*812Since 1995, the State of Indiana has required that, at least eighteen hours before a woman has an abortion, she must be given information provided by the State about, among other things, the procedure, facts about the fetus and its development, and alternatives to abortion. That information is meant to advance the State's asserted interest in promoting fetal life. In other words, the State hopes that women who read that information and consider it will opt not to have an abortion, and will, instead, choose to carry the pregnancy to term. After she has received the mandated information, a woman must wait at least eighteen hours before having an abortion, thus, the State hopes, she will use the time to reflect upon her choice and choose to continue her pregnancy. The State also requires that a woman have an ultrasound and hear the fetal heartbeat prior to an abortion although she may decline the opportunity to do one or both, as 75% of women generally do.1
Prior to July 1, 2016, women could, and generally did, have the ultrasound on the same day of the procedure. This was, in large part, because almost all abortions in Indiana occur at one of four Planned Parenthood of Indiana and Kentucky (PPINK) health centers, and only those few PPINK facilities that offer abortion services (most do not) had the ultrasound equipment on site. The Indiana House Enrolled Act 1337 (HEA 1337), however, amended Indiana law and now requires women to undergo an ultrasound procedure at least eighteen hours prior to the abortion. Because of the structure and location of abortion services in Indiana and the population of women seeking abortions, this change-moving the ultrasound from the day of the abortion procedure to at least eighteen hours before-as we will explore, is significant.
PPINK filed suit against the Commissioner of the Indiana State Department of Health and the prosecutors of Marion County, Lake County, Monroe County, and Tippecanoe County (collectively, "the State"), all in their official capacities.2 PPINK claimed that HEA 1337 unconstitutionally burdens a woman's right to choose to have an abortion, and it sought preliminary relief enjoining the provision during the pendency of the litigation. The district court granted the preliminary injunction. We agree with the well-reasoned conclusions of the district court opinion, from which we borrow heavily.
I.
A. Background information
1. The new law
Indiana Code § 16-34-2-1.1 mandates that at least eighteen hours prior to the abortion procedure, the patient must be provided with the following information (among others) both orally and in writing: "that human physical life begins when a human ovum is fertilized by a human sperm;" the probable gestational age of *813the fetus at the time the abortion is to be performed, including a picture of the fetus at certain gestational ages, and other information about the fetus at its current stage of development; notice that the fetus can feel pain at or before twenty weeks; information about the risks of abortion and of carrying the fetus to term, and information regarding alternatives to abortion and other support services available. Ind. Code § 16-34-2-1.1(a)(1)-(2). A woman seeking an abortion must also receive a color copy of a brochure, authored and distributed by the Indiana State Department of Health, that contains all of this same information. The State controls every aspect of the information conveyed to patients via this brochure-from the drawings, to the color, information about development, and wording of the risks of the procedures. Neither the brochure nor the informed-consent information has been challenged in this litigation.
Prior to the enactment of the challenged law, Indiana required that "[b]efore an abortion is performed, the provider shall perform, and the pregnant woman shall view, the fetal ultrasound imaging and hear the auscultation of the fetal heart tone if the fetal heart tone is audible," unless the pregnant woman certified in writing, on a form drafted by the Indiana State Department of Health, that she declined to view the ultrasound or hear the fetal heart tone. Ind. Code § 16-34-2-1.1(b) (2011). In other words, the provider must offer the ultrasound, but a woman may affirmatively decline. Prior to 2011 the provider did not have to offer the ultrasound, but only had to provide one if specifically requested by the woman. P.L. 193-2011, Sec. 9. In fiscal year 2016, only approximately 25% of women seeking abortion services chose to view their ultrasound images and only approximately 7% chose to listen to the fetal heart tone. Most importantly for this litigation, before 2016, the statute did not mandate when the ultrasound must occur, other than prior to the abortion. As a practical matter, however, the ultrasound procedures were performed just before the abortion. Ultrasound equipment is expensive and scarce. Not all PPINK locations have it, but, at the time the new law was enacted, the four locations that perform abortions had the equipment. Although patients can receive their informed-consent consultations at any one of the seventeen PPINK health centers throughout Indiana, abortions are performed only at four locations throughout Indiana (surgical abortions are available only at three locations). Therefore, to prevent women from having to travel far distances eighteen hours apart, providers performed the ultrasound on the day of the abortion procedure at one of the four facilities that had ultrasound and performed abortions.
The new statute, however, prevents this practice. It requires the following:
At least eighteen (18) hours before an abortion is performed and at the same time that the pregnant woman receives the information required by subdivision (1), the provider shall perform, and the pregnant woman shall view, the fetal ultrasound imaging and hear the auscultation of the fetal heart tone if the fetal heart tone is audible unless the pregnant woman certifies in writing, on a form developed by the state department, before the abortion is performed, that the pregnant woman:
(A) does not want to view the fetal ultrasound imaging; and
(B) does not want to listen to the auscultation of the fetal heart tone if the fetal heart tone is audible.
Ind. Code Ann. § 16-34-2-1.1(a)(5). PPINK argues that this requirement unduly burdens a woman's right to an abortion. Because PPINK's argument is based on the factual context, it is critical first to understand *814how PPINK health centers operate and where they are located.
2. PPINK facilities
At the time the law suit began, PPINK operated twenty-three health centers in Indiana. Due to financial constraints, that number has since dwindled to seventeen. Only four of the centers offer abortion services-Bloomington, Indianapolis, Merrillville and Lafayette-the latter of which offers only non-surgical abortions using medication. Of the centers that offer abortion services, the times these services are available are exceptionally limited. In Indianapolis, abortion services are only available three days per week; in Merrillville, a day and a half per week; and in Bloomington and Lafayette, only one day a week. R. 24-1 at 6. PPINK will perform a surgical abortion only until thirteen weeks and six days after the first day of a woman's last menstrual period. It will provide medication abortions only until sixty-three days after the first day of a woman's last menstrual period. Women who are pushing up against the time deadline may not be able to wait until a provider is available at the facility closest to them, but may need to travel to a more distant facility where a timely appointment can be made. The thirteen PPINK health centers that do not provide abortion services provide well-women examinations, screening for cancer and sexually transmitted diseases, treatment for sexually transmitted diseases and other preventative health care.
There are no clinics in Indiana that perform abortions past these dates. The only providers of abortion services after these dates are hospitals and surgical centers (all of which happen to be located in Indianapolis) and those facilities generally only provide abortions that are medically indicated because of a fetal anomaly or a threat to a woman's health, and these are quite rare. Out of the 7,957 abortions performed in Indiana in 2015 (the year before enactment of the new law), only 27 occurred in a hospital or surgical center. Only eighteen occurred after thirteen weeks. Indiana State Department, Terminated Pregnancy Report-2015, at pp. 7, 17, 18 (released June 30, 2016). Available at https://www.in.gov/isdh/files/2015% 20TP% 20Report.pdf [Last visited June 19, 2018].
PPINK has attempted to expand its health services throughout Indiana, but it operates only seventeen centers spread across a large state and only four that provide abortions. This means that some women must travel great distances to obtain an abortion. For example, Indiana's second largest city, Fort Wayne, had a PPINK health center until July 9, 2018, but it did not provide abortion services. Now it has none.3 The closest center providing such services is 115 miles away in Lafayette (a more than two hour drive).4 There are also no out-of-state abortion clinics that are close to Fort Wayne.
Prior to the enactment of the challenged law, women seeking abortions could have their state-mandated informed-consent session at any one of the seventeen centers across the state. At this appointment, which usually only lasted about fifteen *815minutes, health care providers also calculated the gestational age of the fetus based on the length of time from the first day of the last menstrual period. To make it more convenient for patients, PPINK allowed parents to bring children to these appointments. Women could then have an ultrasound, as required by then-existing state law, on the day of the procedure at the health center providing the abortion. PPINK would use that ultrasound information to verify that the pregnancy was intrauterine (and not ectopic) and to verify the gestational age to insure that the abortions are being performed within the required limits. An ultrasound is not medically necessary prior to an abortion, but the state requirement to perform an ultrasound is not challenged in this case, just the timing of it. Allowing the informed consent to be performed at any of the PPINK centers made it practical for women who live a long distance from the few centers that offer abortion services, by eliminating the need for multiple visits.
Once the new law was enacted, requiring that the ultrasound take place at least eighteen hours prior to the abortion, the barriers for many women increased significantly. Because ultrasound machines were only available at the four PPINK centers that provide abortion services, women who lived a significant distance from one of those centers were faced with two lengthy trips to one of those facilities or an overnight stay nearby. PPINK attempted to ease that burden by purchasing one additional ultrasound machine for one health center that does not offer abortion services, and trained a staff member to use ultrasound equipment at another. Those expenditures exacted a heavy toll on the finances of the organization, and still did not ease much of the burden. The ultrasound machines PPINK uses cost approximately $25,000 and must be operated by trained technicians. National Planned Parenthood policies, which are designed to align with generally accepted medical standards, require that an ultrasound image be interpreted by a physician or an advanced practice nurse. The nurse-practitioners at PPINK do not have the requisite training and PPINK asserts that it can afford neither the cost nor time to enroll nurses in the four-week training program.
3. Population served
The majority of women who seek abortion services at PPINK (and for that matter, the rest of the nation) are poor. The table below demonstrates the income level of patients relative to the federal poverty line (FPL).5
?
R. 24-1 at 14. These women often have precarious employment situations and generally are not paid for days they do not work. Many of them already have one or more children. In 2016, 33.73 percent of PPINK patients reported that they had children living with them. R. 24-1 at 4-5.
4. The district court's decision
The district court carefully weighed the burdens identified by PPINK against the *816benefits the State hoped would accrue to its citizens-the protection of both fetal life and the mental health of women. It concluded that:
The new ultrasound law creates significant financial and other burdens on PPINK and its patients, particularly on low-income women in Indiana who face lengthy travel to one of PPINK's now only six health centers that can offer an informed-consent appointment. These burdens are clearly undue when weighed against the almost complete lack of evidence that the law furthers the State's asserted justifications of promoting fetal life and women's mental health outcomes. The evidence presented by the State shows that viewing an ultrasound image has only a "very small" impact on an incrementally small number of women. And there is almost no evidence that this impact is increased if the ultrasound is viewed the day before the abortion rather than the day of the abortion. Moreover, the law does not require women to view the ultrasound imagine [sic] at all, and seventy-five percent of PPINK's patients choose not to. For these women, the new ultrasound [law] has no impact whatsoever. Given the lack of evidence that the new ultrasound law has the benefits asserted by the State, the law likely creates an undue burden on women's constitutional rights.
Planned Parenthood of Ind. & Ky., Inc. v. Comm'r, Ind. State Dep't of Health , 273 F.Supp.3d 1013, 1043 (S.D. Ind. 2017).
B. The legal standard
We review the district court's grant of a preliminary injunction in this case for an abuse of discretion, reviewing legal issues de novo, factual findings for clear error, and giving deference to the district court's weighing of the evidence and balancing of the equities. Whitaker By Whitaker v. Kenosha Unified Sch. Dist. No. 1 Bd. of Educ. , 858 F.3d 1034, 1044 (7th Cir. 2017).
To obtain a preliminary injunction, PPINK must establish that it has some likelihood of success on the merits; that it has no adequate remedy at law; that without relief it will suffer irreparable harm. City of Chicago v. Sessions , 888 F.3d 272, 282 (7th Cir. 2018). If that burden is met, the court must weigh the harm that the plaintiff will suffer absent an injunction against the harm to the defendant from an injunction, and consider whether an injunction is in the public interest. Id. Our court employs a sliding scale approach, "The more likely the plaintiff is to win, the less heavily need the balance of harms weigh in his favor; the less likely he is to win, the more need it weigh in his favor." Valencia v. City of Springfield , 883 F.3d 959, 966 (7th Cir. 2018) (internal citations omitted).
The district court correctly noted that the need for and propriety of a preliminary injunction of this law would depend mostly on the likelihood of success on the merits. It therefore focused most of its attention, as do we, on that factor.
II.
A. The test articulated in Casey and Whole Women's Health
The basic premise from which we must begin our review of the district court opinion is that the Supreme Court has recognized and affirmed "the right of the woman to choose to have an abortion before viability and to obtain it without undue interference from the State ... [without] the imposition of a substantial obstacle to the woman's effective right to elect the procedure." Planned Parenthood of Se. Penn. v. Casey , 505 U.S. 833, 846, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992). But yet, "[t]he fact that a law which serves *817a valid purpose, one not designed to strike at the right itself, has the incidental effect of making it more difficult or more expensive to procure an abortion cannot be enough to invalidate it." Id. at 874, 112 S.Ct. 2791. How then, do we determine whether a law's effects are incidental or unconstitutionally limiting? The Casey court set forth an undue burden test which declared that a state may not establish a regulation that "has the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a non-viable fetus." Id. at 877, 112 S.Ct. 2791. As the Casey court explained,
A statute with [an improper] purpose is invalid because the means chosen by the State to further the interest in potential life must be calculated to inform the woman's free choice, not hinder it. And a statute which, while furthering the interest in potential life or some other valid state interest, has the effect of placing a substantial obstacle in the path of a woman's choice cannot be considered a permissible means of serving its legitimate ends.
Id . Recently, the Supreme Court reiterated this test noting that Casey held that a law is unconstitutional if it imposes an "undue burden" on a woman's ability to choose to have an abortion, meaning that it "has the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus." Whole Woman's Health v. Hellerstedt , --- U.S. ----, 136 S.Ct. 2292, 2324, 195 L.Ed.2d 665 (2016), (citing Casey , 505 U.S. at 877, 112 S.Ct. 2791 ).
Importantly, both Whole Woman's Health and Casey stress that the undue burden test is context specific. Id. at 2306 ; Casey , 505 U.S. at 885, 112 S.Ct. 2791. An abortion statute valid as to one set of facts and external circumstances can be invalid as to another. Whole Woman's Health , 136 S.Ct. at 2306.
The State argues that the test for weighing abortion regulations differs depending on the purpose of the statute and that Casey and Whole Women's Health establish different tests depending on the nature of the regulation. The State claims that under Casey , an informed-consent and waiting period law will only be invalidated if the regulations "impose a 'substantial obstacle in the path of a woman seeking an abortion.' " Appellant's Brief at 26 (citing Casey , 505 U.S. at 877, 112 S.Ct. 2791 ). This standard, it argues, is somehow different than the undue burden test of Whole Women's Health which, the State says, is only appropriately applied to regulations that ostensibly promote women's physical health. Appellant's Brief at 17. The State claims that the balancing test is not appropriate here because, unlike in Whole Women's Health , the parties' stated interests are fundamentally opposed-the plaintiffs' goal is to help women carry out their decisions to terminate a pregnancy and the State's goal is to persuade a woman to reconsider that decision. Regulations that address informed-consent and waiting periods, the State argues, are subject only to "demonstration that they will cause a significant decline in abortions unrelated to the persuasive impact." Appellant's Brief at 22.
The State is incorrect that the standard for evaluating abortion regulations differs depending on the State's asserted interest or that there are even two different tests-the undue burden test of Whole Women's Health and a less-exacting "substantial obstacle" test (as the State argues) derived from Casey . To the contrary, Casey described the undue burden test as "a standard one of general application," and equated the "substantial obstacle" with "undue burden" noting that "[a] finding of an undue burden is shorthand for the conclusion that a state regulation has the purpose or effect of placing a substantial *818obstacle in the path of a woman seeking an abortion of a non-viable fetus." Casey , 505 U.S. at 876-77, 112 S.Ct. 2791 (emphasis ours). In fact, in Casey's seminal iteration of the undue burden test, the Court applied it to all of the regulations at issue in that case, including those that the state claimed affected women's health (record keeping and reporting), but also to spousal notification and parental involvement, which the state asserted were related to its interest in potential life. Id. at 887-99, 112 S.Ct. 2791. In other words, the Casey Court applied the same undue burden test to all of the regulations at issue in that case without regard to the state's asserted interest. In fact, Casey made clear that "a statute which, while furthering the interest in potential life, or some other valid state interest , has the effect of placing a substantial obstacle in the path of a woman's choice cannot be considered a permissible means of serving its legitimate ends." Id. at 877, 112 S.Ct. 2791 (emphasis added).
Nor is there anything in the Court's decision in Whole Women's Health to suggest that it applied a different standard than the undue burden test articulated in Casey . Rather, the Whole Women's Health Court clearly states to the contrary. When discussing "undue burden" it starts with the sentence, "We begin with the standard, as described in Casey " and then goes on to note how it will apply that standard: "The rule announced in Casey [ ] requires that courts consider the burdens a law imposes on abortion access together with the benefits those laws confer." Whole Woman's Health , 136 S.Ct. at 2309. And in fact, in announcing this rule, the Court cited specifically to the balancing the Casey court did for provisions not justified by a concern for women's health-those related to spousal notification and parental consent. Id. (citing Casey , 505 U.S. at 887-98, 899-901, 112 S.Ct. 2791 ).
Not only does Whole Woman's Health confirm that courts must apply the undue burden balancing test of Casey to all abortion regulations, it also dictates how that test ought to be applied. Citing Casey , the Whole Woman's Health Court emphasized that the undue burden test requires courts to "retain[ ] an independent constitutional duty to review factual findings where constitutional rights are at stake." Whole Woman's Health, 136 S.Ct. at 2310. In other words, a court cannot merely depend on legislative statements and findings in evaluating the constitutionality of laws regulating abortion. Id. The proper standard is for courts to consider the evidence in the record-including, expert evidence. Id. And, as we discuss next, this is precisely what the district court did below.
B. The evidence of burdens and benefits
1. Burdens
Noting the Supreme Court's mandate to consider the evidence in the record and then weigh the asserted benefits against the burdens, the district court did just that; it made findings and evaluated the persuasiveness of the evidence regarding the burdens and benefits created by the new ultrasound law. 273 F.Supp.3d at 1021 (citing Whole Woman's Health, 136 S.Ct. at 2310 ). Beginning with the burdens, the district court considered the burdens as presented by PPINK, focusing first on the proper population to consider, and then considering how the new regulations impact finances, employment, child care, and the safety of women in abusive relationships.
As the district court noted, "[t]he proper focus of constitutional inquiry is the group for whom the law is a restriction, not the group for whom the law is irrelevant." Id. at 1021 (citing *819Casey , 505 U.S. at 894, 112 S.Ct. 2791 ). In this case, as the district court correctly determined, the new ultrasound law is a restriction primarily for women for whom an additional lengthy trip to a PPINK health center for their informed-consent appointment acts as an impediment to their access to abortion services. The district court found specifically that this group consisted of low income women who do not live near one of PPINK's six health centers where ultrasounds are available. We agree with the district court on this point, but also note that the concerns about confidentiality in employment situations and abusive spouses that we address further below, can create impediments that span income levels. Nevertheless, our analysis, like the district court's, does not rely upon this larger group.
All of the burden in this case originates from the lengthy travel that is required of some women who have to travel far distances for an ultrasound appointment at least eighteen hours prior to an abortion. Recall that before the enactment of the new ultrasound regulation, all women seeking an abortion had to travel some distance to the nearest PPINK facility at least eighteen hours prior to an abortion in order to participate in an informed-consent information session. Because the law did not require that women have an ultrasound until just before the abortion procedure, however, they could participate in that informed-consent meeting at any of the twenty-three PPINK facilities spread throughout Indiana.6 Now, however, they must travel on the day prior to the abortion, to one of six PPINK facilities that has ultrasound equipment. As we noted above, this means that some women must travel great distances twice in order to receive an abortion. For example, women in the second largest city in Indiana, Fort Wayne, must now travel approximately 400 miles over two days to obtain an abortion, as the closest ultrasound machine is 87 miles away in Mishawaka (174 miles round trip) and the nearest abortion-providing health center is 115 miles away in Lafayette (230 miles round trip). R. 24-1 at 3, 13-14. Previously, when Fort Wayne still had its non-abortion-providing health clinic, women in Fort Wayne could have their fifteen-minute-long informed-consent appointment right at the PPINK health center in Fort Wayne.
Although the travel distance is the origin of the burden, the district court found that the strain of the law extends into the realm of finances, employment, child care, and domestic safety. The district court considered the testimony of PPINK's expert in gender studies, poverty, and low-wage labor markets, Dr. Jane Collins, who explained the impact of the new law on these interconnected stressors and on the already precarious financial lives of poor women seeking an abortion. R. 24-2. She analyzed the family budgets of low-income women and assessed how the additional costs associated with the new ultrasound law would impact these women and their families. Her testimony confirmed what common sense suggests. Many low-income women do not have employment that pays them when they miss a day of work or they may have precarious job situations in which they could be fired for excessive absences. A second lengthy trip for an ultrasound appointment likely requires a second missed day of work. And women with young children who could previously bring them along to an informed-consent session must leave them behind for the ultrasound, as PPINK's policies prohibit children from being present during an ultrasound. (And as we discuss below, both safety and common sense support such a policy). The new ultrasound law therefore *820requires women to arrange child care for an additional day.
Dr. Collins calculated that the additional cost posed by the new ultrasound law for a woman living in Fort Wayne, Indiana who has children, no car, and would lose a day's wages would be between $219 to $247. R. 24-2 at 18. Many low-income families have a discretionary monthly budget of approximately $40. Id. The additional expenses of over $200 constitute roughly 25% of their entire monthly budget. Id. These expenses are above and beyond the cost of the abortion itself which was, at the time of the hearing, $410 for the abortion and $100 for the ultrasound. R. 24-1 at 8; R. 35-5 at 35. Dr. Collins explained that to cover the costs associated with abortions, many women (about one third) will delay or stop paying basic bills in order to afford an abortion. R. 24-2 at 21. Up to 50% of women borrow money from family and friends. R.24-1 at 20. The district court concluded that, "for many women faced with the already high costs of an abortion and a lack of means to afford them, the additional expenses of lengthy travel, lost wages, and child care created by the new ultrasound law create a significant burden." 273 F.Supp.3d at 1028.
The cost of the ultrasound rule is measured not only in dollars but in time and access as well. Surgical abortions are available at PPINK health centers until thirteen weeks and six days after the last menstrual period. In fiscal year 2016, approximately 22% of all abortions and more than 34% of surgical abortions performed at PPINK took place in the three weeks before the deadline. R. 24-1 at 7-8. Women often push up against the deadline because they are gathering the necessary funds, making logistical arrangements or because they failed to promptly recognize the signs of pregnancy. R. 24-1 at 7-8. (Most women cannot know they are pregnant until at least 4 weeks following their last menstrual period, thus reducing the time they have to discover the pregnancy, explore their options and discuss them with a partner, family or doctor, arrange for missed work and child care, and secure two appointments-to only nine weeks, 6 days for a surgical abortion and thirty-five days, for a medical abortion).
Before the new ultrasound law, PPINK could usually accommodate women imminently facing the deadline by scheduling an informed-consent appointment at the nearest PPINK health center and then, the next day, she could travel the further distance, if necessary, to a PPINK facility that offered abortion services. After the enactment of the new law (and before the district court issued a preliminary injunction), the PPINK health centers with ultrasound machines became so overwhelmed with appointments that PPINK could not adequately respond to women who contacted PPINK near the end of the allowable time period. As a result, PPINK had to double book appointments causing further delays for women and longer wait times for women who were already missing work time and needing to arrange child care. Even with overscheduling, appointment availability grew scarce and women had to wait longer to have an abortion. This precluded the option of medication abortions for some women and any abortion choices for others. Abortion appointments were already scarce in Indiana given that physicians are only available at the four health centers offering abortions at limited times: Indianapolis (3 days/week); Bloomington (1 day/week); Merrillville (1.5 days/week); and Lafayette (1 day per week). R. 24-1 at 6. With such limited availability, delays in getting an ultrasound appointment might mean having to wait an entire week longer before a physician is available at the closest PPINK center, or travelling to the health center where a physician is on duty. Moreover, the new *821law causes other problems related to delay. Although the informed-consent process only took approximately fifteen minutes before, after the enactment of the new law, the process took as long as seventy-five minutes. This added to the cost of child care, missed work time, and made it harder to hide visits from abusive partners.
The district court credited the evidence that the demands on the PPINK staff trying to accommodate so many additional ultrasound appointments during the period of time the law was enforced were unsustainable. The additional quantity of appointments required staff to stay late and took away resources from the many non-abortion services that PPINK provides such as cancer screening, well-women health screening, family planning, and preventative services. According to PPINK's CEO, requiring staff to work at this pace and level of intensity is not workable over the long term, and ultimately leads to high staff turnover, exacerbating the problem further. R. 24-1 at 11.
Finally, the district court found that the new regulation has an impact on victims of domestic violence. The district court noted that one national study showed that 13.8% of women who had an abortion had been in an abusive relationship within a year before the abortion. 273 F.Supp.3d at 1026. Instead of stealing away for a fifteen-minute informed-consent session at a nearby PPINK health center, abused women trying to keep their choice confidential have to arrange to be away for all or most of two days.
The district court also considered the anecdotal evidence submitted by PPINK about nine women who could not obtain an abortion due to the burdens imposed by the new ultrasound law. The court considered the following narratives collected by PPINK from women who described their experiences as follows:
• The nearest PPINK health center to a woman seeking an abortion was over an hour away, and due to the fact that she has two young children and difficulty with transportation, she was unable to schedule the two lengthy trips during the thirteen week, six day timeframe in which an abortion is available.
• A woman from the Fort Wayne area did not schedule an abortion because of the two lengthy trips necessary. She was eleven weeks, four days pregnant when she contacted PPINK, but could not miss work twice within the short timeframe remaining.
• A woman who previously had an abortion at PPINK called to schedule another, but ultimately said she could not schedule one after she was informed she would have to make two trips to the PPINK health center in Bloomington, Indiana.
• A woman living in a shelter with two young children decided not to schedule an abortion appointment because of the transportation and childcare difficulties two appointments would cause.
• A woman who recently started a new job after a year of unemployment stated that she could not drive the three-hour roundtrip to a PPINK health center on two separate occasions due to the combination of work, childcare, and transportation expenses, in addition to her concerns regarding the confidentiality of the abortion.
• A woman who did not learn she was pregnant for ten weeks faced a long delay before she could have her informed-consent appointment that required travel to a PPINK health center, and by the time of her appointment she was one day beyond the deadline for an abortion.
• A woman from Fort Wayne who had a previous abortion at PPINK called to *822schedule another, but once she was informed that she would have to make two lengthy trips to a PPINK health center, she said she could not afford to do so and did not schedule an abortion.
• A woman living an hour north of Fort Wayne who has special needs children declined to schedule an abortion after learning that she would have to make two lengthy trips for each appointment, as she could not afford to be away from her children for that long on two occasions.
• A woman from Fort Wayne who was approaching the deadline to have an abortion declined to schedule an appointment due to the required travel and risk of missing the deadline by the time she could schedule both appointments.
273 F.Supp.3d at 1029-30 (citing R. 24-1 at 16-17; R. 38-1 at 1-2).
Before the district court, and again on appeal, the State argued that PPINK's examples were unreliable as they were passed on to a PPINK staff member and then to the declarant without PPINK taking any action to verify the information. The district court considered the reliability issue but, noting that a court could base a preliminary injunction on less formal procedures and less extensive evidence than a trial on the merits (citing Dexia Credit Local v. Rogan , 602 F.3d 879, 885 (7th Cir. 2010) ), the district court found the evidence to be sufficiently reliable for the purpose at hand. The court reasoned that the reports reflected a plausible, if not likely, consequence of the new ultrasound law. Moreover, the court recognized that, as we explore more fully below, the State's "only evidence that the law furthers its interest in promoting fetal life is from a woman whose testimony was admitted into evidence through the declaration of her physician." 273 F.Supp.3d at 1030 (emphasis in original). In a good-for-the-goose-and-gander way, the district court pointed out that if, for purposes of the preliminary injunction, the court ignored all evidence not directly from its source, "the State would be left without any evidence directly supporting its position." Id. The district court's comparison was apt and its conclusions reasonable. We cannot say that this was an abuse of discretion to consider the anecdotal evidence on both sides.
The State argued that PPINK could mitigate these burdens by making different medical and business decisions, primarily by outfitting all of its health centers with less expensive ultrasound equipment and by putting more resources toward abortion services.7 PPINK's director of abortion services explained that the $25,000 machine that PPINK ordinarily purchases comes with an extended warranty, includes planned maintenance, replacement parts, software updates, support, and a guaranteed 24-hour response time if there are any problems with the machine. R. 38-1 at 3-4. It also integrates with PPINK's electronic record system which is critical when the ultrasound and abortion appointment occur at different health centers.
*823Id. at 4. And, as PPINK points out, even if it could afford to buy the machines, it would still be limited by space and personnel. The district court rejected the State's mitigation argument, noting that the "undue burden inquiry does not contemplate re-examining every pre-existing policy or practice of abortion providers to see if they could further mitigate burdens imposed by a new abortion regulation." 273 F.Supp.3d at 1023. In general, courts do not micromanage an entity's business decisions. See Riley v. Elkhart Cmty. Sch. , 829 F.3d 886, 895 (7th Cir. 2016) (noting that, in the Title VII context, courts do not "second-guess[ ] employers' business judgments"). Provided PPINK set forth a reasonable explanation for purchasing these particular ultrasound machines-and it has, indeed, done so-the district court was entitled to defer to PPINK's justifiable business decisions and consider the burdens of the new ultrasound law within the context of the reality that exists for both PPINK in operating its business and for the patients it serves. We agree with the district court's assessment and its deference to PPINK's reasonable business decisions. In addition, we note that it also would be reasonable for PPINK to make decisions about its medical equipment needs based not only on economic concerns, but also on its ability to provide the best medical care for its patients, to attract certain medical professionals, for the safety of its technicians, to prevent malpractice claims, or for any number of other legitimate reasons.
The same can be said of PPINK's staffing decisions. The State thought that PPINK also could mitigate burdens by training nurse practitioners to interpret ultrasounds. This training, however is both expensive and requires four weeks away from clinical work. PPINK rationally could determine that it was not the best allocation of its resources. The State argues that the right to an abortion does not insulate PPINK from making difficult decisions about the allocation of resources. That may be true, but neither is it appropriate for an opposing party or a court to dictate the best use of resources for a business, provided its choices are within the range of reasonableness-but particularly in the case of a non-profit agency with limited funding seeking to provide the most efficient health care services to a mostly poor population.
The district court credited the attestation by PPINK's president and CEO that PPINK was unable to supply each center with the equipment and staff it needed to provide ultrasounds. In response to the new law, PPINK did buy one new ultrasound machine for one of its non-abortion-providing health centers and trained a staff member to use ultrasound equipment at another, indicating its commitment to providing as much service as it could despite the burdens of the new law. The State's argument about PPINK merely needing to shift resources to afford the ultrasound machines is both odd and unworkable. Only 7% of PPINK's patients receive abortion services, so in theory PPINK could shift resources for the 93% of its other services to abortion services. It seems illogical for a state with an asserted interested in protecting fetal life to be encouraging PPINK to shift all of its resources from other healthcare, such as pregnancy prevention and cancer screening, to abortion services. It is unworkable because, as we noted, neither the State nor the courts has the authority to rewrite PPINK's mission and dictate how it must allocate its limited resources. PPINK operates in a world where limited health care dollars for mostly poor women must be allocated in an efficient way, and in a way that provides the greatest care for the greatest needs.
*824The fact that courts are bound by the reality in which the laws operate is reflected in other abortion cases. In Whole Woman's Health , the Court found that the requirement that all abortion facilities meet the standards for ambulatory surgical centers would reduce the number of abortion facilities in Texas from forty to seven and thus unconstitutionally burden the right to an abortion. Whole Woman's Health , 136 S.Ct. at 2301, 2318. The Court looked at the cost a facility would have to incur to meet the requirements-$1-$3 million-and assumed that the facilities would close rather than be able to meet the requirements, despite the fact that each facility could, in an alternate universe where resources were unlimited, simply make the changes. Id. at 2318.
Similarly, in Schimel , this court looked at the burden imposed by the proposed abortion law requiring physicians who provide abortion services to have admitting privileges at a hospital within thirty miles of the abortion clinic. Planned Parenthood of Wisc., Inc. v. Schimel , 806 F.3d 908, 918 (7th Cir. 2015). In granting an injunction, we recognized that the delays caused by the new law might cause some women to lose the chance to have an abortion within the time period that Planned Parenthood allowed. No one in that case suggested that Planned Parenthood provide later term abortions. Id. Instead this court spent most of the opinion examining the reality of what an emergent situation might look like in the abortion context and how a patient in such a situation might receive care. Id. at 912-16. Courts must consider the impact of the new ultrasound law based on the reality of the abortion provider and its patients, not as it could if providers and patients had unlimited resources.
The State's arguments about mitigating child care burdens similarly miss the mark. The State suggests that women simply could bring along their children to the ultrasound-most of which are performed transvaginally at these early stages. See R. 35-5 at 27. But Planned Parenthood's policy prohibits children at ultrasound appointments, and with good reason. One wonders at what age a child could appropriately sit through such a procedure? A woman undergoing a transvaginal ultrasound must lie still while the transducer is inserted into her vagina and used to view the fetus and her organs. See https://www.healthline.com/health/transvaginal-ultrasound. She would have no way to soothe a crying baby or monitor a toddler running through the exam room. Neither the person performing the ultrasound nor the patient is in a position to monitor the safety of the child in a medical examination room, and PPINK submitted evidence of its concerns about the "serious risk of distraction" for the doctor performing the procedure. R. 35-5 at 26. Nor would most women wish to undergo such a procedure with a pre-teen son or daughter in the room, even with, as the State suggests, a sheet draped over her legs. Like all women, poor women deserve a level of dignity and choice about the confidentiality of their healthcare. Moreover, this is a perplexing argument from a State that wants women to seriously "reflect upon compelling evidence of fetal humanity," and form a bond with the fetus "while viewing this live, moving image of their baby, with arms and legs." Appellant's Brief at 2, 4. It seems likely that having children in the room would significantly decrease the ability for serious reflection in the bulk of situations.
2. Benefits
Balanced against these substantial burdens, the district court considered the intended benefits of the new law. As we just noted, the State wishes to "encourage women to reflect upon compelling evidence *825of fetal humanity," and to persuade a woman to reconsider her decision to have an abortion. Appellant's Brief at 2, 18, 20, 23. The State argues that ultrasounds have a unique impact on a pregnant woman because they allow her to see her own fetus rather than a photograph or illustration of a generic fetus, and this, the State hopes, helps "create a bond that leads them to continue their pregnancy." Appellant's Brief at 4.
The new ultrasound law encourages women to carry pregnancies to term in two ways, the State argues. First, it gives them information about their particular fetus and, second, it gives them time to reflect upon that information before they make their final decision. According to the State, "A woman offered the chance to view an ultrasound 18 hours before an abortion may well have a different mindset than a woman who has already made a final decision and presents herself at the clinic to carry it out." Appellant's Brief at 25. To support its claim that ultrasounds matter, the State introduced a study demonstrating that for the 7% of women who seek abortions and have medium to low "decision certainty," (presumably meaning that they are not very certain about their choice), those who viewed an ultrasound image had a 95.2% rate of proceeding with an abortion compared to 97.5% rate for women with high decision certainty who viewed an ultrasound. Mary Gatter et. al. Relationship Between Ultrasound Viewing and Proceeding to Abortion , 123 Obstetrics & Gynecology 81, 83 (2014). This evidence, however, is focused on the benefits of an ultrasound and not the benefits of an ultrasound eighteen hours before an abortion.
The State's strongest evidence that the eighteen-hour requirement provides some benefit, however, came from the testimony of Dr. Christina Francis, a board-certified obstetrician-gynecologist who testified that she had a patient who would have benefitted from the new law. The patient had a medication abortion at PPINK in Indianapolis and underwent the required ultrasound that day, as required by the law at the time. She told Dr. Francis, that she regretted having the abortion
and feels that an ultrasound waiting period would have given her more time to consider her decision and change her mind.... She underwent the ultrasound on the day of her abortion, immediately prior to receiving the medication. She chose not to view the ultrasound image because she felt that if she saw an image of her baby it would cause her to change her mind. She told [Dr. Francis] that she did not want to be persuaded not to abort because she was already at the clinic, had paid for the abortion, and felt pressured by those circumstances to go through with it. [She] told [Dr. Francis] that had she undergone the ultrasound the day before the abortion, she likely would have viewed the image and she does not think she would have come back the next day to proceed with the medication abortion.
R. 35-1 at 5. This is the State's strongest evidence because it is the only evidence that the eighteen-hour waiting period matters for women seeking abortions, as opposed to the ultrasound itself.
The State also argued that voluntary waiting periods are common for other procedures where physicians give patients the opportunity to weigh the costs and benefits of various options and think of additional questions or concerns. As evidence, the state presented the declaration testimony of Dr. Francis, who explained her preference to "give patients time to reflect on the information they have received, weigh the possible risks and benefits of the procedure, discuss the procedure with loved ones, and ask questions of the doctor." R.
*82635-1 at 2-3. She stated that for "life altering" procedures, she provides informed-consent information one to four weeks prior to the procedure. R. 35-1 at 3. The State did not argue that a waiting period is mandatory for any of these procedures.
Finally, the State argues that the ultrasound law advances important state interests in the psychological health of women considering abortion. For this proposition it relied on a controversial and much maligned (see below) study by Priscilla K. Coleman which concluded that "quite consistently ... abortion is associated with moderate to highly increased risks of psychological problems subsequent to the procedure." Priscilla K. Coleman, Abortion and Mental Health: Quantitative Synthesis and Analysis of Research Published, 1995-2009, 199 British Journal of Psychiatry, 180-86 (2011). Moreover, the State argued, the earlier ultrasound ensures that a woman does not become psychologically committed to having an abortion only to arrive for the procedure and learn that she has waited too long.
The district court unequivocally accepted the State's asserted interests as legitimate. Indeed, Casey instructs that "the State has a legitimate interest in promoting the life or potential life of the unborn." Casey , 505 U.S. at 870, 112 S.Ct. 2791. And, of course, no one would argue that protecting maternal psychological health is not a legitimate state interest.
3. Weighing
After this thorough compilation of the burdens and benefits, the district court turned its attention to resolving the ultimate question-whether, after considering the burden the law imposes on abortion access, together with the benefits those laws confer, the new ultrasound law has "the effect of placing a substantial obstacle in the path of a woman's choice" to have an abortion. Whole Woman's Health , 136 S.Ct. at 2309 (citing Casey , 505 U.S. at 877, 112 S.Ct. 2791 ). But before the court could weigh the benefits and burdens, it had to answer two baseline questions: first, what group of women should the court consider when weighing the burdens imposed, and second, on what aspect of the law should the court focus its benefit and burden weighing analysis-in other words, what is the relevant question presented by this case. The Court in Whole Woman's Health made the answer to the first question clear by explaining that a court must look specifically at "those women for whom the provision is an actual rather than an irrelevant restriction." Id. (citing Casey , 505 U.S. at 895, 112 S.Ct. 2791 ). In this case, the district court determined that the relevant group consisted of low-income women who live a significant distance from one of the six PPINK health centers offering informed-consent appointments.
As for the question of which benefits and burdens the court must weigh, the district court emphasized that the question it was required to consider was "whether the ultrasound law provides the asserted benefits as compared to the prior law. " 273 F.Supp.3d at 1031 (emphasis in original) (citing Whole Woman's Health , 136 S.Ct. at 2311 ). In other words, the only relevant burdens and benefits to consider as a court weighs one against the other are the burdens imposed by the requirement to have an ultrasound at least eighteen hours before an abortion, and the benefits of having the ultrasound at least eighteen hours before the procedure (not the burdens or benefits of the ultrasound itself). PPINK did not challenge the requirement that a patient undergo an ultrasound at some point prior to the abortion. Nor was it challenged when it was enacted. See Appellant's Brief at 4. Therefore the benefits of having an ultrasound at some time prior to an abortion (without regard to the "eighteen hour prior" requirement) are irrelevant.
*827It is the burden of travelling twice which becomes the obstacle to access.
Having determined the proper focus of the inquiry, the district court could turn to its ultimate task of determining whether the burdens of the law's requirements were "disproportionate, in their effect on the right to an abortion" compared "to the benefits that the restrictions are believed to confer." Schimel , 806 F.3d at 919. To determine whether a burden is undue, the court must "weigh the burdens against the state's justification, asking whether and to what extent the challenged regulation actually advances the state's interests. If a burden significantly exceeds what is necessary to advance the state's interests, it is 'undue,' " and thus unconstitutional. Schimel , 806 F.3d at 919.
The district court found that the burdens were significant: additional travel expenses, childcare costs, loss of entire days' wages, risk of losing jobs, and potential danger from an abusive partner. 273 F.Supp.3d at 1037. Increased travel distance, the Whole Woman's Health Court instructed, constitutes a concrete hardship that can ultimately contribute to the burden being undue. Whole Woman's Health , 136 S.Ct. at 2313. See also Schimel , 806 F.3d at 919 (noting that the 90-mile, one-way trip from Milwaukee to Chicago might not cause a significant burden to a person who can afford a car or train ticket, but was indeed an undue burden for the large percent of women seeking abortions who live below and far below the poverty line). These are just the types of burdens, the district court concluded, that prevent women from exercising their right to have an abortion.
The funneling of all informed-consent appointments to the six PPINK health centers with ultrasound equipment imposed other burdens. It required PPINK to double-book appointments which increased wait times for appointments and elongated the duration of those appointments. These were the kinds of incremental burdens that the Supreme Court considered in Whole Woman's Health as well, when it noted that "[t]hose closures meant fewer doctors, longer waiting times, and increased crowding," and that those burdens, along with increased driving distances were the type of incremental burdens, which, when taken together adequately support an "undue burden" conclusion. Whole Woman's Health , 136 S.Ct. at 2313.
On the other side of the scale, the district court found that the State's many arguments regarding the benefits of the ultrasound missed the mark by addressing the utility of the ultrasound itself as opposed to the period of reflection. But even considering the merits of that data submitted by the State, the district court noted that the ultrasound effect study relied upon by the State described the potential impact of viewing an ultrasound to have a "very small" effect on a potential pool of only about 7% of women seeking abortions-those who had low or medium decision certainty, and no impact on women with high decision certainty-those who make up 93% of women seeking abortions. 273 F.Supp.3d at 1032-33 (citing Gatter, Obstetrics & Gynecology, Vol. 123 at 83). And although the study states that women with low decision certainty who viewed an ultrasound image had a 95.2% rate of proceeding with an abortion compared to 97.5% rate for women with high decision certainty who viewed an ultrasound, (Gatter, Obstetrics & Gynecology, Vol. 123 at 83) the State does not tell us how many women with low decision certainty changed their minds even without seeing an ultrasound image. For the ultrasound to have any impact, the women must actually view the ultrasound, and only approximately 25% of PPINK patients chose to do so (We *828do not know whether that number differs between low and high decision-certainty patients because the State presented no evidence on that point.). This means that if there is any chance that this "very small" impact will succeed it will do so only for the pool of women consisting of the 7% of abortion seekers with low or medium decision certainty and only on whatever percentage of that 7% who actually choose to also view the ultrasound, but likely only 25% of that 7% or 1.75%. Nor can we tell if these low decision-certainty patients might have changed their minds even without the ultrasound. In general, the study that both parties cite of over 15,000 women seeking abortions at a Planned Parenthood in Los Angeles demonstrated that most visits end in abortion-98.8%. Gatter, Obstetrics & Gynecology, Vol. 123 at 82. For the whole population of women in that study who viewed an ultrasound, 98.4% had an abortion. Id. It seems from the study that increasing gestational age of the fetus (something that can be determined without ultrasound), had more to do with the decision not to proceed to abortion than viewing of an ultrasound. Id. The district court concluded that if viewing the ultrasound has little to no impact, then "[i]t is simply not a reasonable assumption ... that further time to deliberate on an image that has nearly no impact at the time, would create a meaningfully stronger impact after eighteen hours." 273 F.Supp.3d at 1034.
We agree with the State that it is entitled to try to persuade women not to have an abortion even if the impact is minimal. Nevertheless, in weighing the benefit of the particular measure at issue, a court may consider the minimal putative effects of the State's action. The more feeble the state's asserted interest, "the likelier the burden, even if slight, to be 'undue' in the sense of disproportionate or gratuitous." Planned Parenthood of Wisconsin, Inc. v. Van Hollen , 738 F.3d 786, 798 (7th Cir. 2013).
More importantly, even if the ultrasound does influence a very small percentage of women to alter their decision, all of that is irrelevant, because, as the district court explained, "[e]vidence that some women's decisions as to whether to have an abortion are impacted by viewing the ultrasound is not evidence that doing so at least eighteen hours before the abortion, rather than on the day of the abortion, has any additional persuasive impact." 273 F.Supp.3d at 1032.
The State's argument that the additional eighteen hours gives women time for deeper reflection and to absorb information, actually does address the question at issue in the case, but its argument is unsupported by anything other than Dr. Francis' one anecdote. Moreover, one could just as easily infer that the impact of viewing the ultrasound image, for some women, dissipates over the eighteen hours before the abortion. The State asks us to infer that some women who choose not to view the ultrasound do so because they are under a time pressure and because they have arrived at the health center having already made up their minds, but the State offers no evidence for this. Recall that even under the old law, women who arrived at the health center on the day of their abortion had already received copious information from the State designed to alter their decision to abort, and had plenty of time-at least eighteen hours-to digest and consider their options while not under an acute time pressure.
The only relevant evidence the State submitted to support the proposition that the eighteen-hour requirement increases a woman's ability to reflect more seriously on her decision came from the testimony of Dr. Francis who stated that one patient reported to her that she might have opted *829to view the ultrasound and then might have continued her pregnancy had she been given the option to view the ultrasound eighteen hours before her procedure, as opposed to at the time of the procedure. The district court noted that this was indeed some evidence that women may change their minds if they have more time to reflect on the decision, but it also found this singular example to be exceedingly speculative. "She can only say that she 'likely' would have viewed the ultrasound, if it was offered a day earlier" (which currently only about 25% of women do), and "she 'likely' would not have returned for an abortion the next day." 273 F.Supp.3d at 1035. The district court, when weighing how much weight to give this evidence, concluded that this was "far from compelling evidence that the new ultrasound law would have the impact desired by the State, and as such, it must be given diminished weight in the balancing process." Id. We see no reason to disrupt the district court's vast discretion in weighing this evidence.
Dr. Francis' other testimony-that some of her pregnant patients have told her that "viewing an ultrasound image of their baby caused them to decide not to have an abortion" (R. 35-1 at 4)-does not add anything to the consideration of whether viewing the ultrasound eighteen hours prior to the abortion alters the calculus in any way.
The State's argument that doctors commonly use informed-consent waiting periods to give patients time to consider important medical decisions does come closer to the relevant question in the case-the benefit of a waiting period between acquisition of knowledge and a medical procedure itself. Dr. Francis testified about the importance of giving patients time to reflect, weigh risks and benefits, and think of questions. See R. 35-1 at 3. The district court noted, however, that Dr. Francis does not provide abortion services and therefore could not attest to the utility of a waiting period after an informed-consent process preceding an abortion. The district court instead gave more weight to PPINK's argument that abortion procedures are different than other procedures where doctors give information long before a procedure, because unlike in the context where a doctor is providing a previously unknown diagnosis to a patient and then detailing various options, a woman visiting PPINK to have an abortion knows her diagnosis (she is pregnant), as well as her options-she may continue the pregnancy or have an abortion. Moreover, the law already requires that she be informed of her options and wait eighteen hours until the procedure. The only issue is whether having the ultrasound eighteen hours before alters the calculus. Finally, there is a qualitative difference between a state-mandated waiting period, which the State requires only for abortions, and other optional waiting periods, for all other procedures, where a doctor and her patient may decide together whether time for reflection would be optimal or whether, for example, waiting would cause the patient anxiety, inconvenience, or deter her from having the desired procedure at all.
Moreover, as the district court discussed, there are many office procedures that gynecologists might perform immediately after discussing the procedure and asking for consent, such as colposcopies and LEEP procedures. 273 F.Supp.3d at 1034. And there are many times that doctors might need to perform an emergent procedure immediately after providing informed consent, or times in which patients might opt for immediacy even in a non-emergent situation for the sake of convenience or because they are certain of their decision. Unlike for a hysterectomy or tubal ligation, waiting a few weeks for an abortion is not an option because abortions *830are not available at PPINK after 13 weeks and six days post last menstrual period, and the most vigilant woman will not know she is pregnant until about four weeks after her last menstrual period. In short, abortions are far more time sensitive than most other elective procedures.
Ultimately, the district court's conclusion that the new ultrasound law posed an undue burden was solidified by the fact that the State had almost no evidence that the additional time to reflect advanced its interests. Almost all of the State's evidence on the benefits of the new eighteen-hour ultrasound law focused on the benefits women might receive from having an ultrasound, and not the benefits from having to wait eighteen hours after having an ultrasound to obtain the procedure. The district court noted this and therefore dismissed much of this evidence as irrelevant to the discussion at hand. But before the district court did so, it went above and beyond its duty and thoroughly evaluated the merits of the evidence nevertheless. For example, the district court rejected the State's evidence regarding women's mental health noting that the science behind Dr. Coleman's studies, described above, had been nearly uniformly rejected by other experts in the field. 273 F.Supp.3d at 1036. The district court chose to credit instead two mental health organizations that conducted a comprehensive review of studies on mental health and abortion and concluded that "on the best evidence available ... [t]he rates of mental health problems for women with unwanted pregnancy were the same whether they had an abortion or gave birth." R. 38-3 at 3. A task force of the American Psychological Association similarly reviewed studies and concluded that, "the most methodologically sound research indicates that among women who have a single, legal, first trimester abortion of an unplanned pregnancy for non-therapeutic reasons, the relative risks of mental health problems are not greater than the risks among women who deliver an unplanned pregnancy." Id. at 2-3. That report specifically criticizes Dr. Coleman's 2002 report as being "characterized by a number of methodological limitations that make it difficult to interpret the results." Id. at 4. The district court, using its substantial discretion weighed the competing evidence and determined that PPINK's evidence was "significantly more persuasive on this issue, especially given that Dr. Coleman's studies are the subject of significant criticism." 273 F.Supp.3d at 1036. This type of evidence weighing is well within the district court's province, and we see no reason to disturb its thoroughly reasoned findings.
More importantly, the court below noted, Dr. Coleman's study failed to address the relevant question: whether having an ultrasound eighteen hours before an abortion leads to more favorable psychological outcomes. Id. And, as we have discussed extensively, this was the ultimate question that the district court had to address.
The district court considered all of the following together: the fact that over a third of surgical abortions occur within three weeks of PPINK's deadline for performing abortions; the difficulty of making two lengthy trips in quick succession; the over-booking of informed-consent appointments; and the fact that physicians only provided abortions on limited days in each health center. Adding these burdens together, the district court concluded, "it would be surprising if the new ultrasound law did not prevent a significant number of low income women from obtaining an abortion." 273 F.Supp.3d at 1038. And indeed the evidence provided by PPINK from nine women who were, in fact, severely burdened and impeded in their attempts at obtaining an abortion in the short time that the law was in effect, confirms this prediction.
*831The district court did not err by concluding that the ultrasound law "imposes significant burdens against a near absence of evidence that the law promotes either of the benefits asserted by the State." Id. at 1039. See Whole Woman's Health , 136 S.Ct. at 2318 (striking down the challenged abortion restrictions because the law "provides few, if any, health benefits for women" and "poses a substantial obstacle to women seeking abortions."). "A statute that curtails the constitutional right to an abortion ... cannot survive challenge without evidence that the curtailment is justifiable by reference to the benefits conferred by the statute." Schimel , 806 F.3d at 921.
The State would like to simplify the court's complex burden and benefit weighing to a more cookie cutter approach and have us conclude that Casey paved the way for an almost per se approval of all reasonable waiting periods. Appellant's Brief at 20-22, 38. The Supreme Court in Casey upheld a twenty-four hour, informed-consent waiting period despite the fact patients would need to make two sometimes lengthy trips in order to obtain an abortion. Casey , 505 U.S. at 885-87, 112 S.Ct. 2791. And we followed suit in A Woman's Choice-E. Side Women's Clinic v. Newman , 305 F.3d 684, 691 (7th Cir. 2002), upholding Indiana's eighteen-hour waiting period after an in-person informed-consent meeting. But one of the primary lessons of Whole Woman's Health is that the burden and benefit weighing is context-specific. In Whole Woman's Health , the court based its conclusions about undue burden on the 280,000 square miles of Texas territory, the number of abortion-offering facilities that could operate after the enactment of the contested law, the number of patients each remaining facility would have to accommodate (1,200 per month), the distance women would have to travel to get to a clinic and the population numbers for women who would have to travel this far, the rate of deaths and complications from abortions in Texas, and the cost to clinics of coming into compliance with the new regulation. Whole Woman's Health, 136 S.Ct. at 2301-03. The Court spent much time discussing the importance of these facts in assessing the constitutionality of the contested law, noting that a statute valid as to one set of facts may be invalid as to another. Id. at 2306 (citing Nashville, C. & St. L.R. Co. v. Walters, 294 U.S. 405, 415, 55 S.Ct. 486, 79 L.Ed. 949 (1935) ).
The Court in Casey noted that "in theory at least, the waiting period is a reasonable measure to implement the State's interest in protecting the life of the unborn," and went on to analyze whether such a "waiting period is nonetheless invalid because in practice it is a substantial obstacle to a woman's choice to terminate her pregnancy." Casey , 505 U.S. at 885, 112 S.Ct. 2791 (emphasis ours). This was the exact reasoning we adopted later, in light of Casey , where we noted that "[w]hile a twenty-four hour waiting period that requires two trips to an abortion provider has been found not to impose an undue burden on Pennsylvania women based on the circumstances of that state at the time the Court decided Casey, a similar provision in another state's abortion statute could well be found to impose an undue burden on women in that state depending on the interplay of factors"-factors such as "the number of physicians who perform abortions, the number of abortion facilities, the distances women must travel in order to reach an abortion facility, and the average income of women seeking abortions." Karlin v. Foust , 188 F.3d 446, 485 (7th Cir. 1999). The language in these cases reflects that the facts and context rule the day when evaluating waiting periods. This is far from being a blanket stamp of approval on them.
*832Analyzing the regulation in light of the reality of the facts in Indiana is precisely what the district court did in this case. A court cannot assess the law in a world where PPINK has unlimited resources to open dozens of clinics, each with the ability to provide ultrasound and abortions along with unlimited access to other health care needs, or in a world where all women have paid sick days, and reliable child care and transportation. The court must take the facts as they are presented before it and compare the burdens against the weight of the evidence of the benefits specific to the proposed law. Whole Woman's Health , 136 S.Ct. at 2310. The district court did just that and concluded that the evidence of benefits was exceptionally slight if any, and the burden imposed by the double travel requirement great. As the district court concluded, "the new ultrasound law creates significant financial and other burdens on PPINK and its patients, particularly on low-income women in Indiana who face lengthy travel to one of PPINK's now only six health centers that can offer an informed-consent appointment. These burdens are clearly undue when weighed against the almost complete lack of evidence that the law furthers the State's asserted justifications of promoting fetal life and women's mental health outcomes." 273 F.Supp.3d at 1043.
The State argues to this court that the district court's findings are clearly erroneous. We cannot agree. Under the clear error standard we can reverse a district court's factual findings only if "based on the entire record, we are left with the definite and firm conviction that a mistake has been committed." United States v. Orillo , 733 F.3d 241, 244 (7th Cir. 2013). The district thoroughly addressed each of the burdens and benefits asserted by the parties and engaged in a painstakingly thorough weighing. Its factual findings were not clearly erroneous and are entitled to our deference.
C. The remaining preliminary injunction considerations
That conclusion about the likelihood of success on the merits does not end the inquiry, although it certainly puts the heaviest weight on the scale. PPINK must also show that it is likely to suffer irreparable harm in the absence of preliminary relief and that it has no adequate remedy at law. City of Chicago v. Sessions , 888 F.3d 272, 282 (7th Cir. 2018). "If those burdens are met, the court must weigh the harm that the plaintiff will suffer absent an injunction against the harm to the defendant from an injunction, and consider whether an injunction is in the public interest." Id.
For PPINK and its patients who lose the opportunity to exercise their constitutional right to an abortion, the irreparability of the harm is clear. Even an extended delay in obtaining an abortion can cause irreparable harm by "result[ing] in the progression of a pregnancy to a stage at which an abortion would be less safe, and eventually illegal." Planned Parenthood of Wis. , 738 F.3d at 796. The evidence suggests that the new ultrasound law has already prevented some women from exercising their constitutional rights. It has caused delay to others. Because we, like the district court, have concluded that there is no substantial evidence that the law furthers its stated interest, any harm to the State is minimal, at worst. And the State certainly has myriad remaining methods to persuade women to carry a pregnancy to term in order to promote the State's interest in promoting fetal life.
The State argues that PPINK can avoid some of the harm by expending more resources on abortion services, buying more ultrasound machines, but again, the court must take the record as it finds it and not base its finding on what the facts might look like if the court could devise a different *833business or care model for PPINK. See, e.g., Whole Woman's Health , 136 S.Ct. at 2318. In any case, PPINK presented evidence that it has already shifted resources and tried to mitigate harm to the best of its ability. Some of these changes are unsustainable long term. Others cannot be made at all.
Balanced against the harm to PPINK patients is the State's claim of irreparable harm. The State faces the same harm any State faces when a democratically enacted law is enjoined. It also claims that it would prevent it from furthering its goal of promoting fetal life. This we think, is a minimal, potentially temporary harm in this case, compared with the burdens on the women that the district court identified. This is particularly true when we consider that-even crediting the State's asserted benefit of the law-the potential ability to alter any woman's decision in a manner that protects fetal life would be minimal, at best.
As for the public interest, the district court found that upholding constitutional rights serves an important public interest and we see no reason to add to or upset this finding. And because the State had not demonstrated that its interest would be served by the law, neither could the public's interest.
III.
The State asserts that its reason for this new eighteen-hour ultrasound requirement is to persuade women not to have an abortion. There is no doubt that this is a legitimate position for a state to take. But it is also true that women have the right to choose to have an abortion, albeit with some limitations. Casey , 505 U.S. at 846, 112 S.Ct. 2791. Women, like all humans, are intellectual creatures with the ability to reason, consider, ponder, and challenge their own ideas and those of others. The usual manner in which we seek to persuade is by rhetoric not barriers. The State certainly is entitled to use these rhetorical tools to persuade women not to have an abortion. It has chosen to do so by requiring an informed-consent process-the required contents of which it has designed and mandated-and an ultrasound and fetal heart beat requirement. It also requires every woman to receive a brochure about abortion, the contents of which the State controls in toto-from how it will present the images of fetuses to the decisions about which medical risks it includes and which it omits (for example, the brochure which a woman takes home and is supposed to ponder for eighteen hours, does not speak of the risk to the fetus from drugs and alcohol that a woman may have consumed prior to knowing about an unplanned pregnancy). Moreover, it states as fact that "human physical life begins when a human ovum is fertilized by a human sperm"-a proposition debated among scientists, religious leaders, and medical ethicists. The State has vast power to use the information that it provides to persuade women not to have an abortion. But the requirement that women have the ultrasound eighteen hours prior to the abortion places a large barrier to access without any evidence that it serves the intended goal of persuading women to carry a pregnancy to term. Instead, it appears that its only effect is to place barriers between a woman who wishes to exercise her right to an abortion and her ability to do so. Rhetoric and persuasion are certainly legitimate methods for a state to assert its preference, but it cannot force compliance with its otherwise legitimate views by erecting barriers to abortion without evidence that those barriers serve the benefit the state intended. "Until and unless Roe v. Wade is overruled by the Supreme Court, a statute likely to restrict access to abortion with no offsetting medical benefit cannot be held *834to be within the enacting state's constitutional authority." Schimel , 806 F.3d at 916. In light of the evidence of substantial burdens imposed by the law and without evidence that the additional eighteen hours following an ultrasound has any legitimate persuasive effect on decision-making, the law constitutes an undue burden on those seeking an abortion without any known benefits to balance it. The opinion of the district court is AFFIRMED in all respects.
Kanne, Circuit Judge, concurring in the judgment.
Our decision today is compelled by long-standing Supreme Court precedent. See Roe v. Wade , 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973) ; Planned Parenthood of Se. Pa. v. Casey , 505 U.S. 833, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992).
The State's reason for the new 18-hour ultrasound requirement is to persuade women not to have an abortion. As the opinion notes, "[t]here is no doubt that this is a legitimate position for a state to take." Majority Op. at 833. This, of course, is weighed against the fact that "women have the right to choose to have an abortion, albeit with some limitations." Id.
In this case two evidentiary factors lead me to conclude that the 18-hour requirement imposes an undue burden on a woman's right to choose, which requires affirming the decision of the district court. The first factor is the additional travel necessitated by the availability of only six ultrasound imaging sites located in Indiana at PPINK health centers.* The second factor is that the State offered little evidence to show that an 18-hour wait following an ultrasound would persuade those seeking an abortion to preserve fetal life.
Based on the foregoing factors, I agree that, in the context presented by this appeal, Ind. Code § 16-34-2-1.1(a)(5) -as written-constitutes an undue burden on women seeking an abortion.
This concurrence extends to the final judgment set forth by my esteemed colleague, Judge Rovner, but does not endorse the propriety of the ancillary findings of the district court.