Friday, July 25, 2014

Fearlessness in the Mundane: Yet Another Reason Not to Prescribe Antibiotics Indiscriminately



My stunningly attractive and sophisticated wife happens to be a brilliant GI and liver pathologist (yes, I married up). From time to time, she’ll try to educate me by sharing an article from her field with me. Such was the case about a week ago when she handed me an article titled, Hepatic Histological Findings in Suspected Drug-Induced Liver Injury: Systematic Evaluation and Clinical Associations.  Now, this stuff is typically way over my little pediatrician head, so I told her I’d read it… later. But she wouldn’t relent. “Look at Table 1,” she said. “You need to know this.” Boy, did Table 1 get my attention. 
 
The article was written by members of the Drug-Induced Liver Injury Network (or DILIN) which is an effort involving some of the most prestigious medical centers in the country to better categorize and understand liver injury caused by exposure to drugs. If you had asked me, last week, what drugs are most commonly associated with liver injury, I would have guessed some big-gun, heavy-duty medications such as steroids, immune suppressants such as those used after organ transplantation, chemotherapy drugs, and perhaps some of the anti-fungals. I was only partly right.

Table 1 of the article listed the characteristics of the 249 consecutive patients who were included in the study. The age range of the patients was 7 to 87 years of age. What got my attention were the 8 most common drugs implicated in liver injury.
Turns out I was right only about the steroids. The heading, “Anabolic Agents” was number 6 on the list. They also happened to be the only drug class listed that was not an antibiotic.  Here’s the complete list of the top 8 culprits:

1.      Amoxicillin/clavulanate
2.      Nitrofurantoin
3.      Sulfamethoxazole/trimethoprim
4.      Minocycline
5.      Ciprofloxacin
6.      Anabolic agents
7.      Azithromycin
8.      Levofloxacin

This list should give every physician pause because they are all medications that are prescribed quite frequently. In fact, I suspect that it is the very frequency with which we prescribe them that explains their presence on this list. The likelihood that any one of these medications result in liver injury in an individual patient is extremely low, but when you consider how many courses of antibiotics are prescribed, it’s not surprising that this very small percentage ends up accounting for the majority of incidents. To make an analogy, many more people die in automobile accidents than aviation accidents. Yet more people have a fear of flying than a fear of driving.

Of course, this does not mean we should stop using antibiotics altogether: they are a necessary and useful tool when indications call for their use. But this is yet another warning that the injudicious, indiscriminate use of antibiotics is fraught with danger. No patient should suffer complications from a medication or surgical procedure that was not needed in the first place.

Friday, June 27, 2014

Rational Foolishness: mis-communication, mis-perceptions and the over-prescription of antibiotics


According to the Centers for Disease Control and Prevention, more than 2 million people in the United States suffer infections due to bacteria that are resistant to antibiotics each year, and 23,000 die as a result of these infections. These statistics are worrisome and sobering.  Particularly since this problem is, to a large part, a direct consequence of the unnecessary over-prescribing of antibiotics by physicians.  This begs the question: why do doctors prescribe antibiotics when they are not needed?
I’ll offer my answer to this question at the end of this article but first consider this: a recent survey of clinicians and patients shed some light on this phenomenon.  Just 5% of physicians surveyed said they never prescribe antibiotics when they are not absolutely sure that they are not necessary, and 11% said that “antibiotics won’t hurt if not needed and could help the patient sooner if I’m correct.”
Twenty-eight percent of clinicians surveyed stated that the patient’s request for an antibiotic was the reason for prescribing it when they are not certain that one is necessary, and 10% will prescribe an antibiotic to satisfy a patient’s request when they know it is not indicated.
Satisfying perceived expectations of patients is one of the most often cited reasons pediatricians give for dispensing unneeded antibiotics. Yet, the perception of parental expectations on the part of pediatricians can be completely unfounded as shown by a landmark study published in Pediatrics in April of 1999 (Mangione-Smith et al – The Relationship Between Perceived Parental Expectations and Pediatrician Antimicrobial Prescribing Behavior).  Some of the findings of this study were as follows:
  •  About half of the parents reported that they expected to be prescribed an antibiotic   when they brought their child in with symptoms of the common cold.  
  •  The physician’s perception that the parent expected an antibiotic was the only reliable predictor of prescribing an antibiotic for a viral condition (in which an antibiotic is not expected to be of any use).  Doctors prescribed antibiotics 9% of the time for viral infections when they did not think the parent wanted an antibiotic, but did so 52% of the time when they believed the parents expected one.
  • Surprisingly – or maybe not so surprisingly, if you read my book, Suffer the Children  -- when pediatricians perceived that parents wanted an antibiotic they were more likely to diagnose a bacterial infection.  Let me repeat that.  The physician’s perception of parental expectations drove the diagnosis rather than the patient’s presentation and physical exam findings.  An ear infection was diagnosed 49% of the time if they thought the parent wanted an antibiotic and only 13% of the time when they felt the parent didn’t want one.  Overall, pediatricians made a bacterial diagnosis in 70% of children with upper respiratory symptoms if they thought the parent wanted an antibiotic compared to 31% when they felt the parent did not desire one.
  •  But doctors were often wrong in their assessment of parental expectations.  Twenty-seven % of the time they believed a parent wanted an antibiotic when the parent did not desire one, and they correctly surmised that a parent wanted an antibiotic only 41% of the time when the parent did expect a prescription.
  • Though the desire to satisfy patients is often quoted by physicians as a reason for prescribing unnecessary antibiotics, in this study, failure to provide antibiotics when the parent expected a prescription did not affect satisfaction.  The only predictor of dissatisfaction was poor communication.
Good communication between physician and patient is a cornerstone of high-quality medical care.  Which is why some proponents argue that the best way to reverse the trend of over-prescribing antibiotics is to improve physician communication skills.  One approach is to ask the patient direct questions in regards to their expectations. Yet physicians balk at this, claiming that doing so may exacerbate direct confrontation if the patient were to cling to an unreasonable demand.
“You’re not in a negotiating stand,” a general practitioner interviewed for one study said with regards to when a patient has a high expectation for an antibiotic when one is not warranted. (Mustafa et al, Managing Expectations of Antibiotics for Upper Respiratory Tract Infections: A Qualitative Study)  “You are very much in a direct conflict stand, which doesn’t help anyone.” As a result, practitioners surveyed for this study disclosed that they resorted to indirect ways of eliciting a patient’s expectations, contrary to the advice of many communication’s efforts.
What can parents do to avoid receiving an unnecessary prescription for an antibiotic?  Try to do your best to communicate your expectations with your physician.  Rather than offering “candidate diagnoses” (“I’m worried Johnny has a sinus infection”) try to stick to describing the actual symptoms.  The mention of candidate diagnoses by parents increased the likelihood of antibiotic prescription in one study.  If the doctor determines your child has a viral infection that will not improve with an antibiotic, it’s perfectly appropriate to ask her how she came to this conclusion, what to expect in the coming days, what you can do to help alleviate the symptoms, and what to look for as far as worrisome symptoms that might require follow-up.  Disputing the diagnosis, on the other hand, will often lead physicians to cave in and prescribe a medication your child will not need.
I told you earlier that I would offer my own explanation as to why physicians are so prone to prescribing antibiotics when they know they are unlikely to help.  Isn’t it irrational for highly trained professionals to prescribe treatments that are counter to accepted guidelines of care?  
Actually, the unnecessary prescription of antibiotics is perfectly rational from their point of view of the physician – what one might term, rational foolishness -- and can best be understood in terms of potential risks and rewards.  Risks and rewards for the physician, that is, not for the patient.
Though on a population basis the over-prescribing of antibiotics constitutes a public-health risk, with regard to the individual patient, it is hard for the physician to envision any imminent harm (unless the patient suffers an unanticipated allergic reaction to the medication).  Thus, the way the doctor sees it, the risk of not prescribing a medication (missed diagnosis, lawsuit, unsatisfied patient, loss of time and unpleasantness resulting from a discussion of why an antibiotic is not needed) outweighs the risk of prescribing (rare allergic reaction, unlikely that MY patient will get a resistant bacterial infection from just one teeny antibiotic script).  Besides, we doctors don’t have to pay for the medication.
   

Monday, April 7, 2014

Does the United States have the best health-care in the world?



Does the United States have the best health-care in the world?

That was the question posed by a noted professor of medicine at a recent conference I attended.  Of the physicians and other health professionals that made up the audience, not one raised their hand.  Not one!  How could that be?

Was no one in the audience aware that the United States has among the highest survival rates for cancer treatment in the world?  That our outcomes for complicated surgery are the envy of the globe?  That trauma care is not equaled in any other country? 
Was the audience composed of a bunch of anti-American radicals intent on destroying the best system of medical care the world has ever seen?  After all, look at all the wealthy foreigners that flock to our hospitals to receive the latest innovations in treatment from the best and brightest doctors in the world!

One can say, with a fair degree of confidence that the United States is the best country in the world to be in if you are very sick (definitely where you want to be if you get shot, stabbed or if you’re involved in a high-impact motor-vehicle accident – yes, we’re the kings of trauma care!)  But the objective of a health-care system goes beyond caring for the ill:  it has the responsibility to help preserve health among the non-infirm.  And that’s where we don’t do so well.

This is not a function of the quality of the doctors in our country but of the organization of the preventive services we have available.  The key concept at play is that improving the medical care physicians provide will have a minimal impact on broader outcomes of health in our country.
This can be easily understood when we break down the determinants of one’s health listed here from highest to lowest importance:

Behavioral patterns (smoking, diet/exercise, use of seatbelts, etc.):     40%
Genetic predisposition:            30%
Social circumstances (income, socio-economic status):   15%
Medical care received:    10%
Environmental exposure:   5%

As you can see, medical care received is near the bottom of the list.  So access to great physicians, laboratory tests and newest medications is relatively unimportant in terms of an individual’s overall health compared to habits, genetic make-up and socio-economic status.

Yet, we continue to invest disproportionately in medical care; in fact we spend way more per capita on medical services than any other country in the world.  To make things worse, the traditional model of physician re-imbursement has always favored performing as many procedures (often of dubious value) on the individual patient as possible while not rewarding efforts towards improving health on a community level.


The professor giving the talk offered the following solution, very much tongue-in-cheek:  “If you’re really sick, come to the United States.  If you’re healthy and want to stay that way, go to Norway!”

A better solution would be to re-examine our health-care priorities in this country:  to change physician compensation so that it rewards preventive care instead of offering perverse incentives; to address social determinants of illness and eliminate disparities in delivery of care; and to engage patients as partners in health to promote healthy behaviors.  Until we do so, a large proportion of our health-care dollars will go up in smoke.