concurring in part and dissenting in part:
To the extent that the majority opinion upholds the Virginia DNA testing procedure as applied to violent felons, and holds that the statute qualifies as an ex post facto law with respect to its effect on the mandatory release of prisoners convicted prior to its effective date, I concur in the decision. But I must respectfully dissent from the majority’s determination of the constitutionality of the statute as applied to prisoners convicted of non-violent crimes. Prisoners do not lose an expectation of privacy with regard to blood testing, and the Commonwealth’s articulated interest in the testing of non-violent felons does not counter-balance the privacy violation involved in the procedure.
I.
First, although I concur in the result as applied to felons convicted of violent crimes, I am compelled to address the majority’s strikingly truncated view of the Fourth Amendment protections afforded to a convicted felon, which the majority provides as an alternative basis for its conclusion that the statute, as applied to violent felons, is valid. Having held that the Commonwealth’s knowledge of the identity of those convicted of violent felonies justifies application of the statute to felons of that kind, the majority obliterates — in my mind unjustifiably — the distinction between violent and non-violent felons.
The majority, citing Supreme Court precedent, holds that a prisoner loses “some, if not all, rights to personal privacy otherwise protected by the Fourth Amendment.” Maj.Op. at 308. Although it has been established that incarcerated individuals, particularly because of a partial loss of an expectation of privacy, must carry on their affairs under a significantly limited umbrella of protections against most searches, there exists no blanket authorization of searches involving intrusions under the skin, for which no individual, whether in prison or out, loses a reasonable expectation of privacy.
*312Prisoners most assuredly do give up specific aspects of their reasonable expectation of privacy because of practical concerns relating to living conditions, and because of the necessities involved in ensuring prison security. However, in the present case, appellants have not forfeited their expectation of privacy with respect to blood testing, and no practical penal concern justifies the departure involved in the DNA procedure. Accordingly, the Commonwealth’s DNA testing procedure should be reviewed under the standard applied to a search of any individual when such a search is not based on individualized suspicion: the privacy interest of the prisoner in remaining free of bodily invasion should be balanced against the state interest in carrying out the search.
The majority cites the Supreme Court’s opinion in Bell v. Wolfish, 441 U.S. 520, 99 S.Ct. 1861, 60 L.Ed.2d 447 (1979), to support its sweeping conclusion about the extent of prisoners’ loss of privacy rights. Although Bell allows for certain invasive search procedures, Bell does not suggest that probable cause detainees have abrogated the entire panoply of privacy protections. To the contrary, the Bell Court reaffirms the conclusion that privacy interests of detainees or prison inmates remain, and that these privacy interests must be balanced against the “significant and legitimate security interest of the [penal] institution.” Id. at 560, 99 S.Ct. at 1885.
The majority today implicitly holds that, whether violent or nonviolent, a prisoner loses a reasonable expectation of privacy which justifies the search in the present case. Citing Hudson v. Palmer, 468 U.S. 517, 104 S.Ct. 3194, 82 L.Ed.2d 393 (1984), which held that a prisoner has no reasonable expectation of privacy in his prison cell, the majority indicates that prisoners lose the aspect of their right to privacy that protects them from routine searches. However, the majority goes a step further by applying the Hudson standard to the Commonwealth’s DNA procedure and implicitly holding that the lack of a reasonable expectation of privacy in a prison cell extends to permit unjustified searches of bodily fluids of every felon, violent and non-violent alike.
A prisoner certainly cannot stake a claim to the kind of sanctity of dominion to a cell that a non-incarcerated individual can to a home. Moreover, the Supreme Court has indicated that there are situations in which even a private citizen’s expectation of privacy diminishes, or disappears altogether.1 It is apparent, however, that the search involved in the present case, blood testing, violates a privacy interest that even a prisoner, living in close quarters under constant security surveillance, reasonably can expect to enjoy.
No precedential justification exists for the majority’s holding that convicted felons, violent or non-violent, solely because of past criminal activity, lose the expectation that their bodily fluids will be free of unjustified search. As the Supreme Court stated in Skinner v. Railway Labor Association, 489 U.S. 602, 616, 109 S.Ct. 1402, 1412, 103 L.Ed.2d 639 (1989), “it is obvious that this physical invasion [a blood test] penetrating beneath the skin, infringes an expectation of privacy that society is prepared to recognize as reasonable.” Although Skinner involved the testing of free citizens, its determination that an individual has a reasonable expectation of privacy within one’s own body applies equally to prisoners, unless the prisoner’s privacy right is incompatible with the objectives of incarceration.
The majority, by relying on Hudson to justify its conclusion that prisoners lose a right to privacy from routine searches, indicates that the loss of a prisoner’s expecta*313tion of privacy arises from the significant, if not complete, loss of constitutionally mandated privacy rights that occurs incident to incarceration. Maj.Op. at 308. The Supreme Court, however, while permitting restrictions on prisoners’ privacy rights, has repeatedly reaffirmed the principle that “prisons are not beyond the reach of the Constitution,” Hudson, 468 U.S. at 523, 104 S.Ct. at 3198, and that “prisoners [must] be accorded those rights not fundamentally inconsistent with imprisonment itself or incompatible with the objectives of incarceration.” Id. The Court has justified restrictions on prisoners’ rights, specifically their expectation of privacy, by relying on practical considerations requiring additional authority on the part of prison officials to control potentially dangerous prison communities. Id. at 524, 104 S.Ct. at 3199. There is no such practical concern, at least not one relating to “the objectives of incarceration,” involved in the Commonwealth’s .DNA testing program.2 Accordingly, the majority’s reliance on the Hudson expectation of privacy standard is misplaced.
Additionally, the majority cites Griffin v. Wisconsin, 483 U.S. 868, 107 S.Ct. 3164, 97 L.Ed.2d 709 (1987), to support the proposition that unincarcerated convicted felons “lose their right to privacy against searches of their homes pursuant to an established program to ensure rehabilitation and security.” Maj.Op. at 308. Griffin, however, is inapplicable to the case at hand. In Griffin, a Wisconsin regulation allowed any probation officer to search a probationer’s home without a warrant as long as the officer’s supervisor approved the search, and as long as “reasonable grounds” existed to believe that contraband would be present. Although the search authorized in that case did not require a showing of probable cause, it did require individualized suspicion relating to the person searched based on a “reasonable grounds” standard. Id. at 871, 107 S.Ct. at 3167. As both parties concede, the DNA testing involved in the present case is not based upon any individualized suspicion of the persons tested beyond the statistical assertion that, as a group, convicted felons are somewhat more likely to commit future crimes than members of the non-felonious general population. The Griffin case simply does not support the majority’s conclusion that all convicted felons, both violent and non-violent, may be made subject to random, nonparticularized searches merely because of their past illegal acts. Justification for searches of these individuals must be based, as must all searches of citizens in a free society still clinging to disappearing Fourth Amendment protections, on a balancing of the privacy interest involved against the state interest in the search to determine which interest is more compelling. See Michigan Dept. of State Police v. Sitz, 496 U.S. 444, 110 S.Ct. 2481, 2485, 110 L.Ed.2d 412 (1990). It is with a reasoned application of the Sitz standard, and not with a disturbing restriction of the Fourth Amendment protections afforded to the nation’s prisoners, that the DNA testing procedure, as applied to felons convicted of violent crime, may be justified.
II.
The DNA testing of felons convicted of non-violent crimes is not justified, given the limited and non-compelling state interest-in including them in the testing procedure. The only state interest offered by the Commonwealth for including non-violent felons is administrative ease. I cannot conclude that the government interest in administrative ease suffices to outweigh a prisoner’s expectation of privacy in not having blood withdrawn from his body when that prisoner is not significantly more likely to commit *314a violent crime in the future than a member of the general population.
The Commonwealth, in its Report of the Joint Subcommittee Studying Creation of a DNA Test Bank, Senate Report No. 29 (January 1, 1990), concluded only that “the recidivism data supported the inclusion of plaintiffs convicted for felony sex offenses, assault, capital murder, first and second degree murder, voluntary manslaughter, larceny and burglary.” Nonetheless, it also recommended the testing of all remaining felons, not because such testing would be likely to help solve crimes, but only because their inclusion would make the data bank “more efficient and cost effective.” 3
Thus, the Commonwealth’s expressed, primary legitimate interest in testing nonviolent felons is not based on law enforcement concerns, but on a priority for saving the limited extra funds that would be necessary to determine which of the incarcerated prisoners were appropriate for DNA testing. Such a justification for testing nonviolent felons is based on a tenuous state concern that could just as readily justify the testing of any citizen, as long as the inclusion of that individual served to lessen the Commonwealth’s administrative workload.
The Commonwealth’s DNA testing procedure, as applied to nonviolent felons, will have a significantly limited effect in actually solving future crimes. The district court misapprehended the potential effectiveness of the DNA testing procedure as related to non-violent felons. The court indicated its belief that the testing procedure could be used to facilitate perpetrator identification in up to 30% of cases because a significant specimen of blood or other bodily fluids is likely to be left at 30% of violent crime scenes. Yet both government and independent studies in the record show that, overall, remarkably few crime scenes contain specimens from the perpetrator capable of facilitating DNA matching, and that very few types of crimes are likely to produce sufficient information. One study, cited in the majority opinion, Maj.Op. at 310, found that 194 of the 200 cases (97%) in which DNA evidence was available to link the defendant with the crime involved murders or rapes.
The record supports appellants’ contention that there is an extremely tenuous link connecting persons convicted of non-violent felonies to the commission of future violent crime. It, therefore, contains nothing to substantiate a theory that DNA testing of non-violent felons would assist in solution of future crimes. United States Justice Department statistics provided in the record show that only %0.4 of non-violent felons are later arrested on rape charges, and only %0.8 are later arrested on murder charges. One might assume non-violent drug offenders would be more likely to commit violent crime subsequent to release than other non-violent felons; yet, only %0.4 of them are later arrested for rape, and %0.3 for murder.
The record does not provide similar percentage statistics for the general population. It can be readily inferred that the testing of all citizens, regardless of criminal record, would create a DNA data bank with a similar statistical likelihood of solving future crime as is provided by the testing of non-violent felons. Additionally, the testing of other discrete populations, e.g., racial minorities or residents of underprivileged areas, might produce significantly better statistics than %0.4.4
Lacking a significant statistical likelihood to justify the inclusion of non-violent felons, the Commonwealth is forced to justify the inclusion of these individuals based *315on its amorphous concern for administrative efficiency. In order to withstand appellants’ challenge to the constitutionality of the inclusion of non-violent felons in the DNA testing procedure, the Commonwealth is required to articulate an important state interest justifying the privacy intrusion.5 The majority, without any reference to the Commonwealth’s actual expressed interest, offers an alternative means to support the inclusion of non-violent felons: “if DNA technology becomes more common ...” it will become more likely that DNA clues will be “left as a result of crimes other than murder or rape.” Maj. Op. at 310-11. The majority’s proffered justification denies the Commonwealth’s own conclusion that the recidivism data did not support inclusion of anyone other than violent criminals, and directly ignores the actual stated reason for the inclusion, namely that testing all felons would be “more efficient and cost effective.” Therefore, the majority opinion, by implicitly approving the' Commonwealth’s stated justification, leads me to a deep, disturbing, and overriding concern that, without a proper and compelling justification, the Commonwealth may be successful in taking significant strides towards the establishment of a future police state, in which broad and vague concerns for administrative efficiency will serve to support substantial intrusions into the privacy of citizens.
We should be extremely careful, of course, when we endeavor to replace our judgment for that of elected officials. In the case at hand, the relative lack of situations where DNA evidence obtained from any felon will be useful in solving crimes suggests the possibility that the DNA testing program will prove perhaps, ill-advised as a public policy matter. Yet the determination of the value of any law enforcement program rests almost exclusively with the legislature. To the extent .that the Commonwealth’s testing program, which causes a violation of individual privacy rights, also serves to promote an important government interest with at least a reasonable claim to potential effectiveness, we have no choice but to uphold the practice. See Sitz, 110 S.Ct. at 2488. In the present case, to the extent that the statute applies to felons convicted of violent crimes, there is a sufficiently reasonable connection between the DNA testing procedure and the government interest in better identifying future criminals to withstand a constitutional challenge to the statute. Accordingly, I concur with the result reached by the majority as to the statute’s application to felons convicted of violent crimes after the effective daté of the statute.
However, to the extent that the statute applies to non-violent felons, the Commonwealth’s own report and express justification, as well as statistical evidence in the record, readily show that the specific rationale for testing such individuals is administrative ease. This questionable government interest6 is not sufficient, in my view, to justify the intrusion involved here. For the foregoing reason, I would hold the statute constitutionally invalid as applied to those appellants, and others similarly situated, who are felons convicted of non-violent crimes.
. The Supreme Court has held that individuals have a lessened expectation of privacy in regard to searches of cars and other mobile, public property because a "car ordinarily is not used as a residence or repository for one’s personal effects, and its passengers and contents are generally exposed to public view.” Florida v. Jimeno, — U.S. —, 111 S.Ct. 1801, 1805, 114 L.Ed.2d 297 (1991). But the Court has continually restricted the curtailment of the expectation of privacy to situations in which the property searched has a similarly public character, disallowing the use of the justification to support searches of closed containers that demonstrate an individual’s intent to keep the enclosed material private. Id.
. The majority, in two separate instances, asserts that one of the Commonwealth’s interests in instituting a DNA data bank is to identify persons who are incarcerated. Maj.Op. at 308. Nowhere in the record on appeal has the Commonwealth suggested that its intent in the creation of the data bank is related to the identification of prisoners for any institutional penal purpose. The only stated justification for creation of the bank is the facilitation of future law enforcement. Therefore, the search involved here in the blood testing of potential future criminals is significantly more akin to the gathering of evidence to support conviction at trial than to the activity of identifying felons within a .prison facility.
. That approach would permit DNA testing of all citizens at birth, though violence had not yet manifested itself, as justification for the consequent invasion of privacy.
. An argument can be made that there is little likelihood that a statesponsored program to create a DNA data bank of all racial minorities could survive an equal protection challenge, but it is clear that a similar program targeted at underprivileged citizens, having a disparate impact on these same minorities, would not be subject to equal protection analysis, given the non "suspect” nature of wealth-based classifications. San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 28, 93 S.Ct. 1278, 1294, 36 L.Ed.2d 16 (1973).
. Though not conclusively determined, recent federal court precedent indicates that "in deciding whether an intrusion into an individual’s privacy is justified [the facts to be considered should include] ... whether there is an express statutory mandate, articulated public policy, or other recognizable public interest militating towards access.” Fraternal Order of Police, Lodge 5 v. Philadelphia, 812 F.2d 105, 110 (3rd Cir.1987). In the absence of such an express mandate or articulated public policy, I find it inappropriate for the majority, or any appellate court, to inject a "recognizable" interest where none is offered by the government or available in the record on appeal, particularly when, as here, the state interest that is proffered and promoted does not justify the invasion of privacy-
. I consider the interest questionable, additionally, because it is unlikely that a significant administrative burden will result if the Commonwealth is forced to separate the prisoners convicted of violent felonies from those convicted of non-violent felonies prior to conducting the DNA testing.