This case requires us to consider whether the federal government is immune from damages for injuries its agents caused to an active-duty servicewoman’s baby during childbirth. Our resolution of the issues is controlled by the Supreme Court’s decision in Feres v. United States, which found that military service members were barred from bringing claims against the government under the Federal Tort Claims Act (FTCA) for injuries incident to their military service. In the many decades since its inception, criticism of the so-called Feres doctrine has become endemic. That criticism is at its zenith in a case like this one — where a civilian third-party child is injured during childbirth, and suffers permanent disabilities.
Under the Feres doctrine, federal courts lose their subject matter jurisdiction over claims like this because we conclude the injured child’s in útero injuries are unmistakably derivative of an injury to her mother, an active duty Air Force captain, who gave birth at a Fort Carson Army Base hospital. To be sure, the facts here exemplify the overbreadth (and unfairness) of the doctrine, but Feres is not ours to overrule. Applying controlling law, the government is not liable under the FTCA for the claims of negligence in this case.
Exercising jurisdiction under 28 U.S.C. § 1291, we AFFIRM the district court’s decision.
I. Background
Captain Heather Ortiz was an active-duty service member in the United States Air Force. In March 2009, Captain Ortiz was admitted to Evans Army Community Hospital1 for a scheduled Caesarean sec*819tion. Complications caused by the medical staffs administering of drugs in preparation for the surgery caused a precipitous drop in Captain Ortiz’s blood pressure, leading to hypotension. As a result of Captain Ortiz’s hypotension, her baby, “I.O.,” was deprived of oxygen in útero, leading to severe injuries.
Plaintiff here is George Ortiz; Captain Ortiz’s husband, who, as the parent of I.O., filed a lawsuit against the United States, seeking compensation for I.O.’s injuries, her long-term medical care, and her life-care needs. According to the complaint, in advance of the Caesarian section, one of the hospital’s nurses gave Captain Ortiz the trade drug Zantac, which is commonly used to prevent aspiration of gastric acid during labor or surgery. But as was apparent in Captain Ortiz’s hospital records, she was allergic to this drug and she suffered an allergic reaction to the provided dose soon after it was administered. To counteract the allergy, a doctor ordered that Captain Ortiz receive a dose of Bena-dryl. The Benadryl caused an immediate drop in Captain Ortiz’s blood pressure, resulting in hypotension, an injury that occurs when blood flow is inadequate to perfuse the uterus and the placenta. Captain Ortiz’s hypotension resulted in severe injuries to I.O., including brain trauma that caused cerebral palsy.
In addition, the complaint alleges that personnel at the hospital were negligent in failing to scrutinize the fetal monitoring strips following Captain Ortiz’s allergic reaction and the attendant consequences. Fetal monitoring strips refer to the graphical representation of the fetus’s heart rate during labor. Captain Ortiz alleges that had the hospital personnel timely reviewed the monitoring strips, they could have prevented I.O.’s injuries.
Finally, the complaint alleges that the hospital staff members were “negligent with regard to their care and treatment of Heather Ortiz and I.O.” See App. at 19; see also id. at 19 ¶ 65 (“The medical care providers ... did not appropriately provide care and treatment concerning Heather Ortiz’s blood pressure problem.”); id. at 21 ¶ 77 (“These Defendants deviated from the standard of care and were negligent in ... failing to properly monitor and treat I.O. [sic] condition.”); id. at 19 ¶ 71 (“These providers ... were negligent with regard to their care and treatment of Heather Ortiz and I.O.”).
The government filed a motion to dismiss for lack of subject matter jurisdiction, raising the bar to claims under the FTCA first established in Feres v. United States, 340 U.S. 135, 71 S.Ct. 153, 95 L.Ed. 152 (1950). The government also moved to stay discovery pending the district court’s decision on the motion to dismiss the complaint because the Feres doctrine is classified as jurisdictional and thus a ruling could dispose of the case without any discovery.
The district court agreed with the government that plaintiffs claims were precluded by Feres. In doing so, the district court recognized limited authority with respect to handling third-party Feres claims, especially those related to fetal injuries. Nevertheless, applying each of the several standards adopted in other circuits, and reaching the same result regardless of which it applied, the district court found that Feres barred plaintiffs claims related to both the negligent dispensation of the Zantac and the Benadryl, and the observation of the fetal monitoring strips.
II. Analysis
We address first whether the district court correctly found that it lacked subject *820matter jurisdiction over plaintiff’s claims because of the Feres doctrine. We find that the district court did not err, and, over the course of evaluating this jurisdictional question, we further explain this circuit’s case law on the Feres doctrine, particularly as it applies to third-party claims brought by civilians.
A. Subject Matter Jurisdiction
Because the question of subject matter jurisdiction is partially intertwined with an aspect of the merits of plaintiffs claims here, we proceed under a summary-judgment standard. Holt v. United States, 46 F.3d 1000, 1003 (10th Cir.1995). Thus, our review is de novo, and plaintiff “must present ... evidence sufficient to establish the court’s subject matter jurisdiction by a preponderance of the evidence.” Robinson v. Union Pac. R.R., 245 F.3d 1188, 1191 (10th Cir.2001) (quoting United States ex rel. Hafter v. Spectrum Emergency Care, Inc., 190 F.3d 1156, 1160 n. 5 (10th Cir.1999)). We can only affirm if there is no genuinely disputed issue of material fact concerning jurisdiction. See Pringle v. United States, 208 F.3d 1220, 1223 (10th Cir.2000).
The question of whether sovereign immunity exists “is jurisdictional in nature.” FDIC v. Meyer, 510 U.S. 471, 475, 114 S.Ct. 996, 127 L.Ed.2d 308 (1994). “Absent a waiver, sovereign immunity shields the Federal Government and its agencies from suit.” Id. Although the FTCA constituted an expansive waiver of the federal government’s sovereign immunity for torts committed by government actors, several exemptions apply. Ali v. Fed. Bureau of Prisons, 552 U.S. 214, 218, 128 S.Ct. 831, 169 L.Ed.2d 680 (2008). When an exemption applies, we revert to the baseline application of the government’s sovereign immunity, which deprives the federal courts of jurisdiction. See Garcia v. U.S. Air Force, 533 F.3d 1170, 1175 (10th Cir.2008). The Feres doctrine is one such exemption.
1. The Feres Doctrine
Since 1950, courts have analyzed tort lawsuits brought by service members against the government under the standards set forth in Feres v. United States, 340 U.S. 135, 71 S.Ct. 153, 95 L.Ed. 152 (1950). In that case, the Supreme Court considered several lawsuits seeking recovery for injuries or deaths of servicemen caused by the negligence of fellow military employees acting on behalf of the government. Each serviceman-plaintiff contend-ed that the broad waiver of sovereign immunity under the FTCA provided their recourse toward recovery, even though the injuries were sustained during activities associated with their service. Id. at 137, 71 S.Ct. 153.
The Supreme Court, however, found the FTCA does not - apply in such cases. Instead, the Court crafted a limited judicial exception to the federal government’s broad waiver of sovereign immunity under the FTCA. Id. at 146, 71 S.Ct. 153. In particular, the Court exempted from the scope of that statute “injuries to servicemen where the injuries arise out of or are in the course of activity incident to service.” Id. (emphasis added). This limitation is now known as the Feres doctrine.
In these cases, the crucial question involved in examining whether a service member is barred from recovery under the FTCA is resolving whether the injury was “incident to service.” The ineident-to-ser-vice test applies “consistently to bar all suits on behalf of service members against the Government based upon service-related injuries.” United States v. Johnson, 481 U.S. 681, 687-88, 107 S.Ct. 2063, 95 L.Ed.2d 648 (1987). The incident-to-service test, however, is neither self-defining *821nor readily discernible from the language of Feres or Johnson.
In our application of Feres, we have explained that, regardless of its precise contours, the incident-to-service test applies broadly:
In recent years, the Supreme Court has broadened Feres to the point where it now encompasses, at a minimum, all injuries suffered by military personnel that are even remotely related to the individual’s status as a member of the military. Courts applying the Feres doctrine have given a broad reach to Feres ’ “incident to service” test and have barred recovery by members of the armed services for injuries that at first blush may not have appeared to be closely related to their military service or status. Practically any suit that implicates the military’s judgments and decisions runs the risk of colliding with Feres.
Pringle, 208 F.3d at 1223-24 (emphasis added) (quotations, citations, and alterations omitted). Our general view is the dominant understanding of Feres.
Notwithstanding • a consensus on the Feres doctrine’s potency as a general matter, the Supreme Court and lower courts alike have consistently wrestled with the mechanics of its application to particular facts. In the early years after its adoption, courts focused on the underlying purposes for the doctrine to determine whether application to a given fact pattern furthers those reasons. See Ricks v. Nickels, 295 F.3d 1124, 1128 (10th Cir.2002). Courts articulated several rationales — often called “special factors” — for Feres’s existence. Id. Based loosely on Feres; courts looked to three general policies: “(1) the distinctly federal nature of the relationship between the government and members of its armed forces; (2) the availability of alternative compensation systems; and (3) the fear of damaging the military disciplinary structure.” Walden v. Bartlett, 840 F.2d 771, 773 (10th Cir.1988) (internal quotation marks and citation omitted).2
In many jurisdictions, the extent to which these factors still bear on the Feres analysis is an open question. In some cases, a gradual chipping away at factors (1) and (2) left a solitary reason at the heart of Feres: whether allowing the lawsuit to proceed would disrupt the unique hierarchical and disciplinary structure of the military. See United States v. Shearer, 473 U.S. 52, 57-58 & n. 4, 105 S.Ct. 3039, 87 L.Ed.2d 38 (1985); see also Ritchie v. United States, 733’ F.3d 871, 874 (9th Cir.2013), cert. denied, — U.S. ——, 134 S.Ct. 2135, 188 L.Ed.2d 1124 (2014) (“Although the Supreme Court has offered inconsistent guidance about how Feres should be applied ... we have consistently emphasized the third rationale.”).3 But *822that constriction has not at all been uniform, and even very recent cases have stressed that the Supreme Court’s guidance requires that all the Feres factors, must still be analyzed to determine whether the claims are prohibited. See, e.g., Purcell v. United States, 656 F.3d 463, 465-66 (7th Cir.2011) (recognizing that all three Feres rationales still apply).
Our circuit has simplified the equation, concluding that all the special factors “effectively merged ... with the incident to service test.” Ricks, 295 F.3d at 1130. We credited the Supreme Court’s decision in United States v. Stanley, 483 U.S. 669, 107 S.Ct. 3054, 97 L.Ed.2d 550 (1987), as articulating “that the incident to service test rests squarely on the third ‘special factor’ of preserving the military’s disciplinary structure and Congress’ prerogative in regulating intramilitary affairs.” Ricks, 295 F.3d at 1130 (citing Stanley, 483 U.S. at 683, 107 S.Ct. 3054). But we also stressed that the incident-to-service test does not permit an inquiry into how military discipline is “affected in a particular case,” because that inquiry itself would be an impermissible intrusion into military affairs. Id. In the end, we essentially rejected any focus on the special factors, finding them unduly redundant, and returned the analysis to the inquiry that Feres originally set forth: whether the injury was “incident to service.” Tootle v. USDB Commandant, 390 F.3d 1280, 1282 (10th Cir.2004) (“Rather than focusing on the presence or absence of the Feres rationales ... the relevant question is whether
[plaintiffs] alleged injuries arose ‘incident to service.’ ”).
With all of this confusion and lack of uniform standards, it comes as no surprise that the Feres doctrine, while the law of the land, has received steady disapproval from the Supreme Court on down. An early and vociferous critique came from Justice Scalia in dissent in Johnson, where he stated that “Feres was wrongly decided and heartily deserves the widespread, almost universal criticism it has received.” Johnson, 481 U.S. at 700, 107 S.Ct. 2063 (Scalia, J., dissenting) (internal quotation marks omitted). Justice Scalia (joined by three other Justices) found the Feres special factors lacking in textual support in the FTCA and devoid of logic. See id. at 694-700, 107 S.Ct. 2063. This dissent has provided the cornerstone in the foundation for Feres detractors, and it has been widely praised. See, e.g., Taber v. Maine, 67 F.3d 1029, 1044 (2d Cir.1995) (applauding Justice Scalia’s dissent as a persuasive rebuttal to the Feres doctrine).4
Because the dissent in Johnson and many other appellate-court decisions survey the doctrinal concerns Feres presents, we do not do so here. Suffice it to say that when a court is fprced to apply the Feres doctrine, it frequently does so with a degree of regret. See, e.g., Ritchie, 733 F.3d at 874 (“For the past sixty-three years, the Feres doctrine has been criticized by countless courts and commentators across the jurisprudential spectrum.” (internal quotation marks and citations omitted)); Purcell, 656 F.3d at 465 (“The Feres doctrine, while currently viable, is *823certainly not without controversy ... and has also been widely criticized.”); Regan v. Starcraft Marine, LLC, 524 F.3d 627, 633 (5th Cir.2008) (“This [Feres ] doctrine has been much-criticized.”); McMahon v. Presidential Airways, Inc., 502 F.3d 1331, 1341 (11th Cir.2007) (“The Feres doctrine has been controversial, even as applied to the government ... [b]ut ... Feres remains the law.”); Costo v. United States, 248 F.3d 863, 869 (9th Cir.2001) (“[W]e apply the Feres doctrine here without relish. Nor are we the first to reluctantly reach such a conclusion under the doctrine. Rather, in determining this suit to be barred, we join the many panels of this Court that have criticized the inequitable extension of this doctrine to a range of situations that seem far removed from the doctrine’s original purposes.”); Hinkie v. United States, 715 F.2d 96, 97 (3d Cir.1983) (“We are forced once again to decide a case where we sense the injustice of the result but where nevertheless we have no legal authority, as an intermediate appellate court, to decide the case differently.” (quotation marks and footnote omitted) (alteration incorporated)).
This circuit long ago expressed the same sentiment: “Although many courts have expressed reservations about the continuing validity of the broad Feres doctrine, only the United States Supreme Court can overrule or modify Feres.... Therefore, once again we are constrained to follow the Feres doctrine, notwithstanding its harsh result.” Labash v. U.S. Dep’t of Army, 668 F.2d 1153, 1156 (10th Cir.1982); see also Tootle, 390 F.3d at 1282-83.
In summary, based on our decision in Ricks, the incident-to-service test as originally enunciated in Feres (and independent of the other so-called Feres rationales) is the standard that we apply here. To fully resolve this case, however, we must move beyond the traditional Feres claim because Captain Ortiz is not a plaintiff; I.O., a civilian, is. And so, we next analyze the framework for addressing claims asserted by civilian third parties that arise out of injuries to a service member.
2. Third-Party Feres Claims
From its inception, Feres has applied to cases beyond those brought by service members to recover for their own injuries. In fact, Feres itself dealt with two wrongful-death actions brought by the widows of servicemen. But questions lingered regarding the extent to which Feres could uniformly bar recovery for injuries to third parties in circumstances where the genesis of the third-party injury was intimately associated with injuries to service members.
Stencel Aero
The Supreme Court analyzed those questions in Stencel Aero Engineering Corp. v. United States, 431 U.S. 666, 97 S.Ct. 2054, 52 L.Ed.2d 665 (1977). In that case, the Court found that Feres applied equally to a claim against the United States brought by the manufacturer of an ejection seat, the malfunction of which had caused a serviceman’s injuries. The serviceman had sued Stencel, the manufacturer, and the United States for permanent injuries that occurred when the ejection seat in his military plane failed during a midair exercise. Stencel responded by cross-claiming the United States for indemnity, and the United States raised Feres as a jurisdictional bar. The Court sided with the government, finding that “the third-party indemnity action in this case is unavailable for essentially the same reasons that the direct action by [the serviceman] is barred by Feres.” Id. at 673, 97 S.Ct. 2054.
We have applied this rule to civilian third-party claims beyond a garden-variety *824indemnification suit. Thus, for example, we have denied jurisdiction where the civilian wife of a serviceman brought claims against the government on account of her husband’s negligent vasectomy. Harten v. Coons, 502 F.2d 1363 (10th Cir.1974). Similarly, we have prohibited the daughter of a serviceman from bringing claims for injuries deriving from her father’s military service in Vietnam. See Heffington v. Dep’t of Defense of the U.S., 248 Fed.Appx. 952 (10th Cir.2007).
Other courts have also interpreted Sten-cel Aero broadly to “preclude[ ] suits by third parties that derive — directly or indirectly — from injuries to service members incident to military duty.” Brown v. United States, 462 F.3d 609, 612 (6th Cir.2006). In fact, on the heels of Stencel Aero, every other circuit has acknowledged that Feres "applies to third-party claims brought by civilians beyond the standard indemnification claim.5
The analysis in Stencel Aero became what is now known as the “genesis test.” Application of the genesis test sought to provide a reasoned standard for the question common to all third-party Feres claims; namely, to what extent are injuries to civilian third parties that arise out of circumstances concerning the service member’s relationship with the military also barred under the FTCA?
To this end, the genesis test asks whether the civilian injury has its origin in an incident-to-service injury to a service member. See, e.g., Ritchie, 733 F.3d at 875. If it does, then Feres applies as a bar to the third-party claim, just as it would to a claim by the service member for his or her injuries. See Minns v. United States, 155 F.3d 445, 449 (4th Cir.1998) (“Under this test, if a nonserviceman’s injury finds its ‘genesis’ in the injury suffered by a serviceman incident to service, then the Feres doctrine bars the non-serviceman’s suit.”).
The Genesis Test’s Injury Focus
Despite the near-universal adoption of the genesis test, our court has yet to apply it to the.type of third-party claims we find here. As we explain, in light of our obligation to follow Feres and Stencel Aero, the genesis test is the appropriate mechanism to "apply to third-party Feres claims involving derivative third-party injuries, including those occurring in útero. While we recognize that other circuits have not so inclusively utilized the genesis test for every third-party Feres claim, we are convinced that Stencel Aero dictates our decision in this case.
In addition, we find it persuasive that courts across the circuits find the genesis test helpful to adjudicate a wide array of third-party Feres claims.
First, circuit courts have employed the genesis test when considering whether children can recover for genetic injuries stemming from alleged government negligence in exposing their service-member fathers to radiation or chemical weapons. See Minns, 155 F.3d at 449; Hinkie, 715 F.2d at 98; Laswell v. Brown, 683 F.2d 261, 269 (8th Cir.1982); Monaco v. United States, 661 F.2d 129, 133 (9th Cir.1981). In Minns, for example, the Fourth Circuit explained why Feres precluded the claims *825of the wives and children of servicemen for genetic birth defects resulting from the military’s negligent inoculation of the servicemen in an effort to immunize them from possible biological or chemical attacks during the Persian Gulf War. The court said that the government’s conduct toward the servicemen “was the ‘genesis’ and the ‘but for’ cause of the injuries to the wives and children.” Minns, 155 F.3d at 450. Relying on Feres’s military-discipline rationale, the court found that “[i]f allowed to proceed, their suits would place the courts in exactly the position that the Feres doctrine was designed to avoid.” Id. Other genetic-defect cases have likewise relied on this rationale as the reason for extending Feres to bar the third-party claims. See, e.g., Monaco, 661 F.2d at 134 (dismissing third-party claims where “the court still must examine the Government’s activity in relation to military personnel on active duty”).
Second, an even more clear-cut application of the genesis test occurs when the relative of a service member sues for loss of consortium or mental anguish when the government negligently causes the service member’s death or injury. See Kendrick v. United States, 877 F.2d 1201, 1206-07 (4th Cir.1989); De Font v. United States, 453 F.2d 1239, 1240 (1st Cir.1972). By their very legal nature, these claims are “ancillary or derivative to an injury to a serviceman incident to military service” and thus cannot proceed under Feres. Lombard v. United States, 690 F.2d 215, 226 (D.C.Cir.1982) (internal quotation marks omitted).
But courts have inconsistently described the threshold or starting point for scrutiny of a civilian plaintiffs claims under the genesis test — sometimes underscoring the government’s negligent conduct and other times the service member’s injury.6 While the courts are not united, we appreciate that only the latter approach conforms with Feres and Stencel Aero. Indeed, this injury-focused approach asks first whether there was an incident-to-service injury to the service member, Feres, 340 U.S. at 144, 71 S.Ct. 153, and second whether the injury to the third party was derivative of that injury, Stencel Aero, 431 U.S. at 673-74. A negative answer to either question permits a complaint to survive dismissal for lack of subject matter jurisdiction under Feres.
*826We thus apply the injury-focused approach, finding that it is more faithful to Feres and Stencel Aero, the precedential cases binding on this court. By the same token, we reject the object of the negligence or the target of the medical treatment as linchpins in analyzing third-party Feres claims.7
Our task in this case would be difficult enough if we simply had to navigate the conflicting interpretations of Feres and the genesis test. But an additional (and significant) variable requires further explanation: the fact that I.O.’s injuries were suffered in útero. With that in mind, we next discuss the nature of the in útero cases.
3. In Utero Cases
As a threshold matter, claims for fetal injuries brought by the offspring of servicewomen ordinarily do not pose substantial doctrinal questions under Feres’s genesis test. After all, such causes of actions are third-party civilian claims against the government, typical of the sort that invite a Feres inquiry into whether the third-party injury derives from a service person’s incident-to-service injury.
But to counterbalance the harsh results often associated with the application of Feres to third parties, some courts have resisted the genesis test in prenatal cases. See, e.g., Brown, 462 F.3d at 616; Mossow v. United States, 987 F.2d 1365, 1369-70 (8th Cir.1993); Romero v. United States, 954 F.2d 223, 226 (4th Cir.1992); Del Rio v. United States, 833 F.2d 282 (11th Cir.1987); Lewis v. United States, 173 F.Supp.2d 52, 56-57 (D.D.C.2001), vacated in part on other grounds, 290 F.Supp.2d 1 (D.D.C.2003). These courts justify an “in útero exception” because there is either no injury to the mother or, in fewer cases, injuries to the mother and to the fetus are unrelated. In either scenario, the fetus’s injury must be independent — that is, it Cannot be derivative of the service member’s nonexistent or wholly separate injury.
But this rule is not really an “exception” at all; rather, the upshot of the genesis test in such situations is that the third-party injury did not have its genesis in an incident-to-service injury to the service person.8 By its very terms, the genesis test under Feres will never bar a third party’s claim absent a service-member injury. And where there is a clear sever-ability between the injuries to the fetus and those to the mother, the genesis test will likewise not preclude recovery.
The Nature of In Utero Injuries
A number of cases have analyzed in útero injuries under the Feres doctrine. And like with the genesis test more generally, these courts have struggled to ascertain the crux of the genesis test for in útero injuries, disagreeing about whether the starting point is an injury to the service member or simply treatment directed *827at or for the benefit of the service member.
Plaintiff here oscillates between these approaches, urging that we consider a narrow application of Feres for claims by infants injured during childbirth under either standard. Based on the cases from other circuits, plaintiffs hedging is understandable, and the Sixth Circuit’s decision in Brown is a useful illustration of the interaction between the competing principles. See 462 F.3d at 611-16. In that case, the court was forced to grapple with its circuit’s own precedent, Irvin v. United States, 845 F.2d 126 (6th Cir.1988), which had broadly adopted the genesis test for in útero injuries because “[t]he treatment accorded [to a fetus’s] mother is inherently inseparable from the treatment accorded ... [to] a fetus in his mother’s body.” Brown, 462 F.3d at 617 (quoting Scales v. United States, 685 F.2d 970, 974 (5th Cir.1982)). To break away from this authority, the Brown court cited with approval the collection of in útero cases, “concluding] that the child’s FTCA claim was not barred by Feres because the child had sustained an independent injury.” Id. at 613 (“[S]ome of the cases are factually similar to the case before this Court in that they allege a negligent injury only to the child, rather than an injury to both the child and the-service member.”). It further .distinguished Irvin because “the death of the Irvin infant was caused by (and was therefore derivative of) an injury suffered by the child’s mother.” Id. at 614. This made all of the difference in the world because, as the court repeatedly underscored, “Deborah Brown sustained no physical injury whatever from the effects of the negligent prenatal treatment, from her pregnancy, or from Melody’s birth.” Id. at 611 (emphasis added). As we read the opinion, this focus on the injury permitted the Sixth Circuit to apply the genesis test, but escape the weight of Irvin’s seemingly broad holding and permit the military dependent to recover for her injuries.
Brown’s articulation of the service person’s injury as the place to start is sound, and other courts have gravitated toward an injury-focused approach for cases involving fetal injuries. For example, in Romero, the Fourth Circuit found that “[i]n our view the relevant inquiry in a genesis analysis is whether a service member was injured, not whether the negligent act occurred during active duty service.” Romero, 954 F.2d at 226. The court in Romero was persuaded in part by the fact that the military mother “suffered no physical injury as a result of the allegedly negligent conduct.” Id. at 224. Similarly in Mossow, the Eighth Circuit noted that “[b]ecause [the- baby’s] injury is not derivative of an injury to a service member, we find his cause of action for legal malpractice is not barred by the genesis test under Feres.” 987 F.2d at 1370.
And finally, the Ninth Circuit’s recent decision in Ritchie convincingly explains why the only sustainable way to approach the genesis test is through a focus on the service member’s injury itself. Ritchie, 733 F.3d at 877-78. In that case, the court applied the genesis test to dismiss claims for fetal injury brought about because the fetus’s servicewoman mother was forced, against doctor’s orders, to participate in physical training while pregnant. The court first found that the in útero exception was inapplicable based on Ninth Circuit precedent, instead applying a version of the genesis test to find that examining the fetus’s claim would involve second-guessing military orders and thus could not be maintained. Id. at 875-76. Addressing the plaintiffs urging that the court adopt the in útero exception, the court commented in dicta that the result would be unchanged. Indeed, the court *828found that “[o]nly where a fetus alone suffers injury can the claim survive Feres.” Id. at 877-88 (“[A] civilian fetus’s claim may only escape Feres if its servicewoman mother suffered no injury from the purportedly negligent acts.”).
Other Approaches
The injury-focused approach, however, has not gained universal acceptance in the in útero cases. For example, the district court in Lewis found a different result flowed from Romero, concluding that “[t]he crucial issue ... is whether the negligent medical treatment' leading to [the baby’s] injury was provided to him or to his mother.” Lewis, 173 F.Supp.2d at 57.9 But, at the very least, the internal inconsistencies in Romero are evidence as to why the treatment-focused approach is unworkable:
Admittedly, in satisfying its duty of care to Joshua, proper prenatal treatment would have involved his mother’s body. The sole purpose of the treatment, however, would have been directed at Joshua.
Mrs. Romero suffered from a congenital cervical weakness. This condition apparently placed Joshua at risk of injury. It did not, however, affect Mrs. Romero’s health. Presumably her state of health would have been the same whether the physician placed the sutures or not. If the treatment had been administered, its sole purpose would have been directed at preventing injury to Joshua. The failure to place the sutures during the prenatal period and to cut them immediately preceding birth was the direct cause of the injuries to Joshua, a civilian. Because the purpose of the treatment was to insure the health of a civilian, not a service member, Feres does not apply.
Romero, 954 F.2d at 225. As this suggests, the difficulties in ascertaining the beneficiary of the treatment are never more pronounced than in a case involving mother and fetus. It is difficult to comprehend how courts are well-positioned to determine whether a particular negligent act was directed at the mother, at the fetus, or at both.10
Faced with the dilemma of the inherently-inseparable nature of prenatal and neonatal treatment, the Romero court strained to demonstrate that the government’s conduct was directed at, and intended to benefit, the fetus in order to avoid Feres’s application. But Romero’s analysis is unpersuasive, especially to the extent that it sought to differentiate in útero cases from other third-party Feres claims that required application of the genesis test. Id. at 226. (“We are persuaded that a genesis analysis is inappropriate here.”).
In any event, this distinction was unnecessary because as the court ultimately concluded, the fetus’s “injury did not derive from any injury suffered by a service member.... Because no service person was injured [the fetus’s] claim is not Feres-barred.” Id. at 226. Bending over backwards to discern the target and beneficiary *829of the treatment only served to confuse what could have been a straightforward injury-focused approach that would have led to the same result.
Injury-Focused Approach
Regardless of the split in authority pertaining to the primary concern in the in útero cases, we are convinced that the injury-focused approach is the one required by Supreme Court precedent. As we have said, at a doctrinal level, we need not look much further than Stencel Aero to confirm that the heart of the genesis test is a service member’s incident-to-serviee injury. Even more fundamentally, the service member’s injury — and by extension, the derivative nature of a third-party injury — is the touchstone of all Feres claims. See Madsen v. United States ex rel. U.S. Army, Corps of Engineers, 841 F.2d 1011, 1012 (10th Cir.1987) (stressing that Feres’s incident-to-service test is fundamentally concerned with the relationship between the injury and the service member’s status, not the negligence itself).
Along the same lines, it is worth reiterating that Feres is a jurisdictional doctrine that directs our focus to the injury as a threshold matter, largely independent of the viability of plaintiffs negligence or other tort claims. Remember that if Feres applies, then the government is immune to lawsuits notwithstanding the FTCA’s broad waiver. And although we proceed .on summary judgment,11 this preliminary analysis of jurisdiction limits our full-fledged consideration of whether the government owed a duty to any party or whether that duty was breached. But the treatment-focused approach takes us away from the typical Feres question and requests that we purely investigate the merits of the plaihtiff s negligence claim, asking us to analyze the existence of a duty, whether that duty was breached, whether the government caused the alleged injury and so forth. In the end, “[t]he mere fact that the cause of action is not derivative ... but is an original and distinct cause of action ... does not remove it from the prohibition of Feres.” De Font, 453 F.2d at 1240 (emphasis added).
In many ways, the treatment-focused approach requests that we put the cart before the horse. The Feres doctrine has always operated as an antecedent jurisdictional hurdle — that is, it activates an inquiry into our ability to even consider the merits of the tort alleged against the government. The injury-focused approach appreciates this prefatory concern, deferring any substantial merits discussion related to the actions of the government vis-a-vis the third party. In order to reach those questions, we must proceed past the Feres bar and to do so, we consider the relationship between the third-party injury and the service member’s injury as a threshold matter.12
*830And the treatment-focused approach could produce anomalous results. For example, in many cases treatment is simultaneously provided for the benefit of both mother and fetus. Oftentimes, only one of the two is injured by medical negligence. In addition, a common scenario might involve treatment aimed solely to benefit the fetus, but which results in an injury to the mother. If the baby suffers an obviously derivative injury as a result of the accidental injury to the mother, then Feres should bar the claim regardless of the fact that the treatment was originally intended to benefit only the baby. By the same token, we can envision a scenario where treatment was provided solely for the mother’s benefit, but negligence in providing such treatment injured, the baby alone. In those cases, Feres would not operate as a bar because the injury could not be derivative of the non-existent injury to the mother.
One final point: we do not intend our rule here to convey that the source and scope of the treatment is irrelevant or necessarily incompatible with the inquiry under Feres. An examination of the treatment may, for instance, be crucial in determining whether the service person was injured at all, see Brown, 462 F.3d at 612-13, or whether that injury was in fact incident to military service, see Monaco, 661 F.2d at 133. But any review of the treatment can only be a means to an end— the end being whether the third-party injury is derivative of the service person’s incident-to-serviee injury. Phrased another way, our point is simply that looking to the target of the alleged negligence — or the treatment — is not the applicable test. It is not the case that deciphering the target of treatment will always resolve the inquiry. But we leave open the possibility that identifying a separate negligent act may help to establish a direct and non-derivative injury to the fetus.13
For those reasons, we disagree with the concurrence. A treatment- or conduct-focused approach has its own virtues, as the concurrence explains. But it also has its own vices, and we think the injury focus is most consistent with Stencel Aero and the complexities of fetal injury cases.14 After all, the FTCA is a broad waiver of sovereign immunity and the Feres doctrine excepts a class of service member claimants who are injured incident to military service. Feres is concerned not so much with the government’s conduct per se but with whether the claimant’s injuries arose from that service. We share some of the con*831currence’s concerns about application of the genesis test, but those concerns are not enough to adopt a categorical bar for any fetal injuries arising from obstetric care.
Conclusion
At bottom, the in útero cases do not require a unique standard. The genesis test supplies the relevant rule, and its purpose is not undercut simply because the plaintiffs injuries arose in útero. As it has always been, the test is whether the civilian (fetal) injury had its genesis in a service-related injury to a service person (mother).
4. Application
Turning to the facts of this case, the focal point of our examination is whether I.O.’s injury was derivative of (i.e., had its genesis in) an injury to Captain Ortiz.15 Given this standard, plaintiffs allegations cannot overcome Feres. When we weigh all of the competing evidence, and view it in the light most favorable to plaintiff, we find that the injuries that 1.0. sustained forming the basis of the complaint are derivative of the injuries to her mother, Captain Ortiz.
First, plaintiff failed to allege in the complaint and the briefing below that Captain Ortiz was not injured by the hospital personnel’s actions. See, e.g., App. at 18-19 (Complaint) (repeatedly referencing Captain Ortiz’s “allergic reaction,” “blood pressure problems,” and “hypotension”); id. at 10, 26 (mentioning failure to “treat [Captain] Ortiz’s condition”); id. at 143 (Response to Motion to Dismiss) (“I.O.’s injuries are not derivative of Ms. Ortiz’s.”); id. at 146 (“As these resulting injuries were not derívate of Ms. Ortiz’s injuries ... ”); see also Aplt. Br. at 6, 25. Although on appeal plaintiff argues that Captain Ortiz was uninjured, see, e.g., Aplt. Reply at 6-8, that argument not only comes too late, it misstates the record. Accordingly, only if I.O.’s injuries are truly separate from Captain Ortiz’s injuries can plaintiff overcome Feres.
But no such theory is present on this record. Indeed, the allegations in the complaint and the affidavits offered as evidence tell a tragic but straightforward story that precludes recovery. I.O.’s oxygen loss was the result of Captain Ortiz’s drop in blood pressure, which was caused by the negligent administration of Benadryl following Captain Ortiz’s allergic reaction to the Zantac. Even plaintiffs expert, a board-certified obstetrician-gynecologist, essentially conceded this theory in an affidavit attached to plaintiffs opposition to the government’s motion to dismiss:
Because Ms. Ortiz was given Zantac, she was given diphenhydramine (Benadryl), which is routinely given during times of allergic reaction to prevent anaphylactic response. Anaphylaxis in a pregnant woman could lead to maternal hypoxia and/or hypotension which could, in turn, harm an unborn child. Accordingly, diphenhydramine was given in order to prevent such an allergic reaction and to benefit and prevent injury to both Ms. Ortiz and her unborn child, 1.0.
App. at 129 (emphasis added).
At bottom, the source of I.O.’s ultimate brain injury was her servicewoman moth*832er’s blood-pressure problems. It does not matter that Captain Ortiz does not allege legal injuries16 within the complaint; the point is that she was injured incident to service, which would bar her claims and anything derivative of them. The plain fact is that Captain Ortiz’s service-related injury led to an injury to her civilian daughter. Even construing those facts in plaintiffs favor, the Feres doctrine — in conjunction with the genesis tesi&wkey;bars I.O.’s claims related to her brain injury.
With respect to the alleged failure to scrutinize the fetal monitoring strips, there is no independent injury that would support this claim. Thus, even though the defendant had a duty to review the fetal monitoring, this alleged injury to I.O. is still derivative of the injuries to Captain Ortiz. On this point, the district court succinctly (and correctly) stated “[b]ut for the administration of Zantac and Benadryl to Captain Ortiz, failure to review the fetal monitoring strips in this case would not have caused I.O.’s injuries.” App. at 234-35. This conclusion is not overcome by the fact that, according to plaintiff, the monitoring of the fetal heart rate provides no benefit to the mother. As we highlighted above, the target of the treatment is not our focus in Feres cases.
Even if we were to adopt a treatment-focused approach, however, the result would be the same because the complaint unambiguously alleged that the treatment that caused the injury was provided to both Captain Ortiz and I.O.17 Thus, even under a treatment-focused approach, we see no reasoned way to sever the injury-inducing treatment and afford I.O. relief.
Plaintiff criticizes this approach as a sort of blanket rule that would bar all “pregnancy-related claims of children of active duty mothers, even where the mother has no injury.” Aplt. Reply at 3. But where the mother is truly not injured at all, Feres does not apply under our rule. See, e.g., Brown, 462 F.3d at 611; Romero, 954 F.2d at 226. While we focus on the injury to the mother as our starting point, our analysis does not end there. As required by the genesis test, the injury to the fetus must trace back to that injury in order for Feres to render the court without jurisdiction. And certainly there are also cases where the mother and the fetus are injured, but their injuries are truly unconnected — that is, the fetus’s injury is not derivative of a service-related injury to the mother. But that is not the case here, and plaintiffs own allegations and concessions are decisive on this score.18
In sum, the Feres doctrine applies to the injuries alleged here. We wish, frankly, that were not the case. But in faithfully *833applying Supreme Court authority, Tenth Circuit precedent, and the persuasive decisions from other circuits, the incident-to-service and genesis standards require such an outcome. The district court properly found that it lacked subject matter jurisdiction and granted summary judgment.
B. Discovery Order
In addition, the plaintiff contends the district court erred by not allowing additional time and discovery to respond to the government’s Rule 56 motion. Under Rule 56, a party must “state with specificity how the additional material will rebut the summary judgment motion.” Libertarian Party v. Herrera, 506 F.3d 1303, 1308-09 (10th Cir.2007) (internal citation omitted). Plaintiff supplied two reasons. One, that additional evidence would prove that the Benadryl was provided for both Captain Ortiz and I.O.’s benefit. Two, that the presentation of evidence concerning the fetal monitoring strips would prove an independent injury not subject to Feres. The district court denied the motion.
We find no abuse of discretion in that decision. As an initial matter, the government provided plaintiff with a complete set of medical records related to Captain Ortiz’s treatment. Plaintiff does not deny as much. The secondary documents and testimony that plaintiff requested, therefore, could not add any serious value to plaintiff’s ability to defeat summary judgment, and plaintiff did not convince the district court otherwise.
Moreover, plaintiff has difficulty escaping the fact that he alleged in the complaint that Captain Ortiz suffered from hypotension and related blood-pressure problems as a result of the government employees’ alleged negligence. Even in the light most favorable to plaintiff, those allegations outline an injury to Captain Ortiz, which ultimately resulted in the harm to I.O. Accordingly, there is no merit to plaintiffs argument regarding the district court’s abuse of discretion in refusing to reopen discovery.
Similarly, on the fetal-monitoring issue, it appears that plaintiff had all of the relevant documents and information and cannot therefore allege the district court abused its discretion in not allowing further discovery. In any event, as we explained above, we do not credit that argument as providing an alternative avenue to escape Feres’s application. For this reason, the relevance of any additional discovery on the fetal monitoring strips is obviated, and the district court correctly forestalled the discovery requests.
III. Conclusion
We AFFIRM the district court’s decision granting summary judgment in favor of the government.
. The hospital, which is owned and operated by the government, is located in Colorado *819Springs on the Fort Carson Army base and provides medical services and benefits primarily to military personnel and retirees, and their families.
. In Feres, the Court also acknowledged that the absence of parallel private liability supported excluding service-member injuries from the FTCA waiver. See Feres, 340 U.S. at 146, 71 S.Ct. 153. This rationale did not last five years. See United States v. Brown, 348 U.S. 110, 112, 75 S.Ct. 141, 99 L.Ed. 139 (1954) (rejecting the parallel-private-liability argument).
. The Ninth Circuit further paved its own course, latching onto the Supreme Court's warning "that the Feres doctrine "cannot be reduced to a few bright-line rules; each case must be examined in light of the statute as it has been construed in Feres and subsequent cases.” Shearer, 473 U.S. at 57, 105 S.Ct. 3039. Bearing this in mind, the Ninth Circuit constructed a series of factual predicates that clue to whether a given injury was incident to service: "In cases where the existence of a Feres bar is not clear, we have looked to four factors to determine whether an activity is incident to military service: (1) the place where the negligent act occurred; (2) the duty status of the plaintiff when the negligent act occurred; (3) the benefits accruing to the *822plaintiff because of his status as a service member; and (4) the nature of the plaintiff's activities at the time the negligent act occurred.” Dreier v. United States, 106 F.3d 844, 848 (9th Cir.1997) (paragraph format altered).
. Recently, in fact, Justice Thomas dissented from the denial of certiorari in a case that asked the court to overrule Feres, suggesting that, “at a bare minimum, it should be reconsidered.” See Lanus v. United States, - U.S. -, 133 S.Ct. 2731, 2732, 186 L.Ed.2d 934 (2013) (Thomas, J., dissenting from the denial of certiorari).
. See Minns v. United States, 155 F.3d 445 (4th Cir.1998); Mossow v. United States, 987 F.2d 1365 (8th Cir.1993); Smith v. United States, 877 F.2d 40 (11th Cir.1989); Irvin v. United States, 845 F.2d 126 (6th Cir.1988); In re Agent Orange Litig., 818 F.2d 201 (2d Cir.1987); West v. United States, 744 F.2d 1317 (7th Cir.1984) (en banc); Hinkie v. United States, 715 F.2d 96 (3d Cir.1983); Gaspard v. United States, 713 F.2d 1097 (5th Cir.1983); Lombard v. United States, 690 F.2d 215 (D.C.Cir.1982); Monaco v. United States, 661 F.2d 129 (9th Cir.1981); DeFont v. United States, 453 F.2d 1239 (1st Cir.1972).
. In Minns, for instance, the Fourth Circuit focused on the target of the government’s negligent conduct as the key consideration. According to that court, where the negligent act or behavior was directed at the service person, then any injuries that had their genesis in that negligence, even those to third parties, were barred by Feres. See Minns, 155 F.3d at 450 ("This negligence in implementing and administrating the program to the servicemen thus was the "genesis” and the "but for” cause of the injuries to the wives and children.”); see also Matthew v. United States, 452 F.Supp.2d 433, 442 (S.D.N.Y.2006) (“The relevant inquiry is not, therefore, who is injured and when the injury becomes manifest, but rather the time and nature of the negligent act alleged.”).
But in Lombard, the D.C. Circuit focused instead on whether there was an actual injury to the serviceman that, based on typical causation principles, ultimately generated an injury to the third party. There, the plaintiffs claimed that government negligence during its oversight of the "Manhattan Project” in New Mexico allowed Mr. Lombard, a member of the Army, to handle radioactive substances, which negatively altered his genetic makeup. As a result, his children were born with physical defects. Relying on Stencel Aero, the court found that “Feres has also been held to bar cases brought by third parties ... where the claim originates with or derives from an injury to a serviceman incident to military service.” Lombard, 690 F.2d at 219 (emphasis added); see also Adkins v. United States, 869 F.2d 593, at *2 (4th Cir.1989) (unpublished table decision).
. There is no reason to categorically exclude consideration of these factors when divining the source of the third-party injury. But, contrary to the comprehensive treatment-focused approach adopted by some circuits, we emphasize the primacy of identifying an injury to a service member and tracing the third-party injury to it as the crucial elements.
. Plaintiff alleges that the in útero exception is an entirely separate rule from the genesis test that applies to cases involving fetuses. Due to some conflicting case law, this classification makes some sense but is irrelevant in any event. Because the totally independent injury to the fetus is a raison d’etre for the in útero rule, it serves to negate the derivativeness that is required for the genesis test. As we have said, properly construed, the in útero exception is bettqr described as a particular class of cases under which application of the genesis test reveals an independent third-party injury not barred by Feres.
. Lewis and the Eleventh Circuit’s decision in Del Rio are outliers to the extent that they found no Feres bar even when the fetus's injuries stemmed from an injury to the servicewomen mother. These cases are unfaithful to Feres for all of the reasons we discuss herein.
. Even some cases that ultimately applied Feres as a bar to third-party claims misguidedly stressed the importance of the target of the negligence. See Scales, 685 F.2d at 974 ("In cases that allow the dependents of servicemen to sue the government, the negligent conduct is directed to the dependent alone and does not involve any decisions by the military toward enlisted personnel.”).
. We note that on summary judgment, we consider the merits of the case to the extent that they are intertwined with the question of subject matter jurisdiction. See Pringle, 208 F.3d at 1222. And while this procedural posture changes both the standard and the evidence available for our consideration, it is not an invitation to forego the jurisdictional question to reach the merits. In other words, we consider certain “aspect[s] of the substantive claim,” id. at 1223, that bear on our jurisdictional analysis, see id. at 1223 n. 3 (explaining how important jurisdictional facts under Feres “overlap with the merits of the FTCA claim”).
. One might question whether discussion of the service person's injury and the causal link between the service member's injury and the third-party injury impermissibly ushers in negligence principles. But our examination of those elements exists on a jurisdictional, rather than a merits-based, plane. To the extent that we discuss injury or causation, we do so not with an eye toward whether those elements might state a prima facie tort *830against the government, but out of fidelity to Feres and Stencel Aero to decide whether we have jurisdiction.
. The injury-focused approach is not without its flaws, as the concurrence well points out. But the injury focus fittingly accounts for the different outcomes in cases involving fetal claims. Compare Brown, 462 F.3d at 611 ("Deborah Brown sustained no physical injury whatever from the effects of the negligent prenatal treatment.”); Romero, 954 F.2d at 226 ("In our view the relevant inquiry in a genesis analysis is whether a service member was injured.”); Utley v. United States, 624 F.Supp. 641, 645 (S.D.Ind.1985) ("The only party injured by the alleged malpractice is a civilian infant.”), with Scales, 685 F.2d at 974 (holding that servicewoman mother "sought relief for injury to herself”); Irvin, 845 F.2d at 127 (finding that the baby's injuries were derivative of an injury to the mother). To be sure, this clean categorization is not entirely true to type. See Lewis, 173 F.Supp.2d at 54 (describing injuries to both mother and child but finding Feres inapplicable); Del Rio, 833 F.2d at 286 (same). But the injury-focused approach is the superior alternative.
. For instance, under the concurrence’s approach, government conduct aimed at the child but injuring the mother and then deri-vately injuring the child would be actionable. This situation would squarely raise Feres doctrine concerns.
. We reject plaintiffs apparent contention that I.O.’s injuries do not ''arise[] out of activity incident to service.” Aplt. Br. at 39. We decline to be misdirected from our starting point for the incident-to-service test: Captain Ortiz’s injury. Were this a case in which a military member was uninjured, we would not pursue the Feres course in the first place. Here, we are faced with a service-related injury of a service member that potentially resulted in a derivative injury to a third party. Under those circumstances, we look to whether the service member's injury was incident to service, not the third party’s. There is no genuine dispute that Captain Ortiz’s injury was incident to her military service.
. A prudent lawyer would never allege a legal injury to the service person in a third-party Feres case. Accordingly, it is incumbent on the court to review the factual record to determine whether the facts themselves describe a factual injury to the service person (and whether the third-party injury is derivative of that injury).
. In addition to the board-certified obstetrician-gynecologist, plaintiff also submitted an affidavit from a registered nurse in support of his opposition to the government’s motion to dismiss. Both indicated that the Benadryl was given for the benefit of both Captain Ortiz and I.O. See App. at 120, 129-30.
.Ñor do we dispute that medical personnel-including those at military, hospitals— owe duties of care under state law to fetuses during and in connection with prenatal and neonatal care. Pizza Hut of Am., Inc. v. Keefe, 900 P.2d 97, 101 (Colo.1995) ("Colorado, like other jurisdictions, recognizes a child’s right to bring a cause of action for prenatal injuries.”). Our focus on the service member’s injury, however, does not vitiate the duty in any way. As stressed, the Feres inquiry says nothing of the merits of plaintiff's claim for, in this case, medical malpractice; it is a separate and antecedent jurisdictional question.