We write this opinion to explain our reasons for denying defendant General Motors Corporation’s motion for a protective order. *476This is a products liability case stemming from an automobile accident in which the plaintiff, Debra Buon-giovanni, was severely injured. Plaintiff alleges that the seat back of her General Motors vehicle was defective, causing her injury. Other “seat back failure” claims have been filed and litigated against General Motors in recent years.
Buongiovanni has served General Motors with several requests for production of documents, including requests for studies that General Motors’ engineers have conducted regarding the design and safety of seats in rear end impacts. Specifically, the plaintiff requested: “documents reviewed and/or generated by General Motors employees concerning their engineering analysis of the circumstances surrounding seat back litigation brought against GM. [This request does not include communications prepared by GM counsel.]”
General Motors filed this motion for a protective order, asserting that those documents, which General Motors refers to as the “litigation study,” are not discoverable as they are protected by the attorney-client privilege. We denied the motion for a protective order because the documents as a whole are not protected by the attorney-client privilege. However, selected documents may be protected on an item-by-item basis.
The Pennsylvania Rules of Civil Procedure define the scope of discovery as “any matter not privileged which is relevant to the subject matter involved in a pending action, whether it relates to the claim or defense of any other party.” Pa.R.C.P. 4003.1. The scope of discovery in Pennsylvania is broad and the limitations placed on discovery are to be narrowly construed. Szarmack v. Welch, 456 Pa. 293, 318 A.2d 707 (1974).
Generally, Pennsylvania law allows the discovery of relevant information even though it was prepared in anticipation of litigation. Pa.R.C.P. 4003.3. However, an exception to this rule is that “the mental impressions of a party’s attorney or his conclusions, opinions, memo-randa, notes or summaries, legal research or legal theo*477ries” are not discoverable. Id. This is commonly known as the “work product” exception.
The General Motors documents in question do not as a whole fall under the work product exception. General Motors, through its legal staff attorney, Drema M. Kopcak, asserts that the “litigation study” was prepared by GM engineers at the request of GM counsel. However, General Motors engineer Ted Bertacini provided the following sworn deposition testimony regarding the documents at issue in another seat back case:
“Q: Let’s go through this. First, the typewritten information it says on the top seat safety research & development team, and then it has purpose. Then underneath the purpose also in typewritten words it says quote ‘to develop recommendations to the seating technical committee regarding seat performance in rear impact.’ Did I read that correctly?
A: Correct. . . .” (Dep. p. 251.)
“Q: In your handwritten note it says and I quote ‘the issue of seat back strength in rear impact has been coming up more and more over the past year.’ Can you tell us what you meant by that?
A: I meant that from different areas of the corporation, the issue was being brought up both from the legal staff as well as from different engineers who were voicing an opinion that it may be appropriate to revisit this issue. So it was coming from at least a couple of different areas . . . .” (Dep. p. 253.)
“Q: Do you know what their concerns with regarding the issue, seat back strength during rear end impacts?
A: I think the general consensus was to evaluate or reevaluate how current seats at the time were doing.
Q: During rear end impacts?
A: During rear end impacts.
Q: Out in the field?
A: Out in the field. . . .” (Dep. p. 254.)
“Q: If there was no major problem then what was the reason for investigating the issue of seat back strength?
*478A: The issue was twofold. One, to assist with a rationale for litigation defense, and two, and probably more important to the engineers, to reassess whether or not the current strength level was performing adequately. And if not, whether or not reasonable increases in strength would be required. . . .
Q: What was the purpose of going the next step to try and reevaluate to see if a different design or if a new design standard would be prudent?
A: One, we were being asked to at least document an engineering rationale for our current design practice, and there were others, other than myself, that thought that we should, that also wanted to look at the data to decide as a group whether or not reasonable or large increases might be required. . . .” (Dep. p. 293.)
“Q: I apologize, this is a little difficult to read, but this is how we received it from General Motors in the production of documents in this case. The date of this document is February 20, 1991, correct?
A: I believe so.
Q: It is from Drema Kopcak, correct?
A: Correct.
Q: And it is to you, correct?
A: I’m one of the addressees, yes.
Q: The other addressee is Karen Wehrly, correct?
A: Correct.
Q: I think you already told us that Karen Wehrly was working with the engineering analysis group?
A: Yes, she was.
Q: That was a group that was helping legal counsel both inside and outside General Motors in defending cases?
A: Correct.
Q: The subject of the document is seat back litigation, correct?
A: Correct.
Q: Let’s look at the first paragraph. It says at our January meeting of the seating safety research development team, the team asked legal staff to review recent litigation related to seat back performance in rear im*479pacts. With your assistance, referring to you and Miss Wehrly, we provided an initial review to the committee on February 14. At that meeting, the team asked for further analysis. Did I read that correctly?
A: I believe you did.
Q: The meeting that was being referred to was this January 14, 1991 meeting, correct?
A: I believe so.
Q: It doesn’t say in this correspondence from Miss Kopcak to you, Mr. Bertacini, that she came to the team and requested this study, does it, in the paragraph I just read?
A: No, it does not.
Q: It says the team asked the legal staff to review recent litigation related to seat back performance in rear impacts, doesn’t it?
A: Yes, it does.
Q: Does that refresh your memory?
A: Yes, it does.
Q: What is your testimony?
MR. BURKHOLDER: Pardon me. Testimony with regard to what?
MR. LEOPOLD: If that refreshes his memory, I want to know what his memory is with respect to who requested the litigation study to be performed, whether it was the team or Drema Kopcak?
THE WITNESS: The initial presentation was made by Drema Kopcak to the seating technical committee. The seating technical committee assigned the research & development team to investigate or address the issue. Drema came to the meeting and the seat safety research & development team thought that it may provide, shed some light on how seats are performing in the field to look at litigation data. And, therefore, we asked Drema if she could go back and perform such a study. And she brought apparently her initial review to the February 14 meeting of the seat safety research & development team. Apparently the team believed that further analysis was required. . . .” (Dep. p. 308.)
*480It is clear from the testimony given at this deposition that the seat back study was done primarily for the purpose of improving the safety of General Motors vehicles’ seat backs, rather than in preparation for litigation. If, however, specific documents from within the study are either shown to be solely the work product of GM attorneys, or subject to attorney-client privilege, those will not be discoverable.
In order for a party properly to claim attorney-client privilege, each document must be specifically designated and described and precise reasons must be given for preserving the confidentiality of each document. Int’l Paper Co. v. Fibreboard Corp., 63 F.R.D. 88 (D.Del. 1974). The fact that a legal department has taken part in an engineering study does not automatically make the entire study undiscoverable work product. See SCM Corp. v. Xerox Corp., 70 F.R.D. 508 (D.Conn. 1976) (“[ljegal departments are not citadels in which public, business or technical information may be placed to defeat discovery and thereby ensure confidentiality.”).
Because the “litigation study” documents are not, as a whole, protected under the work product exception, General Motors’ motion for a protective order is denied.
As one considers the matter, it is a bit curious for General Motors to take the position that its interest in seat safety began with concern for litigation rather than safety itself. For the above reasons, we enter the following:
ORDER
And now, July 10, 1996, the motion for a protective order is denied. The materials are currently in the court’s chambers. Their review will be scheduled by counsel at the convenience of counsel. The court will be present from time to time in the course of selection (initially by defense counsel) to review the bona fides of that selection process and to resolve differences or questions should they arise.