Laurence, J.
David Pare was convicted in December, 1995, of three counts of indecent assault and battery on a child under fourteen years of age, in violation of G. L. c. 265, § 13B. The [567]*567child, John,1 who was the son of Fare’s girlfriend, was ten in June, 1992, when he first accused Pare of sexual abuse. That accusation came over three years after the last alleged incident and John’s last contact with Pare. No physical evidence supported the allegations. Pare denied them and argued that John either fabricated the allegations because John believed Pare responsible for introducing his mother to drugs (which had led to her incarceration for selling drugs), or made them as the result of improperly suggestive influences. John’s credibility was the critical issue before the jury.2
By pretrial motion Pare sought access to a variety of records, arguably privileged, relating to John’s allegations of sexual abuse and his treatment and counseling, pursuant to the decision in Commonwealth v. Bishop, 416 Mass. 169, 181-183 (1993), which outlines a five-stage process governing a defendant’s access to, and the admissibility at trial of, privileged records.3 [568]*568Among the materials requested were records relating to an evaluation of John in 1992 by a “sexual information and trauma team” (SITT) at the Center for Family Development (CFD) in Beverly. 4 It was immediately following the SITT evaluation that the first official child abuse report against Pare was filed, which culminated in his indictment and trial. The motion judge was satisfied with Fare’s detailed submission explaining his reasons for seeking access to the records. See Bishop, 416 Mass, at 181-182, having determined that the records were privileged, id. at 181, he examined them in camera for their relevance to issues in the case. Id. at 182. Thereafter, the judge allowed a few “relevant” portions to be disclosed to counsel, but withheld the bulk of them, including all SITT-related materials, as “irrelevant.”5
Fare’s principal contention on appeal is that the motion judge erred in applying Bishop so as to deny his counsel access to the CFD records relating to the SITT evaluation. Stressing the suspect circumstances of John’s disclosure of abuse that emerged from the SITT evaluation, Pare claims that those records must have been relevant to his defense. He also assigns as error the trial judge’s subsequent refusal to reconsider the motion judge’s ruling as to those materials6 while nonetheless allowing the prosecutor to elicit testimony regarding various [569]*569aspects of the SITT process. That testimony included statements by John’s social worker, Joan Matteuzzi, to the effect that she had observed but not participated in the SITT interviews, that the interviews had been “comfortable and supportive” and “nonleading”, and that it was as a result of the SITT evaluation that she obtained sufficient information to file with the Department of Social Services (DSS) the report of Fare’s suspected sexual abuse of John which instigated the criminal charges against Pare. Having examined the SITT, CFD and related documents and found a good many of them relevant to issues in the case, Bishop, supra at 182, we agree with Pare that the motion judge’s refusal to allow his counsel access to those documents created an undue risk of an erroneous conviction and entitles him to a new trial.
1. Background. As of October, 1989, John, then seven years old, lived with his five and three year old sisters and their mother at the Lynn home of Pare, then the mother’s boyfriend. After an investigation by a DSS social worker, Janine Brummer, of the children and their living conditions at the Pare house,7 they were removed and placed in a series of foster homes over the next three years. John did not see Pare thereafter. In January, 1990, John was separated from his sisters. John’s mother was imprisoned from January, 1991, to February, 1992, for selling drugs to teenagers.8
Throughout the period of his foster placements, John “acted out” and exhibited escalating behavioral and emotional problems, including accusations of sexual molestation against a foster brother and a foster sister (which he later recanted) and “vulgar,” “sexualized” language and conduct. By the spring of 1992, John’s behavior so concerned his foster mother and one of his social workers that the social worker requested a sex abuse evaluation by a SITT at the Center for Family Development “to assess [John’s] behaviors, the escalation of the behaviors, and to determine if he needed services pertaining to those behaviors.” John was referred to a particular individual at the CFD, Ginny Catalfamo, who conducted the SITT evaluation [570]*570over six separate sessions “from March of 1992 until sometime in the summer of 1992.” Upon the conclusion of the SITT process, the investigation of Fare’s alleged sexual abuse of John began, resulting in his indictment and conviction. Shortly thereafter, John, who began manifesting suicidal behavior, was sent to a restrictive facility for disturbed children, where he resided at the time of trial. While there, John elaborated on the charges of abuse by Pare to investigating police officers.9
2. The Privilege Determination.10
The motion judge determined generally that the materials sought by Pare were privileged pursuant to several enumerated statutes but did not specify which privilege applied to the SITT report and related CED records. Pare has questioned the correctness of the judge’s privilege determination. His criticism has merit.
The judge made no subsidiary findings, nor is there anything in the record, establishing11 that the individuals involved in the SITT process met the eligibility requirements of any of the statutes relied on by the judge. The only evidence regarding the CED representative who ran the SITT process (Catalfamo, who did not testify) describes her variously as a “sexual abuse clinician,” a “sexual abuse counsellor,” a “therapist,” and an “evaluator.” Such general labels are insufficient to meet the [571]*571specific professional requirements of the relevant privilege statutes.12 No other SITT participants were identified, nor their professional qualifications described, in the record.13
The Bishop protocols have not altered the traditional recognition that testimonial privileges, which have the effect of inhibiting full disclosure of the truth, are exceptional and to be strictly construed. See Three Juveniles v. Commonwealth, 390 Mass. 357, 359-361 (1983); Cronin v. Strayer, 392 Mass. 525, 532-533 (1984). In balancing the social benefits of the asserted privilege against the defendant’s constitutional right to a fair trial, we must “resolv[e] any doubt in favor of disclosure.” Bishop, 416 Mass, at 177.
Consequently, we conclude that the record before the motion judge did not support his determination that all the SITT-CFD materials sought by Pare were privileged. Nonetheless, that error itself does not advance the appeal in this case. Even assuming the materials were not privileged, Pare would have to demonstrate their relevance — it is axiomatic that all evidence must be relevant to be eligible for admissibility. See Proposed Mass. R. Evid. 402. Here, the motion judge properly addressed the issue ánd ruled that the SITT records and related documents [572]*572were irrelevant. It is that ruling which affords Pare a substantial basis for reversal.
3. The “Relevancy Determination,”14 We recall that the resolution of relevancy issues lies within the discretion of the trial judge, whose exclusion of proffered evidence as irrelevant is to be reviewed under the deferential “abuse of discretion” standard. Commonwealth v. Gonzalez, 22 Mass. App. Ct. 274, 281 (1986). Because of the unique circumstances of this case, however, we would, even without actually examining the SITTrelated materials, be inclined to deem them of such inherent and probable relevance to the central issue in the case — John’s credibility — as to entitle Fare’s counsel to review them under Bishop in order to determine whether their use at trial was required to assure Pare a fair trial. See Bishop, 416 Mass, at 182. Cf. Commonwealth v. Jones, 404 Mass. 339, 341 (1989) (construing the statutes applicable to the social worker privilege and child abuse investigations to “allow a criminal defendant to obtain the investigation and evaluation reports which' ultimately led to his indictment and criminal prosecution”). Having examined them, we confidently conclude that they satisfy the standard of actual relevance. Accordingly, the motion judge’s denial of Fare’s request for their disclosure to defense counsel, under appropriate protective orders, constituted an abuse of discretion requiring a new trial. See Commonwealth v. O ’Brien, 27 Mass. App. Ct. 184, 187-188 (1989). Cf. Commonwealth v. Martin, 392 Mass. 161, 163 (1984) (“As a general proposition, a judge does not have discretion to exclude relevant evidence”).
Our analysis rests on two basic considerations. The first is the breadth of the idea of relevance in our law. Evidence is generally relevant so long as it has “a ‘rational tendency to prove an issue in the case,’ ” or makes a “desired inference more probable than it would be without” the evidence. Commonwealth v. Fayerweather, 406 Mass. 78, 83 (1989), quoting from, respectively, Commonwealth v. Chretien, 383 Mass. 123, 136 (1981), and Commonwealth v. Copeland, 375 Mass. 438, 443 (1978). The desired evidence “need not establish directly the proposition sought; it must only provide a link in the chain of proof.” Commonwealth v. Gordon, 407 Mass. 340, 351 (1990), quoting from Commonwealth v. Tobin, 392 Mass. 604, 613 (1984). Indeed, evidence is to be considered relevant if it only “throwfs] light,” Commonwealth v. Palladino, 346 Mass. 720, [573]*573726 (1964), or “shed[s] light on an issue,” Adoption of Carla, 416 Mass. 510, 513 (1993), or, “in connection with other evidence, it helps [the fact-finder] a little.” Commonwealth v. Tucker, 189 Mass. 457, 467 (1905). It is relevant if it “could have been helpful” for a jury in determining whether a complainant was telling the truth. Commonwealth v. Fayerweather, supra at 83 (a child sexual abuse case). So long as evidence possesses any of these probative tendencies, even if it “is of marginal significance, we cannot say that it [is] irrelevant.” Cohen v. Liberty Mutual Ins. Co., 41 Mass. App. Ct. 748, 752 (1996).15
The second analytical factor supporting disclosure is the contextual nature of relevancy, whether actual or likely. Relevancy rulings are not abstract propositions but are case and fact specific. They derive from the relationship of the particular facts desired to be introduced in evidence, the particular facts sought to be proved or disproved, and the particular issues in the case to which those facts are claimed to relate. See Liacos, Massachusetts Evidence 112 (6th ed. 1994); Young, Pollets & Poreda, Massachusetts Evidentiary Standards, Evidentiary Standard 401, at 48 & n.l (1995 Supp. to Hughes, Evidence). The particular circumstances of this case not only point to the appropriateness of measuring relevancy relatively leniently, but also, even without reviewing the materials, make the conclusion that the SITT and related records would be both relevant and “useful to [Fare’s] defense,” Commonwealth v. Two Juveniles, 397 Mass. 261, 269 & n.7 (1986), virtually irresistible as a matter of logic.
[574]*574Initially, we note that the facts of this case do not implicate the principal public policy rationale which prompted the Bishop procedures, i.e., minimizing the potential reluctance of rape and sexual abuse victims to seek needed therapy. See Bishop, 416 Mass, at 176; Fuller, 423 Mass, at 221-222. In cases in which children of tender years, like John, are subjected to diagnostic evaluations at the behest of an adult without the necessity of their assent, the concern that disclosure of the records of that evaluation will create a “chilling effect. . . preventing those in need from seeking . . . help,” Commonwealth v. Collett, 387 Mass. 424, 428 (1982), seems of much reduced significance.16
More specifically, the unique factual constellation here pointed to the propriety of disclosure of the SITT and related records to defense counsel as a matter of common sense and “practical necessity.” Bishop, 416 Mass, at 181, quoting from Commonwealth v. Collett, 387 Mass, at 438. The charges against Pare depended entirely on the credibility of a ten year old child who had endured a severely dysfunctional family situation, including his mother’s drug addiction and imprisonment, and whose behavioral and emotional condition so deteriorated over the relevant period that he became potentially suicidal, requiring his removal from foster' homes and placement in a restric[575]*575tive facility for three years. The child had made several charges of sexual misconduct against others which he later acknowledged were false. Compare Commonwealth v. Bohannon, 376 Mass. 90, 94-95 (1978). From those incidents, the child learned that adults in authority took allegations of sexual abuse seriously and that a person who touches another “in a bad way” gets into trouble. See Commonwealth v. Fuller, 423 Mass. 216, 226 (1996) (“a credible showing that a complainant previously had fabricated allegations of sexual assault” would constitute “material evidence” with respect to the issue of the defendant’s guilt).
The child delayed reporting Fare’s abuse for over three years after the last alleged incident and his last encounter with Pare. That unexplained and overly long delay made the testimony about the child’s reported accusations ineligible as “fresh complaint” (as the Commonwealth concedes). See Commonwealth v. Fleury, 417 Mass. 810, 813, 815 (1994); Commonwealth v. Traynor, 40 Mass. App. Ct. 527, 531-533 (1996).17 Moreover, a delay of such magnitude inevitably “provokes unease,” Commonwealth v. Hyatt, 31 Mass. App. Ct. 488, 491 (1991), about the possibility of “invention or distortion of an event by mistake, twist of memory, fantasizing, contrivance, etc.” Commonwealth v. Dion, 30 Mass. App. Ct. 406, 413 (1991). See Fuller, 423 Mass, at 226 (“credible information tending to suggest that the complainant has difficulty distinguishing fantasy from reality” would be “material evidence” on the issue of the defendant’s guilt).
Compounding those sources of potential inaccuracy was the child’s admitted anger over Fare’s relationship with his mother, which John believed had led to her drug abuse and incarceration. Such anger created the reasonable likelihood of both bias and motive to lie on the child’s part, with specific reference to Pare as their probable object. See ibid, (a showing of bias against the defendant woul4 be material exculpatory evidence).
Further supporting the probable relevance of the SITT records were the inconsistencies in the child’s pretrial stories and trial testimony with respect to the nature and extent of Fare’s alleged abuse. (See notes 2 and 9, supra.) Given the protracted character of the SITT evaluation, and the promptness of the filing of a sexual abuse charge against Pare hard on its heels, it [576]*576was a not unreasonable inference that the SITT materials might well reveal similar inconsistencies which, “however minor, if explored in cross-examination, might have affected the victim’s credibility . . . [which] was crucial to the outcome of the case.” Commonwealth v. O’Brien, 27 Mass. App. Ct. at 188.
Finally, the circumstances of the SITT evaluation process itself, even to the limited extent revealed by the record, made the relevance of its documentation to Fare’s defense rationally overwhelming. Prior to that evaluation, Pare does not appear to have been a focus of suspicion for abusing John.18 Nothing attributable or ostensibly traceable to Pare prompted the referral for the evaluation. The evaluation was conducted by a “team,” possibly consisting of several persons, headed by Catalfamo, the “sexual abuse clinician” or “counselor.” It took place over the course of at least three months and six separate meetings with John. At the conclusion of the process and as a direct “result of information . . . obtained during this evaluation,” Matteuzzi, who had observed all of the sessions, filed a report of Fare’s suspected abuse of John. On these few facts it was overwhelmingly inferable that during the SITT process John made his first allegations of abuse against Pare and that they were not spontaneous but rather in response to prolonged questioning. Contrast Commonwealth v. Allen, 40 Mass. App. Ct. 458, 462 (1996).
In light of the well-established “reality of children’s susceptibility to suggestive interview practices,” United States v. Rouse, 100 F.3d. 560, 572-574 (8th Cir. 1996), and our explicit acknowledgment that “leading or coercive [or suggestive] questioning [and interview techniques] can distort a child’s memory,” Commonwealth v. Allen, 40 Mass. App. Ct. at 462, the factual background suggestive of false, inaccurate, or subtly induced allegations by John made the specifically identified SITT and related CFD records prima facie relevant. Who was present, how questions were posed, what responses were given and how they comported with later allegations were all matters that defense counsel should have been given the opportunity to ascertain under Bishop, if only to assess why John’s attitude [577]*577had dramatically changed toward Pare, whom he had previously said he loved and wanted to live with, and why it took numerous sessions and several months to extract the initial accusations against Pare.19 It is difficult, on this record, for us to conclude otherwise than that the SITT and related records would be presumptively material to Fare’s defense.
Our in camera review of the entire body of the withheld documents confirms that presumption. Without excessively and inappropriately disclosing the precise contents of the impounded records, we glean from the SITT-related materials themselves that prior to the SITT evaluation John had made no inculpatory allegations against Pare; that for over two months and through five sessions John leveled no sexual abuse charge against Pare; that John’s eventual accusation was not voluntary or spontaneous; and that it did not emerge until what Matteuzzi herself described as “the very end of the evaluation.” Moreover, the notes and reports relating to SITT interviews20 would appear to present to competent defense counsel numerous fruitful opportunities to contest the fairness and reliability of the process. Among other points of potential challenge that the documents arguably reveal are: the use of a unique, copyrighted protocol directed at uncovering sexual abuse, devised by Catalfamo, that may or may not have accorded with accepted professional standards; the use of leading questions and the prodding of John for particular responses; relatively persistent and repeated inquiries consciously aimed at overcoming John’s prolonged reluctance to make accusations and his perceived lack of [578]*578cooperation; cajoling John by repeated assurances that neither he nor his mother would get into trouble on account of his revelations of sexual abuse; terminating John’s visitations with his mother in the midst of the evaluation out of concern that she might be hindering John’s “cooperation” (a situation that could well have exacerbated John’s ire against Pare); the arguably suggestive and criticizable use of anatomically correct dolls in questioning John and having him demonstrate alleged abuse (see Commonwealth v. Allen, 40 Mass. App. Ct. at 464); explicitly mentioning charges of sexual abuse against Pare made by John’s siblings; and the arranging of a confrontation of John by his siblings, who, in the face of John’s denials of abuse, urged him to support their charges.
In short, defense counsel would be able to find in the materials a relevant basis for calling an expert to opine regarding the suggestive or coercive propensities of the techniques used in the SITT.interviews. Compare Commonwealth v. Allen, 40 Mass. App. Ct. at 462-464 & nn.8 & 11. Such a presentation might well have negatively affected the jury’s assessment of the credibility of John’s accusations, see Commonwealth v. O’Brien, 27 Mass. App. Ct. at 188, and consequently “might have had a significant impact ... on the outcome of the trial.” Commonwealth v. Fayerweather, 406 Mass, at 84, quoting from Commonwealth v. Grieco, 386 Mass. 484, 491 (1982).
Moreover, other impounded documents, peripheral or not directly related to the SITT evaluation itself, display highly pertinent facts about John’s ability to perceive, remember and recount, because of his mental and emotional state during the relevant time period. Revealed in a number of expert reports and diagnoses are problems associated with memory and perception capacity; moderate intellectual impairment; erratic and spotty powers of recollection; frequent lying; and instances of fantasizing, “seeing things that aren’t there,” confused and distorted thinking, flashbacks, dissociative statements, “homicidal ruminations,” distorted thoughts about authority figures, and a “pattern” of blaming and accusing those who frustrated or angered him. Some of this material also indicates that John’s mother may have sexually molested him but that he blamed Pare for her acts because of his idealized conception of her. It is an extreme understatement to observe that, “if explored in cross-examination, [such facts] might have affected [John’s] credibility . . . [which] was crucial to the outcome of the case.” Commonwealth v. O’Brien, 27 Mass. App. Ct. at 188.
[579]*579Thus, unlike the situation in Commonwealth v. Reed, 417 Mass, 558, 562 (1994), lack of access to the impounded documents prevented Fare’s counsel from conducting as effective a cross-examination of the key witness as he could have otherwise. Inspection of the withheld documents shows that counsel was similarly disadvantaged, by being denied access, in his cross-examination of the two witnesses who most significantly bolstered John’s trial testimony. Matteuzzi, who portrayed herself at trial as a passive observer of the SITT process who asked no questions but was present only for John’s moral support, in fact appears to have intervened to press John with leading questions at several critical moments. Her testimonial description of the conversations at the SITT interviews as “pretty much . . . instigated by [John]” is substantially belied by the SITT evaluation report, which demonstrates a reluctant and uncooperative child from whom the sexual abuse charge against Pare was ultimately coaxed, utilizing leading questions and confrontational techniques, only after five fruitless sessions. Matteuzzi’s CFD report on John’s status, prepared in the middle of the SITT sessions, explicitly stated that, as part of John’s “treatment plan,” she would “[e]ncourage him to discuss the sexual abuse that he has spoken of with David Pare.”21
Being foreclosed from bringing out such information in the cross-examination of Matteuzzi was particularly harmful to Fare’s case because the prosecutor was allowed, over strenuous objection, to present testimony by Matteuzzi concerning various aspects of the SITT interviews. Her description effectively conveyed to the jury that the SITT sessions with Matteuzzi present provided a previously unavailable “safe and comfortable and supportive setting for [John] to make a disclosure”; that the interviews were conducted in “essentially a nonleading kind of situation”; that as a result John was able to unburden himself by making his allegations of Fare’s sexual abuse; and that the experienced professionals hearing his charges were so impressed by their credibility that they immediately reported them to authorities.
[580]*580Without access to the withheld documents — which depict a far less neutral and volitional process than did Matteuzzi’s testimony — Fare’s counsel was unable to challenge, much less undermine, such inferences favorable to the prosecution. He could not point to evidence of bias and improper interviewing techniques by the evaluators; evidence of resistance, equivocation or qualification by John in making his accusations; evidence that John may have been coaxed or badgered into making the charges; evidence that it took six interviews and intervention by others because John was reluctant to make allegations; or any other evidence that arguably could have indicated that the inferences which the Commonwealth wished the jury to draw from Matteuzzi’s testimony were inaccurate.22
Without the unavailable documents, defense counsel similarly could mount no effective cross-examination of the Commonwealth’s lead witness, DSS social worker Brummer. She testified about having initiated John’s removal from Fare’s house and placement in a foster home, being assigned as his ongoing DSS social worker, and being responsible for referring John to CED and Catalfamo for a sexual abuse evaluation in March, 1992. She further stated that from the time John was moved to the facility for behaviorally disturbed children in 1992 sometime after his SITT evaluation, he greatly improved and was “able to tell [her] disclosures, albeit hard disclosures but truthful disclosures.” Despite defense counsel’s objection and motion to strike, the judge neither struck the remark nor instructed the jury to disregard it. In the context of the trial, it was unmistakably clear that the “disclosures” Brummer believed to be truthful were John’s disclosures of Fare’s sexual [581]*581abuse.23 Her earnest description of John’s truthfulness was likely to have been given considerable weight by the jury. See Commonwealth v. Powers, 36 Mass. App. Ct. 65, 66, 68, 70 (1994) (jury likely to perceive a DSS social worker with years of investigating sexual abuse “as especially qualified to judge the credibility of children”).24
The damaging effect of Brummer’s testimony could, however, have been significantly diluted if defense counsel had been able to question her about information contained in the nondisclosed documents revealing that she had unsuccessfully sought a SITT evaluation in November, 1989 (less than a month after John had come to DSS attention) based solely upon her unconfirmed suspicions of Fare’s sexual abuse of John, then repeated her request in July, 1991 — an effort that led to John’s actually undergoing a sexual abuse evaluation “which [the records state] was inconclusive” and during which John said nothing against Pare. Such persistence in apparently seeking to provoke John to reveal abuse by Pare, against a background of no specific evidence or accusations of such abuse from October, 1989, through May, 1992, could well be construed to be the product of a biased campaign against Pare by Brummer, who seems to have become particularly close to John’s mother (viewed by [582]*582John and possibly Brummer to have been victimized by Pare) as a result of accompanying John to his monthly visitations with his mother. In the midst of John’s 1992 SITT evaluation, however, Brummer recommended that such visitations should cease (which they did) because the mother was regarded as the source (so the records can fairly be interpreted) of John’s ongoing uncooperative attitude toward the SITT process and its participants, including Matteuzzi. In her reports during that period, Brummer expressed the view that it was necessary to end visitation so that John (and his siblings) “can disclose what is necessary.” Without access to these reports, defense counsel’s effort to mitigate the harmful effect of Brummer’s expert and apparently objective testimony through cross-examination was foredoomed.
In summary, we agree with Pare that the documents reflecting John’s allegations of sexual abuse and his related counseling and treatment were presumptively relevant; further, we find they were actually relevant to his defense. We also agree that it was reversible error to deny his counsel access thereto in accordance with the Bishop protocol; that the Superior Court’s denial of access created a reasonable, if not substantial, risk of an erroneous conviction, see Bishop, 416 Mass, at 17725; and that the defendant is entitled to a new trial. Accordingly, we reverse Fare’s judgments of conviction and set aside the verdicts.
So ordered.