dissenting:
In asking the Mississippi Supreme Court to answer how the discovery rule under Miss.Code. Ann. § 15-1-36 applies to the facts of this case, the majority states that there is confusion as to “which of the Mississippi Supreme Court decisions control a case such as this.”1 I agree that something is amiss, but it is not in the statute we have been asked to apply or in the decisions of the Mississippi Supreme Court.
The history of the statute of limitations defense defendants advance before this court is important. “The defendants’ statute-of-limitations defense was not included as a contested legal or fact issue [in the pretrial order],” it appeared in the order only under the heading “additional matters to aid in the disposition of [the case].”2 The defendants requested no jury instruction regarding limitations and did not argue before the jury that the Husses had sufficient knowledge to trigger the running the statute of limitations; rather, they argued causation, that not even the defendant doctors could have known whether Terbutaline caused, or could have caused, Huss’s condition — an idiopathic phenomenon. As the magistrate judge noted in rejecting defendants’ post-judgment motion, which raised the statute of limitations defense, “ ‘[the] defendants failed to establish the approximate date on which the statute of limitations began to run’ because ‘there was no proof of the date by which plaintiff knew or should have known [that Terbutaline was probably the cause of her injury and that her physicians should not have given her the drug].’ ”3 Having chosen not to pursue the statute of limitations at trial and, therefore, having failed to develop evidence on the defense, defendants make the tendentious request that this court conclude as a matter of law that the defense, which is by its very nature a *249fact driven inquiry, bars Huss’s claim. I respectfully dissent.
I
As an initial matter, I note that certification is simply not appropriate for two reasons. Mississippi Rule of Appellate Procedure 20(a) explains that certification is available when “it shall appear ... to any United States Court of Appeals that there may be involved in any proceeding before it questions or propositions of law of this state which are determinative of all or part of that cause and there are no clear controlling precedents in the decisions of the Mississippi Supreme Court....”
First, Rule 20 suggests that it is appropriately invoked when Mississippi law is unclear, and not when a United States Court of Appeals simply wishes to have the Mississippi Supreme Court apply the law to a set of facts for it. As the Mississippi Supreme Court explained, “Rule 20(a) is thus subject to the discretion of the Court, and, although this Court generally strives to limit our inquiry to issues of law, we have clearly done otherwise in the past.”4
Indeed, I find this situation parallel to that in Boardman v. United Services Automobile Association, where the Mississippi Supreme Court chided this court for certifying questions that the court considered to be “in the nature of law application questions”: “Though subscribing to no theory of mechanical jurisprudence, we would think it ordinarily within the competence of a federal judge, given stipulated facts and eight decisions of this Court declaring and refining the applicable legal principles, to decide the case substantially the same, so far as legal rules determine the outcome of a litigation, as would the appropriate state court.”5 As I understand this case, the issue before this court is applying law to facts, and I would not shift that task to the Mississippi Supreme Court.
Second, Rule 20(a) states that certification is not available unless “there are no controlling precedents in the decisions of the Mississippi Supreme Court.” There is controlling precedent here, and as discussed below, I find no ambiguity in the case law that warrants certification.
II
We are called upon to “apply[ ] the unambiguous language of Miss.Code. Ann. § 15-1-36(2),”6 which provides, in pertinent part, that the statute of limitations in Mississippi for medical malpractice claims runs “from the date the alleged act, omission, or neglect shall or with reasonable diligence might have been first known or discovered.” The Mississippi Supreme Court described how the statute’s discovery rule operates in Sutherland v. Ritter:
Thus, in medical negligence cases, we must focus our inquiry on when a plaintiff, exercising reasonable diligence, should have first discovered the negligence, rather than the injury.... Furthermore, in the medical malpractice context, the discovery rule may apply in cases where the injury is not latent at all, but where the negligence which caused the known injury is unknown. For instance, a patient who undergoes a medical procedure may develop serious complications which are clearly known. However, if the patient has no reason to know that the doctor’s negligence in performing the procedure caused the eom-*250plications, the discovery rule will apply, even though the injury itself is not latent at all.7
As Sutherland makes clear, the application of the discovery rule is necessarily a case-by-case factual determination — a conclusion fully consistent with the Mississippi Supreme Court’s earlier cases.
A reading of the case law discussed by the panel majority leads to the rather unremarkable conclusion that the facts of each case were different and, therefore, the statute of limitations began to run at different times. Sutherland says nothing different; it purports only to resolve confusion as to whether the statute of limitations is triggered by actual or presumptive knowledge of (1) just the injury or, (2) the injury, professional negligence, and causal connection between the two. Sutherland and the other case law reflect that all three factors are involved when applying the discovery rule: when the patient knew or should have known of the injury, the cause of the injury, and the relationship between the injury and the cause, as guided by an overarching reasonableness standard.8 I describe the facts in each case at some length to dispel the notion that they are in tension.
The Mississippi Supreme Court addressed the discovery rule under the Mississippi Tort Claims Act (MTCA) in Wright v. Quesnel.9 The discovery rule under the MTCA operates in all parts pertinent like the statute here. The plaintiff was pregnant, and during the eighth month of pregnancy, twice visited a doctor within a week, presenting high blood pressure and symptoms of pre-eclampsia. Both times the doctor merely ordered bed rest. Six days after the first visit, she again went to the doctor with severe pain. The baby had died in útero. On these facts, the court concluded that the plaintiff “had enough information at the time of death such that she knew or reasonably should have known that negligence had occurred.”10 The court specifically pointed out that “Wright did not offer any evidence that she could not have discovered the injury within the applicable statute of limitations.”11 This is a straightforward conclusion based on the facts: a patient complains of the same symptoms to a doctor on multiple occasions within a week, the doctor orders no treatment, and the patient in a short time experiences an adverse outcome related to the symptoms she presented. A reasonable person would not need an expert to connect the injury, the negligence, and the relationship between the two.
Powe v. Byrd presents a similar situation, although under § 15-1-36.12 The decedent had for two years received treatment from his doctor for gastritis and hemorrhoids. As it turned out, he had cancer in his colon and lungs. However, it took the plaintiff more than two years after the decedent’s cancer diagnosis to file a malpractice claim. The plaintiff argued that she could not have known of the negli*251gence until she received an expert’s opinion. The court found that argument to be “disingenuous and without merit.”13 It is a rather unsurprising conclusion that the plaintiff did not need an expert report to suspect physician error when, after the decedent received the wrong diagnosis and treatment for years, it turned out he had colon cancer and not hemorrhoids. Once again, the adverse outcome was directly related to the symptoms the patient presented with and the mistreatment by the physician.
The dictum in PPG Architectural Finishes Inc. v. Lowery describing Powe is consistent.14 The court explained that the plaintiff in Powe did not need an expert to tell her there was a problem. To the extent that PPG discusses the injury and not the negligence, the statute of limitations at issue in PPG is Mississippi’s residual statute of limitations, § 15-1-49, which focuses on latent injuries and not negligence.
Nor does Barnes v. Singing River Hospital Systems create tension.15 The plaintiff had been diagnosed as having rheumatoid arthritis and received treatment beginning in 1989. In August 1995, the plaintiff went to her doctor and complained of pain and swelling in her knee— a symptom ostensibly consistent with rheumatoid arthritis. Nine days later, she went to the hospital and complained of fever, vomiting, and pain in her right elbow. She was diagnosed with acute bronchitis, gastritis with dehydration, and a fractured right elbow — again, what appears to be a plausible diagnosis — and discharged. However, the plaintiff had sepsis, a serious infection;16 she eventually had to have multiple amputations. The Mississippi Supreme Court concluded that “[wjhile the Barneses may have been aware of Lisa’s injuries before the one year time limit was up, they could not reasonably have known that Singing River was responsible for those injuries until their medical expert notified them of the possible negligence.”17 This was a case where, even though there was an obvious injury, the negligence that caused the injury was not necessarily obvious. There was no facially obvious connection between the injury and the error, or the relationship between the two. This is the type of situation that Sutherland clarifies when it explains that the discovery rule covers cases of known injury but unknown negligence.
The panel majority’s discussion of Wayne General Hospital v. Hayes18 creates some tension, but the case itself does not. The decedent was originally admitted to the hospital for observation for pneumonia; however, her condition worsened and she was transferred to another hospital. Once there, her doctors determined that she needed peritoneal dialysis. During that procedure, the doctor perforated the decedent’s bowels, which resulted in peritonitis. She developed a serious blood infection. She was then transferred to another hospital, but eventually died. Her death certificate listed cardiomyopathy, congestive heart failure, and sepsis as *252causes of death. The court concluded that the plaintiffs should have known of the negligence by the time of the decedent’s death:
Moreover, the plaintiffs, at the time of [decedent’s] death, had enough information such that they knew or reasonably should have known that some negligent conduct had occurred, even if they did not know with certainty that the conduct was negligent as a matter of law. Since the death certificate included sepsis as one of the causes of death, it should have been apparent to the plaintiffs that some negligent conduct had occurred. Additionally, [decedent] was hospitalized at Arkansas Children’s Hospital subsequent to the bowel perforation which allegedly occurred at the University of Mississippi Medical Center. This should have alerted her survivors of possible problems with her medical treatment.19
It is not hard to see why the court concluded that plaintiffs knew or should have known of the negligence: they knew that a surgical procedure had punctured the decedent’s bowel; that she developed peritonitis; that she developed a blood infection; and that she then had to be transferred to another hospital. Her death certificate listed sepsis — a serious infection generally in the blood stream20 — as a cause of death. That is a direct chain of events — similar to Wright and Powe— where the causal connection between the alleged negligence and injury was reasonably apparent.
In Neglen v. Breazeale the decedent entered the hospital with an abdominal aortic aneurysm, and he underwent surgery during which “the affected blood vessel is replaced with a graft made of synthetic material.”21 However, following surgery, decedent complained of severe abdominal pain and experienced post-surgical bleeding. The decedent was bleeding extensively from the graft, and, after spending three days on a ventilator, died. The Mississippi Supreme Court held that summary judgment on the statute of limitations question was not appropriate “since questions of fact exist[ed].’22 The decedent’s doctors had misinformed the plaintiff about some issues and did not tell her about other issues concerning what happened. Considering the doctors’ role in shaping the information the plaintiff knew, the court could not conclude as a matter of law on the record before it that plaintiff acted unreasonably in relying on what the doctors told her and waiting to pursue her claims.
Finally, in Sutherland, the plaintiff complained that his doctor committed malpractice by prescribing Zyprexa. Sutherland developed a number of side effects, and decided to stop taking the drug without consulting his doctor. However, he resumed taking it, and eventually checked himself into the hospital. In his deposition, Sutherland stated that he checked himself in because, he said, “Zyprexa was destroying my life.”23 When asked if he believed Zyprexa was the cause of his problems, Sutherland answered that “[i]t was not a belief, it was knowing.”24 When he was discharged from hospital, his discharge summary stated that “the Zyprexa has been discontinued and the patient reports that he feels less flat and ‘zombie’ *253like.”25 He later told another doctor that Zyprexa had “caused a lot of bad side effects.”26 That doctor diagnosed Sutherland’s condition as Tardive Dyskinesia Syndrome (TDS). Sutherland stopped taking Zyprexa in April 2001, but did not file his claim until January 2004.
Sutherland argued that his claim was not time-barred under § 15-1-36, as he had a “latent injury,” and that he did not know he had TDS until January 2002. The Mississippi Supreme Court found Sutherland’s argument unavailing: “By his own admission, Sutherland knew who, when, how, and by what he had been injured soon after receiving treatment and the Zyprexa prescription from Dr. Ritter, and certainly, no later than the date of his discharge from St. Dominic. Considering Sutherland’s action, we conclude that Sutherland knew that Dr. Ritter’s prescribing him Zyprexa had caused him to suffer an injury”27 To recapitulate the circumstances in Sutherland: the plaintiff admitted knowing that the drug prescribed by his doctor was the source of the injury of which he complained. He knew of the injury, the alleged negligence, and the causal connection between the two. He did not need to know the name of his condition to know those things, much like the plaintiffs in Wright and Powe did not need experts to alert them to the possibility of negligence.
I highlight that Sutherland does not cast doubt on Wright, the case the panel majority “previously concluded ... was the most analogous to the Husses’ case and governed their claims.” In both cases, the Mississippi Supreme Court simply held that the plaintiff had enough information to know of the injury, the alleged negligence, and the causal connection. Admittedly, the conclusion in Wright was circumstantial and inferential, while plaintiffs knowledge in Sutherland was direct; but that alone cannot cause “the precedential effect of Wright [to] no longer [be] clear.” In both cases it was clear that the plaintiff knew or reasonably should have known of the injury, negligence, and causal relationship between the two.
In short, all of these cases flow naturally from the design of Mississippi’s discovery rule: it is a fact intensive, three-element inquiry. In some instances the negligence is quite obvious, in others not so. Facial similarities — such as sepsis being involved in both Barnes and Hayes create the appearance of inconsistency when results differ, but the truth is that those similarities belie greater differences that a careful examination of the facts in each case reveals. These cases all turn on questions relating to what the plaintiff knew and when. The Mississippi Supreme Court’s decision in Sutherland, even to the extent it may create other problems, says nothing different.28
*254III
If anything, Sutherland adds clarity to this case, as it addresses the exact issue here: when Huss should have known that the doctors’ negligence caused her injury. Moreover, Sutherland lends support to my earlier dissent.29 The Husses’ situation falls precisely into the hypothetical described in Sutherland — a person who knows that he is ill or injured, but is unaware of any negligence, much less any connection between the negligence and injury:
Furthermore, in the medical malpractice context, the discovery rule may apply in cases where the injury is not latent at all, but where the negligence which caused the known injury is unknown. For instance, a patient who undergoes a medical procedure may develop serious complications which are clearly known. However, if the patient has no reason to know that the doctor’s negligence in performing the procedure caused the complications, the discovery rule will apply, even though the injury itself is not latent at all.30
Huss, who was pregnant, went to her doctor complaining of cramping, pressure, and contractions; when the first attempts at treating her contractions failed, her doctor ordered that she be given Terbutaline. Her contractions stopped, and her doctors ordered that she continue to take Terbuta-line. The injuries that Huss suffered, however, were not pregnancy related; rather, the injuries were cardiomyopathy, pulmonary edema, and congestive heart failure. Other than the sequence of events — pregnancy complications, treatment, diagnosis of injury — there was nothing to clue Huss into the source of her injury. Indeed, these types of ailments could have been naturally occurring or could have resulted from life-style choices.31
This case thus differs from those like Wright, Powe, and Wayne where there was a direct injury-negligence-causation nexus. Nor is this case factually similar to Sutherland where the plaintiff admitted knowing “who, when, how, and by what he had been injured.”32
Of course, what happened is that the defendants faced a choice at trial between different defenses that were facially inconsistent: (1) Huss knew enough of the causal connection between her condition and Terbutaline to trigger the statute of limitations; or (2) there was no causal connection between Huss’s condition and Terbu-taline, and that the doctors could not know of any connection. To have argued both to the jury would have been awkward to say the least. Rather than do so, the defendants made a tactical choice that arguing causation was the better defense. That entailed the implicit assertion that if the physicians could not have known of any nexus, then Huss, a lay person, certainly could not have known either. They lost, and then came hat in hand to this court. We sit to correct errors made by the courts, not to correct the parties’ trial strategies or to impose judicial will — here, medical tort reform — upon a jury in frustration of its fundamental role in governance.
*255rv
With deference, the error in the question the panel majority certifies is taking from the jury what is quintessentially a fact question — what did the plaintiff know and when — and presenting it to the Mississippi Supreme Court as a question a law. The majority has picked but three from among the many factors from the case here and determined that, as a matter of law, one is, standing on its own, going to be determinative of when the statute of limitations began to run. But the answer to the majority’s question is: “It depends.” It depends on, inter alia, the sequence of events; how obvious the connection between the negligence and the injury is; what other conditions and illnesses the plaintiff may have; what independent knowledge the plaintiff had; how confident a diagnosis the plaintiff received; whether the plaintiff received conflicting diagnoses; and what the negligent doctor(s) told or represented to the plaintiff, the credibility of the doctor(s) involved, and the facial plausibility of what the doctor(s) said. Sorting through these types of factual inquiries is not subject to the precise line drawing that the majority’s question to the Mississippi Supreme Court suggests. This is not to say that a jury always must decide if and when the statute of limitations began to run; as Sutherland demonstrates there will be occasions when it is clear as a matter of law. This, however, is not such a case.
Even assuming that Sutherland is inconsistent with the Mississippi Supreme Court’s other discovery rule case law, certification is still inappropriate. Sutherland is an authoritative explanation of the law; to the extent other cases seem inconsistent, our task is simple: to follow and apply the court’s most recent elucidation of the law. The court implied as much when it explained that it was “tak[ing] this opportunity to clarify the law.”33 To these eyes, the Mississippi Supreme Court helpfully described how courts are to approach the factual situation — known injury but unknown negligence and unknown causal connection — that is involved here. To the extent there are inconsistencies, the Mississippi Supreme Court has already told us what we are to do.
With all respect for my colleagues, I must dissent.
.To be clear, the decision to certify the question was made months before the Mississippi Supreme Court decided Sutherland v. Ritter, 959 So.2d 1004 (Miss.2007).
. Huss v. Gayden, 465 F.3d 201, 204 (5th Cir.2006).
. Id.
. McIntyre v. Farrel Corp., 680 So.2d 858, 860 (Miss.1996) (emphasis added).
. 470 So.2d 1024, 1030-31 (Miss.1985).
. Sutherland, 959 So.2d at 1008.
. Id. at 1008-09.
. See, e.g., Neglen v. Breazeale, 945 So.2d 988, 990 (Miss.2006) ("In other words, statute of limitations begins to run when the patient can reasonably be held to have knowledge of the injury itself, the cause of the injury, and the causative relationship between the injury and the conduct of the medical practitioner.”). There is some question, discussed infra note 28, as to whether Sutherland requires an exclusive focus on discovery of the negligence; however, that question is not implicated here.
. 876 So.2d 362 (Miss.2004).
. Id. at 367.
. Id. (emphasis added).
. 892 So.2d 223 (Miss.2004).
. Id. at 228.
. 909 So.2d 47 (Miss.2005).
. 733 So.2d 199 (Miss.1999).
. See Lawson v. Dallas County, 112 F.Supp.2d 616, 622 n. 11 (N.D.Tex.2000) ("Sepsis is a systemic blood infection in which pathogens and poisonous products infect the blood stream.”). According to Sted-man's Medical Dictionary, 28th ed., sepsis is "[t]he presence of various pathogenic organisms, or their toxins, in the blood or tissue.”
. Barnes, 733 So.2d at 206.
. 868 So.2d 997 (Miss.2004).
. Id. at 1001.
. See supra note 16.
. 945 So.2d at 989.
.Id. at 991 (emphasis added).
. 959 So.2d at 1006.
. Id.
. Id.
. Id.
. Id. at 1009.
. As the Sutherland dissent points out, the majority's focus on knowledge of the negligence rather than the injury suggests that the statute of limitations would start to run against a plaintiff who knows a doctor erred but is unaware that the error caused any injury. See id. at 1010-11 (Diaz, J., dissenting); see also Jackson Clinic for Women, P.A. et al. v. Henley, Nos.2005-IA01833-SCT, 1999-IA-01286-SC, 2007 WL 2265136, at *8 (Miss. Aug. 9, 2007) (Diaz, J., concurring in the judgment). This is in tension with the black letter principle that no cause in tort lies where the plaintiff does not have an injury. This criticism of Sutherland, however, does not apply here: Huss knew of the heart condition but the purported negligence and causal connection to the injury were not obvious. In other words, this is not a case where the plaintiff knew only of the negligence but not the injury.
. See Huss, 465 F.3d at 209-11 (Higginbotham, J., dissenting).
. 959 So.2d at 1008-09 (emphasis added).
. For example, Stedman’s Medical Dictionary lists five types of cadiomyopathy and multiple etiologies for each. The etiologies include familial/genetic, idiopathic, alcohol, and toxic substances.
. 959 So.2d at 1009.
. 959 So.2d at 1007.