The mother of a minor child appeals from a judgment of the Probate and Family Court, dispensing with the need for her consent to the adoption of her son, arising out of a proceeding sponsored by the Boston Children’s Services Association (BCSA). See G. L. c. 210, § 3 (1986 ed.). The mother raises five issues on appeal: (1) that the probate judge erred in excluding certain evidence which addressed facts already determined in a care and protection case; (2) that the probate judge improperly relied upon the deterioration of the family’s relationship caused by court-ordered separation; (3) that the probate judge’s findings fail to demonstrate consideration of the contribution of the mother’s boyfriend to the family; (4) that the probate judge improperly relied upon certain expert testimony; and (5) that certain findings of fact are unsupported by the evidence. We transferred the case from the Appeals Court on our own motion. We now affirm.
*3In March of 1985, when the boy was approximately three years old, temporary custody of him was awarded to the Department of Social Services (DSS). He was placed in a foster home with his older brother. About a year later, in April of 1986, the Boston Juvenile Court adjudicated the boy in need of care and protection and permanently committed him to DSS, but did not issue its findings until April of 1987, more than one year after its adjudication. The mother did not appeal.1 A motion to file late a notice of appeal, brought by the mother’s new attorney, was denied after hearing. The mother’s counsel then sought review by this court, pursuant to G. L. c. 211, § 3, of the Juvenile Court judge’s refusal to allow her to file a late appeal. Relief was denied by a single justice. No appeal was taken from this denial. See Mass. R. A. P. 15(c), 365 Mass. 860 (1974).
In 1986, BSCA, pursuant to G. L. c. 210, § 3(h), filed a petition to dispense with parental consent to the child’s adoption. The mother objected and the court appointed an attorney to represent her and one to represent the child. The hearing which is the basis of this appeal spanned five days.2 The probate judge found, on January 8, 1988, by clear and convincing evidence, that the mother and the biological father were unfit to assume parental responsibilities, and that the BCSA’s plan for adoption was in the best interests of the child.3 The probate judge allowed the petition and dispensed with the need for the parents’ consent for adoption. The probate judge also found that visitation between the mother and the child was harmful to the child and thus terminated the visits. The mother filed a notice of appeal and a motion to continue visitation pending appeal. The motion was denied. We shall note additional facts as we examine the issues raised by the mother on appeal.
1. Issue preclusion. The mother’s first argument addresses the correctness of the probate judge’s ruling regarding the ex*4tent to which the mother would be permitted to introduce evidence on issues previously addressed by the Boston Juvenile Court judge in his findings of fact. The mother argues that principles of collateral estoppel should not apply in this instance because she was twice denied leave to file a late appeal. See supra at 3.
The probate judge stated that her understanding of the law was that she would have to exclude “ . . . evidence that could have been introduced when those facts were litigated in the lower court.” When the mother’s counsel stated that he would not present evidence because he believed the judge’s ruling precluded it, the judge corrected him on several occasions, emphasizing that he was permitted to present evidence and that she would rule on it as it was offered and consider any objections that were made. The record reflects that the probate judge did just that, and the only evidence that was excluded addressed facts specifically found by the Juvenile Court judge or exhibits introduced in evidence at the care and protection hearing. The mother’s attorney made neither an offer of proof nor indicated in any other manner how the mother’s case was prejudiced by reliance on the Juvenile Court judge’s findings.
In a care and protection hearing, the court’s function is to determine who shall have “custody” of the child. Custody includes the power “(1) to determine the child’s place of abode, medical care and education; (2) to control visits of the child; (3) to consent to enlistments, marriages and other contracts otherwise requiring parental consent.” G. L. c. 119, § 21 (1986 ed.). Because the interest of the child is best served by a stable, continuous environment with his or her own family, the State is justified in interfering only when parents appear unable to provide for the child’s care and protection. Custody of a Minor (No. 1), 377 Mass. 876, 882 (1979). The critical inquiry is “current parental unfitness.” Id. at 883. In a proceeding on a petition to dispense with consent to adoption under G. L. c. 210, § 3, the focus is whether “the natural parents are currently fit to further the welfare and best interests of the child.” Bezio v. Patenaude, 381 Mass. 563, 576 (1980). This analysis also considers whether, in the best interests of the child, the *5legal relationship with the parents should be terminated. This result has been accurately described as an “extreme step.” Petition of Dep’t of Social Servs. to Dispense with Consent to Adoption, 391 Mass. 113, 119 (1984). In both proceedings, a finding of parental unfitness must be supported by clear and convincing evidence. Custody of a Minor (No. 2), 392 Mass. 719, 725 (1984).
The question presented by the mother’s first argument is the extent to which she was bound by and precluded from relitigating, in the consent to adoption proceeding, facts necessarily determined in the care and protection proceeding. Issue preclusion limits a party to one opportunity to try a case on the merits in order to promote judicial economy and protect the soundness of judgments. Home Owners Federal Sav. & Loan Ass’n v. Northwestern Fire & Marine Ins. Co., 354 Mass. 448, 455 (1968). This principle is of particular importance in child welfare cases where public policy encourages prompt and final resolution of custody issues. Custody of a Minor, 389 Mass. 755, 764 (1983). Issue preclusion is not limited to the judgment but also encompasses all facts necessary to reaching it, Olsen v. Olsen, 294 Mass. 507, 509 (1936), as well as those findings treated as essential to the earlier action that resulted from “full litigation and careful decision.” Home Owners Federal, supra at 455.
In Petition of Dep’t of Social Servs. to Dispense with Consent to Adoption, 384 Mass. 707, 710 n.9 (1981), we stated that, if a party fails to appeal from findings issued in a care and protection hearing, then that party will be precluded from collaterally attacking and attempting to relitigate those findings at a subsequent proceeding to dispense with consent to adoption. The findings from the earlier proceeding, we concluded, are binding on the parties. Id. Upon further consideration, however, we are concerned about the potential unfairness that may result from binding parties by findings of fact from the care and protection hearing.
We first note that there is often a time lapse of a year or more between the care and protection hearing and the trial on the petition to dispense with consent to adoption. These find*6ings, we fear, may in some instances be out of date and, although possibly relevant to issues in the subsequent proceeding, they should not be conclusively determinative of the issues. See Aetna Casualty & Surety Co. v. Niziolek, 395 Mass. 737, 745-747 (1985) (fairness to litigants decisive consideration in determining whether collateral estoppel should apply).
Moreover, in Petition of Catholic Charitable Bureau to Dispense with Consent to Adoption, 392 Mass. 738, 742-743 (1984), we noted that different interests attach at each proceeding. For instance, a parent may agree with the State and admit in a care and protection hearing that he or she is unable to provide for the child at that particular time and, in the best interests of the child, that the child should be removed from the parent’s custody. Once, however, a petition is filed under G. L. c. 210, § 3, to dispense with consent to adoption, “the cooperative effort between parent and the intervening agency or individual necessarily ceases, and the relationship between the parent and government acquires an adversary character.” Id. at 743. Accordingly, the parent should not be bound by the earlier findings because those findings did not carry the same finality as the petition to dispense with consent. Because the parents’ incentive to litigate the custody issue and to appeal from findings of the earlier hearing may have been much less, those findings should not be binding on the parties in the subsequent proceeding on the petition to dispense with consent. See generally Restatement (Second) of Judgments § 28 (1982).
Because we now choose not to follow Petition of Dep’t of Social Servs. to Dispense with Consent to Adoption, 384 Mass. 707, 710 n.9 (1981), and determine that the parties should not have been bound by the earlier proceeding, we must now consider the effect of the judge’s ruling in this case. The probate judge incorporated and adopted only twelve of the 118 findings made by the Juvenile Court judge. Some of these twelve findings concerned the child’s behavior which were supported by a report of a physician’s evaluation made before the physician received the findings of the Juvenile Court judge. The balance of the twelve findings dealt with the child’s satisfactory adjustment to the foster parents. In no instance did the probate judge *7bar evidence of the child’s condition or of parental efforts since the time of the Juvenile Court findings. Moreover, the probate judge’s findings were fully supported by evidence introduced at the subsequent proceeding. Accordingly, we conclude that any error resulting from allowing collateral estoppel to apply was harmless.
2. Deterioration of the family relationship. The second issue raised by the mother on appeal is whether the probate judge improperly relied upon the deterioration of the mother-son relationship caused by the child’s placement in a foster home, where visitation with the mother was supervised and limited to a single one-hour weekly session. The mother’s argument is without merit. In Petitions of Dep’t of Social Servs. to Dispense with Consent to Adoption, 399 Mass. 279, 289-290 (1987), we indicated that a lengthy separation between parent and child, the growth of emotional ties between the child and a different custodian, and the effect of a forced return of the child to the biological parents are all factors that should be weighed by the judge, with no one factor being determinative.
The evidence showed that, even from the inception of his foster care, the boy did not ask for his mother or manifest signs of concern about being separated from her. There was evidence that before weekly visits with his mother he would express his hatred for both the mother and the visits. During the visits the child eschewed any contact with the mother and, to avoid interaction with her, pretended to be a dead or sick cat or dog. The record also reflects testimony of social workers present at the visits that the mother suffered from an inability to engage her son in active play or conversation and, because of this inability, that the mother and child actually interacted for about fifteen minutes of the hour-long visit.
Evidence introduced at trial further showed that the boy was anxious to end the visits and anxious to return to his foster mother. If the foster mother was not waiting in the same place, the boy became apprehensive and persistently asked for her.
At the care and protection proceeding, the trial judge found that the child had been emotionally deprived and suffered developmental delays including speech inappropriate to his age. *8The findings from the Boston Juvenile Court also show that the child was hyperactive and self-abusive. The record from the consent to adopt proceeding demonstrated that the child had made significant strides in these areas while in foster care. The court-appointed psychiatrist testified, however, that, if the boy were returned to his mother, his behavior would regress.
All of these factors were properly considered by the judge. There was no error.
3. Contribution by the mother’s boyfriend. The mother next asserts that the probate judge’s findings fail to demonstrate that the judge considered the contribution of the mother’s boyfriend to the family. Generally, the probate judge’s assessment of the weight of the evidence and credibility of the witnesses is entitled to deference, Petition of Dep’t of Social Servs. to Dispense with Consent to Adoption, 397 Mass. 659, 670 (1986), and, absent clear error, should be left undisturbed. Petition of Dep’t of Pub. Welfare to Dispense with Consent to Adoption, 371 Mass. 651, 656 (1976). The mother’s boyfriend testified that he worked at two jobs and was not at home during the day. The mother testified that the boyfriend often worked at night. The judge reasonably could have concluded that the boyfriend was not available to assist in caring for the child.4 There was also evidence to suggest that the boyfriend’s presence did not enhance the mother’s ability as a parent. For example, during weekly visits with the child, the boyfriend’s interaction with the child did not influence the mother’s behavior in engaging the child in play or conversation. In this instance, the findings of the probate judge adequately support the ultimate conclusion.
4. Expert testimony. The fourth issue raised by the mother on appeal is that the probate judge improperly relied upon the expert testimony of the psychiatrist designated by the probate court to evaluate the family. The mother asserts that the probate judge, in reaching her conclusion that the mother was unfit to assume parental responsibility, placed heavy emphasis upon the. *9mother’s “supposed intellectual shortcomings.” The mother further asserts that the expert’s opinion5 was based on conjecture, thus rendering it inadmissible. If the expert’s opinion is excluded, the mother avers, no evidence exists to support the probate judge’s conclusion. Again, the mother’s argument is without merit. The expert testified that the basis for his opinion was the mother’s “presentation” during the evaluation. Specifically, the expert testified that the mother had minimal understanding as to why she was involved in the proceedings, minimal ability to conceptualize her son’s state of mind, and was slow to perceive events that happened around her.
Mental disorder is relevant only to the extent that it affects the parents’ capacity to assume parental responsibility, and ability to deal with a child’s special needs. Petition of Dep’t of Social Servs. to Dispense with Consent to Adoption, 392 Mass. 696, 701 (1984). Parental fitness as to one child does not render a parent fit with respect to a different child if the other child has special needs. Richards v. Forrest, 278 Mass. 547, 553 (1932). There was no error in the admission of the expert testimony.
In addition to the evidence previously discussed regarding the mother’s inability to engage the child on a consistent basis in play or conversation, at least two witnesses testified that during the weekly visits the mother could not focus on more than one child at a time.6 Moreover, those witnesses testified that the mother could not set appropriate limits to control the child’s behavior during the one-hour visits. Evidence was admitted in the consent to adoption proceeding that the boy was hyperactive and self-abusive.7 The record also reflects testimony that the mother provided no positive reinforcement for good behavior, could not maintain eye contact with the child, and did not nurture the child in any way. The judge did not err.
*105. Findings unsupported by the evidence. The final issue raised by the mother is that certain findings of fact are unsupported by the evidence and that, absent this evidence, the judge’s ultimate conclusion as to the mother’s inability as a parent could not stand.
First, the mother challenges the finding that the child is hyperactive and tends to be aggressive and self-abusive. The findings of the Boston Juvenile Court judge, as well as testimony from the termination proceeding, support this determination. Despite progress made with the foster parents as evidenced by the boy’s calmness before visits with his mother, the record reflects testimony that the boy regressed to hyperactive and aggressive behavior during and after the visits. The child would throw toys and spit at and kick his mother. He would struggle to be free from unwanted embraces from his mother and either ignore her or act defiantly towards her. One social worker testified that after visits with the mother the boy was hyperactive and extremely anxious to be reunited with the foster parent. Accordingly, there is adequate evidence to support the judge’s conclusion.
The mother next challenges the judge’s finding that she suffers from a continual inability to integrate, at home, parental skills taught to her in early intervention programs. While the mother quotes portions of the transcript to support this argument, at least two witnesses testified that, despite their visits to the mother’s home, they could not actually testify as to whether the mother was integrating these skills at home. One witness, however, did testify that although she had seen toys appropriate to the child’s age in the mother’s apartment, the mother was not using them in an appropriate manner to stimulate her daughter. Some of the evidence favoring the mother on this point was offered by a social worker who maintained an ongoing relationship with the mother that might be harmed by harsh testimony. The trial judge was in the best position to assess the credibility of the witnesses’ testimony and determine the weight afforded it. Petition of New Bedford Child & Family Serv. to Dispense with Consent to Adoption, 385 Mass. 482, 488-489 (1982). The record supports the judge’s conclusion as to this finding.
*11The third finding challenged by the mother is that she has made no progress in providing a safe environment for her children. The ability to provide a safe environment appears to encompass the mother’s ability to protect the children from physical harm by preventing them from reaching and ingesting poisonous fluids in the household and to protect them by setting limits and establishing a daily routine in the home. Conflicting evidence was given on each aspect. One witness testified that there were no locks on the cabinets to prevent access to cleaning fluids. The mother’s boyfriend testified that they had installed locks on the cabinets. One social worker testified that she often had to set limits for the boy during the visits when the mother failed to do so, for example, by instructing him not to throw toys. Another social worker testified that the mother had made some progess in providing a safe environment. Ample testimony existed demonstrating that the concern for a safe home environment still exists today. The trial judge is in the best position to assess the credibility of the witnesses and we again determine that the record supports the judge’s conclusion as to this finding.
Finally, the mother challenges the finding that her three year old daughter, presently living with her and her boyfriend, suffers from developmental delays. While the mother relies on expert testimony of the psychiatrist who performed the court-ordered evaluation to support her position, she ignores testimony provided by other witnesses. One social worker, for example, testified that the daughter’s gait was unstable, that she had tantrums, and that she suffered language deficits. Another testified that the daughter was clumsy and had problems articulating and planning movement. This witness also testified that the daughter had problems with gross motor skills such as spatial relations and hand-eye coordination. This witness testified that the daughter spoke in sentences of two or three words. A third witness testified that at two and a half years of age, the daughter’s speech was still unintelligible. The judge’s findings, therefore, are supported by the evidence presented at trial.
*126. Conclusion. Although we decide today that parties should not be bound by findings of fact determined in care and protection hearings that may be out of date, or the product of a proceeding where the parent may not have a compelling incentive to litigate, we believe that those findings should not be completely removed from the judge’s consideration in the later proceeding in determining the parents’ current unfitness. The findings from the care and protection proceeding may be introduced in evidence at the trial on a petition to dispense with consent to adoption to the extent they are both relevant and material. Sufficient evidence existed on the record from the consent to adoption proceeding to demonstrate that neither the mother’s attitude toward child rearing nor her actions in interacting with the child had changed between the time of the care and protection hearing and the proceeding on the petition to dispense with consent. Moreover, the evidence also showed a weakening of the mother-son relationship, the growth of a relationship between the son and the foster parents, and that returning the child to the mother would not be in the child’s best interest. Accordingly, despite our conclusion that the parties should not be bound by the findings in the earlier proceeding, we see no reason to disturb the grant of permission to dispense with consent to adoption.
Judgment affirmed.
The mother was represented by different counsel in the consent to adoption proceeding.
June 25, 29, and 30, 1987, and September 14 and 15, 1987.
The child’s biological father did not participate in this proceeding or in the earlier care and protection proceeding.
The Boston Juvenile Court judge, in the care and protection proceeding, considered the boy Mend’s role and found him unfit to assume parental responsibility for the child. The boyfriend is not a party to this action.
The expert testified that his impression was that the mother has “borderline intellectual ability” and, although psychological testing had not been completed, it was his opinion that she would score “somewhere in the range between normal and retarded.”
The mother’s three year old daughter was often present at the visits.
The record does reflect, however, that, with specialized foster care, the child’s behavior was improving.