ORDER DENYING UNIVERSITY OF SOUTHERN CALIFORNIA’S MOTION FOR REVIEW AND RECONSIDERATION OF MAGISTRATE JUDGE’S ORDER FOR PRODUCTION OF DOCUMENTS
PAEZ, District Judge.
I. Introduction
Plaintiff Roselyn Wolpin has brought a suit in the Southern District of Florida against defendant Philip Morris, Inc., alleging that she has suffered health problems as a result of occupational exposure to secondhand tobacco smoke. As part of that action, defendant Philip Morris sought the production of certain raw data relating to an epidemiological study entitled “Lung Cancer in Nonsmoking Women” (hereinafter “Fontham Study”). The Fontham Study focused on whether passive smoking and exposure to secondhand smoke increase the risk of lung cancer in lifetime nonsmoking women. The conclusions of the Study apparently indicate a potential link between secondhand smoke and lung cancer.
The Fontham Study was conducted by five institutions: one among these was the University of Southern California (USC). Dr. Anna Wu, Ph.D., served as USC’s principal investigator for the Study. On or about October 23,1997 Philip Morris served a subpoena on USC seeking a subset of the raw data collected for the Study that is in the possession of Dr. Wu.
Philip Morris subsequently brought a motion to compel production of these documents. This motion was heard by the assigned magistrate judge on September 9, 1998. On the same day the magistrate judge issued an order granting defendant’s motion, subject to certain conditions (“Sept. 9, 1998 Order”). USC has now filed the instant motion for review and reconsideration of the magistrate judge’s order.
The Court heard oral argument on USC’s motion on November 16, 1998. The Court then issued a minute order on December 7, 1998, requiring the parties to submit addi[422]*422tional documentation. In particular, the Court required defendant Philip Morris to provide justification for the document production sought in light of a pending motion for summary judgment by defendant in the Southern District of Florida. It also required USC to provide (in camera) a hardcopy sample of the data that USC would be required to produce under the magistrate judge’s Order.
The parties have submitted the requested docurñentation. Additionally, in the interim, the court in the Southern District of Florida has denied the summary judgment motion and set a pretrial schedule in the underlying civil action. (January 13, 1999 Order in Case No. 96-1781-CIV-DIMITROULEAS (S.D.Fla.).) The Court therefore proceeds to consider USC’s motion for reconsideration of the magistrate judge’s ruling.
II. Discussion
A. Standard of Review of Magistrate Judge’s Findings
Under 28 U.S.C. § 636(b)(1)(A), a district court may reconsider a magistrate judge’s determination of non-dispositive pretrial matters if the magistrate’s order is “clearly erroneous or contrary to law.” Similarly, Federal Rule of Civil Procedure 72(a) provides that magistrate judges’ rulings on nondispositive motions may be set aside or modified by the district court only if found to be “clearly erroneous” or “contrary to law.” Fed.R.Civ.P. 72(a); see also Grimes v. City and County of San Francisco, 961 F.2d 236, 241 (9th Cir.1991).1 The “clearly erroneous” standard applies to the magistrate judge’s findings of fact; legal conclusions are freely reviewable de novo to determine whether they are contrary to law. See United States v. McConney, 728 F.2d 1195, 1200-01 (9th Cir.1984), overruled on other grounds, Estate of Merchant v. C.I.R., 947 F.2d 1390 (9th Cir.1991). To conclude that a magistrate judge’s decision is clearly erroneous, the District Court must arrive at a “definite and firm conviction that a mistake has been committed.” Federal Sav. & Loan Ins. Corp. v. Commonwealth Land Title Ins. Co., 130 F.R.D. 507 (D.D.C.1990).
The standard of review with respect to discovery requests presents a special ease, however. Where the magistrate’s decision concerns an evidentiary question of relevance, “the Court must review the magistrate’s order with an eye toward the broad standard of relevance in the discovery context. Thus, the standard of review in most instances is not the explicit statutory standard, but the clearly implicit standard of abuse of discretion.” Geophysical Sys. Corp. v. Raytheon Co., Inc., 117 F.R.D. 646, 647 (C.D.Cal.1987) (reversing magistrate judge’s order denying motion to compel discovery); see also In re Application for an Order for Judicial Assistance in a Foreign Proceeding in the High Court of Justice, Chancery Division, England, 147 F.R.D. 223, 225 (C.D.Cal. 1993) (citing Geophysical, 117 F.R.D. at 647). A judge abuses his discretion “only when his decision is based on an erroneous conclusion of law or where the record contains no evidence on which he rationally could have based that decision.” Premium Serv. Corp. v. Sperry & Hutchinson Co., 511 F.2d 225, 229 (9th Cir.1975).
However, the language of Geophysical Systems implies that the abuse of discretion standard does not apply to a review of all discovery disputes, only those that involve a question of relevance. Geophysical Sys., 117 F.R.D. at 647. Judge Tashima reasoned that it is precisely because the issues at hand were issues of relevance traditionally left to the discretion of the trial court that the district court should employ an implicit standard of abuse of discretion with respect to the magistrate judge’s finding. By implication, when a court reviews a discovery dispute that is not concerned with relevance, the abuse of discretion standard does not apply. Here, the Court is not confronted with a factual question of relevance. Rather, the magistrate judge’s decision hinged primarily upon which choice of law rules and what privilege law the court should apply and the scope of the privilege set forth [423]*423under the applicable state law. Because the Court finds that there is no issue of relevance at stake in this case, we decline to employ the implicit abuse of discretion standard. Instead, the Court looks to determine whether or not the magistrate judge’s determination was clearly erroneous or contrary to law.
B. Choice of Law Issues
Federal Rule of Evidence 501 provides that “in civil actions and proceedings, with respect to an element of a claim or defense as to which State law supplies the rule of decision, the privilege of a witness ... shall be determined in accordance with State law.” Fed.R.Evid. 501. However, the rule does not specify which state law should be applied. Here, the underlying action has been brought in Florida and Florida law will determine the substantive issues of the ease. However, USC argues that, since this motion to compel disclosure of the data has been brought in California, the Court should apply the law of privilege of California.
Although our Circuit has not addressed this issue, persuasive district court cases indicate that where the court hearing the discovery dispute and the court hearing the underlying action differ, the court hearing the discovery dispute must apply the choice of law rules of its forum. See, e.g., Tartaglia v. Paul Revere Life Ins. Co., 948 F.Supp. 325, 326 (S.D.N.Y.1996) (applying New York choice of law principles where proceeding to compel production was brought in New York, although underlying breach of contract action was pending in Ohio); Roberts v. Carrier Corp., 107 F.R.D. 678, 685-86 (N.D.Ind.1985) (noting that the Seventh Circuit has applied the rule that the court making the discovery decision applies the law of its forum, not the forum of the underlying lawsuit). See also Connolly Data Sys., Inc. v. Victor Technologies, Inc., 114 F.R.D. 89, 91 (S.D.Cal.1987) (“a district court is usually bound by its own privilege law when compelling testimony in a deposition for use in another forum” (but noting that there was also a choice of law clause present in the agreement in question selecting California law as the applicable law)); 3 Weinstein’s Federal Evidence § 501.02[3][c][ii] (2d ed.1998) (noting that courts usually apply the law of the state where the deposition is being taken where the state of the deposition and state of the underlying action differ).2 Therefore the Court applies California choice of law rules, and not those of Florida.
1. California Choice of Law Rules
Under California choice of law rules, the Court must apply a “government interest analysis.” Hurtado v. Superior Court, 11 Cal.3d 574, 579-80, 114 Cal.Rptr. 106, 522 P.2d 666 (1974). Under this approach, courts examine the underlying policies of the state laws involved to determine the respective government interests at stake. Generally speaking the forum will apply its own rule of decision “unless a party litigant timely invokes the law of a foreign state. In such event he must demonstrate that the latter rule of decision will further the interest of the foreign state and therefore that it is an appropriate one for the forum to apply to the case before it.” Hurtado, 11 Cal.3d at 581, 114 Cal.Rptr. 106, 522 P.2d 666 (1974) (citations omitted). Therefore, unless defendant Philip Morris can demonstrate a Florida state interest in this data, California law must apply.
The State of California has a strong interest in the confidentiality of the Fontham Study data. Approximately eighty percent of the study participants were women residing in California; none of the study participants resided in Florida. (USC’s Mot. at 9.) The database employed includes data eollect[424]*424ed by the California Department of Health Services,3 and the registry of data concerning cancer patients maintained by USC is subject to the confidentiality provisions of the California Health and Safety Code. California therefore has a strong governmental interest in the application of its rule of law. Accordingly, the Court agrees with USC that, under California choice of law principles, California privilege law should apply.
2. Florida choice of law rules
Even if the Court’s analysis is incorrect and Florida conflict of laws rules apply, such an analysis also leads to an application of California privilege law. Florida applies the Restatement (Second) of Conflict of Laws standard in tort cases as set forth in Section 139(2): “Evidence that is privileged under the local law of the state which has the most significant relationship with the communication but which is not privileged under the local law of the forum will be admitted unless there is some special reason why the forum policy favoring admission should not be given effect.” Restatement (Second of Conflict of Laws), § 139(2) (1989). The state with the most significant relationship is generally the one' where the communication in question took place, here, California. Id. at cmt. e. Here, since Florida is the forum state, and there is no privilege concerning this information in Florida, the information must be admitted unless this policy is outweighed by countervailing considerations. “Among the factors the forum should consider in determining whether or not to admit the evidence are: (1) the number and nature of contacts between the forum state and the parties and the transaction involved, (2) the relative materiality of the evidence that is sought to be excluded, (3) the kind of privilege involved, and (4) fairness to the parties.” Anas v. Blecker, 141 F.R.D. 530, 532 (M.D.Fla.1992) (citing Restatement (Second) of Conflict of Laws, § 139 cmt. d).
In Anas v. Blecker, the Florida district court addressed the question of whether plaintiff should be allowed to depose a representative of the Appraisal Institute of Illinois and to compel the production of confidential Institute files and records. The underlying ease was a Florida state law fraud action. The district court found that it would be against public policy to apply Florida law in deciding issues relating to the Illinois institution’s internal rules and policies. Anas, 141 F.R.D. at 532. The Court also found that if the privilege from the other forum was relied upon by persons involved, the privilege should bear more weight, particularly where the person asserting the privilege is a non-party. Id. at 533 (“If it is a privilege asserted by a nonparty, the forum ‘will be more inclined to recognize the privilege and to exclude the evidence.’ ”). Therefore the Court applied Illinois privilege law.
As in Anas, the Court here considers the actions of a California entity subject to California regulations. The statements made to USC and the Department of Health were made in reliance upon the confidentiality provisions of the study and of California law. And finally, the party asserting the privilege here is a nonparty to the action. Therefore, even under Florida choice of law rules, California law of privilege should still apply.
The Court finds that, to the extent the magistrate judge relied upon Florida privilege law to resolve this dispute, this choice was contrary to law. Under either a California choice of law or Florida choice of law analysis, California privilege law should apply. The Court thus proceeds to a contemplation of the applicable privilege law in California.
C. The Law of Privilege in California
The magistrate judge found that Florida privilege laws should apply and that because there is no cognizable privilege under Florida privilege law, USC could be compelled to release the data. As addressed above, the Court finds that this conclusion was contrary to law. However, the magistrate judge also found that even if California law applied, the information sought by Philip Morris is not privileged. (Transcript of Proceedings Be[425]*425fore the Honorable R.J. Groh, Jr., Sept. 9, 1998 (“Transcript”) at 000016-000020.)
USC argues that the information in this case is governed by California Health and Safety Code sections 100330 and 103885. Section 103885 provides for the establishment of a statewide cancer registry, which is subject to the confidentiality provisions of section 100330. Cal. Health & Safety Code § 103885 (1998).4 The data in question here was obtained in part through this cancer registry.
Sections 100330 and 103885 must be read in conjunction with section 1040 of the California Evidence Code. Section 1040 provides:
(b) A public entity has a privilege to refuse to disclose official information, and to prevent another from disclosing official information, if the privilege is claimed by a person authorized by the public entity to do so and:
(1) Disclosure is forbidden by an act of the Congress of the United States or a statute of this state; or
(2) Disclosure of the information is against the public interest because there is a necessity for preserving the confidentiality of the information that outweighs the necessity for disclosure in the interest of justice; but no privilege may be claimed under this paragraph if any person authorized to do so has consented that the information be disclosed in the proceeding. In determining whether disclosure of the information is against the public interest, the interest of the public entity as a party in the outcome of the proceeding may not be considered.
Cal.Evid.Code § 1040(b) (1998). The section has been read to establish two different privileges: an absolute privilege if disclosure is forbidden by a federal or state statute, and a conditional privilege in all other cases where the court determines that disclosure is against the public interest. Shepherd v. Superior Court, 17 Cal.3d 107, 123, 130 Cal. Rptr. 257, 550 P.2d 161 (1976). USC contends that the information sought by Philip Morris is protected by an absolute privilege as established by state statute, specifically California Health and Safety Code § 100330; alternatively, USC argues that disclosure is against the public interest.
1. whether USC may assert the privilege
Although Dr. Wu and USC are not public employees, the information collected in this study was obtained not only by Dr. Wu and USC, but also by the California Department of Health Services. The Fontham Study information originated with the California Tumor Registry, a public institution, and USC acted jointly with DHS in obtaining the data. Therefore, the Fontham Study raw data is arguably official information. Additionally, the privilege of Section 1040 may be invoked either by the public entity itself or its authorized representative. Cal. Evid.Code § 1040 (1995 Comments). Finally, California Attorney General’s office, on behalf of DHS, has chosen to claim the statutory privilege, prohibiting USC from complying with the subpoena originally served on USC by disclosing records of interviews, written reports, and statements obtained in the Fontham Study. (Hertz Decl.) Under these circumstances, the Court has little difficulty concluding that the privilege of section 1040 has been properly invoked.
2. California Health and Safety Code section 100330 X
Section 100330, which USC invokes as the basis for its assertion of absolute privilege, states in relevant part:
[426]*426All records of interviews, written reports, and statements procured by the department or by any other person, agency, or organization acting jointly with the department, in connection with special morbidity and mortality studies shall be confidential insofar as the identity of the individual patient is concerned and shall be used solely for the purposes of the study.
Cal. Health & Safety Code § 100330 (1998). Just what the California Legislature meant by both the terms “confidentiality” of information and the “purposes of the study” is not clear. The Legislature gave no indication of whether it intended to bar disclosure completely for any purposes other than use in covered scientific studies, or whether the data might be sought during the course of litigation so long as confidentiality is adequately preserved.
The transcript of the proceedings before the magistrate judge indicates that he believed that “purposes of the study” should be construed broadly, to encompass Philip Morris’ request: “at bottom, the intended use of the data is to determine the accuracy, the completeness and the validity of the study. So it is being used — intended to be used for the purpose of the study.” (Transcript at 000019.) The magistrate judge’s findings rely heavily on the Supreme Court case, Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Daubert explores the standard for the admission of expert scientific testimony under Federal Rule of Evidence 702. The question in Daubert was whether the requirement that a scientific technique must be “generally accepted” in the scientific community, as articulated by the D.C. Circuit in Frye v. United States, 293 F. 1013, 1014 (D.C.Cir.1923), was still applicable after the adoption of the Federal Rules of Evidence. The Supreme Court found that the Frye test was superseded by the Federal Rules. Therefore a trial court must evaluate scientific evidence, not on the basis of whether it is generally accepted in the scientific community, but based upon whether it is relevant and reliable. Daubert, 509 U.S. at 589, 113 S.Ct. 2786. The district court must determine “whether the expert is proposing to testify to (1) scientific knowledge that (2) will assist the trier of fact to understand or determine a fact in issue. This entails a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether that reasoning or methodology properly can be applied to the facts in issue.” Id. at 592-93,113 S.Ct. 2786.
The magistrate judge, extrapolating from the Supreme Court’s opinion, reasoned that since a trial judge is allowed to evaluate the scientific validity of a study, focusing in particular on the principles and methodology used to obtain such evidence, the underlying data of the study must be the subject of discovery. (Transcript at 000018-19.)5 However, Daubert does not reach the question of what information is discoverable for the purposes of evaluating other proffered scientific testimony or evidence. It only speaks to the standard that district courts must use in order to determine whether scientific evidence is admissible. In short, Daubert is not dispositive of whether the raw data requested here by defendant Philip Morris is protected by a state law privilege.
The purpose of the statewide cancer registry pursuant to California Health and Safety Code section 103885 is to “establish a statewide system for the collection of information determining the incidence of cancer.” [427]*427Cal. Health & Safety Code § 103885(a) (1998). Such information is intended to advance the study of causes of cancer and to promote medical innovations regarding its cure and prevention. Id. at § 103885(g) (“the department and any regional cancer registry designated by the department shall use the information to determine the sources of malignant neoplasms and evaluate measures designed to eliminate, alleviate, or ameliorate their effect.”) The plain language of the statute indicates that the contemplated use of the raw data for the Fontham Study by Philip Morris in civil litigation is not strictly within the “purposes of the study.”
However, this does not end the Court’s inquiry. First, section 100330 indicates that the records in question must be “confidential insofar as the identity of the individual patient is concerned.” This language supports the magistrate judge’s findings that it is the identity of the subjects that must be kept confidential, and not necessarily all of the information gathered. The magistrate judge’s order provides that “names and addresses of the subjects of the study shall be redacted from the documents produced, and defendant is expressly prohibited from attempting in any way, directly or indirectly, to ascertain the identity of any such subject.” (Sept. 9, 1998 Order at 112.) Thus it seeks to protect the confidentiality of individual subjects.
Furthermore, the statutory declaration of confidentiality does not create an absolute bar on disclosure in judicial proceedings. In Davies v. Superior Court the California Supreme Court considered the intersection between confidentiality safeguards and the liberal requirements of litigation-related discovery. Davies v. Superior Court, 36 Cal.3d 291, 204 Cal.Rptr. 154, 682 P.2d 349 (1984). The petitioner in Davies sought traffic collision reports of accidents and certain collision diagrams prepared in conjunction with California Highway Patrol investigations of accidents occurring in the same location. Davies, 36 Cal.3d at 295, 204 Cal. Rptr. 154, 682 P.2d 349. The State refused to produce these documents on the grounds that accident reports of the California Highway Patrol are confidential under the relevant statute,6 and the diagrams requested encompassed information from those reports. The California Supreme Court emphasized that confidentiality concerns and extensive pretrial discovery must be accommodated so as to further both purposes to the extent possible. Id. at 299, 204 Cal.Rptr. 154, 682 P.2d 349. “In construing a statute which accords confidentiality to information sought by discovery liberality in allowing discovery is the rule unless statutory or public policy considerations clearly prohibit it.” Id. at 300, 204 Cal.Rptr. 154, 682 P.2d 349. Courts must balance the purposes of liberal discovery rules, which include expedition of litigation and efficacious fact-finding, against the purpose of confidentiality protections, which is to encourage persons involved to give a full and accurate account.
The court in Davies ultimately found that the statute did not provide that data generated from accident reports should be kept confidential, because important public policy concerns supported revelation of information which might reveal a defective condition of public property. Id. at 300, 204 Cal.Rptr. 154, 682 P.2d 349. The court observed that: [428]*428“this data easily can be made available without indication of the identity of the reporting person or persons and can be disclosed without compromising the confidentiality of identifying information in the reports themselves.” Id.
This is not a situation identical to Davies, insofar as Philip Morris seeks the raw data itself. However, the language of the court in Davies indicates that California courts consider a confidentiality-based privilege to not be an absolute bar to the discovery of such information.7 Since the existence of a confidentiality provision is not an absolute privilege, then, the Court turns to the second type of privilege envisioned by section 1040: the conditional privilege.
3. conditional privilege
Section 1040(b)(2) of the California Evidence Code requires courts to limit disclosure only when “disclosure of the information is against the public interest because there is a necessity for preserving the confidentiality of the information that outweighs the necessity for disclosure in the interest of justice.” Cal.Evid.Code § 1040(b)(2).
The magistrate judge determined that: “the necessity for preserving the confidentiality does not outweigh the necessity for the disclosure of the information for purposes of this litigation, particularly under the conditions in which it is going to be granted, one of which, of course, also is that it is to be used solely for purposes of this instant litigation.” (Transcript at 000024.) Second, the magistrate judge found that there was a strong public interest in scientific reliability and in subjecting scientific studies to critical inquiry.
The Court agrees with the magistrate judge’s findings. Section 100330 explicitly states that nothing in that section shall prohibit the publishing of compilations of data that do not identify individual eases and sources of information. Cal. Health & Safety Code § 100330. Thus, so long as the proper privacy concerns are met and data is disclosed without revealing a subject’s identity, the statute’s mandate for confidentiality has been met. Nor should production of the data under the terms of the magistrate judge’s protective order serve to chill cancer patients’ willingness to provide information crucial to ongoing mortality studies.
The Court’s conclusion that a statement of confidentiality does not automatically preclude such information from discovery is in harmony with Ninth Circuit case law. In Zambrano v. INS, for example, the Ninth Circuit observed that there are three types of confidentiality provisions in federal law: those which expressly bar disclosure in legal proceedings, those which expressly allow for such disclosure, and those which are silent on the issue of disclosure. Zambrano v. INS, 972 F.2d 1122, 1125 (9th Cir.1992), rev’d on other grounds, Reno v. Catholic Social Servs., 509 U.S. 43, 113 S.Ct. 2485, 125 L.Ed.2d 38 (1993). The court noted that general statutory prohibitions against disclosure of information do not bar judicial discovery “absent an express prohibition against such disclosure.” Id. (citing St. Regis Paper Co. v. United States, 368 U.S. 208, 82 S.Ct. 289, 7 L.Ed.2d 240 (1961)). In the absence of express prohibitions against production in a legal proceeding, confidentiality provisions, at least in federal statutes, should be narrowly construed.
In addition, the California legislature has elsewhere adopted specific safeguards protecting confidential medical information from discovery. For example, information concerning persons with AIDS is not discoverable in any administrative, civil, or criminal proceeding. See Cal. Health & Safety Code §§ 120975, 121025 (1998).8 Sections 100330 and 103885 have no such explicit prohibitions.
[429]*429Finally, this case presents a strong need for disclosure “in the interest of justice.” Obtaining this information may assist defendant Philip Morris in determining the validity of the Fontham Study results and thereby will allow for a thorough inquiry into the underpinnings of the scientific evidence presented at trial. Circuits that have explicitly considered the application of confidentiality provisions to scientific research have reached a similar result. In Deitchman v. E.R. Squibb & Sons, Inc., 740 F.2d 556 (7th Cir. 1984), the Seventh Circuit found that, in order for the pharmaceutical company requesting scientific research information to analyze the accuracy and methodology of the study in question, access to the underlying registry data was essential. Id. at 563. “[Wjhere substantial need is shown, even confidential matter may be revealed.” Id. at 565. Likewise, the Second Circuit, in a case quite similar to the one before the Court, found that the district court properly denied a motion to quash a subpoena requiring a third-party research institution to supply data underlying a published study regarding exposure to asbestos and smoking. In re American Tobacco Co., 880 F.2d 1520 (2d Cir.1989). The Second Circuit observed that, even if New York law did recognize a qualified privilege, the district court could allow disclosure of the data under an appropriate protective order. American Tobacco Co. 880 F.2d at 1531. The Seventh and Second Circuits have indicated that the balance between patient and subject confidentiality concerns and the need for public disclosure of data for purposes of discovery may tip in favor of disclosure.
Thus the Court finds that section 100330 of the California Health and Safety Code does not, in fact, absolutely prohibit disclosure of the information requested. In addition, the magistrate judge, in conducting the public interest analysis required by California Evidence Code section 1040(b)(2), determined that the balance tipped in favor of disclosure, with the appropriate protections in place. The Court finds that the magistrate judge’s conclusion was not clearly erroneous. Accordingly, it declines to reconsider or set aside the magistrate judge’s ruling on this ground.
D. Privacy Protections under the California Constitution
Finally, USC argues that confidential personal and medical information is protected by Article 1, Section 1 of the California Constitution and therefore Philip Morris is required to show a compelling need for the data sought.
Article 1 of the California Constitution contains an explicit protection of individual privacy: “All people are by nature free and independent and have inalienable rights. Among these are enjoying and defending life and liberty, acquiring, possessing, and protecting property, and pursuing and obtaining safety, happiness, and privacy.” Cal. Const, art. 1, sec. 1. California courts have observed that “[a] person’s medical profile is an area of privacy infinitely more intimate, more personal in quality and nature than many areas already judicially recognized and protected.” Board of Med. Quality Assurance v. Gherardini, 93 Cal. App.3d 669, 677, 156 Cal.Rptr. 55 (1979). However, this privacy right also is not absolute. Binder v. Superior Ct., 196 Cal.App.3d 893, 242 Cal.Rptr. 231 (1987) (“Even though the inalienable right of privacy guaranteed by the California Constitution is a fundamental interest of our society, other state interests such as facilitating the ascertainment of truth pertaining to legal proceedings compete with this right.”) Where the constitu[430]*430tional right of privacy is involved, the party seeking discovery of the private matter must demonstrate a compelling need for discovery so strong as to outweigh the privacy right when these two competing interests are carefully balanced. Lantz v. Superior CL, 28 Cal.App.4th 1839, 1854, 34 Cal.Rptr.2d 358 (1994). If an intrusion on the right of privacy guaranteed by the California Constitution is deemed necessary in any particular instance, it should be structured so as to be minimally intrusive while still achieving its objective. Id. at 1855, 34 Cal.Rptr.2d 358 (referring to Wood v. Superior Ct. 166 Cal. App.3d 1138, 212 Cal.Rptr. 811 (1985)).
The magistrate judge succinctly found that “the patient’s interest in the confidentiality of this information is outweighed by what I consider to be the compelling need for this highly relevant information.” (Transcript at 000025.) The magistrate judge indicated that he had already reviewed the competing confidentiality and public utility concerns in his discussion of California privilege law. The Court will not disturb this factual finding that defendant Philip Morris’ need is compelling, since the magistrate judge’s finding is not clearly erroneous.
E. Superior Court Action
Finally, USC draws the Court’s attention to a Superior Court order of July 31,1997, in which the court found that “Health and Safety Code Section 100330 and federal laws forbid the disclosure of the sought after raw data. Further disclosure is contrary to state law.” (Request for Judicial Notice, Ex. 4 (Super.Ct. Order Case No. 987453, filed July 31, 1997).) This order was filed in response to a subpoena from Philip Morris for production of the Fontham Study raw data that was served upon the California Department of Health Services. The court found that not only was production barred by the section 100330, but also redaction of names was insufficient to protect the confidentiality of study participants. Id. Also, the California Court of Appeal has previously ordered the Superior Court to quash an identical subpoena in a different action; however, in that instance, Philip Morris simply withdrew its subpoena. (Request for Judicial Notice, Exs. 7 & 8.)
These decisions are not binding on the Court. USC is correct that a federal court, in interpreting state law, is bound by the decision of the state’s highest court. In the absence of such a decision, the district court must “predict how the highest state court would decide the issue using intermediate appellate court decisions, decisions from other jurisdictions, statutes, treatises, and restatements as guidance... .where there is no convincing evidence that the state supreme court would decide differently, a federal court is obligated to follow the decisions of the state’s intermediate appellate courts.” Nelson v. City of Irvine, 143 F.3d 1196, 1206-07 (9th Cir.1998) (citing In re Bartoni-Corsi Produce Inc., 130 F.3d 857, 861 (9th Cir.1997)). This does not indicate, however, as USC suggests, that this Court must follow the sole state court decision available on this issue. It simply means that the Court must conduct a thorough analysis of all relevant sources of law and attempt to approximate the decision of the California Supreme Court. After conducting such an analysis, the Court still finds that the magistrate judge’s Order should stand.
USC also raises the concern that, even under the magistrate judge’s order, certain subjects would be identifiable. The Court finds that the magistrate judge’s requirement that Philip Morris make no attempt to discover the identity of the subjects in question is sufficient to address these concerns.
III. Conclusion
For the reasons given above, USC’s motion to reconsider the magistrate judge’s order granting defendant’s motion to compel production of documents is denied. USC is ordered to comply with the magistrate judge’s order within 20 days of entry of this Order, subject to all protective measures imposed by the magistrate judge. This Order shall constitute the Court’s final ruling on defendant’s motion. Thus, the Clerk shall close this file.
IT IS SO ORDERED.