The defendant, Henri Paul Tallard, Jr., appeals the Superior Court’s (Smith, J.) denial of his motion to vacate, modify or amend his sentence on the ground that it is unconstitutional under the rule announced in Apprendi v. New Jersey, 530 U.S. 466 (2000). We affirm.
The following facts are supported by the record. After a jury trial in July 1996, the defendant was convicted of, among other things, assault by a prisoner. See RSA 642:9 (1996) (amended 2000). He was sentenced to an extended term of ten to thirty years on the assault charge pursuant to RSA 651:6, 1(h) (1996) (amended 1998), which authorizes an extended sentence if the court finds that the defendant knowingly committed the assault “where he knows the victim was, at the time of the commission of the [assault], a law enforcement officer acting in the line of duty.” The defendant appealed the sentence, contending that a county correctional officer is not a law enforcement officer under RSA 651:6, 1(h). We affirmed. See State v. Tallard, 143 N.H. 228 (1998).
In June 2001, the defendant filed in superior court a motion to vacate, modify or amend his sentence. He relied upon Apprendi, which held that “[o]ther than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.” Apprendi, 530 U.S. at 490. The defendant argued that his sentence violated the Apprendi rule because the State failed to allege in the indictment and prove to the jury the factual predicate for his enhanced sentence; namely, that the defendant knew the assault victim was a law enforcement officer acting in the line of duty. The defendant’s substantive argument has been approved by this court in State v. Ouellette, 145 N.H. 489, 491 (2000), in which we indicated that under Apprendi, “the sentencing enhancement factor — that the defendant knew that the victim was a law enforcement officer acting in the line of duty — is an element of the charged offense[]” that must be charged in the indictment and proved to the jury beyond a reasonable doubt.
The trial court denied the defendant’s motion. The court noted that the defendant’s case became final more than a year before the Supreme Court *185issued the Apprendi decision. It then found that the Supreme Court’s decision in Teague v. Lane, 489 U.S. 288, 300 (1989), in which a plurality of the Court sought to “clarify how the question of retroactivity [of new constitutional rules of criminal procedure] should be resolved for cases on collateral review,” prevented it from applying Apprendi retroactively to the defendant’s case.
On appeal, the defendant contends that this court should not apply Teague, or at least should not apply it as strictly as the federal courts, because it sets forth a rule of federal habeas corpus law that is “driven by policy considerations that do not apply with equal force in this state, or under the circumstances of this appeal.” We acknowledge that there is some confusion in the case law over the applicability of Teague in state courts. It has been said that “[s]tate courts hearing claims for collateral review ... are free to set their own retroactivity rules independent of Teague.” State v. Mohler, 694 N.E.2d 1129, 1132 (Ind. 1998). We believe, however, that the issue is more complicated.
On the one hand, state courts clearly can determine the retroactivity of their own decisions on state law issues. See, e.g., Meadows v. State, 849 S.W.2d 748, 754 (Tenn. 1993). On the other hand, the retroactivity of federal law pronouncements appears to be subject to federal retroactivity rules. See id.
The determination whether a constitutional decision of [the United States Supreme] Court is retroactive — that is, whether the decision applies to conduct or events that occurred before the date of the decision — is a matter of federal law. When questions of state law are at issue, state courts generally have the authority to determine the retroactivity of their own decisions. The retroactive applicability of a constitutional decision of [the Supreme] Court, however, is every bit as much of a federal question as what particular federal constitutional provisions themselves mean, what they guarantee, and whether they have been denied. In order to ensure the uniform application of decisions construing constitutional requirements and to prevent States from denying or curtailing federally protected rights, [the Court has] consistently required that state courts adhere to [its] retroactivity decisions.
American Trucking Assns., Inc. v. Smith, 496 U.S. 167, 177-78 (1990) (plurality opinion) (quotation and citations omitted); see also Michigan v. Payne, 412 U.S. 47, 57 (1973) (reversing state supreme court’s retroactive application of United States Supreme Court decision). Thus, because Apprendi was based upon the Federal Constitution, see Apprendi, 530 *186U.S. at 476, we may well be required to apply Teague to determine whether Apprendi applies retroactively on state collateral review. See State v. Sepulveda, 32 P.3d 1085, 1086-87 (Ariz. Ct. App. 2001).
Even if not required to do so, however, we are inclined to follow Teague here. We note that a number of state courts have adopted the Teague analysis even as to the retroactivity of their own decisions. See, e.g., People v. Flowers, 561 N.E.2d 674, 682 (Ill. 1990). But see, e.g., Meadows, 849 S.W.2d at 755 (declining to apply Teague to new state constitutional rule). One stated reason for doing so is the similarity of the purposes behind the state collateral proceeding and federal habeas review. See, e.g., Daniels v. State, 561 N.E.2d 487, 489 (Ind. 1990). In addition, as noted in Flowers, 561 N.E.2d at 682, Teague dealt with finality issues that are of similar concern in both federal and state criminal contexts. Finally, the court in State v. Slemmer, 823 P.2d 41, 49 (Ariz. 1991), noted pragmatic reasons for adopting Teague:
The law regarding retroactivity is complex enough without requiring counsel and trial judges to apply different retroactivity rules, depending on whether the substantive decision is grounded on state or federal constitutional principles — especially when many decisions are grounded on both. Given the supremacy of the United States Supreme Court on federal issues and its current explication of the law, we think public policy presently requires that we adopt and apply the federal retroactivity analysis to decisions of state constitutional law.
Of course, we need not decide whether to adopt Teague with respect to state constitutional decisions, as that issue is not before us. We conclude, however, that we should apply Teague to determine whether the defendant may invoke Apprendi on collateral review.
In Teague, a plurality of the Court set forth the following rule governing retroactivity of cases on collateral review: “Unless they fall within an exception to the general rule, new constitutional rules of criminal procedure will not be applicable to those cases which have become final before the new rules are announced.” Teague, 489 U.S. at 310. The Court then recognized two exceptions. The first is “that a new rule should be applied retroactively if it places certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe.” Id. at 311 (quotation omitted). The second exception is for a new rule that “requires the observance of those procedures that are implicit in the concept of ordered liberty.” Id. (quotation and ellipsis omitted). The second exception is “reserved for watershed rules of criminal procedure.” Id.
*187The trial court found that Apprendi announced a new rule of criminal procedure that would not apply retroactively unless it fell within one of the Teague exceptions. The court found the first exception inapplicable because Apprendi did not place assault on - a correctional officer beyond the criminal law-making authority’s power to proscribe. It then found the second exception inapplicable because Apprendi did not announce a watershed rule of criminal procedure. On appeal, the defendant challenges only the court’s failure to treat Apprendi as a watershed rule.
The United States Supreme Court has not yet addressed whether Apprendi set forth a watershed rule for purposes of the Teague analysis. In addition, the defendant acknowledges that among the federal and state courts to have addressed the issue, his interpretation appears to be the minority view. He nevertheless urges us to adopt that position.
The Supreme Court has stated that to fall within the second, “watershed,” Teague exception, “a new rule must meet two requirements: Infringement of the rule must seriously diminish the likelihood of obtaining an accurate conviction, and the rule must alter our understanding of the bedrock procedural elements essential to the fairness of a proceeding.” Tyler v. Cain, 533 U.S. 656, 665 (2001) (quotations omitted). We conclude that Apprendi does not meet this exception.
First, Apprendi altered the procedure under which an enhanced sentence can be imposed. It did not affect the procedure for obtaining an accurate conviction on the underlying offense. Cf. United States v. Sanchez-Cervantes, 282 F.3d 664, 669 (9th Cir.), cert. denied, 123 S. Ct. 48 (2002). “In this light, it seems arguable whether Apprendi increases the reliability of the guilt-innocence determination at all because the rule does not protect the innocent from conviction, it instead limits the sentencing exposure of those who have been validly convicted.” United States v. Moss, 252 F.3d 993, 999 (8th Cir. 2001), cert. denied, 534 U.S. 1097 (2002).
The defendant contends, however, that it is erroneous to view Apprendi as simply a sentencing case and argues that because Apprendi redefined certain sentence enhancement factors as elements of the crime, it constitutes a watershed rule of criminal procedure. We disagree. In Neder v. United States, 527 U.S. 1, 15 (1999), the Supreme Court held that a trial judge’s refusal to charge the jury on an element of the offense was subject to a harmless error analysis. The Court reasoned that “[u]nlike such defects as the complete deprivation of counsel or trial before a biased judge, an instruction that omits an element of the offense does not necessarily render a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence.” Id. at 9. Accordingly, we cannot conclude that a rule requiring certain factual elements to be proved *188to the jury is one that, if infringed, would “seriously diminish the likelihood of obtaining an accurate conviction” or one that implicates “the bedrock procedural elements essential to the fairness of a proceeding.” Tyler, 533 U.S. at 665 (quotations omitted). As the United States Court of Appeals for the Fourth Circuit has noted, the application of harmless or plain error analysis to Apprendi claims by the majority of federal appellate courts “[f]urther supports] the view that Apprendi does not rise to the level of a watershed change in criminal procedure.” United States v. Sanders, 247 F.3d 139, 150 (4th Cir.), cert. denied, 534 U.S. 1032 (2001); cf. United States v. Cotton, 535 U.S. 625, 631-32 (2002) (applying plain error analysis to Apprendi claim).
In sum, we are persuaded by the above considerations, and the reasoning of the majority of courts to have addressed the issue, that Apprendi does not meet the Teague requirements for retroactive application to a case on collateral review. We need not address the defendant’s contention at oral argument that he was entitled to relief under State habeas corpus law because that issue was not briefed. See State v. Sprague, 146 N.H. 334, 338 (2001).
Affirmed.
Brock, C.J., and Broderick, Dalianis and Duggan, JJ., concurred.