The Carroll County Juvenile Court terminated the parental rights of Laura J. (the biological mother) to T. J. and to R. G. and terminated the parental rights of R. G.’s father, Steve G. 1 Both Laura and Steve appeal, contending that the evidence was insufficient, that the court improperly admitted certain evidence, and that the court failed to consider placing the children with relatives. As the cases are closely related, we have consolidated the appeals. We hold that the evidence was sufficient, that the contested evidence was either admissible or harmless, and that the appellants waived any review about placing the children with relatives. Thus, we affirm.

1. In a parental rights termination case, the question for the appellate court is whether, under the evidence as construed most favorably to the court’s findings, a rational trier of fact could have found clear and convincing evidence showing (a) parental misconduct or inability and (b) that termination was in the children’s best interests.2 Parental misconduct or inability is shown if (i) the children are deprived, (ii) lack of parental care caused the deprivation, (iii) the cause of such deprivation is likely to continue, and (iv) the continued *92deprivation is likely to cause serious harm to the children.3

The evidence showed that one cold November day in 1997, a caseworker with the Carroll County Department of Family & Children Services (DFACS) found six-year-old T. J. and four-year-old R. G. under the care of an eighteen-year-old half-brother in a home without heat, electricity, or food. The half-brother stated that he did not know where Laura and Steve were, that they had been gone a few days on a trucking trip, that he did not know when they were coming back, and that he had no money or means to care for the children. DFACS took the children into custody and was not contacted by either parent until December 1, over one week later, at which time Laura was unable to give an address for further contact. DFACS set an appointment with the parents to develop a reunification plan. A few days later a deprivation hearing was held, but only Steve attended, explaining that Laura was off her medication and therefore unpredictable. At Steve’s request, the court continued the hearing so that he could obtain counsel.

Neither parent appeared at the December 17 appointment to develop a reunification plan, so DFACS proceeded without them. The first goal concerned the mother’s mental health, as she had a history of emotional instability arising out of a bipolar disorder or a recurring depressed mental state to the point of psychosis. She often did not take her medication. She was to undergo a psychological evaluation, stay on any recommended medications, and maintain emotional stability. Second, she was to complete parenting classes and have consistent bi-weekly visitations with her children. Third, she was to maintain stable and secure housing. Since Steve had not legitimated R. G., he was to legitimate her before a case plan could be developed for him.

On January 7, Laura and Steve met with DFACS and reviewed the goals of the plan. The parents missed a scheduled visitation because they arrived two hours late. They became extremely hostile and abusive and refused to leave until law enforcement personnel removed them. Laura immediately returned, again requiring the presence of law enforcement to protect DFACS personnel. She then made abusive phone calls to DFACS personnel, threatening them and their children.

The next day the parents appeared at the deprivation hearing, which was again postponed at their request. The court later reviewed and incorporated the case plan and ordered the parents to comply therewith. At the postponed deprivation hearing in April 1998, the parents failed to appear, but their counsel did attend, at which time *93the court found clear and convincing evidence that the children were deprived as a result of parental inability and that the parents were uncooperative and had not met the goals of the case plan. The court reemphasized the need for the parents to comply with the case plan, which it appended to its order. The parents did not appeal this order.

DFACS again reviewed the case plan with Laura, and when the parents failed to comply, a panel review was held in May 1998, which the parents did not attend, but Laura’s counsel did. The panel found the parents had failed to comply with the case plan goals and recommended termination. The court in June 1998 once again ordered the parents to comply with the plan and attached a copy of the plan and panel review to its order. Notably, at this point Steve was still not formally part of the plan, as he had not yet made any effort to legitimate R. G.

To determine whether to renew custody to DFACS, the court held a second deprivation hearing in October 1998, which both parents and their counsel attended. The court again found clear and convincing evidence of deprivation caused by parental inability, which order the parents did not appeal. In fact, the parents consented to the renewal of DFACS’ custody. In October the court also entered an order announcing that DFACS intended to seek termination of parental rights.

At this point, the parents suddenly became cooperative in complying with most of the case plan. For the first time they began visiting the children on a consistent basis, and Steve filed to legitimate R. G. They began and completed parenting classes. But they did not obtain stable and secure housing. DFACS petitioned to terminate their rights in January 1999, which petition the court granted after a hearing in July 1999.

(a) Deprivation and Cause. Neither parent appealed the two deprivation orders entered in April and October 1998, which found that the children were deprived as a result of the parents’ inability to care for them. Thus, the determinations are binding on the parents and satisfy the first and second factors.4

Moreover, the evidence showed that for a year or longer prior to the filing of the termination petition, Laura and Steve had failed to develop a parental bond with the children in a meaningful, supportive manner and further failed to comply with a court-ordered plan of reunification, both of which circumstances manifest that the children were without proper parental care and control.5 Regarding parental *94bond, the parents failed to contact DFACS until a week after the children were taken into custody. Over the next eleven months they made thirteen appointments to see T. J. but did not show six times and were one to three hours late on the other seven, resulting in five cancelled visits and in only two brief visits with T. J. They did not have a single on-time full visit with T. J. until November 1998, one year after DFACS had taken custody and shortly after termination proceedings had been announced.

Similarly, they did not contact DFACS about visiting R. G. until March 1998, four months after DFACS took custody in November 1997. Because of their cancellations and tardiness, they had no on-time full visit with R. G. until July 1998, which was two months after the May panel review recommended termination.

Regarding case reunification plans, the evidence showed that Laura failed to obtain a psychological evaluation until five months after the case plan instructed that she do so; she exhibited emotional instability by threatening and abusing caseworkers and by showing up, against instructions, at R. G.’s foster home and getting arrested there; she and Steve did not begin parenting classes until eleven months after the institution of the case plan and until after termination proceedings had begun; Steve did not seek to legitimate R. G. until that same time period; they failed to make the bi-weekly visits to the children; and they failed to obtain stable, secure housing, having lived in five different places (including the cab of a truck) since the plan was devised, with the most recent residence posing an outside safety hazard to the children owing to debris.

This evidence, combined with the evidence set forth earlier in this opinion, is sufficient to sustain findings on the first two factors.

(b) Likely to Continue. “Evidence of past parental conduct resulting in deprivation may be considered in determining whether the deprivation is likely to continue and cause harm to the children.”6 Here the parents’ repeated failure to visit their children and consistent failure to meet the case plan goal of obtaining stable housing authorize the juvenile court’s determination that deprivation would likely continue. Their admirable improvements since DFACS began seeking termination do not demand a finding that the deprivation would not continue. In the Interest of L. S. D. explained:

[A] court may find that a parent’s conduct over the years is a better predictor of future conduct. A few months of partial stability do not establish that the parent is capable of maintaining the progress. In fact, improvement that takes place *95only after the termination petition is filed is often unconvincing. . . . What weight to give recent improvements is a question for the trier of fact. Thus, in considering recent improvement, the trial court, not the appellate court, determines whether a parent’s conduct warrants hope of rehabilitation.7

(c) Serious Harm to Children. Sufficient evidence showed that the continued deprivation would cause serious mental and emotional harm to the children. T. J. is emotionally disturbed and has a bipolar disorder requiring a stable environment to avoid deterioration and in which to receive a complex regimen of potentially toxic medications consistently and regularly. A great amount of structure, supervision, and consistency is mandated to stabilize T. J.’s propensity to violence and uncontrolled behavior. Since being placed in a structured, stable foster home, T. J. is less violent and is thriving. On the other hand, Laura’s repeated failures for months to appear for scheduled visits deeply disappointed T. J., and her later visits ended up upsetting the child greatly.

Similarly, when after eight months Laura and Steve finally visited R. G., the child responded negatively and separated from them easily. R. G. responded negatively to their other visits also. R. G. is now receiving counseling for special needs. R. G. has done so well in foster care that Steve even requested that the foster parents adopt the child if R. G. cannot be returned to him.

(d) Best Interests. Sufficient evidence supported the court’s findings that termination was in the best interests of the children, considering their physical, mental, emotional, and moral condition and needs, particularly their need for secure and stable homes.8 Those same factors proving the parents’ inability to properly raise their children may also show that termination is in the children’s best interests.9 Here the progress of the children in foster care combined with their special needs and their negative reactions to their parents authorizes the court’s determination that termination was in their best interests.10

2. Laura and Steve complain that the trial court erred in admitting the case plans into evidence. They argue that because they were not served with summons, process, and notice for the deprivation actions in which these case plans were approved, the case plans were *96inadmissible. Setting aside the fact that there is no requirement that case plans are only admissible if approved at a properly noticed deprivation hearing, we hold that in any case the matter has been waived. Both parents’ counsel appeared at the two deprivation hearings, and the parents themselves appeared at the second hearing. Neither counsel nor parents objected or otherwise raised the issue of proper notice or service at the deprivation hearings, thereby waiving this issue.11

3. The parents contend that the court erred in admitting a single court-ordered home evaluation into evidence. They argue that under OCGA § 15-11-12 (a), until an adjudication of deprivation is made, the court may not order a home evaluation to be made. They further argue that under OCGA § 15-11-56 (c), such an evaluation is admissible only in dispositional hearings, not in hearings on termination petitions.

First, under what circumstances a court under OCGA § 15-11-12 (a) may order a home evaluation is irrelevant in determining the admissibility of a home evaluation. Moreover, here the court had twice previously adjudicated in unappealed orders that the children were deprived.

Second, In the Interest of M. L. P.12 held that under OCGA § 15-11-56 (c) the juvenile court properly considered certain panel reports in a termination proceeding, even though they were hearsay. M. L. P went on to hold that even if inadmissible hearsay, the reports’ admission did not constitute reversible error where other evidence sufficed to support the findings of the trial judge.13 “When a trial judge considers both admissible and inadmissible evidence, it is presumed that he separates the wheat from the chaff. . . .”14 Here the other evidence of parental inability amounting to unfitness as set forth above sufficed to sustain the termination order.

4. The parents claim that the juvenile court erred when it did not make or require a search for a suitable family member or relative for placement of the children, as required by OCGA § 15-11-103 (a) (1). But neither parent raised this issue below, thereby waiving the matter for appellate review.15

Moreover, even if the issue were preserved for appellate review, the court heard evidence on why available relatives were unacceptable and even invited further evidence on the matter if any parties so desired. Inasmuch as there is no conclusive preference given to relatives,16 the court retained a wide discretion to determine whether to keep the children in their stable foster homes or to place the children with a relative. Under the “any evidence” standard of review,17 we discern no abuse of that discretion here.

Judgments affirmed.