Shannon and Danny NELSON, Individually, and on Behalf of E.N. F/K/A E.N., a Minor, Appellees, v. Lynn M. LINDAMAN, Lynn M. Lindaman, M.D., P.L.C. D/B/A Lindaman Orthopaedic, and Mercy Medical Center—Des Moines, Appellants

APPEL, Justice

(dissenting).

I respectfully dissent.

I first review the factual record presented in the defendants’ motion for summary judgment, making all inferences favorable to the plaintiffs as the nonmoving party. 6 Second, I review the relevant Iowa statutes. In particular, I note the distinction in the immunity provisions of Iowa Code section 232.73 (2009) between the first prong of the statute, involving mandatory reporting, and the second prong of the statute, which extends immunity to those “aiding and assisting in an assessment of a child abuse report” made to the Iowa Department of Human Services (DHS). Third, I examine the manner in which Iowa courts and other jurisdictions have handled motions for summary judgment involving immunity statutes. Finally, I apply principles gleaned from the previous discussion to the unique facts of this case. As will be seen below, I conclude the trial court correctly denied summary judgment in this ease.

I. Overview of the Summary Judgment Record Viewed Most Favorably to the Plaintiffs.

E.N. was brought to the hospital by his parents on June 18, 2009, with a broken arm. His injury was a spiral fracture. A spiral fracture of the bone runs at an angle through the bone rather than evenly across it. Spiral fractures usually require a twisting force to occur.

When asked at the emergency room how the injury occurred, E.N.’s father stated that he was putting E.N. on his bed, that the child put his arm behind his back, and that E.N.’s weight caused the bone to snap. At the time, E.N. was a twenty-two-day-old infant, who weighed eight pounds, eleven ounces.

Dr. Scott Barron, the emergency room physician who first examined E.N., suspected nonaccidental injury. Dr. Barron told E.N.’s father that he was required to report the fracture to DHS. E.N. was admitted to the hospital under the care of Dr. Douglas Selover, who also suspected nonaccidental injury.

Dr. Selover contacted Dr. Lynn Linda-rían, a pediatric orthopedic surgeon, to provide consultation with respect to E.N.’s fracture. When asked by a DHS child protective worker about E.N.’s injuries the following day, Dr. Lindaman responded that the father’s story of how the injury occurred was “plausible.” A contemporaneous DHS record describes the conversation as follows:

*17 This worker questioned as to whether a child, weighing only 8 lbs 11 oz, would have enough force to create this injury. I also provided information that dad had provided a different explanation with how he laid [E.N.] down, with one hand under his head and the other under his butt. I also questioned as to whether a crying child’s arm would go back behind him as he would more likely to be pulling his arms tight in front of him. Through this line of questioning, he stated on several occasions, “the mechanism they described fits the fracture seen.”

When informed of Dr. Lindaman’s statement that the father’s story of how the injury occurred was “plausible,” Dr. Sel-over exclaimed, “You got to be kidding.” Dr. Selover talked with Dr. Lindaman about the cause of the injury during E.N.’s initial hospitalization. At that time, Dr. Selover expressed his view that it was a pretty clear-cut case of nonaccidental injury.

While DHS staff continued to be suspicious of the injury, Dr. Lindaman’s repeated statements that the father’s story was “plausible” and that “the mechanism” described by the father “fits the fracture,” caused them pause. Prior to talking to Dr. Lindaman, DHS planned to seek a no-contact order against E.N.’s father. After talking with Dr. Lindaman, the DHS child protective worker consulted her supervisor. She was concerned the orthopedic specialist treating E.N. had repeatedly emphasized that the mechanism described by the father “fit the injury.” They interpreted his comments as an opinion not supporting the presence of child abuse. As a result, the decision was made not to seek a no-contact order at that time. DHS staff, however, informally urged the mother not to allow E.N. to be with his father alone, but no further immediate action was taken at the time of E.N.’s discharge from the hospital.

DHS continued to have concerns about E.N., however, and the case was presented to a multidisciplinary team for review. The team included five physicians with experience in evaluating cases of child abuse. At a meeting on June 30, all agreed the injury could not have occurred as described by the father. One of the participants, Dr. Resmiye Oral, requested medical records for further review by orthopedic specialists to confirm the unanimous view of team members. Dr. McAu-liff, another physician, was dispatched to confer with Dr. Lindaman.

On July 6, the evaluators at the University of Iowa sent an email advising that the injury could not have happened as indicated by E.N.’s father. On July 7, DHS began working with the county attorney to file a no-contact order.

On July 8, Dr. McAuliff shared with the multidisciplinary team that he had spoken with Dr. Lindaman. Dr. Lindaman was more forthcoming with Dr. McAuliff than he had been with DHS staff earlier in the case. Dr. Lindaman told Dr. McAuliff that he had not seen many infants in his practice and had never seen this type of injury before. In light of his discussion with Dr. McAuliff, Dr. Lindaman agreed that the injury was suspicious.

Unfortunately, on July 8, before DHS served the no-contact order, E.N. arrived at the hospital with head trauma and other very serious injuries.

In June of 2011, the plaintiffs filed suit naming Dr. Lindaman, his professional corporation, and Mercy Medical Center-Des Moines as defendants. A deposition of Dr. Lindaman was part of the summary judgment record. At the deposition, Dr. Lindaman took a very narrow view of his *18 professional responsibilities and the nature of his discussions with DHS.

Dr. Lindaman took the position that his job was the management of the fracture and the concerns of Dr. Barron, that the trauma may have been nonaccidental, was not his concern because it did not impact his management of the fracture. Dr. Lin-daman stated that he did not explore whether the injury was in fact consistent with the father’s explanation because “that’s an investigative function through DHS or to the police, not [ ] medical.” He further stated that “to investigate whether the mechanism happened the way dad explained, that’s not a medical investigation. That’s a legal or criminal investigation.” Dr. Lindaman testified that in his interaction 'with DHS, “[h]e was not providing information on the safety of the child. [He] was providing information only on the humerus fracture.” According to Dr. Lin-daman, his statement to DHS was “merely on [his] orthopedic evaluation of [E.N.]. [H]e was not speaking with them on judging what happened at [E.N.] ’s home.”

Dr. Lindaman also testified in his deposition that while the father’s story “could be consistent” with the injury, it would not commonly occur when putting the child down and would be “a rare kind” of injury. He stated in his deposition that if he had been the emergency room physician on the day E.N. arrived, he too would have reported the injury to DHS the way Dr. Barron did.

Dr. Lindaman also filed an affidavit in connection with the “Motion for Summary Judgment as to the Lindaman Defendants.” In that affidavit, Dr. Lindaman stated that the only opinion he developed was that “if the history the father was providing to [him] was true, that history could possibly be consistent with the type of spiral humeral fracture [he] observed in this child.”

He further stated that he was aware of the opinions of Dr. Selover and Dr. Barron suspecting child abuse prior to his conversation with DHS. He stated he was not surprised that the child abuse investigators seemed to have concluded that E.N.’s fracture was due to child abuse “since spiral humeral fractures in non-ambulatory children are rare.” He also asserted he would have reported the incident to DHS if the case had been reported to him in the first instance.

The records of Dr. Lindaman’s conversations with DHS, however, do not indicate that he advised DHS that he took a very narrow view of his responsibilities and that he was not “judging [assessing?] what happened at E.N.’s home.” He did not advise DHS that spiral humerus fractures in no-nambulatory children are rare, that he would have reported the incident had E.N. been presented to him in the first instance, or that his opinion was limited solely to the biomechanics of the possibility of a fracture occurring if the story told by E.N.’s father were true. He simply repeatedly told DHS that the story was “plausible” and the “mechanism” described “fit the injury.”

The record in the proceedings related to the motion for summary judgment contained the report of one of the plaintiffs’ experts, Dr. Geoffrey Miller, an orthopedic surgeon. In reviewing the file, Dr. Miller stated that while it was possible Dr. Linda-man may have simply made an oversight in his initial opinion,

the orthopedic surgeon did not make the diagnosis of non-accidental trauma in his consult, even as a possibility even though he acknowledged an ongoing workup for that diagnosis. This could be an oversight, but his decisions after-wards make this explanation tenuous at best. It does not appear to have been an oversight with the repeated opportu *19 nities to modify the diagnosis after meeting with DHS and other treaters, as well as his deposition testimony ... where he specifically disagreed with other doctors.

In a supplemental report, Dr. Miller characterized the failure of Dr. Lindaman to detect rib trauma in E.N. as “further evidence of this■ doctor’s inexplicable and' stunning disregard for the suspected child abuse diagnosis made by both of the other treaters.” Another of the plaintiffs’ medical experts opined that Dr. Lindaman “obtained a history that makes no sense as a reasonably certain medical explanation for a cause of a spiral fracture in a 22 day old infant.”

The district court held that there were genuine issues of material fact as to whether the defendant doctor rendered an opinion or not for DHS, whether reliance on that opinion caused injury to the child, whether the doctor’s communications to DHS were in good faith or not, whether the doctor’s conduct was actually aiding or assisting in a child abuse assessment, and whether the doctor’s conduct was entitled to immunity for his conduct. We granted the defendants leave to file an interlocutory appeal.

II. Iowa’s Child Protection Reporting and Immunity Regime.

A. History of Concern Regarding Failure to Report Suspected Child Abuse. For many years, underreporting of child abuse by medical professionals has been recognized as a significant problem. Concern about participation of medical professionals in the child abuse reporting system continues notwithstanding the passage of mandatory child abuse reporting statutes. As noted by one commentator, “fear of legal action is frequently a reason for not reporting.” Marjorie R. Freiman, Note, Unequal and Inadequate Protection Under the Law: State Child Abuse Statutes, 50 Geo. Wash. L. Rev. 243 , 263 (1982). According to an article in the prestigious journal of the American Medical Association, physicians sometimes do not wish to get involved in child abuse reporting situations,. despite the fact that statutes mandate such actions. See John M. Leventhal, The Challenges of Recognizing Child Abuse: Seeing is Believing, 281 J. Am. Med. Ass’n 657, 658 (1999).

According to yet another commentator: [Rjecent studies reveal that physicians admit that they do not report all suspected cases of child abuse and neglect. They offer several justifications for this noncompliance. The most common explanations are concerns about the way child prgtection agencies handle reported cases and beliefs that state involvement ■ often . does not help the child. Some physicians publically admit that they do not want to get involved with the legal system, a sentiment probably held privately by many physicians.

Ellen Wright Clayton, To Protect Children from Abuse and Neglect, Protect Physician Reporters, 1 Hous. J. Health L. & Pol’y 133, 140-41 (2001) (footnotes omitted).

B. Overview of Iowa Statutory Framework. Part two of division III of Iowa Code chapter 232 addresses child abuse reporting, assessment, and rehabilitation. See Iowa Code §§ 232.67 -.77. Iowa Code section 232.67 provides explicit legislative findings. Under this Code provision, the legislature declared “[cjhildren in this state are in urgent need of protection from abuse.” Id. § 232.67. The legislature further stated the purpose of the statutory provisions was “to provide the greatest possible protection to victims or potential victims of abuse through encouraging the increased reporting of suspected cases of abuse, [and] ensuring the thor *20 ough and prompt assessment of these reports.” Id.

In order to achieve the legislative purpose, chapter 232 part two establishes a system of mandatory and permissive reporters of child abuse, a reporting procedure, and a structure for investigation of reports by DHS. Id. §§ 232.69-.77. Knowing and willful violations of reporting obligations are a simple misdemeanor, id. § 232.75(1), as is the knowingly false reporting of child abuse, id. § 232.75(3). A person who knowingly fails to report or interferes with mandatory reporting is civilly liable for damages proximately caused by such failure or interference. Id. § 232.75(2).

While the statute thus imposes affirmative obligations on mandatory reporters, it also contains an immunity provision which is at the heart of this appeal. The immunity provision in Iowa Code section 232.73(1) provides that

[a] person participating in good faith in the making of a report ... or aiding and assisting in an assessment of a child abuse report ... shall have immunity from any liability, civil or criminal, which might otherwise be incurred or imposed.

The statute has two classifications for immunity. The first prong protects persons who “make a report” of child abuse under Iowa Code chapter 232. Id. § 232.73(1). Dr. Lindaman plainly does not fall in this category. The statute also provides, however, that immunity extends to persons who in good faith are “aiding and assisting in an assessment of a child abuse report” by DHS. Id. It is the second prong of the statute that is implicated in this case.

C. Applicable Iowa Caselaw. We have had a few occasions to interpret the immunity provision of Iowa Code section 232.73. The first case is Maples v. Siddiqui 450 N.W.2d 529 (Iowa 1990). In that case, the plaintiffs filed a medical malpractice action against the defendant, Dr. Sid-diqui, claiming their son was placed in foster care because of an improper diagnosis of the cause of the child’s malnutrition. Id. at 529 . The plaintiffs sought to recover for their loss of companionship as a result of the removal of their son from their home. Id. In this case, we held that Dr. Siddiqui, on the facts presented, was entitled to immunity. Id. at 530-31 .

We rejected the notion that the immunity statute did not apply because the action was a medical malpractice action. Id. at 530 . Instead, we focused on “the causal theory of plaintiffs’ loss-of-companionship claim.” Id. We noted the claim was tied to the court-ordered placement “[i]rrespec-tive of the other elements of damage which might have resulted from defendant’s improper diagnosis.” Id. Clearly, Maples does not stand for the proposition that there can be no recovery in a medical negligence claim where a report of child abuse is involved. The immunity in Iowa Code section 232.73 where a good-faith report has been made extends only to the extent that the plaintiffs claim for damages is causally tied to the report itself. See id.

In Maples , we also considered whether alleged negligence is sufficient to defeat the good-faith requirement of the immunity statute. Id. We concluded that a showing of negligence does not defeat good faith. Id. If negligence alone was sufficient to defeat good-faith immunity, we reasoned, the immunity statute would be deprived of its bite. Id. We did not hold, of course, that immunity applies to all cases where negligence was involved, but only that a showing of negligence was not sufficient to deprive a defendant of an immunity defense if good faith under one of the prongs of the immunity statute could be established. See id.; cf. Whaley *21 v. State, 90 Wash.App. 658 , 956 P.2d 1100, 1106 (1998) (holding immunity in child abuse reporting statute extends only to damages caused by the making of a child abuse report).

Our second case dealing with the immunity provisions of Iowa Code section 232.73 is Garvis v. Scholten, 492 N.W.2d 402 (Iowa 1992). In Garvis , the plaintiff asserted the defendants improperly disclosed certain confidential medical information in the course of a child abuse investigation. Id. at 402 . The fighting issue in the case was whether the good-faith standard in the statute was objective or subjective. Id. at 403-04 .

We held the standard for good faith was subjective. Id. We declared “[g]ood faith in section 232.73 rests on a defendant’s subjective honest belief that the defendant is aiding and assisting in the investigation of a child abuse report.” Id. at 404 . We quoted a case from the commercial context noting that good faith means only honesty in fact, colorfully described as including situations involving ‘“the pure heart and empty head.’ ” Id. (quoting Jackson v. State Bank of Wapello, 488 N.W.2d 151, 156 (Iowa 1992)). Because the “subjective good faith in aiding and assisting the investigation went unchallenged,” we declined to disturb the district court’s ruling sustaining the defendants motion for summary judgment. Id.

The takeaway points from Maples and Garvis are important but narrow. First, Maples establishes that the mere presence of negligence is plainly insufficient to defeat immunity. 450 N.W.2d at 530-31 . Second, Maples stands for the proposition that immunity applies to damage claims causally related to the reporting or aiding and assisting in an assessment of a child abuse report or a child abuse investigation. Id. Third, Garvis held that the standard for evaluating the making of a report or aiding and assisting a child abuse investigation is “subjective honest belief’ in making a report or in “aiding and assisting in the investigation of a child abuse report.” 492 N.W.2d at 404 . In neither of these cases did we address the question of the proper standards to be applied in a motion for summary judgment based on the immunity provision. To that I now turn.

III. Standards for Summary Judgment of Immunity Claims.

Courts considering immunity defenses in the context of motions for summary judgment have taken a variety of approaches. In some cases, courts have determined that immunity issues should be decided by the court in advance of trial in order to achieve the policy purposes that underlie immunity. See, e.g., May v. Se. Wyo. Mental Health Ctr., 866 P.2d 732, 738-39 (Wyo.1993). At the other extreme, some courts have held that questions of subjective good faith always involve questions of fact. See, e.g., de Abadia v. Izquierdo Mora, 792 F.2d 1187, 1191 (1st Cir.1986) (noting that on the “issue of subjective good faith, there might always be a question of fact [and that] it is difficult to think there could ever be summary judgment”); Sabia v. Neville, 165 Vt. 515 , 687 A.2d 469, 473 (1996) (rejecting a subjective good-faith standard because “a material issue of fact would always be present, precluding summary judgment” (internal quotation marks omitted)). In between the two poles, some courts have employed a shifting burden of production approach where once a defendant makes a prima facie case for immunity, the burden shifts to the plaintiff to produce at least some evidence from which an inference of lack of good faith can be drawn. See, e.g., S.G. v. City of Monroe, 843 So.2d 657, 662 (La.Ct.App. 2003).

*22 Even when summary judgment for the defendant is not precluded in subjective good-faith immunity cases, however, the courts recognize there is rarely direct evidence of subjective good faith, and as a result, reasonable inferences that can be drawn from circumstantial evidence are sufficient to generate a fact question on the issue. See United States v. Sullivan, 406 F.2d 180, 186 (2d Cir.1969) (noting intent is rarely susceptible of direct proof and must be established by legitimate inferences from circumstantial evidence); Van Nattan v. United States, 357 F.2d 161, 162 (10th Cir.1966) (intent is seldom shown by direct evidence and “in most cases must be proved by inference from the facts and circumstances of the particular case”); 7 Synthon IP, Inc. v. Pfizer Inc., 472 F.Supp.2d 760, 779 (E.D.Va.2007) (noting that intent rarely can be proven by direct evidence).

In Iowa, we have not yet considered the proper approach to summary judgment when the plaintiffs contest the defendant’s claim of good-faith immunity under Iowa Code section 232.73. In Maples , the issue was not how to approach the question of subjective good faith in the context of summary judgment, but only whether the presence of negligence defeated immunity under the statute. 450 N.W.2d at 530 . In Garvis , although summary judgment was granted, no one contested the subjective good faith of the defendants and thus the result in the case does not help us in this particular situation when subjective good faith is contested. 492 N.W.2d at 404 .

We have, however, applied our summary judgment framework in other immunity contexts when subjective good faith has been at issue. In Hlubek v. Pelecky, 701 N.W.2d 93, 94 (Iowa 2005), we considered whether area education agency (AEA) officials were entitled to summary judgment in a ease in which the plaintiff charged they tortiously interfered with his contractual and prospective business relations and intentionally inflicted emotional distress by' investigating charges of sexual abuse. The applicable statutes provided immunity for AEA personnel who participated in good faith and acted reasonably in such investigations. Id. at 96-97 (citing Iowa Code sections 280.27 and 613.21 (2001)). We held the evidence showed the defendants had acted in good faith and the plaintiff “ha[d] presented no contrary evidence on the issue.” Id. We applied a similar approach in Green v. Racing Association of Central Iowa, 713 N.W.2d 234, 245-46 (Iowa 2006).

If the approach in Hlubek and Green were applied under Iowa Code section 232.73, a declaration of subjective good faith by a defendant might ordinarily be sufficient to require the plaintiff to produce evidence from which legitimate inferences of lack of good faith could be shown. Yet, even under this type of approach, as observed in a case cited by the majority, questions of subjective good faith are “ordinarily” not resolvable upon summary judgment. 8 Rite Aid Corp. v. Hagley, 374 *23 Md. 665 , 824 A.2d 107, 119 (2003) (internal quotation marks omitted); 9 see also Miller v. Dep’t of Corr., 36 Cal.4th 446 , 30 Cal. Rptr.3d 797 , 115 P.3d 77, 97 (2005) (noting “issue of a plaintiffs subjective; good faith belief involves questions of credibility and ordinarily cannot be resolved on summary judgment”); Hoefer v. Wis. Educ. Ass’n Ins. Trust, 470 N.W.2d 336, 340 (Iowa 1991) (whether statements were expressions of insincere opinion intended to deceive or mislead was “ordinarily a jury question” (internal quotation marks omitted)).

The consequence of the HlubeJc-Green type approach is that subjective good faith is determined through an examination of the circumstances in each particular casé, and “proof of intent or state of mind is rarely established as fact by direct evidence, but may be inferred from the facts regarding the individiiars actions or other circumstances.” S.G., 843 So.2d at 662 . In the summary judgment context, of course, all legitimate inferences are made in favor of the nonmoving party. Wallace v. Des Moines Indep. Cmty. Sch. Dist. Bd. of Dirs., 754 N.W.2d 854, 857 (Iowa 2008).

IV. Application of Principles.

I now turn to the application of' the above principles to the facts of this case. There is no question under our caselaw, and under the majority of caselaw from other jurisdictions, that the mere fact that Dr. Lindaman may have been negligent in his evaluation of E.N. is not sufficient to escape the application of the immunity provided by Iowa Code section 232.73. See Garvis, 492 N.W.2d at 404 . And, while it is true the immunity provision applies only with respect to damages caused by a report or by a bad faith failure to aid and assist, Maples, 450 N.W.2d at 530 , the plaintiffs make no suggestion on appeal that damage to E.N. was caused by anything other than the failure of the state to intervene under chapter 232.

Turning to the specific language of the immunity statute, it is important to recognize that this case does not involve a situation in which a mandatory or permissive reporter triggers a DHS investigation by *24 making a report protected by the immunity provision of Iowa Code section 232.73. Dr. Lindaman filed no report with DHS. While the legislature plainly desired to encourage reports of suspected child abuse when it enacted the child protection provisions of the Code, id. § 232.67, these policy reasons are not at work in this case. This case does not involve protecting a mandatory reporter who stepped forward to report suspected child abuse and is entitled to immunity in order to encourage others to do the same.

Instead, this case involves the second or the aiding-and-assisting-in-an-assessment-of-a-child-abuse-report prong of Iowa Code section 232.73. And, this case involves a claim of failure to aid and assist in good faith, not an arguably overzealous report of suspected child abuse. The question is not whether Dr. Lindaman did too much, but whether he did too little.' Specifically, the question raised in this case is whether Dr. Lindaman was in “good faith” engaged in “aiding and assisting” DHS “in an assessment of a child abuse report.” Id. § 232.73.

There is evidence in the record that Dr. Lindaman was in fact avoiding aiding and assisting in an assessment of a child abuse report. He testified that in his conversations with DHS, he was not “judging what happened at E.N.’s home” and that he was “not providing information on the safety of the child.” This does not sound like aiding and assisting in an assessment of a child abuse report by any standard, objective or subjective. Further, if he were acting in subjective good faith, and aiding and assisting in an assessment of a child abuse report, surely he would have disclosed that spiral fractures in infants are “very rare” and that if he would have been the physician during intake, he would have filed a child abuse report too. He disclosed these views when defending a lawsuit against him, but he did not offer them to DHS when it was assessing the report of child abuse involving E.N. The fact he did not offer these views to DHS suggests he did not see his role as aiding and assisting in an assessment of the child abuse report. He was not judging, and he was not providing information relating to safety.

In addition, a jury could conclude that Dr. Lindaman’s minimal and cramped response to DHS was designed to further his own interest in not getting drawn into a potentially controversial matter. Surely, a subjective desire of not wanting to get involved or to pass the buck to someone else would be an ulterior motive that would defeat subjective good faith under the statute. Indeed, the existence of such motivations among professionals was one of the reasons for the enactment of child abuse reporting statutes in the first place.

There is other evidence that supports inferences against Dr. Lindaman’s claim of entitlement to immunity. According to the plaintiffs’ view of the evidence, the notion that a twenty-two-day-old baby, after placed in bed by his father, put his arm behind his back and then suffered a spiral fracture under his own weight is obviously suspect. Indeed, the plaintiffs’ point to Dr. Selover’s contemporary reaction to Dr. Lindaman’s position, namely the exclamation, “You got to be kidding.” Yet, Dr. Lindaman declared that the father’s story was “plausible” and the mechanism “fit the injury.” He defended his responses in his deposition as technically correct as a theoretical matter. However, he further emphasized in his deposition that he viewed the assessment of whether child abuse occurred as someone else’s responsibility as it did not relate to his management of the fracture. Is this “not-my-department” type of response consistent with subjective good faith in “aiding and assisting in an assessment of a child abuse report?” Could a reasonable jury conclude that Dr. Lindaman was attempting to avoid entan *25 glement in a sticky situation rather than aid and assist DHS in its investigation? Instead of cooperating with DHS investigators and others on the medical team in a collaborative fashion, could a reasonable jury conclude that Dr. Lindaman preferred to head for the exit and allow others to take responsibility rather than get involved? Could a reasonable jury conclude that his evasive responses were not aiding and assisting in an assessment of a child abuse report, but really an act of stone walling and hand washing? If so, then Dr. Lindaman was not aiding and assisting in an assessment of a child abuse report in subjective good faith under Iowa Code section 232.73.

The lack of support in the record for his position from every other physician who reviewed the file as part of a child abuse assessment arguably tends to support the inference that Dr. Lindaman just did not want to get involved. The plaintiffs point to Dr. Selover’s statement, “You got to be kidding,” as telling. And as stated by the plaintiffs’ expert, Dr. Lindaman held to his opinion long after it made any sense to do so. He arguably originally decided to give a brusque, incomplete, and even misleading answer to DHS in a verbal game designed to avoid getting drawn into a controversy and then decided to attempt to avoid professional embarrassment by defending it when challenged by Dr. Selover. His attitude toward the DHS assessment could be regarded as not a good faith “how can I help you?” but rather something else, a defensive posture akin to “don’t ask me, I’m just the bone guy, I’m not responsible. Anything is possible. Don’t confuse me with the facts or the opinions of others. I’m busy. Good-bye.” 10

Of course, a jury could well come to a much more favorable conclusion after assessing Dr. Lindaman’s credibility and hearing all the evidence. There is no question that on the evidence presented a reasonable jury could conclude he was expressing his honest opinion and he. was not very knowledgeable about infants. Perhaps, as the saying goes, a jury could conclude this is a case involving a defendant with “ ‘a pure heart and an empty head.’ ” Garvis, 492 N.W.2d at 404 (quoting Jackson, 488 N.W.2d at 156 ). Alternatively, a reasonable jury could conclude the real problem in this case was that DHS investigators misconstrued Dr. Lindaman’s statements and erroneously concluded that his observation that the mechanism described “fit the injury” was by implication a statement of opinion that child abuse did not occur, or at least was not substantiated, and not realize they were receiving a “don’t-ask-me-that’s-not-my-department” type answer. However, the question is whether, on its unique facts, the plaintiffs have enough evidence from which legitimate inferences may be drawn to proceed with the case. I conclude there was enough to do so.

In closing, I note the result today does not promote the policies of the child abuse reporting statutes. This case should not be confused with a reporting case in which a professional takes the sometimes difficult but legally required step of reporting suspected child abuse. In that setting, generous immunity may be appropriate. There, the statute demands the reporter receive the benefit of the doubt and may be deprived of immunity only if not acting in good faith.

*26 Here, however, the question is whether Dr. Lindaman’s minimal and narrow participation in the assessment of the child abuse report was sufficient to entitle him to statutory immunity that requires good faith in the aiding and assisting in an assessment of a child abuse report. The policies underlying immunity can certainly be offended by making it too difficult to obtain, but the policies of the statute are also undermined by extending immunity too far. One must ask whether stretching immunity in Iowa Code section 232.73 to cover the unique circumstances of this case “ ‘provide[s] -the greatest possible protection to victims or potential victims of abuse through encouraging the increased reporting of suspected cases of such abuse [and] insuring the thorough and prompt investigation of these reports.’ ” State v. King, 434 N.W.2d 627, 629 (Iowa 1989) (quoting Iowa Code § 232.67 (1987)). I doubt it. By granting summary judgment on the immunity question in this setting, when the plaintiffs’ claim of lack of subjective good faith in aiding and assisting in an assessment of a child abuse report is at least plausible, to use Dr. Lindaman’s unfortunate term, I fear the purposes of the mandatory reporting statute will not be promoted, but will be undermined. I fear the takeaway from this case will be that evasive and uncooperative responses to DHS child abuse investigators will be regarded as legally protected conduct. If so, our child protection system has lost some of its teeth. I hope I am wrong in that regard.

The undisputed bottom line, however, is that the child abuse reporting system failed E.N. in this case, with tragic results.

HECHT, J., joins this dissent.

. As a preliminary matter, I note the plaintiffs' claim that the defendants' amendment to their answer asserting the affirmative defense of immunity under Iowa Code section 232.73 (2009) should have been denied. The majority notes, in a footnote, that the plaintiffs did not specifically use the term "abuse of discretion” to describe the appropriate standard of review for a district court's decision to allow an amendment to the pleadings in their appellate brief before this court. I would not find the argument waived for failure to state the magic words of the undisputed standard of review. See Rife v. D.T. Comer, Inc., 641 N.W.2d 761, 766, 768 (Iowa 2002) (allowing amendment subject to abuse of discretion). On the merits, however, the motion to amend came four months prior to the discovery and pleading deadlines and five months before the scheduled trial. Further, from the beginning of the litigation, the nature of Dr. Lindaman's participation in the Iowa Department of Human Services investigation was identified as a factual issue. The plaintiffs did not seek an extension of time in the summary judgment proceedings to conduct further discovery with respect to the immunity issue. Under these circumstances, I agree the district court did not abuse its discretion in allowing the amendment in this case.

. The holdings in these cases, of course, provided the impetus for the United States Supreme Court to adopt an objective good-faith test in immunity cases involving alleged government official misconduct. See generally Harlow v. Fitzgerald, 457 U.S. 800, 818-19 , 102 S.Ct. 2727, 2738-39 , 73 L.Ed.2d 396, 410-11 (1982).

. The caselaw repeatedly emphasizes the difference between "objective good faith," which is more amenable to summary judgment, than "subjective good faith," which turns on credibility issues. The leading case is Harlow, in which the Court rejected a subjective good-faith standard in the context of immunity in favor of objective good faith because subjective good faith "rarely can be decided by summary judgment." 457 U.S. at 816-18 , 102 S.Ct. at 2737-38 , 73 L.Ed.2d at 409-10 ; see also Maestas v. Lujan, 351 F.3d 1001, *23 1011 (10th. Cir.2003) (noting that in order "to encourage courts to resolve qualified immunity questions on summary judgment, the Court removed the subjective-good-faith factor”); Rowan County v. Sloas, 201 S.W.3d 469, 474-84 (Ky.2006) (contrasting availability of summary judgment when immunity is objective good faith with lack of availability of summary judgment when a subjective good-faith standard is employed); Sabia, 687 A.2d at 473 (rejecting subjective good-faith standard in immunity context because summary judgment would not be available).

. The majority cites a number of child abuse statutory immunity cases in which summary judgment was granted. Many are distinguishable from the present case in that they do not involve an alleged failure to aid and assist a child abuse investigation, but instead involve a challenge to a report of child abuse. See Wolf v. Fauquier Cnty. Bd. of Supervisors, 555 F.3d 311, 318 (4th Cir.2009) (noting immunity asserted based upon affirmative report of child abuse); Watson v. County of Santa Clara, 468 F.Supp.2d. 1150, 1156 (N.D.Cal. 2007) (noting that California statute grants immunity with respect to mandated or authorized reporting, but not aiding and assisting); S.G. v. City of Monroe, 843 So.2d 657, 659-60, 661 (La.Ct.App.2003) (noting physician moved for summary judgment on ground that she was immune from liability based upon affirmative report of child abuse). In addition, some of the cases cited by the' majority involve a different substantive standard than that applicable under Iowa law. See Wolf, 555 F.3d at 318 (noting Virginia statute employs a strong presumption of immunity and places the burden on person who would overcome immunity); O’Heron v. Blaney, 276 Ga. 871 , 583 S.E.2d 834, 836-37 (2003) (noting that immunity sustained based on "objective” reasonable cause to suspect child abuse has occurred, unlike under the Iowa statute).

. It is apparent from the record that Dr. Lindaman sought to restrict his exposure to a claim of negligence by limiting his role in the treatment of E.N. to the management of the fracture. While this strategy may be an effective defense with respect to limiting the scope of his duty in the underlying negligence claim, it tends to undercut his claim of statutory immunity because he was not aiding and assisting in an assessment of a child abuse report, but was instead focused solely on management of the fracture.