Storms v. Coughlin, 600 F. Supp. 1214 (S.D.N.Y. 1984)
December 19, 1984
Thomas A. COUGHLIN III, James E. Sullivan, Co. O. Brady, Co. O. Merritt, Unknown C.O.'s, Defendants.
United States District Court, S.D. New York.
*1215 *1216 Donald Storms, John D. Cramer, Richard Freggans, Joseph Clayborne, pro se.
Robert Abrams, Atty. Gen., State of N.Y., New York City, for defendants; Eugene Murphy, Asst. Atty. Gen., New York City, of counsel.
MEMORANDUM OPINION AND ORDER
HAIGHT, District Judge:
In an effort to control drug use among state prisoners, New York State prison officials recently instituted a program of daily tests for traces of narcotics and marijuana in the urine of prisoners assertedly selected at random from the whole prison population. The plaintiffs, four inmates at Ossining Correctional Facility, challenge this practice as violative of their Fourth, Fifth, and Fourteenth Amendment rights. They have asked for a preliminary injunction against its continuance. Defendants, various prison officials, move to dismiss pursuant to Rule 12(b) (6), Fed.R.Civ.P.
An evidentiary hearing was held on September 20, 1984, at which plaintiff Storms appeared both to testify and to represent the pro se plaintiffs. Also testifying was Correctional Officer Jerold Brady, who frequently administers the urinalyses at Ossining. The facts adduced at the hearing are summarized below.
Concerned about high levels of drug abuse at Ossining, officials began in December, 1983 a program of "random" urinalyses designed to detect and thereby deter inmate drug use. The program was authorized by and carried out pursuant to Directive # 4937 of the New York State Department of Correctional Services. Prior to December, 1983, a prisoner's urine was tested for the presence of narcotics only if he was actually suspected of drug use. In December, however, prison officials announced a program of randomly testing each day a few prisoners who showed no signs of drug use. The officials apparently assumed that prisoners would be less likely to use drugs knowing that they might be so tested at any time.
The number and identity of prisoners to be randomly tested is determined daily. Prisoners continued to be tested if they are actually suspected of drug use. Each day the testing officer gives first priority to the testing of these suspected inmates' urine. Once he has completed these tests, the officer determines whether he has time to run some random tests and, if so, how many can be run. He then reports to the watch commander and informs him of the number of random tests which can be run. The watch commander chooses the particular prisoners to be tested by picking cards from a board in his office. As described by Officer Brady, the board holds a group of cards, each of which lists the name of an inmate. Apparently all inmates are represented. Those inmates whose cards are picked off the board are ordered to report for urinalysis.
Perhaps recognizing the potential for conscious or unconscious abuse in this method of selecting prisoners, prison officials are currently taking steps to design a computer program which will select prisoners on a truly random basis. This program was not yet in use at the time of the hearing, nor was its estimated time of introduction specified.
Once the prisoners to be tested are selected, they are individually escorted to a large restroom in the hospital section of the prison. There is some dispute about the size of the restroom and its proximity to the hospital emergency room. In any event, there is no door on the restroom, and the entrance is so positioned that female hospital employees pass by with some regularity on hospital business. According to Officer Brady, before he collects a urine sample he clears the restroom of other *1217 occupants and takes the prisoner to a corner stall which cannot be seen from the door of the restroom. At least one prisoner has submitted evidence to the contrary, an unsworn statement contending that other prisoners are on occasion permitted to watch the process. Further, according to plaintiff Storms' testimony, prisoners are aware of the presence of women employees outside the restroom, although it is unclear whether the prisoners are actually in these employees' view.
Once positioned in the stall, the prisoner is directed to urinate into a plastic bottle held by the officer. If the prisoner refuses to give urine, he is disciplined as though he had used drugs. If he is willing to give urine but unable to do so, he is given water to drink. If after three or four hours he is still unable to produce urine, he is sent for medical treatment. If a specimen is collected, the bottle is sealed, taken to the testing room, and refrigerated to await testing. The manufacturer of the testing machine recommends that urine samples which are not to be tested within one day be frozen. Officer Brady testified that this is standard practice, but plaintiff Clayborne has presented evidence that it is not always done, or at least was not always done in the past.
The machine used for testing is known as the Syva EMIT ST drug detection system. Very little direct evidence bearing on the reliability of the machine was introduced at the hearing, but it has been suggested that the machine does not always produce accurate results. The testing itself is done by a correctional officer. Running the test is apparently a relatively simple process, and the testing officers are thoroughly trained. They also receive periodic refresher courses in testing from the machine's manufacturer.
If the test returns a positive result, the testing officer gives the sample to a second officer, who runs a second test on the same machine. Only if the second test is positive is the prisoner subject to discipline. Once two positive results are obtained, a disciplinary hearing is held at which the prisoner may contest the results of the EMIT machine. Because urine sample is not preserved, however, the prisoner cannot obtain a retest should he wish to challenge the test results in this manner.
* * *
Following the conclusion of the evidentiary hearing I denied the application for a preliminary injunction for reasons stated on the record and herein repeated. Defendants now move to dismiss the complaint.
Construed liberally, plaintiffs' complaint poses two substantial questions: first, whether this type of random testing violates the Fourth Amendment ban on unreasonable searches and, second, whether the imposition of disciplinary sanctions based solely on these test results, after the sample on which the test is based is thrown away, violates the plaintiffs' rights to due process. In moving to dismiss, defendants contend that plaintiffs lack standing to assert some of the objections they raise to the testing procedures and that in any event the procedures are in full compliance with constitutional requirements.
The parties are agreed that a urinalysis test constitutes a Fourth Amendment search. For this limited purpose it is indistinguishable from the blood test held to fall within the Fourth Amendment by Schmerber v. California, 384 U.S. 757, 767-768, 86 *1218 S.Ct. 1826, 1833-35, 16 L. Ed. 2d 908 (1966). Because the two "searches" both involve the forced extraction of bodily fluids granted, by different meansthe mode of analysis employed by the Court in Schmerber seems particularly instructive here.
Defendant Schmerber was involved in an automobile accident. The police officer on the scene noticed that he displayed classic symptoms of alcohol intoxication, and after Schmerber had been hospitalized and treated the officer placed him under arrest and directed that he be tested for intoxication. Schmerber objected to the admission of the results of this test on grounds, inter alia, that the test violated his Fourth Amendment rights. The test required extraction of a small amount of blood, a process the Court characterized as involving, for most people, "virtually no risk, trauma, or pain." 384 U.S. at 771, 86 S. Ct. at 1836. The Court nevertheless differentiated it from the more traditional searches of clothing or possessions which may be freely conducted pursuant to a lawful arrest, holding that because of the test's greater intrusiveness, "[t]he interests in human dignity and privacy which the Fourth Amendment protects forbid [searches involving intrusions beyond the body's surface] on the mere chance that desired evidence might be obtained." Id., at 769-770, 86 S. Ct. at 1835-36. The Court required a higher showing of cause"a clear indication that in fact such evidence will be found"before permitting the search. Id., at 770, 86 S. Ct. at 1835. In analyzing the reasonableness of the warrantless blood test, the Court considered the level of likelihood that intoxication would be found, the exigency created by the metabolism of blood alcohol, the reliability of the test performed, the defendant's response to the process of blood extraction, and the manner of blood extraction. Id., at 770-772, 86 S. Ct. at 1835-37. Finding that the defendant's manifest symptoms of intoxication created a strong likelihood of high blood alcohol levels, that exigency existed because blood alcohol levels decline rapidly following ingestion, that the test was reliable, produced no particular anxiety and violated no scruples of the defendant, and that the blood extraction was performed by a physician in an antiseptic environment, the Court affirmed the test's reasonableness. Id.
I find urinalysis analogous to a blood test. Although it involves no forced penetration of body tissues, as does a blood test, it does involve the involuntary extraction of body fluids. In that sense, if not literally, it is an "intrusion beyond the body's surface." Schmerber, 384 U.S. at 769, 86 S. Ct. at 1835. Further, the "interests in human dignity and privacy," id., at 769-770, 86 S. Ct. at 1835-36, which concerned Justice Brennan in Schmerber are plainly implicated when an inmate is forced to perform in the presence of a prison guard what is ordinarily regarded as a private bodily function. In a way in which having blood extracted could never be, being forced under threat of punishment to urinate into a bottle held by another is purely and simply degrading. Thus, this type of search is entitled to at least the level of scrutiny given blood tests by Schmerber.
That said, it is clear that because they are prison inmates these inmate plaintiffs cannot invoke the "clear indication that in fact ... evidence will be found" standard of cause applied to the motorist in Schmerber. 384 U.S. at 770, 86 S. Ct. at 1835. In a series of cases culminating most recently in Hudson v. Palmer, ___ U.S. ___, 104 S. Ct. 3194, 82 L. Ed. 2d 393 (1984) and Bell v. Wolfish, 441 U.S. 520, 99 S. Ct. 1861, 60 L. Ed. 2d 447 (1979), the Supreme Court has indicated that the constitutional rights which would otherwise be accorded prisoners give way when in conflict with the security needs of the prison. In other words, although convicted prisoners do not park all constitutional protections at the prison gates, Jones v. North Carolina Prisoners' Labor Union, 433 U.S. 119, 129, 97 S. Ct. 2532, 2539, 53 L. Ed. 2d 629 (1977), "[t]he fact of confinement as well as the legitimate goals and policies of the penal institution limits these retained constitutional rights." Bell, supra, 441 U.S. at 546, 99 S. Ct. at 1877. One *1219 of these legitimate goals is, of course, maintaining order and discipline within the prison. Pell v. Procunier, 417 U.S. 817, 823, 94 S. Ct. 2800, 2804, 41 L. Ed. 2d 495 (1974). Therefore, prison administrators must be accorded "wide-ranging deference in the adoption and execution of policies and practices that in their judgment are needed to preserve internal order and discipline and to maintain institutional security." Bell, supra, 441 U.S. at 547, 99 S. Ct. at 1878. As described in more detail below, a primary effect of this limitation of rights is to permit prison officials to perform various types of searches on a much slighter showing of "cause" than would be required before state officials could perform the same searches on citizens who are not incarcerated.
In Bell, the Supreme Court fleshed out the boundaries of that deference by analyzing a series of security-oriented policies maintained by the federal jail in New York City. One of the policies particularly relevant here, upheld by a five-person majority of the Court, dictated that all prisoners undergo a body cavity search following every "contact" visit with outsiders. After removing all of their clothes, the male prisoners were forced to spread their buttocks and lift their genitals for visual inspection; female prisoners similarly exposed their anal and vaginal cavities. The search required no physical contact between guard and prisoner.
In approving the procedure, Justice Rehnquist noted:
The test of reasonableness under the Fourth Amendment is not capable of precise definition or mechanical application. In each case it requires a balancing of the need for the particular search against the invasion of personal rights that the search entails. Courts must consider the scope of the particular intrusion, the manner in which it is conducted, the justification for initiating it, and the place in which it is conducted. (citations). A detention facility is a unique place fraught with serious security dangers. Smuggling of money, drugs, weapons, and other contraband is all too common an occurrence. And inmate attempts to secrete these items into the facility by concealing them in body cavities are documented in this record ... and in other cases....
We do not underestimate the degree to which these searches may invade the personal privacy of inmates. Nor do we doubt, as the District Court noted, that on occasion a security guard may conduct the search in an abusive fashion. [United States ex. rel. Wolfish v. Levi] 439 F.Supp.,  at 147. Such abuse cannot be condoned. The searches must be conducted in a reasonable manner. Schmerber v. California, supra, at 771-772 [86 S.Ct. at 1836-37]. But we deal here with the question whether visual body-cavity inspections as contemplated by the [jail's] rules can ever be conducted on less than probable cause. Balancing the significant and legitimate security interests of the institution against the privacy interests of the inmates, we conclude that they can. [footnotes omitted].
The details of this balancing process were left unstated in Bell. Lower courts are left to reason from the ultimate holding: a prisoner's Fourth Amendment expectation of privacy does not extend so far as to shield him or her from a humiliating body cavity search following a contact visit with those outside the prison, whether or not there is a basis independent of the mere fact of the visit to suspect the prisoner of smuggling contraband. In the same decision, the Court also held constitutional a system of random, unannounced cell searches carried out for purposes of security. 441 U.S. at 557, 99 S. Ct. at 1883.
In analyzing the constitutionality of the body cavity searches, the Court did not rigorously apply the Schmerber analysis, although it expressly approved the Schmerber reasonable manner requirement in the text quoted above. The decision does suggest that somewhat more careful scrutiny must be given to such degrading search methods. As Justice Rehnquist wrote, "this practice instinctively gives us *1220 the most pause," 441 U.S. at 558, 99 S. Ct. at 1884, and Justice Powell, who joined the remainder of the decision, was unwilling to approve such intrusive search methods without some cause. 441 U.S. at 563, 99 S. Ct. at 1886 (Powell, J., concurring in part and dissenting in part). Urinalyses are entitled to the same level of scrutiny accorded body cavity searches. It may, perhaps, be debated whether one type of search offends "human dignity and privacy," Schmerber, 384 U.S. at 770, 86 S. Ct. at 1835, more than the other, but the difference is a matter of degree, not kind. Both are degrading. It is this basic offense to human dignity, rather than any particular style of causing offense, which sets this type of search apart from traditional types. See, Bell, 441 U.S. at 563, 99 S.Ct.at 1886 (Powell, J., concurring in part and dissenting in part). By the same reasoning, however, urinalysis is not entitled to higher scrutiny than body cavity searches. Therefore, because Bell permits body cavity searches to be conducted without probable cause or reasonable suspicion, it follows that urinalysis is available on a similar basis to prison officials as a means to prevent inmate misconduct.
Although I cannot say that urinalysis offends human dignity and privacy to a degree so much greater than a body cavity search as to require a higher level of cause, there are two circumstantial differences between the body cavity searches in Bell and these urinalyses which must be explored. First, the body cavity searches occurred only after prisoners' exposure to outsiders. Arguably, then, there was some minimal level of cause to believe that the prisoners possessed contraband, for they had the opportunity to obtain it from the visitor. Granted, whatever level of cause this might be is far below the "clear indication that ... evidence will be found" standard of Schmerber, 384 U.S. at 770, 86 S. Ct. at 1835, and even the familiar "reasonable suspicion" standard. United States v. Brignoni-Ponce, 422 U.S. 873, 882, 95 S. Ct. 2574, 2580, 45 L. Ed. 2d 607 (1975). Yet arguably some cause exists. The challenged urinalyses, by contrast, are wholly random.
Examination of the facts of Bell, however, defeats this distinction. As the dissenting Justices make clear, these body cavity searches were very nearly gratuitous. Prison visitors were searched before entry. During the visits the inmates were required to wear one-piece jumpsuits which zipped up the front, and the visits took place in a glass-enclosed room under the scrutiny of officers. In order to lodge contraband in an anal or vaginal cavity, the prisoner would plainly be required to disrobe from the waist up while under the gaze of guards. Bell, 441 U.S. at 577-578, 99 S. Ct. at 1893-94 (Marshall, J., dissenting), 594-595, 99 S. Ct. at 1902-03 (Stevens, J., dissenting). There was virtually no possibility that contraband would be found because it was highly unlikely that the prisoners could hide objects without detection.
Although no statistical evidence of the prevalance of drug use among Ossining inmates was presented, I take judicial notice that drug use among prisoners is a serious, disruptive problem within American prisons. The level of drug use is high enough in all prisons that it seems much more likely that any inmate picked at random from the Ossining prison population would be under their influence than that a Bell inmate would successfully maneuver contraband into his or her body cavities. It is this level of probability which lies at the heart of "cause." The random urinalyses at Ossining seem more likely to turn up evidence of misconduct than the body cavity searches approved in Bell and thus to be based on a higher level of cause. The fact that there is no articulable "triggering" factor is consequently irrelevant.
It might also be argued that the body cavity searches in Bell were in a sense voluntary in that prisoners could avoid them if they desired by avoiding contact visits. Ossining's inmates, in contrast, have no such option. They must respond when called under threat of punishment. This distinction, however, is illusory. It would hardly comfort the Bell prisoners to *1221 inform them they could avoid body cavity searches simply by foregoing physical contact with the outside world. The contact visits involved not only family and friends but legal counsel. 441 U.S. at 577, 99 S. Ct. at 1893 (Marshall, J., dissenting). Such visits were thus important not only for bringing some touch of affection to sterile prison lives but also, because many Bell inmates were pre-trial detainees, for defending criminal proceedings. Only in the most technical sense were contact visits voluntary, for the body cavity searches could be avoided only at the cost of relinquishing perhaps the most significant privilege given the inmates.
Thus I conclude that Bell requires me to defer to New York State prison authorities. If they choose to carry out these searches and do so in a reasonable manner, the Fourth Amendment will not stand in their way. It remains to examine the reasonableness of the manner of search. The searches occurred as stated above. The Schmerber court, in analyzing reasonableness, looked to the reliability of the test, the defendant's response to it, and the manner of testing. 384 U.S. at 770-772, 86 S. Ct. at 1835-37.
The plaintiffs have challenged both the reliability of the tests and the manner in which they are performed. As to the reliability of the tests, I do not write upon a clean slate. Other state prison systems have had their random urinalysis testing subjected to federal constitutional attack. In Jensen v. Lick, 589 F. Supp. 35 (D.N.D. 1984), a prisoner in the North Dakota state penitentiary challenged the accuracy of the same EMIT urine screening program. The district court observed that "[t]he Center for Disease Control in Atlanta, Georgia, undertook to test the reliability of `Emit' testing procedures and found them to be 97 to 99% accurate," data upon which the manufacturer relied in claiming that "testors can act with a 95% confidence in the accuracy of the test result." 589 F. Supp. at 38. "As used," the district court continued, "the 95% statistical figure, in the field of science and medicine, is recognized to mean almost complete certainty." Ibid. (emphasis in original). The court granted defendants' motion to dismiss the complaint.
The opinion does not particularize the reliability testing procedures followed by the Center for Disease Control.
On the other hand, in Kane v. Fair, Civ. No. 136229 (Super.Ct.Mass., decided August 5, 1983), a Massachusetts trial court considered the EMIT process in the light of the medical evidence before it and concluded:
"No evidence having been introduced that warrants my finding that EMIT has been generally accepted by toxicologists and/or pharmacologists as producing a reliable positive result in the absence of independent confirmation, I find that it does not produce such a result absent such confirmation."
Slip op. at 4.
The court in Kane preliminarily enjoined introduction of a positive EMIT test in a prison disciplinary hearing "unless accompanied by evidence that the positive result was confirmed by an alternative method of analysis." Id. at 9.
It is interesting to note that while the federal Bureau of Prisons has promulgated regulations requiring, inter alia, that "staff shall randomly sample each institution's inmate population during each month to test for drug use," they also require that "staff shall have each positive urine test validated to substantiate the positive result." 28 C.F.R. § 550.30(a), (b). The regulations do not specify what initial tests or validating procedures are to be utilized.
The present plaintiffs submit an affidavit of Leo Gross, Ph.D., verified December 12, 1983. Dr. Gross holds his doctorate in *1222 biophysics, and is research director of the Waldemar Medical Research Foundation. His affidavit quotes a printed statement issued by the manufacturer of the EMIT process:
"Because there are many variables that affect ordinary drug levels, an EMIT-st test result is useful only as an indication.... Results should be confirmed by ... an alternative equally sensitive analytical method when loss of rights or other corrective action is contemplated." Gross affidavit at ¶ 4.
Dr. Gross also discusses the phenomenon of "false positives," ¶¶ 5-7. Their falsity may apparently be detected by alternative testing methods, such as chromatography and mass spectrography. ¶ 8. It appears from other material submitted by plaintiffs that the use of common, non-prescription cold medications may skew the EMIT test results for drugs. Memorandum of Ann Burton, July 1984.
While I do not regard the present plaintiffs' showing as strong enough to justify the issuance of a preliminary injunction, this record and these authorities raise an issue of substance as to the test's reliability. The complaint certainly survives a motion to dismiss. I regard the case as appropriate for assignment of counsel from the Court's civil pro bono panel.
There is more evidence in the present record of the manner of conducting the searches, to which I now turn. Officer Brady testified that urine collection was conducted wholly in private, out of the sight of both other inmates and female prison staff. The plaintiffs presented evidence that this was not the case, at least not consistently. They presented statements and testimony that other inmates and female nurses were occasionally able to view the process of urine collection. I consider this an important question. This type of search raises Fourth Amendment concerns precisely because its execution is particularly degradingthat is, in terms of the Schmerber analysis, because the victim's response to it is so extremely negative. It is important, then, that the conduct of the search be no more degrading than is reasonably necessary to satisfy the legitimate security interests of the institution. Forcing an inmate to urinate in front of others, male or female, significantly enhances the humiliating nature of the test. By presenting testimony that the collection is always conducted in view only of the prisoner and testing officer, the State is in effect admitting that security demands do not require that the collection be conducted in view of others. Cf. Bell, 441 U.S. at 577, 99 S. Ct. at 1893 (Marshall, J., dissenting) (because of time pressures, body cavity searches were sometimes conducted in view of other inmates). To conduct these humiliating procedures in the view of others when there is no legitimate need to do so is thus unreasonable.
Following collection of the urine, the bottle is sealed and entrusted to prison personnel. Plaintiffs assert that this procedure leaves the samples vulnerable to contamination by vindictive guards. Assuming *1223 this is true, it does not make the procedure unreasonable. Bell, 441 U.S. at 560, 99 S. Ct. at 1885. Should such contamination occur, it may, of course, present a specific instance of unconstitutional behavior. It would be impossible to protect against this particular risk, however, and the procedures for handling the samples prior to testing to which Officer Brady testified seem careful and reasonable.
I do find one weakness in defendants' general testing procedure. Because these tests involve both embarassment and potential punishment there is, as plaintiffs point out, the possibility of their abuse for purposes of harassment. In particular, prisoners may be targeted for testing simply to harass them. It is important to insure that when the State chooses to employ such intrusive random searches as these the procedures for selecting the inmates to be tested are truly random. To its credit, the State recognizes this and is moving to adopt computer-guided random selection procedures. The current procedure, however, cannot be certified as random. The watch commander stares at a board containing a card for each prisoner and picks several. The commander would presumably contend that he does choose at random, or at least on whim, picking cards off the board with no thought given to the identity of the prisoner selected. Quite likely this assertion would be in good faith. The potential for abuse, however, is apparent. So long as he is potentially aware of the name of the prisoner he is choosing, the commander may, consciously or unconsciously, steer his choices toward less favored inmates. Certainly the practice lends itself to abuse, although there may be no conscious abuse.
Again, if this system served a legitimate security purpose it might well be acceptable. There is no indication that it does. The selection procedures could be made entirely random by throwing replicas of the cards into a large container and drawing them out blindly. This would be just as simple as the current method, would serve an identical purpose, and would eliminate any potential for abuse. The unjustified potential for abuse in the face of readily available alternatives makes the current procedure unreasonable.
The State contends that Hudson v. Palmer, ___ U.S. ___, 104 S. Ct. 3194, 82 L. Ed. 2d 393 (1984), forbids the type of reasonableness inquiry conducted here. In Hudson, the plaintiff contended that his Fourth Amendment rights had been violated by a search of his cell allegedly conducted for the sole purpose of harassing him. In denying his claim for damages, at least under the Fourth Amendment, the Court held that that amendment does not apply to cell searches. Id., at 3200. In effect, it was held that Fourth Amendment protection stops at the cell door. This, however, is distinctly different from holding that it stops at the prison door. This was a question expressly reserved in Bell, supra, 441 U.S. at 558, 99 S. Ct. at 1884, and Hudson does not purport to address it.
Hudson holds that a prisoner retains no legitimate expectation of privacy in his cell, a position reached by weighing the prisoner's interest in privacy within his cell against the institution's need to maintain security. Hudson, 104 S. Ct. at 3200-3201. In part because of the severely limited potential for privacy within a cell, the Court "str[uc]k the balance in favor of institutional security." Id., 104 S. Ct. at 3201.
The body cavity search holding of Bell, discussed above, implicitly restricted a prisoner's Fourth Amendment expectation of privacy in his or her body, so much so that I have held that that expectation does not forbid forced random urine collection and analysis. However, Bell did not, nor does Hudson, hold that a prisoner retains no legitimate expectation of privacy in his or her body. Unlike the cell which, as Hudson noted, "shares none of the attributes of privacy of a home, an automobile, an office, or a hotel room," 104 S. Ct. at 3201, quoting Lanza v. New York, 370 U.S. 139, 143-144, 82 S. Ct. 1218, 1220-21, 8 L. Ed. 2d 384 (1962), a prisoner's body remains the same whether in or out of prison. There is *1224 an inherent diminution in the privacy interest adhering in an individual's immediate surroundings when he or she exchanges apartment for cell. The body undergoes no such fundamental change when incarcerated. The surroundings become less private; the body remains the same.
Thus the Hudson balancing of interests between the prisoners' interests and those of the institution is not so readily struck wholly in favor of the institution. To be sure, because of the prison's security needs the prisoner's expectation of privacy in his or her body is diminished. The residual strength of that interest, however, dictates that the Fourth Amendment not be declared altogether inapplicable. Prisoners retain the protection from unreasonable searches of their bodies secured by the Fourth Amendment.
Having determined that the Fourth Amendment provides plaintiffs with some, albeit limited, protection from urinalyses conducted in an unreasonable manner, it remains to be decided whether my prior decision to deny plaintiffs a preliminary injunction should be altered in any manner. I conclude that it should. It is uncontradicted that the prisoners are subjected, to a method of selection which unnecessarily exposes them to some risk of harassment. They are entitled to a preliminary injunction against its continuance.
No other modification in my prior order, however, is warranted. Although plaintiffs presented an unsworn statement by inmate Fernandez that he was forced to urinate in the presence of others, the three plaintiffs who submitted evidence concerning the conduct of their own urine collection, Storms, Cramer, and Clayborne, did not mention this. Plaintiff Storms did mention that hospital employees passed outside the open door of the restroom, but was unclear whether they did or were able to watch the collection. Apparently inmates are in view of those outside unless the testing officer is careful to guard against this. Officer Brady testified that he conducted the urine collection in private. He admitted that hospital employees inevitably passed by the door of the restroom left open for security purposesbut stated that he took the prisoners to a corner urinal where they could not be viewed by those passing by in the corridor. I accept the officer's testimony and conclude that plaintiffs have not demonstrated that they are likely to show that they have been subjected to or are threatened with urinalyses conducted with so little regard for human dignity and privacy as to be unreasonable. I do so, however, only because I credit Officer Brady's testimony that the testing officers do take care to avoid unnecessary humiliation. Thus I find no reason to preliminarily enjoin such behavior. Plaintiffs may well be entitled to a permanent injunction if they prove at trial that their version of the facts, rather than Officer Brady's, truly describes the standard method of urine collection.
The threat which is intended to make this testing program an effective deterrent *1225 is that prisoners who are found to have urine samples indicating that they are using drugs may be punished. They are disciplined only after a hearing, but one assumes that the test results will speak very loudly at this hearing. Plaintiffs assert that the possibility that they will be disciplined solely on the basis of these test results violates their constitutional rights to due process.
I have previously adverted to plaintiffs' charge that the EMIT test results are unreliable. Point I at 1221-1222, supra. I reiterate that the issue poses a substantial question. It must be remembered that these tests are random; they are done with no reason to think that the particular inmate is using drugs. The EMIT results are thus ordinarily the sole ground for discipline. The question would not be so close if the prisoner were tested only after exhibiting behavior or an appearance consistent with drug use or after having been found to possess drugs. The two facts together conduct or possession plus positive drug results would persuasively indicate drug use. Without the supporting evidence of behavior, however, unreliable (assuming that they are) test results seem a much weaker reed on which to premise disciplinary action. Officer Brady testified that the prison requires any sample which tests positive to be re-tested by another officer. While this precaution will correct for human error in the testing process, it will not affect machine-generated error. Only testing by a different method can do that. This course is made impossible by the prison's practice of disposing of the urine sample after testing, thus denying those prisoners who wish to obtain an independent test the opportunity to do so. This practice, too, is troublesome.
That said, however, I conclude that plaintiffs are premature in seeking their injunction. None of them have been disciplined for drug use. At least three have been tested, but only plaintiff Clayborne claims, in an affidavit submitted in opposition to the motion to dismiss, to have been the victim of a false positive test. Following the test, Clayborne was called for a disciplinary hearing. At the hearing he vigorously denied his use of cocaine, the drug detected in his urine. He claims to have shown prison officials an affidavit by a medical researcher, Dr. Leo Gross, which stated that all urine samples which are to be EMIT tested must be frozen if they are not to be tested on the day of collection. Failure to freeze results in a greater likelihood of false positives. At the hearing, Clayborne asked whether the testing officers properly froze his specimen. His hearing was "adjourned pending investigation"; ultimately the charges were dismissed because of unexplained "improper procedures." One assumes that the improper procedure was the failure to freeze samples, a practice which, judging from Officer Brady's testimony, has since been corrected. Plaintiffs have also submitted statements by other prisoners who claim to have been the victims of false positives or improper test procedures.
In order to have standing to litigate a constitutional claim, plaintiffs must be able to demonstrate real or threatened injury to themselves, not to third parties. Warth v. Seldin, 422 U.S. 490, 499, 95 S. Ct. 2197, 2205, 45 L. Ed. 2d 343 (1975). If, as they must here, plaintiffs rely on threatened injury, the threat must, at a minimum, be real, not hypothetical, in order to satisfy the case and controversy clause of Article III. United Public Workers v. Mitchell, 330 U.S. 75, 90, 67 S. Ct. 556, 564, 91 L. Ed. 754 (1947). It is clear that plaintiffs' worry that they may be threatened with discipline on the basis of a false positive is not wholly illusory. The prison will continue to test. They may be tested. It seems clear that the machine makes occasional, although possibly infrequent, mistakes and prison testing procedures are apparently not always perfect. False positives may occur.
However, the assertion that in the process of subsequently subjecting plaintiffs to discipline the prison authorities might violate their due process rights is merely speculative. Plaintiff Clayborne's experience is illustrative. The sole evidence *1226 against him was the positive test. He was accorded a hearing at which he was given an opportunity to challenge the reliability of the test. He successfully did so and was not disciplined. Although the prison relied solely on the test, it plainly did not deny Clayborne due process. I conclude that the threat that plaintiffs will be denied due process is not sufficiently real to establish a case or controversy for purposes of Article III of the Constitution. I thus have no power to entertain this claim, and it is dismissed without prejudice. As noted above, however, plaintiffs may still challenge use of the EMIT system under the Fourth Amendment.
I vacate my prior denial of injunctive relief to the following limited extent: defendants are preliminarily enjoined from selecting prisoners for random urinalysis by a method which carries with it an unnecessary risk of harassment. Specifically, defendants may not use a method of selection in which the official who chooses the inmates to be tested is aware of the identity of those prisoners while he is choosing them. Defendants must adopt a system of selection in which the prisoners to be tested are chosen blindly.
Defendants' motion to deny the Fourth Amendment claims is denied. The motion to dismiss the due process and Fifth Amendment claims is granted.
It is SO ORDERED.NOTES
 At the same hearing, I denied plaintiffs' request for class certification on the grounds that plaintiffs, because they appear pro se, cannot adequately represent the class. See Oxendine v. Williams, 509 F.2d 1405, 1406 (4th Cir. 1975).
 One of their contentions, that this testing violates their Fifth Amendment right against self-incrimination, is clearly foreclosed by existing caselaw. Schmerber v. California, 384 U.S. 757, 86 S. Ct. 1826, 16 L. Ed. 2d 908 (1966). Nor does it violate the Fifth Amendment to use their refusal to submit to the test against them at a disciplinary hearing. South Dakota v. Neville, 459 U.S. 553, 103 S. Ct. 916, 74 L. Ed. 2d 748 (1983). The Fifth Amendment claims are dismissed.
 The court in Kane permitted use of a positive EMIT test result as the basis for a disciplinary report, as opposed to sanctions. Slip op. at 9.
 As did the Massachusetts court in Kane, supra, the federal regulations apparently permit a disciplinary report on the basis of an unsubstantiated urine test. 28 C.F.R. § 550.30(b).
 The court in Kane, supra, quoted and relied upon the same manufacturer's statement. Slip op. at 3.
 The plaintiffs also presented evidence that at least one inmate found it impossible to urinate in the view of a correctional officer. After being called for a test, the inmate attempted periodically for four hours to urinate into the proffered bottle. When he was unable to do so, he was sent for psychiatric evaluation. The prison psychiatrist sent back a court note stating that it was wholly understandable that a prisoner might be unable to urinate in the presence of others. No further action was taken. Although that inmate's experience is useful in illustrating the potentially degrading nature of the test and the need for sensitive procedures, it does not make those procedures per se unreasonable. At least two plaintiffs, Storms and Clayborne, have been subjected to the test. Neither had problems urinating. Further, that prisoner, although subjected to possibly unnecessary embarrassment, was not punished for his inability to urinate. Punishment for refusal to urinate is permitted under the Directive which governs urinalysis. Instead of punishing the inmate, prison officials recognized his inability for what it wasinability, not refusaland dropped proceedings once the inmate's claims were certified by the psychiatrist.
 Thus the Hudson declaration that prisons need take no steps to prevent even harassing searches is inapplicable here. Hudson, 104 S. Ct. at 3201. Even if that were not true, there is no worry here that prisoners could "decipher" a pattern of planned random searches. 104 S. Ct. at 3201. The requirement of true randomness will simply assure that this testing program is not used to harass. It will in no way undercut the effectiveness of the program since it affects only how the prisoners are selected, not when or where. It will not help the prisoners to figure out who will be tested or when the tests will occur, the concerns which animated the Chief Justice's opinion. Id.
 Plaintiffs also request damages for this violation. However, because none alleges that he was selected solely for purposes of harassment, plaintiffs allege no compensable injury which flowed from this violation.
 In order to obtain preliminary relief, plaintiffs must demonstrate irreparable injury from the practice contested and either 1) that they are likely at trial to demonstrate a violation of their rights or 2) sufficiently serious questions going to the merits to make them a fair ground for litigation and a balance of hardships tipping decidedly toward plaintiffs. Sadowsky v. City of New York, 732 F.2d 312, 313, 316 (2d Cir.1984).