With the practice of medicine changing at an unprecedented pace, doctors are challenged with making ethical decisions about money, romance, and patient care on a regular basis. More than 7500 physicians from over 25 specialties responded to our survey asking how they tackle some of the toughest ethical questions. This slideshow highlights their answers and insights, and how physician attitudes have evolved over recent years.

The vast majority (78%) of physicians say they would not deny or avoid treatment in the face of financial penalties. That's an increase from 74% in 2014. Many commented—in clear and colorful language—that patients come first and that such behavior would be "unconscionable."

But many also feel that their backs are against a wall. "No," writes a family physician, "and that is why I would be fired." A pathologist says, "Yes. I am a salaried employee and have no choice."

"Yes," writes an orthopedist, "and this will become a more and more common practice as physicians become corporate employees rather than their own person. When this occurs and keeping your job entails toeing the corporate line, ethics will take a back seat."

Still, many doctors note that there are ways to achieve both ends. "There is often a way to save money with a treatment with similar efficacy," suggests a family practice physician, while a neurologist observes, "Not everything that is appropriate is necessary."

Arthur Caplan, PhD, bioethicist at NYU Langone Medical Center, says the idea that some doctors are becoming "bottom-line oriented" is a little unsettling. "Hopefully if they answered 'yes' or 'it depends,' they are thinking of other positive treatment strategies that they could pursue, rather than simply withholding treatment," he says.

The share of doctors saying they would decline a potentially more effective treatment because a patient's insurance wouldn't pay for it, while small, is creeping up. But doctors are quick to take issue with the idea that they are actively "denying" treatment. Many respondents note that patients have the option of self-pay.

"I do not deny but discuss the reason for not recommending the treatment with the patient," writes a neurosurgeon. Says an internist who works in a federally qualified health center, "New drugs are coming out all the time for diabetes. I know that the GLP-1 receptor agonists aren't going to be paid for by insurance, so I usually don't offer them."

Many doctors noted that they simply cannot afford to pick up the tab themselves. "If insurance doesn't pay and the patient cannot pay, the cost falls on the practice," writes a radiation oncologist. "I am happy to do charitable work and waive fees where possible, but sometimes this is not practical or even possible."

Overall, physicians are confident that they could be unbiased in their prescribing habits even if they were paid to speak on behalf of a drug company or accepted meals from drug representatives.

In 2010, about 37% answered no to "Would you agree that you should refuse gifts or perks from pharmaceutical companies because they may influence your medical judgment?"

"It is more than annoying to believe that doctors are so gullible that all it takes is a meal to get them to prescribe a new drug," writes a psychiatrist. "I enjoy the meals and the presentations, because without them—and the questions and comments of other doctors—I get to learn very little regarding new drugs in my field. There's no time to read about them during the day at work, and I'm too busy or too tired to do it after work."

Doctors answering "no" point out that drug companies wouldn't use these tactics unless they worked. "Are you kidding me?" writes an internist. "Drug companies aren't fools and they don't spend all that money for no reason. We all are susceptible to marketing manipulation." A pediatrician agrees: "Anybody who answers yes to this question is delusional or lying."

The percentage of physicians saying they would be willing to upcode a patient's condition to get treatment covered fell by more than half compared with 2014.

In many cases, a physician's "no" was followed by a "but":

"No, but I will paint a tableau of horror and litigation to describe what will happen to their medical director if they fail to heed me," writes a family physician.

"No, but I would use the code to its fullest," comments a neurologist.

"No, but my previous employer encouraged the practice and penalized those who didn't comply," says an HIV/AIDS physician.

"No, but the temptation is always there to do so," notes a cardiologist.

"No, but there are gray areas in presenting information," writes a family practitioner.

Some long-held barriers seem to be breaking down to a degree. The percentage of doctors saying it is okay to become romantically involved with a patient or former patient has increased since 2010, although it is still a minority. While most respondents (70%) say such relationships are absolutely off limits, that's down since 2010. Only a tiny minority (2%) of doctors consider a relationship with a current patient to be permissible, but many say relationships with former patients might be acceptable.

"[It's] consenting sovereign adults doing what they feel is right," writes a dermatologist. Other physicians say a relationship with a former patient is allowable, assuming that the patient is not a psychiatric patient and receives medical care from another provider.

For some, the nature of a doctor's specialty has to be considered. "As a gynecologist, absolutely never. There may be other specialties, such as pathology or radiology, where in certain circumstances such a relationship might be ethically acceptable."

Still, some primary care physicians don't rule out the possibility. "I don't personally think it's ethical to have an intimate relationship with a current patient," writes a family physician, "but someone I once treated, sure."

Female physicians take a dimmer view of doctor-patient romantic relationships versus their male counterparts. One out of 4 male doctors say that physician-patient relationships are acceptable, but fewer than 1 in 5 female doctors agree.

Caplan, the bioethicist, says, "Women are more aware of the imbalance of power." They are more likely to have felt pressured or to have had to fend off someone's unwanted advances, "so they know what it's like to be put in a vulnerable position. They understand that role a little better."

The idea of a relationship between a doctor and a patient's family member (assuming the family member is involved in medical visits) presents a murkier question for physicians. While 70% of doctors oppose relationships with caregiving family members—the same percentage that opposes relationships with patients themselves—19% say a relationship might be acceptable depending on the circumstances. Only 10% of doctors say such relationships do not present an ethical concern.

Among the many physicians who say "it depends," the acceptability of such a relationship with a patient's family member would depend on "type of physician," "location and community size," and "as long as the doctor-patient relationship has ended." One orthopedist notes that relationships with family members are often in place before relationships with patients. "Many of my patients are family members of friends. If I became involved with one of my friends, I don't see a problem with it."

As one family physician writes, "I live in a community of 2000 people and 1500 are patients. What else is there—priesthood?"

Six years ago, 95% of physicians said it is never acceptable to cover up or avoid revealing a mistake that would potentially harm a patient. That percentage dropped to 78% in this year's survey.

Caplan calls the increase, "surprising and disturbing." He says, "The movement to apologize is very strong on the administrative side, and we're seeing a big push to make hospitals and nursing homes safer." Physicians' willingness to cover up mistakes "runs against safety issues and concerns. This is a finding that makes me nervous. The shift is in the wrong direction."

"There is no pleading the Fifth Amendment in the hospital—only in the court room. What doctor wants to self-incriminate?" asks a urologist.

"I wouldn't lie if asked a direct question, but I might not volunteer that I made a mistake if I saw the potential for serious harm to the patient by doing so," says a neurologist.

"We all make mistakes. However, covering up for them is never acceptable, especially when a patient may be compromised," says a radiologist.

Many of the doctors who say "it depends" want to know more about the situation. "How could it cause harm?" "How good is my rapport with the family?" "How much harm was done?"

There's been little or no change in the responses to this question over the years. The large majority of physicians (78%) say they would report an impaired colleague, noting the importance of patient safety and their legal obligation to do so.

Many write that they would only report a colleague after individually approaching him or her first. Some say that they would "try to get him help." Others would "warn him in advance" that they planned to report.

Many respondents clarified that they would only report the situation if their colleague was impaired at work. "A doctor who is not on call and drinks too much at a dinner party should not be reported," notes a psychiatrist. An orthopedist noted that the determining factor is "if it impairs his practice of medicine."

With 41% in favor and 42% opposing, the issue remains deeply divisive. In 2014, 39% favored drug testing, compared with 41% in 2016.

"I think it is a good idea if it truly reflects a physician's ability to provide care at the time of service," writes an internist, but "if a physician has a social drink 2 hours after work and tested positive for it, then he or she should not be penalized for that."

A psychiatrist writes, "For those performing invasive procedures (surgery, interventional radiology, interventional cardiology, ob/gyn) absolutely yes." Many simply commented, "Yes; I have nothing to hide."

Those who are opposed note that a positive test does not necessarily constitute abuse but can ruin a physician's career. "A positive EtOH or positive THC test does not distinguish between use or abuse and could lead to great harm," writes a family physician. An internist says, "Absolutely not. Only if there is probable cause. I want my constitutional rights protected like everyone else's."

Primary care physicians were very slightly more in favor of testing (43%) than were specialists (40%). Several primary care physicians made comments along the line of "Airline pilots do it" and "Would you want surgery from a surgeon under the influence of drugs or alcohol? I wouldn't."

Among specialists, nephrologists were the most supportive of mandatory testing (50%), followed by anesthesiologists (48%), a group considered to be at high risk for abuse. One anesthesiologist called mandatory testing a "deterrent," while another focused on patient safety. "I have worked with impaired physicians and nurses before. It is such a huge risk for everyone."

Urologists (26%) were by far the least likely to back mandatory testing. One urologist commented that it "adds a Big Brother mentality to medical practice."

Nearly two thirds of physicians (63%) say they would not avoid patients with comorbid disease or those who did not adhere to treatment plans if they were being rated on outcomes in a new payment plan.

Those who say they would (17%) or might (20%) avoid such patients lament such practices as "a matter of practice viability," writes one ob/gyn. A radiologist notes that it's "an unavoidable result of penalizing physicians for poor results in comorbid patients," while a family practice physician calls it "as good an opportunity to prune the vine as any."

Indeed, the question hit a nerve with some respondents who are frustrated by quality metrics. One gastroenterologist lashed out, "Bad outcomes are usually due to noncompliance on a patient's side, so physicians should not be penalized for this. The whole idea of quality-based reimbursement is completely absurd and disgusting."

Among specialists, orthopedists and plastic surgeons are the most likely to say that they would avoid treating riskier patients. Not surprisingly, critical care doctors (7%) are the least likely to say they would turn away such patients, noting that it would be "unethical in an ICU environment" and that they are "morally obliged to treat all patients."

Some consider the question through a financial lens. An orthopedist writes, "I can't afford to take complicated cases if it decreases income," while a plastic surgeon says, "I can't help anyone if I'm out of business." But many others say the decision is strictly a matter of patient safety and the efficacy of the treatment. "I do that already and tell patients with significant comorbidities that they have a higher chance of serious complication. I feel it's the right thing to do," writes one orthopedist. "I might decide to avoid procedures that are more risky but would not avoid the patients," writes another.

One endocrinologist acknowledges that the decision would be a struggle. "I like to think that I would have the ethics to avoid this, but if I could not make a living, then I am afraid I would be tempted to do this."

The percentage of physicians supporting the idea that patients who practice unhealthy behaviors should pay more for insurance has declined compared with just 2 years ago.

Among those opposing different rates based on behaviors, some voice an altruistic, humanitarian perspective. "Patients should be supported rather than blamed," writes an ED physician. "It's not ethical," says a gastroenterologist. Others say it's an issue of fairness. "Most health plans have a copay on visits/tests/medications that causes people with unhealthy lifestyles to end up paying more. Why should the premium cost more too?"

Others are pragmatists. "We take patients as they are," wrote an orthopedist. "What characteristics are you going to penalize, and what is genetic and what is behavior?"

Arthur Caplan, a bioethics professor and director of the Division of Medical Ethics at New York University Langone Medical Center, says the dramatic shift likely reflects the fact that tiered payment plans based on behavior "haven't worked very well, and I think doctors know that. It doesn't seem to work in terms of producing better, healthier behavior."

Since 2010, the percentage of respondents saying they would prescribe a placebo has nearly doubled. Many physicians mentioned the pressure of needing to have patients rate them highly on physician-rating sites or in hospital-rating questionnaires.

"You can thank Press Ganey patient satisfaction surveys and employers for that!" laments an internist.

"A lot of medicine involves placebo, such as the recommendation to take acetaminophen for basically everything," writes a pediatrician. "I love this idea," submits a dermatologist. "We should do this far more often," writes an emergency medicine specialist.

Although the shift seems to run counter to the emphasis on patient autonomy, bioethicist Caplan says, "There has been a mountain of data showing the power of the placebo effect, and as a result there has been a shift in the ethical argument. I think we are catching that movement in this question. It is a placebo-specific response."

Doctors are backing away from the idea of performing procedures based on fear of malpractice suits. Only 13% of physicians say they would do so, down from 20% in 2014.

"It's dishonest," writes a radiologist. Writes a family physician, "An unwarranted procedure is malpractice."

"I have definitely ordered tests where I think the likelihood of something being wrong is very low," says an internist. "However, if the test isn't run and something comes up down the road, I would definitely be held liable."

"Stress tests. We need to be 100% accurate. The 'retrospectoscope' feeds the lawyers," says a cardiologist.

"If the procedure were diagnostic and would help me feel more confident about my diagnosis, I would," says an anesthesiologist.

"I would not carry out an unwarranted medical procedure," says an emergency medicine physician.

Specialists were slightly more likely than primary care physicians to say they would perform a medically unwarranted procedure because of malpractice concerns.

Not surprisingly, answers varied widely across specialties, with radiologists saying they would be most likely to perform such procedures.

"Sadly, it happens every day in radiology," notes one. "So many procedures, e.g. MRI, are not necessary but are ordered out of ignorance or fear of litigation or (and this makes me so angry) because the cardiologist owns the CT scanner and is in it for the money. I have to decide whether to go along or fight, and it's just not possible to tilt at all of those windmills."

Another notes, "Unnecessary imaging studies are done routinely to avoid lawsuits. For some, procedures have simply become an entrenched part of practice."

That sentiment resonates with an ob/gyn, who asks, "Ever heard of C-sections?"

More than two thirds of respondents say that doctors who have patient contact should be required to get an annual flu shot.

"Do no harm," writes a pediatrician. "We don't have the right to expose our patients," says a pulmonologist. "We cannot put our patients at risk," comments a cardiologist. "Lead by example," writes one family medicine physician, while another says, "A harmless gesture to protect our patients is a gesture of beneficence—suck it up, gang."

While physicians may understand the importance of flu shots, a quarter of them bristle at the notion that they should be required. Doctors must be afforded the same choices as the rest of society, they insist. "Why should our choice of profession require us to give up our personal liberties?" asks an anesthesiologist.

The vast majority of doctors say that they have never suspected and then failed to report or investigate a case of domestic violence. The percentages were virtually identical among male and female physicians.

Among those who have had this suspicion, many note that their states do not have laws requiring them to report domestic violence involving adult victims. Others say the victims asked them not to report the situation.

"In my state, unless the abuse involves a knife or gun or weapon, the police will not investigate and cannot bring charges," writes a gerontologist. "We are instructed at our courses on domestic violence, mandated for licensure, not to report these cases to the police."

"My state does not mandate reporting or have a reporting mechanism for cases involving competent adults," notes one palliative care physician. "Mostly in spousal abuse where we are not mandatory reporters," writes another.

"In our state, child abuse is a mandated reporting event but adult domestic violence unfortunately is not; therefore, we have to go by the wishes of the individual (although in these situations I obviously try and encourage the adult to seek help)," says an internist.

"I have been requested [to not report it]," writes a neuropsychiatrist. "I gave all appropriate shelter and abuse intervention information."

Just over half of the respondents say they would caution a patient against having a procedure performed by a physician whose abilities they questioned, but as one family physician notes, "this is very tricky."

Many doctors note the need for tact. "Subtle but still clear suggestions are the best way to go," says one. Rather than questioning another doctor's abilities, many note that they would simply recommend that a patient seek a second opinion or would recommend a doctor whose abilities they respected. More than a third (34%) say that it would depend on the situation.

As one orthopedist says, "A doctor practicing outside of their scope of practice? Yes. A doctor that I think sucks (my opinion)? No."

Fewer than 4 out of 10 doctors (38%) say they would drop tight-fisted insurers. That represents a decrease since 2010. As a result, the number of doctors saying they would eliminate such insurers is nearly equal to the number saying they wouldn't.

"Unfortunately, the economics and business of medicine cannot be ignored," writes a pulmonologist. Says a pediatrician, "I already dropped Medicare/Medicaid for just that reason."

"Yes, and long-time patients would drop me if the insurance became too expensive for them," notes a family physician.

Others resoundingly disagree. "I was in this exact situation and did not drop the insurer/patients," writes a pulmonologist. A pediatrician comments, "No; the state insurance is low and slow but the practice is surviving and I sleep well at nights."

Whether they answered "yes," "no," or "it depends," doctors' comments regarding confidentiality indicate that they are largely of one mind: They will follow federal, state, and local mandates requiring them to report specific communicable diseases. The difference lies in whether they perceive such reporting to be a violation of confidentiality.

"Yes, I'd follow the law. HIV and other communicable diseases must be reported," writes an immunologist.

"Yes, that is allowed and the right thing to do," writes a family physician.

"Who are the others? If it is a sexual partner who I am seeing as a doctor, I would encourage a joint meeting," says a family physician.

"No; state-mandated reporting does not constitute a breach," says an endocrinologist.

For the record, Caplan says that where the law requires reporting, there is no confidentiality and therefore no ethical breach.

Ob/gyns are far less likely than primary care physicians as a whole to say that they would breach patient confidentiality (42% vs 59%). This is probably because ob/gyns treat more sexually transmitted infections and are familiar with various federal, state, and local reporting requirements. Because they routinely comply with these mandates as a matter of practice, ob/gyns may be less likely to regard compliance as a breach of confidentiality.

"If by 'breach of confidentiality' you mean 'report it to the department of health or CDC,' then yes, since we are mandated to do so for certain diseases."

"Yes. I work in ob/gyn and frequently get the Department of Public Health involved."

Where the law is silent, however, ob/gyns say they would "strongly encourage the patient to share the information" or would "let the patient know that it is her or his responsibility to prevent the spread of disease."

Doctors overwhelmingly (71%) say that practitioners of alternative medicine should be licensed by the states.

But the very logic that has some doctors insisting upon licensure for alternative practitioners—"there is far too much fraud happening because these providers are unregulated"—has opponents decrying the idea of even legitimizing these practitioners: "Absolutely not; they are charlatans." State licensure of alternative healers "suggests state approval for something very iffy," cautions a pediatrician.

Pragmatists note that the lack of science supporting alternative therapies would make writing standards—and hence creating licensure requirements—extremely difficult. A trauma surgeon says, "I don't see how alternative therapists would be licensed since there are no standards or science."

"I'm not convinced that naturopaths and other alternative practitioners should be allowed to ply their trade at all," writes a family physician.