MEMORANDUM OPINION

JOHNSTONE, District Judge.

Since approximately 1981 the Kentucky Department of Corrections has conducted a urinalysis procedure at the Kentucky State Reformatory and the Kentucky State Penitentiary in an attempt to curtail the use and dissemination of illicit drugs in those institutions. Urine specimens are periodically collected from certain inmates and tested for the presence of drugs with the Syva Emit drug detection system. If the EMIT test indicates that an inmate has drugs in his system, then that inmate is charged with a violation of Category IV, Item 2 of the Correction Department's Infraction and Penalty Code, "being under the influence of drugs or intoxicants," and is subject to a maximum punishment of 45 days in disciplinary segregation and a loss of 60 days good time credits. This action challenges the use of the EMIT test as the sole evidence of an individual's use of contraband drugs.

The action originated as a motion for contempt under the Consent Decree entered in Kendrick v. Bland, 541 F. Supp. 21 (1981). It was consolidated with an individual action filed by James Carl Higgs and other inmates at Kentucky State Penitentiary. Civ. Action No. 83-0265-P (J). All parties demanded a preliminary injunction to enjoin the use of the Syva Emit test to punish inmates for the use of drugs. They contend that that practice constitutes a violation of the constitutional guarantee against loss of liberty without due process of law.

The court referred the action to United States Magistrate W. David King for his report and recommendation. He conducted an evidentiary hearing on the motion for preliminary injunction on May 10, 1984, in Louisville, Kentucky. The hearing continued on May 16 and 17, 1984 at the Kentucky State Penitentiary in Eddyville, Kentucky. On July 30, 1984, the magistrate entered his report on the preliminary injunction. He recommended that the court issue a preliminary injunction providing that:

1. The Syva Emit drug detection system not be used as the sole indication of intoxication in a `Category 4, Item 2 Being under the influence of drugs, intoxicants' proceeding before the adjustment committee. . . .

2. In any instance in which the results of the Syva Emit urine screen is used to support an incident report charging an inmate with being under the influence of drugs or intoxicants; the Department of Corrections shall: (a) implement reasonable procedures in the urine donor's presence to assure the integrity and preserve the freshness of the sample; (b) implement reasonable procedures to record the chain of custody for the urine sample from the time the sample is taken until it returns to the inmate's institution after examination; (c) allow the charged inmate to cross-examine the laboratory technician concerning the validity of the test results either in person or through written interrogatories.

The court remanded the matter to the magistrate on October 1, 1984, for more detailed findings. The second, supplemental report, entered on February 22, 1985, recommended that the same injunction be issued and included additional factual findings. This court, on May 2, 1985, accepted the magistrate's recommendation and granted plaintiff's motion for a preliminary injunction. It enjoined defendants from taking any disciplinary action against plaintiffs based solely upon unconfirmed results of Syva Emit drug detection tests, and stated that a formal finding of fact and conclusion of law would follow its order entering the preliminary injunction. These are those findings and conclusions as promised by the court.

FINDINGS OF FACT The Test Itself

The urinalysis test employed by Corrections is called the enzyme multiplied immunoassay technique (EMIT). It works on a biochemical principle. An antibody which reacts to drugs by binding is added to the urine. The reactions which result produce a substance which can be compared to known values through a device called a photometer which indicates whether drugs are present in the urine. Different combinations of regents are used to produce the substance depending upon which drug is sought. The test measures a by-product of the reaction of the urine. Record of Evidentiary Hearing, May 10, 1984, at 49-52. Thus, the EMIT test does not measure the amount of drugs in the urine directly, but measures instead the reaction of an enzyme to the drug. Id. at 62.

Plaintiff's expert, Thomas A. Regent, Chief toxicologist for Erie County in Buffalo, New York, testified that the EMIT test, without use of a confirmation test, is not sufficiently reliable to be used to identify drugs if punitive action will result from their identification. Id. at 61. He gave several reasons for his opinion. Since the EMIT test does not measure the drug itself, positive tests should be confirmed with a second test to ascertain that urine does indeed contain the drug and that the enzyme reaction was not caused by some other element. Id. at 62. The confirmation test could be gas chromatography or mass spectrometry. Id. at 69. Moreover, various elements in high quantity may cause cross-reactions, the binding of an antibody to legal compounds with a structure similar to illegal compounds. Cross-reactions may create falsely positive results. Drugs such as aspirin, darvon, amitriptylene, methaqualone, and valium are drugs which could cause cross-reactions. Id. at 83.

The primary advantages of the EMIT test are that it is quick, it is relatively inexpensive, and it can be operated by people who are not scientific experts. It will also detect drugs in very low concentrations. Id. at 61, 163. Standing alone, without a confirmation test, it is sufficiently reliable for conducting a statistical survey of a population group to discover whether, for example, teenagers in Ohio smoke marijuana, because absolute accuracy is not necessary. However, "when one is dealing with one's employment, someone's health, someone's goodwill, mental state and so on, any punitive action which takes place, . . . confirmation is necessary." Id.

The literature published by the Syva company which produces and markets the EMIT test recommends back-up test of positive tests. A March 1983 booklet published by the manufacturers of the test, entitled Frequently Asked Questions About Syva and Drug Abuse Testing provides, "[i]t is good scientific practice to confirm a positive result from any test method in cases where a person's rights, privileges, treatment or employment is at stake." Plaintiffs' Exhibit 2 at 6. Defendant's own expert, Ernest Slightom, testified that he was aware of no changes in the scientific procedures for administering the test that would cause it to be more accurate than it was in March, 1983. Id. at 170. He also conceded that the September 16, 1983 issue of the Morbidity and Mortality Weekly Report states that "[r]egardless of which assay kit is used, [EMIT] test results should be interpreted by qualified personnel and positive results should be verified so that there is a very limited possibility of false positive results." Id. at 173.

Slightom testified that the EMIT test is reliable without a back-up test. However, his disagreement with Rejent was one only of degree. He admitted that in March, 1984, he testified at another hearing that the EMIT test alone is not reliable, that any single test should be backed-up with a second test. Id. at 180. This inconsistency is explained by the fact that Slightom believes that a positive test should be confirmed if the individual tested is not incarcerated. However, confirmation is not necessary if the individual is in prison or on parole. Id. at 130-131.

The Procedure

At Kentucky State Reformatory, inmates who give urine samples are chosen in several ways. A totally random sample is prepared by, for example, choosing every fourth inmate on a list. Also, inmates suspected of drug use are often selected by unit managers or shift captains. Specific types of inmates are routinely tested: for example, inmates returning from furlough, inmates residing in a housing unit where drug use is suspected, and inmates who have submitted a positive test in the past. Id. at 218-19. Kentucky State Penitentiary uses a similar method to select inmates. It keeps a drug abuse list of individuals who are "high risk drug offenders" by their background or by their institutional use of drugs. Id. at 277. In addition, the senior captain may choose a random group of inmates to be tested. Id.

The procedure used to collect urine specimens from the chosen inmates is substantially the same at the two institutions. The inmates receive a specimen cup labeled with their name and number. They are observed while giving their specimens by a correctional officer. The sample is covered and sealed with evidence tape and the identification is checked, all in the inmates' presence. The appropriate chain-of-custody forms are then filled out and the sample is logged. At Kentucky State Reformatory the samples are placed in a locked metal container in the captain's office and immediately sent to Luther Luckett Correctional Complex where they are locked in a refrigerator for later testing. Id., at 220. At Kentucky State Penitentiary, the samples are stored in a styrofoam transport box inside of a refrigerator in the laboratory. The laboratory is forbidden to inmates unless they are doing janitorial work, and then they are directly supervised while they are there. Id. at 360.

If an inmate at either prison has a positive urine test, then he is charged with violation of Category IV, Item 2, use of drugs or other intoxicants. The positive test is alone sufficient to convict an inmate of the infraction. No other evidence is necessary. Punishment at Kentucky State Penitentiary generally entails 45 days segregation time and 60 days loss of good time credits. Id. at 282-83. In addition, the effects of a conviction might include: denial of parole, loss of meritorious housing, denial of furlough, and denial of transfer to a less secure institution. Punishment of Reformatory inmates is of like kind, except that they are less likely to receive segregation time.

According to formal guidelines at both institutions, the positive test and chain-of-custody form should be attached to the misconduct report issued to the charged inmate. Although the Consent Decree requires that inmates be allowed to call any witness whose appearance would not be a threat to the security of the institution, no inmate has ever been allowed to call at the adjustment committee hearing the lab technician who conducts the urinalysis tests to determine whether the test was conducted correctly. Id. at 234-35. No reason is given for the denial as is required by the Consent Decree.

CONCLUSIONS OF LAW

In considering whether a preliminary injunction should be issued, a trial court must consider the following four factors:

1. Whether the plaintiffs have shown a strong or substantial likelihood of success on the merits;

2. Whether the plaintiffs have shown irreparable injury;

3. Whether the issuance of a preliminary injunction would cause substantial harm to others; and,

4. Whether the public interest would be served by issuing a preliminary injunction.

Likelihood of Success on the Merits

A disciplinary proceeding which may result in disciplinary segregation and loss of good time must comply with fundamental fairness. Incarceration in a penal institution naturally results in the necessary withdrawal or limitation of many privileges and rights. Pell v. Procunier, 417 U.S. 817, 94 S.Ct. 2800, 41 L.Ed.2d 495. Nevertheless, inmates are not stripped of all of their constitutional rights. They retain the right to minimum fundamental fairness and diminished due process rights. Wolff v. McDonnell, 418 U.S. 539, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974). A part of the due process guarantee is that an individual not suffer punitive action as a result of an inaccurate scientific procedure. See United States v. Brown, 557 F.2d 541 (6th Cir. 1977).

At least two courts have held that the chance of false positives in unconfirmed EMIT test results and the concomitant loss of liberty violates minimum fundamental fairness and the prisoners' due process rights. Johnson v. Walton, unpublished opinion, Docket No. 561-84 Rm (Rutland Superior Court Vermont Feb. 14, 1985); Kane v. Fair, 33 Cr.L. 2492 (Mass.Superior Court, August 5, 1983). In Johnson, the Vermont court found that the authorities presented to it agreed that the EMIT test was not scientifically reliable when used alone, and that another test, mass-spectroscopy, was recognized by all authorities as 100% reliable. The court noted that it did not understand why the Commissioner of Corrections did not use mass-spectroscopy to confirm all positive tests because although that procedure was more expensive, methods should have been available for the inmates to reimburse Corrections. It did not, however, require the defendants to make mass-spectroscopy available. Instead, it found that when EMIT is confirmed by thin-layer chromotography, a positive result indicates presence of the drug with scientific certainty beyond a reasonable doubt. Thus, it required confirmation by either mass spectroscopy or thin-layer chromotography.

The Massachusetts court in Kane v. Fair, supra, made similar findings. It concluded that the state failed to show "that knowledgeable scientists would accept an unsubstantiated EMIT-positive result as evidence of drug use." Therefore, considering the punitive measures that flowed from a positive test and the adverse impact such a test could have on an inmate's chance for parole, the court held that "[n]o positive EMIT test result may be introduced as evidence in any disciplinary hearing unless accompanied by evidence that the positive result was confirmed by an alternative method of analysis."

The evidence presented to this court is similar to that which confronted the Massachusetts and Vermont courts. The plaintiffs' expert was emphatically of the opinion that the EMIT test, standing alone, is not sufficiently reliable to show the presence of drugs if the result was a loss of liberty. Although defendant's expert testified in this hearing to the contrary, he too, in a different hearing only a few months before, stated that the test should be confirmed. Even the manufacturer of the test recommends that positive tests be confirmed by an alternate method of analysis. Moreover, both experts agreed that legal drugs in high doses could interfere with the test and create positive results when there are no drugs in the urine.

The court recognizes that other courts have accepted the EMIT test as sufficiently reliable to stand as the only evidence in a parole revocation hearing, Smith v. State, 250 Ga. 438, 298 S.E.2d 482 (1983), and in a prison disciplinary hearing. Jensen v. Lick, 589 F. Supp. 35 (D.N.D. 1984). Two other courts denied preliminary injunctions to enjoin use of the test, but have not yet tried the merits of the cases. Storms v. Coughlin, 600 F. Supp. 1214 (S.D.N.Y. 1984); Peranzo v. Coughlin, 608 F. Supp. 1504 (S.D.N.Y. 1985). However, in Storms the court refused to grant the preliminary injunction because it found that plaintiffs/prisoners' motion was premature because none of them had yet been disciplined solely on the basis of a positive test result. Storms, 600 F. Supp. at 1225. That is not the case here. The court heard a substantial amount of testimony from plaintiffs who have been disciplined solely on the basis of a positive EMIT test. Peranzo can also be distinguished from the present action. There the New York Corrections officials routinely tested each urine specimen with two EMIT tests. This decreased the risk of false-positive test results. In the present case, the lab technician at Luther Luckett does not confirm the positive tests with any second test.

Therefore, the court finds that the case law and the expert evidence at the hearing indicates a probability of success on the merits. Additionally, the court notes that the testimony included a substantial amount of evidence from inmates that the procedures provided for collecting and preserving the urine specimens were sometimes not followed, and that other inmates had access to the specimens which encouraged tampering with the samples. Thus, the danger of a "clean" inmate receiving disciplinary action for a "dirty sample" is greatly increased, especially since the inmates cannot call the lab technician to testify, and often cannot call other actors in the chain of custody.

Irreparable Injury

The second factor to be considered by the court is whether plaintiffs have shown irreparable injury. Clearly, plaintiffs have suffered and continue to suffer damages from use of the test. Good time credits are forfeited. Inmates are forced to endure disciplinary segregation. Transfer restrictions, loss of privileged housing, and limitations on parole all result from a positive EMIT test. All of these punitive actions stem from the allegedly unconstitutional use of the EMIT test alone to convict inmates of a Code violation, and would not be compensable with monetary damages. When an alleged deprivation of a constitutional right is involved, most courts hold that no further showing of irreparable injury is necessary. See e.g., Planned Parenthood v. Citizens for Community Action, 558 F.2d 861, 867 (8th Cir. 1977); Zepeda v. INS, 753 F.2d 719 (9th Cir. 1983); Doe v. Mundy, 514 F.2d 1179 (7th Cir. 1975). Therefore, the court finds that plaintiffs would suffer irreparable injury if the preliminary injunction was not entered.

Substantial Harm

Defendants argue that issuance of a preliminary injunction will result in substantial harm to defendants. The court does not agree. First, defendants are not prohibited from testing for drugs with the EMIT test. They need only confirm the test with other evidence — either another alternate method test, testimony that the accused was seen with contraband, testimony of symptoms of a person who has used contraband, evidence that drugs were found in his cell, or other demonstrable or circumstantial evidence that the accused inmate used the drug.

Second, the institutions have other methods of combating drug use and trafficking. Warden Rees testified that there is a drug dog which smells out marijuana. Periodically, visitors' vehicles and the visitor's themselves are searched. Corrections officials check in-coming mail and packages for contraband, inmates and their cells are searched, and the financial activities of the inmates are monitored. "It is a multi-faceted approach of which the urinalysis test is only a part." Record 241.

Finally, defendants argue that the cost of confirmation tests would in itself create substantial harm. However, the monetary loss would not be great enough to outweigh the possible constitutional violations.

Service of the Public Interest

In any case where there is a strong likelihood that citizens are being deprived of their constitutional rights, it is presumed that the public interest is served by protecting those rights. Since there are alternative methods of discouraging drugs in the institution and since the defendants are not precluded from using the EMIT test if it is confirmed by other evidence, the court finds that granting the preliminary injunction would not be detrimental to the public interest.

Having considered the preceding four factors, the court determines that it is in the interest of justice to grant a preliminary injunction pending a determination of the issue on its merits. The defendants argue that they have changed the language of a Category IV, Item 2 offense from "being under the influence of drugs" to the "use of drugs;" thus, the injunction is no longer necessary. The court does not agree. The change in language does not affect the reliability of the test or the adequacy of the collection procedures — the reasons for the injunction. The court's decision is based upon the use of the test as the sole evidence presented in a punitive action proceeding whether the proceeding is for an alleged violation of "intoxication" or "use of drugs."

Corrections also maintains that its procedures for collection and preservation of the urine samples are adequate. The court agrees that Corrections' procedures, in theory, protect the integrity of the sample. However, there was sufficient evidence presented at the hearing of irregularities in the procedure to warrant an injunction requiring defendants to implement practices which prevent such irregularities. Therefore, the court rejects defendant's objections and shall maintain the preliminary injunction previously entered.