fees, expert wit ness fees and health risk assessment costs, under CERCLA. We conclude that the district court correctly dismissed these response cost claims. 12

Although the Redland Plaintiffs acknowledge the existence of an ongoing remedial action at the site, they contend that the remedial action “is not even attempting to address the health risks created by the contaminated site,” and therefore their request for expert fees, costs of health risk assessments and other costs are not “challenges” to the remedial action. Brief for Appellants at 17. They also contend their action is an action for response costs, which is not subject to CERCLA’s prohibition against private remedial actions. See 42 U.S.C.A. § 9613(h) (West Supp.1994).

Under CERCLA, there are at least two theories on which a private individual can base an action for response cost. First, a person who has incurred response costs that were necessary and consistent with the national contingency plan (“NCP”), has a private right of action under 42 U.S.C.A. § 9607(a)(4)(B) (West Supp.1994). Second, any person may bring a civil action on his own behalf in the proper district court against any person, including the United States or its agencies, for violations of CERCLA or against the President or the Administrator of EPA for their failure to perform any act or duty arising under CERCLA. See 42 U.S.C.A § 9659(a) (West Supp.1994); see also Key Tronic Corp., — U.S. at -, 114 S.Ct. at 1965-66.

The district court’s jurisdiction over such actions is limited as follows:

No Federal Court shall have jurisdiction under Federal law other than under section 1332 of Title 28 (relating to diversity of citizenship jurisdiction) or under [relevant] State law ... to review any challenges to removal or remedial action selected under section 9604 of this title, or to review any order issued under section 9606(a) of this title, in any action except one of the following:

(1) An action under section 9607 of this title to recover response costs or damages or for contribution.

(4) An action under section 9659 of this title (relating to citizen suits) alleging that the removal or remedial action taken under section 9604 of this title or secured under section 9606 of this title was in violation of any requirement of this chapter. Such an action may not be brought with regard to a removal where a remedial action is to be undertaken at the site.

42 U.S.C.A. § 9613(h)(1), (4) (West Supp. 1994). 13

The absence of a definition of “response costs” has been the source of much litigation since CERCLA’s enactment. The terms “response” and “respond” are defined as “remove, removal, remedy, and remedial action,” including enforcement activities. 42 U.S.C.A. § 9601(25) (West Supp.1994). “ ‘[R]emedial action’ means those actions con sistent with permanent remedy taken instead of or in addition to removal actions in the event of a release or threatened release of a hazardous substance into the environment, to prevent or minimize the release of hazardous substances....” 42 U.S.C.A. § 9601(24) (West Supp.1994) (in relevant part). The terms “remove” or “removal” means:

the cleanup or removal of released hazardous substances from the environment, such actions as may be necessary taken in the event of the threat of release of hazardous substances into the environment, such actions as may be necessary to monitor, assess, and evaluate the release or threat of release of hazardous substances, the disposal of removed material, or the taking of such other actions as may be necessary to prevent, minimize or mitigate damage to the public health or welfare or to the environment, which may otherwise result from a release or threat of release.

42 U.S.C.A. § 9601(23) (in relevant part).

We do not believe the Redland Plaintiffs’ litigation costs are “response costs” under any of these definitions. The health risk assessment and expert testimony were designed to assess, for litigation purposes, what health risks, if any, the plaintiffs were exposed to while using the Park for recreation. The expert assessments were conducted long after the Park was closed to recreational use and have nothing to do with any remedial or response action at the Park itself. Moreover, under section 9607, plaintiffs may only recover response costs which are necessary and consistent with the NCP. See 42 U.S.C.A. § 9607(a)(4)(B). The heart of these definitions of removal and remedy are “ ‘directed at containing and cleaning up hazardous releases_ [T]herefore[,] ... ‘necessary costs of response’ must be necessary to the containment and cleanup of hazardous releases.” United States v. Hardage, 982 F.2d 1436, 1448 (10th Cir.1992) (health assessments conducted by experts retained for litigation purposes not response costs under section 9607(a)(4)(B)), cert, denied, Advance Chemical Co. v. U.S., — U.S. -, 114 S.Ct. 300, 126 L.Ed.2d 248 (1993); see Cook v. Rockwell Int’l Corp., 755 F.Supp. 1468, 1476 (D.Colo.1991) (“A plaintiff who has incurred no costs, except for litigation expenses, prior to the filing of a CERCLA action has incurred no ‘necessary costs of response’ under § 9607(a).”); Ambrogi v. Gould, Inc., 750 F.Supp. 1233, 1246 (M.D.Pa.1991) (health assessments not recoverable response costs under CERCLA). Therefore, we do not believe the district court erred in determining that plaintiffs’ costs are not response costs because they are not “monies ... expended to clean up sites or to prevent further releases of hazardous chemicals.” Redland Soccer Club Inc. v. Dep’t of Army, 801 F.Supp. 1432, 1435 (M.D.Pa.l992). 14

VI. Summary Judgment Against Elliotts

Finally, we consider the Elliotts’ claims. They challenge the district court orders dismissing their FTCA negligence claim against the Army in which they seek damages for medical monitoring, past and future medical expenses, pain and suffering and emotional distress. The district court held that the Elliotts, like the other plaintiffs, had produced no evidence they were actually exposed to chemicals at the Park. We concluded in Section IV.A. that there is enough evidence to create a genuine issue of material fact as to whether persons using the Park for thirty-three hours or more had enough exposure to carcinogens and non-carcinogenic toxins to create a significant risk to their health. Nevertheless, we affirmed the district court’s order granting summary judgment to the government on the claims of the Township Workers, the soccer players and the Neighbors for medical monitoring because they had failed to show that their exposure made any medical monitoring tests or examinations necessary or desirable other than those that health care professionals recommend for people who have had no exposure to any of the toxic substances that might be found at the Park.

The Elliotts’ case differs in a striking respect from the other plaintiffs in this action: Tracey Elliott suffers from leukemia and Todd Elliott suffers from enlarged lymph nodes. None of the other Redland Plaintiffs show any signs of physical injury from their exposure. Accordingly, the Elliotts have shown harm because the illnesses of their children demonstrate an immediate need for medical monitoring beyond that which is recommended for the general population. Still, neither their common law tort claim nor any of their statutory claims can survive unless they establish their exposure is the cause of their increased medical needs. The district court held that Dr. Wright’s statement that the Elliott children’s illnesses were “related to” their exposure to the health hazards at the Park that resulted from the Army’s deposit of toxic substances there was insufficient to show causation under applicable Pennsylvania law. We believe the conclusion that the Elliotts have failed to show causation, as a matter of law, should not have been made at this stage of the proceeding on the record before the district court.

In order to establish a traditional tort claim for negligence under Pennsylvania law, the Elliotts must establish that the Army’s failure to exercise reasonable care towards them and any breach of its duty exposed them to an elevated risk of foreseeable harm, which resulted in injury. 15 Mohler v. Jeke, 407 Pa.Super. 478, 595 A.2d 1247 (1991). On this record, we think the question whether the Elliotts can successfully establish all these elements of their claim is a question of fact, not law. At this stage, we look only to see if the Elliotts have introduced enough evidence to create a genuine issue of material fact as to each of these elements, including causation, the only one still at issue. For the following reasons, we conclude they have.

We note initially that “cause in fact,” or physical cause, is not the same as “proximate cause,” or legal cause, and that both must be shown. See Bell v. Irace, 422 Pa.Super. 298, 619 A.2d 365, 367 (1993); Novak v. Jeannette Dist. Mem. Hosp., 410 Pa.Super. 603, 600 A.2d 616, 618 (1991). Causation in fact is normally a question for the jury, but proximate cause poses questions of law which require the court to determine whether the defendant’s negligence was so remote that, as a matter of law, he cannot be held liable for the harm which subsequently occurred. See Bell, 619 A.2d at 367; Novak 600 A.2d at 618. Of course, under Pennsylvania law, the exposure must appear to contribute “substantially” to the Elliotts’ injuries, but this is an issue of degree that is usually a question for the factfinder. See Paoli II, 35 F.3d at 761 n. 31 (citing Hamil v. Bashline, 481 Pa. 256, 392 A.2d 1280, 1284) (1978) (discussing Pennsylvania’s “substantial” factor requirement on proving causation).

The district court believed that Dr. Wright’s statement that the Elliott children’s Alnesses were “related to,” rather than “caused by,” any assumed exposure faded to show a causal connection between Tracey’s exposure and her leukemia. It relied on Novak v. United States, 865 F.2d 718 (6th Cir.1989), in rejecting Dr. Wright’s report and thus analyzed Dr. Wright’s report in terms of proximate cause, rather than “but for” cause or causation in fact. In Novak, an expert testified that plaintiffs death was “related” to a swine flu vaccination he received. Id. at 720. This medical opinion was based on the expert’s assumption that the disease plaintiff suffered was caused by a virus. Tellingly, “no witness for the plaintiff could say with scientific or medical certainty that the particular vaccine at issue ... caused [plaintiffs] disorder.” Id. at 722. Consid ered along with the defendant’s expert testimony that there was no scientific evidence that the vaccine was related to the cause of plaintiffs illness, the United States Court of Appeals for the Sixth Circuit concluded that the district court clearly erred in upholding a finding of causation and entering judgment for the plaintiff. Id.

On this appeal, however, we are reviewing the Elliotts’ claims at the summary judgment stage, and on that basis alone Novak is distinguishable. Whether Dr. Wright’s testimony will persuade the factfinder that leukemia is caused by toxins of the type found at the Park remains to be seen. For summary judgment purposes, however, we believe that the Elliotts have introduced sufficient evidence to establish a genuine issue of material fact regarding causation. Dr. Wright testified:

It has been stated in a standard text of oncology that irradiation and exposure to toxic chemicals are the most studied environmental factors that predispose to leukemia. In another standard text of cancer epidemiology, it has also been stated that the probable causes of leukemia include a range of factors, acting singly or in combination, some involving intrinsic host mechanisms and others reflecting environmental exposures, including radiation, chemicals and others.

App. at 2190a. Dr. Wright continued:

Some of the chemicals found on [NCAD] including] ... but not restricted to arsenic, cadmium, trichloroethylene, chloroform, DDT, PCBS, carcinogenic PAHs, hexachlo-robenzene, radium, and pentachlorophenol. Lymphoreticular malignancies in humans have been linked to exposure to benzene, chloroform, chlorophenols, and trichloroe-thylene. Animal studies have also supported a causal relationship between exposure to trichloroethylene, DDT, and benzene and lymphoreticular malignancy in animals. A positive statistical association of childhood leukemia and well water contaminated with chlorinated organics, including trichloroethylene chloroform has been reported. An increased leukemia mortality has been attributed to occupational exposure to chemicals, including organic solvents, and chlorophenols, all chemicals known to be present at Marsh Run Park.

Id. at 2191a. In contrast to Novak, Dr. Wright also testified, “based upon a reasonable degree of medical certainty,” some of the chemicals at the Park cause cancer, including acute leukemia, and Tracey Elliott’s disease is related to her exposure at the Park. Id. at 2238a. 16

Under Paoli II, 35 F.3d at 750-52, the requirement of expert testimony on issues of the type involved here is a matter of substantive law governing a plaintiffs burden of proof. However, Pennsylvania caselaw on causation does not require that expert testimony include any “magic words” such as “caused by,” rather than “related to.” To the extent that “magic words” have any significance in the Pennsylvania cases, they seem merely to reflect Pennsylvania’s sensible requirement that the expert speak “with a reasonable degree of medical certainty.” In Gradel v. Inouye, 491 Pa. 534, 421 A.2d 674, 679 (1980), for example, the Supreme Court of Pennsylvania held:

Expert medical opinion on causation need not be unqualified and absolute, i.e., stated in ‘categorical terms;’ ordinarily, it must establish that the injury was, to a ‘reasonable degree of medical certainty,’ caused by the alleged negligence.

[When] the complexities of the human body place questions as to the cause of pain or injury beyond the knowledge of the average layperson ... the law requires that expert medical testimony be employed. In addition to its bearing on whether or not the defendant’s conduct was negligent, such testimony is needed to establish that the injury in question did, with a reasonable degree of medical certainty stem from the negligent act alleged.

Id. (quoting Hamil v. Bashline, 481 Pa. 256, 392 A.2d 1280, 1285 (1978) (emphasis omitted from original and added)). We believe Dr. Wright spoke with the “reasonable degree of medical certainty” required by Pennsylvania caselaw.

The district court also concluded that Dr. Wright’s report did not sufficiently “rebut” the affidavit submitted by the Army’s expert, Dr. Jandl, stating that leukemia is not caused by the types of chemicals found at the Park and that Tracey’s leukemia and Todd’s enlarged lymph nodes therefore cannot be causally linked to their exposure. Dr. Wright’s report stated that Tracey Elliott’s leukemia is related to her exposure to the chemicals found at the Park and that Todd Elliott has an increased risk of cancer due to his exposure to the same chemicals. It also states that exposure to certain chemicals, including those found at the Park, is among the risk factors associated with leukemia. Dr. Wright and Dr. Jandl simply reached different conclusions regarding the cause of the Elliott children’s injuries after reviewing the Elliotts’ medical records. Because their opinions conflict as to the fact of causation, there remains a genuinely disputed issue of material fact on the issue of causation, which is for the factfinder to resolve. It is up to the jury to decide whether the chemicals at the Park were a substantial contributory cause of the Elliott children’s illnesses. Viewing the evidence in the light most favorable to the Elliotts, the nonmoving party, we believe Dr. Wright’s report is enough to permit the Elliotts to survive the United States’ motion for summary judgment on the issue whether Todd and Tracey Elliotts’ present injuries, including any need for special medical monitoring, have been caused by their exposure to any toxic substances the Army may have deposited in the landfill under the Park in which they played.

VII. Discovery Disputes

Because we will reverse the district court’s order granting summary judgment to the United States on the Elliotts’ claims, we must consider the discovery issues the appellants raise concerning the district court’s decision that the deliberate process privilege enabled the United States to withhold discovery of certain documents that could be relevant or likely to lead to the discovery of relevant information. The plaintiffs appeal three discovery related orders, issued by the district court, which denied their motions to compel production of documents. 17

A. The Deliberative Process Privilege

First, the plaintiffs allege that the district court’s order, dated August 13,1992, denying their motion to compel production of one hundred thirty-nine documents based upon the defendants’ deliberative process privilege, was an abuse of discretion. The deliberative process privilege permits the government to withhold documents containing “confidential deliberations of law or policy-making, reflecting opinions, recommendations or advice.” In re Grand Jury, 821 F.2d 946, 959 (3d Cir.1987) (citing NLRB v. Sears Roebuck & Co., 421 U.S. 132, 150-54, 95 S.Ct. 1504, 1516-18, 44 L.Ed.2d 29 (1975); EPA v. Mink, 410 U.S. 73, 89 & n. 16, 93 S.Ct. 827, 836-37 & 16, 35 L.Ed.2d 119 (1973)), cert. denied sub. nom., Colafella v. United States, 484 U.S. 1025, 108 S.Ct. 749, 98 L.Ed.2d 762 (1988). 18

“[T]he ultimate purpose of this long-recognized privilege is to prevent injury to the quality of agency decisions.” Sears, Roebuck & Co., 421 U.S. at 151, 95 S.Ct. at 1516. It recognizes “that were agencies forced to operate in a fishbowl, the frank exchange of ideas and opinions would cease and the quality of administrative decisions would consequently suffer.” First Eastern Corp. v. Mainwaring, 21 F.3d 465, 468 (D.C.Cir.1994) (quotations and internal ellipses omitted). The deliberative process privilege does not protect factual information, even if such information is contained in an otherwise protectable document, as long as the information is severable. See In re Grand Jury, 821 F.2d at 959. In addition, it does not protect “Communications made subsequent to an agency decision.” United States v. Farley, 11 F.3d 1385, 1389 (7th Cir.1993).

The privilege, once determined to be applicable, is not absolute. First Eastern Corp., 21 F.3d at 468 n. 5; Farley, 11 F.3d at 1389. After the government makes a sufficient showing of entitlement to the privilege, the district court should balance the competing interests of the parties. The party seeking discovery bears the burden of showing that its need for the documents outweighs the government’s interest. This Court has previously stated that “the party seeking disclosure may overcome the claim of privilege by showing a sufficient need for the material in the context of the facts or the nature of the case ... or by making a prima facie showing of misconduct.” In re Grand Jury, 821 F.2d at 959 (internal citations omitted). The United States Court of Appeals for the District of Columbia, recently determined that a district court, in balancing the interests, should consider at least the following factors: “(i) the relevance of the evidence sought to be protected; (ii) the availability of other evidence; (iii) the ‘seriousness’ of the litigation and the issues involved; (iv) the role of the government in the litigation; [and] (v) the possibility of future timidity by government employees who will be forced to recognize that their secrets are violable.” First Eastern Corp., 21 F.3d at 468 n. 5.

Thus, a party’s assertion of the deliberative process privilege requires a two-step review in the district court. First, it must decide whether the communications are in fact privileged. Second, the court must balance the parties’ interests. Because the district court did not sufficiently explain its rationale in either respect, we will vacate its order of August 13, 1992 denying the plaintiffs’ motion to compel discovery of the 139 documents which the Army claimed the deliberative process privilege. 19 On remand, the district court should demonstrate its adherence to the process.

The initial burden of showing privilege applies is on the government. See Schreiber v. Society for Savings Bancorp., 11 F.3d 217, 221 (D.C.Cir.1993). To meet it, the government must present more than a bare conclusion or statement that the documents sought are privileged. Id. Otherwise, the agency, not the court, would have the power to determine the availability of the privilege. Id.

The United States, in support of its assertion of the deliberative process, initially gave the district court a list and description of fifty-nine requested documents which plaintiffs had requested with supporting affidavits. There it states that all the documents fall within one of three categories: “1) [d]raft documents intended only for internal review; 2) comments on draft documents; and 3) internal, pre-decisional notes and memoranda recommending courses of agency action.” App. at 1900. The affidavit then states in general terms that the documents in each category are within the purview of the privi lege. 20 The listing provided some benefit to the district court in its description of the documents, but the detail given in the various descriptions varies. The description given for most of the documents withheld provides little more than general information indicating which of the three general categories the documents fall into. For example, one document is described as “August 1990 draft New Cumberland Army Depot Remedial Investigation/Feasibility Study of- the AMSCS.” Approximately two months after the filing of the initial affidavit and document list, the defendants presented the district court with a second affidavit, verbatim except for the different date, and a listing of an additional 82 documents for which the defendants wished to invoke the privilege. The descriptions of these documents were similarly con-clusory.

Before this Court, both the United States and the Elliotts dispute the reasoning behind the district court’s order. The Elliotts rely on the district court’s use of the phrase “compelling reason” in its order to argue that the court erred by applying a “compelling reason test.” The United States points to the court’s use of the word “outweighs” to argue that the district court applied the correct balance in the exercise of its discretion. The district court stated:

It is understandable that plaintiffs may wish access to this material but they state no specific or compelling reason to obtain any particular item. Based upon our review of the arguments made and authorities cited we believe the need to protect the predecisional, deliberative process in government decision making outweighs the general desire of plaintiffs to view this material as part of discovery.

App. at 2402.

We are reluctant, on this record, to decide just what the district court meant by the use of any particular word or phrase. Instead, we think the district court should, on remand, apply the balancing test as we have outlined it. It should also make any findings of fact that may be needed to support its implicit conclusion that the documents sought fell under the deliberative process privilege, if it so decides on remand. See Comptroller of the Currency, 967 F.2d at 636 (Although the district court used the language of balancing, “[njeither the order nor the ... hearing that preceded it indicates with any clarity, [ ] the factors that persuaded the court to [reach] this conclusion.”); In re Grand Jury, 821 F.2d at 959.

The district court may elect to perform a preliminary in camera review of the documents in question before balancing the competing interests and exercising its discretion. In In re Grand Jury, we referred to the Supreme Court’s statement in Kerr v. United States District Court, 426 U.S. 394, 406, 96 S.Ct. 2119, 2124-25, 48 L.Ed.2d 725 (1976), that “in camera review is a highly appropriate and useful means of dealing with claims of governmental privilege.” In re Grand Jury, 821 F.2d at 959. 21

In considering the United States’ assertion of privilege, the district court should keep in mind the fact that Federal Rule of Civil Procedure 26 authorizes broad discovery into “any matter, not privileged, which is relevant to the subject matter involved in the pending action, [see Federal Rule of Civil Procedure 26(b)(1) ] but the deliberative process privilege, like other executive privileges, should be narrowly construed.” See Coastal States Gas Corp. v. Dep’t of Energy, 617 F.2d 854, 868 (D.C.Cir.1980); Cooney v. Sun Shipbuilding & Drydock Co., 288 F.Supp. 708, 716 (E.D.Pa.1968) (collecting cases).

B. Waiver of the Privilege

The plaintiffs also attack the district court’s order, dated January 29, 1993, denying their motion to compel discovery of five documents. In this respect, they claim, even if the documents were privileged, that the United States has waived its privilege. Waiver is based on the Army’s disclosure of these five documents in the course of a subsequent response. The district court concluded that this disclosure was “inadvertent” and did not qualify as a “voluntary” waiver. App. at 2621. See Transamerica Computer Co. v. International Business Machines Corp., 573 F.2d 646, 651 (9th Cir.1978). The district court did not err when it rejected the plaintiffs’ waiver argument.

We also reject the plaintiffs’ contention that the importance of the documents should be factored into the determination of whether the government waived its privilege. The importance of the documents is relevant to the balancing of interests, but the plaintiffs refer us to no cases and we find none stating that it is relevant to waiver. Moreover, we believe the importance of the documents is immaterial to whether their disclosure was voluntary.

C. Relevance

Finally, the plaintiffs challenge the district court’s order, dated January 14, 1991, regarding their attempt to compel discovery against the Army’s contention that the Privacy Act, 5 U.S.C.A. § 552a(b)(ll) (West 1995), covered some of the plaintiffs’ requests. In this order, the district court held that the Privacy Act did not protect the discovery sought, but denied some of the requested discovery as overbroad or burdensome. The plaintiffs contend that the court failed to apply the standards this court requires in deciding whether a discovery request is over-broad or burdensome. See Josephs v. Harris Corp., 677 F.2d 985, 992 (3d Cir.1982). The United States responds with a footnote in its brief, stating:

The basis for appealing this order is unclear given the fact that the district court required the United States to supply plaintiffs with information which otherwise would have been protected by the Privacy Act, 5 U.S.C. § 552a(b)(ll). This Court should reject the appeal because plaintiffs have failed to show how the district court abused its discretion.

Brief of Appellees at 41 n. 22.

In Josephs, we stated “the mere statement by a party that the interrogatory was overly broad, burdensome, oppressive and irrelevant is not adequate to voice a successful objection to an interrogatory.” Josephs, 677 F.2d at 992 (internal quotations omitted). Instead, “the party resisting discovery must show specifically how each interrogatory is not relevant or how each question is overly broad, burdensome or oppressive.” Id. (citations, internal ellipses and internal quotations omitted).

The record before us has only the district court’s order which states its conclusion that the discovery requests were “overbroad and burdensome.” App. at 1507. On remand the district court might wish to set forth its recognition and use of the Josephs standards in support of its conclusion that the plaintiffs’ request for some documents be overturned,

VIII. Conclusion

The order of the district court dismissing the claims of the Neighbors, the Soccer Plaintiffs and the Township Workers will be affirmed. Its order granting the United States summary judgment on the Elliotts’ claims will be reversed, and their case will be remanded for further proceedings consistent with this opinion.

The parties shall each bear their own costs.

1 . Basically, the named plaintiffs sought to include all persons exposed to any toxic substances the Army had deposited in the affected lands.

2 . We may summarily dispose of two of the arguments appellants raise on appeal. We hold that their argument that the district court abused its discretion when it denied class certification lacks merit and so will affirm the order denying class certification. We also reject appellants' argument that the district court erred in denying “Plaintiffs’ Motion for Emergency Relief,” relating to a discovery motion filed in a separate case, O’Neal v. United States, No. 1:CV-90-1073 (order filed Nov. 16, 1993). The district court had consolidated O’Neal with the instant case for discovery purposes. It held that counsel for the Depot correctly sought to preclude plaintiffs’ counsel from contacting current or former Depot employees about the case without first complying with 32 C.F.R. § 516.35(d). This regulation requires an individual seeking information from present or former employees of the Army to make the request in writing to appropriate Department of Defense personnel. It is known as the Touhy provision after United States ex rel. Touhy v. Ragen, 340 U.S. 462, 71 S.Ct. 416, 95 L.Ed. 417 (1951). The United States argues this order is not appealable because it was entered in a separate case not before the Court in these appeals. Because O'Neal was consolidated with Redland for discovery purposes, the United States also argues that the motion was untimely because discovery had ended and summary judgment had been entered in Redland on all issues but one when the court denied "Plaintiffs’ Motion for Emergency Relief.” Assuming that the order denying discovery in O'Neal is properly before us, we nevertheless hold that this argument also lacks merit.

3 . This issue is the focus of plaintiffs' argument that the district court erred in holding that the government's deliberate process privilege justified the Army's refusal to disclose or discuss certain internal records. We discuss that issue infra in Part VIII of this opinion, where we conclude that the district court did not adequately explain the reasons for its ruling. We note there, however, that this discovery related not to the nature or toxicity of the substances the Army deposited in the landfill, the subject of other discovery and extensive testing, but rather to the Army’s knowledge of their presence. That knowledge, or lack of it, seems to us to have little relevance to the plaintiffs’ medical monitoring claim. See also infra footnote 20. We are thus satisfied that any error in this respect, if indeed the order refusing discovery, as opposed to the failure fully to explicate its rationale, is errone-oxis, does not materially affect the parties’ medical monitoring claims, which are the subject of the claims of all plaintiffs except the Elliotts, who also claim standard tort damages, including damages for pain and suffering.

4 . A United States Army Environmental Hygiene Agency Interim Final Report (Draft) dated December I4-Í7, 1987, called Groundwater Contamination Survey at New Cumberland Army Depot, listed "damaged canned goods” as the major item for disposal along with "damaged, out-of-specification or empty containers from such materials as napalm thickener (aluminum naphthalate soaps), decontaminating agent noncorrosive, decontaminating solution DS-2) (sic), bleaches, and clothing impregnating compounds (acetylene tetrachloride or chlorinated aniline in a chlorinated paraffin binder).” Appellants' Ap pendix ("App.") at 1013a. A United States Environmental Protection Agency Report dated July 9, 1988 speculates that acids, solvents, fuels and plating solutions may also have been disposed of in the landfill.

5 . The Elliotts are also plaintiffs in an action against the Three Mile Island nuclear plant and allege their illnesses were caused by its release of radiation in March of 1978. See Brinser v. Metropolitan Edison Co., No. 481-S-88 (Pa. Commw.Ct. filed Feb. 1988).

6 . Because the parties on appeal do not dispute the admissability of any of the expert reports, we do not address the experts’ qualifications or the reliability of their techniques or data, as otherwise required under our decisions In re Paoli Railroad Yard PCB Litigation, 35 F.3d 717, 742-49 (3d Cir.1994) ("Paoli II"), cert. denied, General Electric Co. v. Ingram, - U.S. -, 115 S.Ct. 1253, 131 L.Ed.2d 134 (1995), and In re Paoli Railroad Yard PCB Litigation, 916 F.2d 829, 855-859 (3d Cir.1990) (“Paoli I").

7 . No one points to any demographic, epidemio-logic or any other type of scientific data, nor to any risk-utility analysis that supports EPA's million-fold regulatory factor as demonstrating the presence of a hazard, nor does this threshold appear in the regulatory or statutory history. Nevertheless, the million-fold factor seems ubiquitous in regulatory risk-utility determinations despite its indeterminate pedigree. We will assume that it has some rational basis and thus represents a regulatory determination to which we must defer in deciding plaintiffs’ statutory claims. Chevron U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984); Federal Labor Relations Authority v. Dep’t of Navy, 966 F.2d 747 (1992). For purposes of simplicity, we will also use it to assess the tort claims. We note, however, that a common law court may still be free to apply standard tort risk-utility analysis to the problem of defining the threshold at which a toxic substance becomes a hazard.

8 . The injury requirement is particularly important before a remedy such as medical monitoring is provided because the plaintiff's injury is only an increased possibility of harm rather than actual harm. Paoli II's requirement of "special” medical monitoring implicitly recognizes the longstanding requirement in all tort cases other than those based on the old "intentional” common law torts for various forms of trespass that a plaintiff must prove an injury before he may recover anything from a defendant. See Gideon v. Johns-Manville Sales Corp., 761 F.2d 1129, 1136 (5th Cir.1985) ("An actionable tort, whether based on negligence or strict liability consists of two elements: a failure to act in accordance with the standard of care required by law and a resultant injury.... However egregious the legal fault, there is no cause of action for negligence ... until there is 'actual loss or damage resulting to the interests of another.' ") (citations omitted). Otherwise, a polluter would become a health care insurer for medical procedures routinely needed to guard persons against some of the ordinary vicissitudes of life. It would convert toxic torts into a form of specialized health insurance. Imposition of liability on this basis seems to go beyond current tort theories of negligence or strict liability by requiring a polluter to pay for medical procedures that the general population should receive. Thus, Paoli II requires plaintiffs to show not only that their exposure to toxic substances is greater than normal background levels, but that the increased risk of inju ry from such exposure warrants medical monitoring against future illness beyond that which is recommended for everyone. See Paoli II, 35 F.3d at 788.

9 . We again note, however, that Dr. Cronce's report is not based on any study quantifying the actual effects of this hypothetical migration. We also note again that EPA's basis for its use of the one-in-a-million lifetime ratio to judge significant exposure is not readily apparent. Nevertheless, we will assume a rational basis for EPA’s one-in-a-million standard in defining "significant” risk. See supra footnote 7.

10 . Dr. Daum does not distinguish between the different groups of Redland Plaintiffs. Rather, she refers simply to them as the “individuals” exposed to contaminants at the Park. App. at 3006a.

11 . The Redland FTCA Plaintiffs also argue the district court erred in dismissing their FTCA request for remedial action under CERCLA and HSCA. This remedy is not available against the United States under FTCA, and we will affirm the district court's order dismissing it. See, e.g., Hatahley v. United States, 351 U.S. 173, 182, 76 S.Ct. 745, 752, 100 L.Ed. 1065 (1956) (district court did not possess power under FTCA to enjoin United States); Moon v. Takisaki, 501 F.2d 389, 390 (9th Cir.1974) (per curiam) (FTCA does not submit United States to injunctive relief). Accordingly, we need not, and do not consider whether the district court correctly denied the Redland FTCA Plaintiffs' motion for summary judgment on their claim of negligence per se against the United States.

12 . Initially, we note that the Redland plaintiffs also contend that the district court erred in dismissing their claim for medical monitoring under CERCLA. We believe that the elements of a claim for medical monitoring under CERCLA and HSCA are the same as the elements for a common law medical monitoring claim set out in Paoli I and Paoli II. Because of our conclusion that the Redland FTCA Plaintiffs failed to introduce sufficient evidence to survive a summary judgment motion on their FTCA medical monitoring claim, we need not and do not address whether they could recover medical monitoring costs from the United States as a "response cost” under CERCLA or HSCA. We also note that the United States Supreme Court recently held that attorneys fees are not recoverable as response costs under CERCLA. See Key Tronic Corp. v. United States, - U.S. -, -, 114 S.Ct. 1960, 1967, 128 L.Ed.2d 797 (1994). Thus, as the Redland Plaintiffs concede, we must affirm the district court’s order dismissing their request for attorneys fees under CERCLA.

13 . The Redland Plaintiffs argue the district court incorrectly limited its analysis to section 9613(h)(1), (4) in dismissing their CERCLA claims. Nevertheless, when we look to section 9659, we see that the issues still to be decided on this appeal do not involve either of the kinds of claims that can support a citizen’s suit alleging a removal or remedial action undertaken or secured by or through EPA. The Redland Plaintiffs do not contend otherwise. Therefore, we focus our analysis on the first exception to section 9613(h).

14 . The Redland Plaintiffs also argue that the district court erred in dismissing their citizen suits for injunctive and remedial relief, as well as attorneys fees, expert fees and health risk assessment costs under HSCA. In view of our disposition, we need not decide that issue. We note, however, that the test of HSCA is markedly different from that of CERCLA. We express no opinion, however, about any claim for medical monitoring under state law.

15 . The parties agree that Pennsylvania law governs all state law issues in this case. We believe causation is an issue that should be determined by state law on all the theories of recovery the Elliotts assert. Paoli II, 35 F.3d at 717. Like the district court, we recognize that violation of a statute or administrative regulation may be negligence per se, and therefore a higher degree of care may be required in handling dangerous or toxic materials. The principles of negligence per se aid plaintiffs in establishing a breach of duty, but they do not avoid the issue of causation. The basic elements of any negligence claim remain a duty, breach of the duty, actual loss or harm and a causal connection between the breach and the harm. See Casey v. Geiger, 346 Pa.Super. 279, 499 A.2d 606, 612 (1985).

16 . Of course, if it could be shown, by cross-examination or otherwise, that Dr. Wright used the term "relation" to mean "correlation” in the statistical sense instead of cause in either the medical or legal sense, the force of his testimony could be significantly affected.

17 . The plaintiffs also challenge the district court’s order, dated March 4, 1993 granting the defendants’ motion to require the plaintiffs to travel to the residence or place of business of certain witnesses who were sought to be deposed. The plaintiffs contend that it was an abuse of discretion to require them to send an attorney there to conduct the depositions and that the district court should have ordered the depositions to occur at the site of the litigation. In the end, the plaintiffs deposed these individuals by phone and so did not incur the travel expense to which they objected. The plaintiffs now claim this was unfair because they were unable “to evaluate the appearance and conduct of [the] witnesses." Reply Brief of Appellants at 22. We cannot say, based on this generalized complaint, that the district court's order was an abuse of discretion. See Marroquin-Manriquez v. I.N.S., 699 F.2d 129, 134 (3d Cir.1983), cert. denied, 467 U.S. 1259, 104 S.Ct. 3553, 82 L.Ed.2d 855 (1984) (a district court has broad discretion in determining the manner in which discovery is conducted.).

18 . Our discussion of the deliberative process privilege is based, in part, on interpretations of the bank examination privilege. The two privileges are similar and precedent concerning them is often relied upon interchangeably. See, e.g., Schreiber v. Society for Savings Bancorp, Inc., 11 F.3d 217, 220-22 (D.C.Cir.1993); In re Subpoena Served Upon the Comptroller of the Currency and the Secretary of the Board of Governors of the Federal Reserve System, 967 F.2d 630, 634 (D.C.Cir.1992).

19 . Our remand on these discovery related orders does not affect our affirmance of summary judgment against the plaintiffs who were unable to show an injury. The discovery that was denied sought information on violations of law, the defendants’ knowledge, overall contamination, etc. See App. at 2334-37. These issues go to breach of duty, or violation of law, not to the special monitoring that might be necessary from the exposure involved here. See also supra footnote 3. Unlike the other plaintiffs, the Elliotts and their children seek damages beyond medical monitoring and have produced evidence showing present illness, not just future risk of harm.

20 . The entire argument the United States presented to the district court follows:

An essential element in the effective management of the Army’s environmental program at Army installations is the assurance that the Army, its contractors, and other reviewing agencies may engage in free and candid discussions while formulating Army positions, arriving at Army decisions, and preparing final versions of Army documents. Draft documents and the comments on draft documents listed in attachment 1 are an integral part of the deliberational, predecisional processes that results in a final Army decision. Editorial changes that would be apparent ff [o]m a comparison of the draft to the final document reflect the personal opinions and mental impressions of the drafting and editing staff. Disclosure of draft documents and comments on drafts would stifle the editing process and impair the frank presentation of ideas that accompanies the drafting and finalization of Army documents. Disclosure would, therefore, result in an identifiable harm to the public interest, namely an impairment of the decision making process would result in prejudice to the Army’s goal of making the best possible decisions and producing the best possible documents.

App. at 1900-01.

21 . See also In re Franklin Natl. Bank Securities Litig., 478 F.Supp. at 582 (“Given this clash of strong competing interests, the official information privileged usually requires examination of documents in camera.”); Northrop v. McDonnell Douglas Corp., 751 F.2d 395, 405 (D.C.Cir.1984) ("The litigant’s need for the information cannot be balanced against its sensitive and critical role in the government's decision making process without any indication of what that information is.").

The plaintiffs also contend that the district court erred by not considering whether the government committed illegal acts, the fact that the government is alleged to be a tortfeasor, that the individual who characterized the documents could not objectively evaluate their deliberative content, that the government failed to persuasively show the potential harm from disclosure and that the government failed to adequately describe the documents. We believe these tests tend to beg the privilege question.