It isn’t much of a case, but it created questions in my mind.
A mom brings her 8 year old daughter to the hospital for a nonproductive cough. No fever. No runny nose. Just a cough. The patient had started school again this week, and so the microbiome in her nasal passages had thus begun mixing with all of the other microbiomes on school lunch tables, desks, and childrens’ shirt sleeves. The end result was that now she was coughing for a couple of days – like a majority of other children in the school.
The child looked fine. I told the mother that she likely had a “head cold” and that it would have to run its course.
The mother wasn’t convinced.
“How do you know she doesn’t have the flu”?
“Well, I don’t know for sure, but even if she did have the flu, it wouldn’t change the management right now. It would still have to run its course.”
“I want her tested for the flu.”
“Influenza testing really won’t help us much. The test itself has a high false negative rate, meaning that even if the test is negative, a large percentage of people still end up having influenza.”
“I want her tested for the flu.”
Fine, I thought. It’s your money.
Forty five minutes later, results from the influenza swab came back negative.
“So like I was saying before, this is something that will have to run its course.”
“You said that the test wasn’t accurate. There are a lot of kids in her school with influenza right now. I want her to get Tamiflu.”
“Tamiflu isn’t going to help. It is only effective within the first 48 hours. You said the cough started two days ago.”
“You know, I have always had good things to say about this hospital and I’ve never had this many problems with a doctor in the emergency department before. It hasn’t been 48 hours and I want her to get the Tamiflu.”
So I gave the woman a prescription. Let her spend the $120 for a drug that won’t work.
Then I began thinking. Would some bean counting clipboard carrier claim I provided unnecessary care?
So what do you think? Knowing that rapid influenza testing has a significant false negative rate and knowing that influenza is widespread in the country, but also knowing that a patient has mild symptoms …[poll id=”8″]
By the way, after the visit, a pharmacist called and said that the patient was requesting a different medication since the patient was on state “insurance” and the “insurance” didn’t cover the cost of the medication.
Unfortunately, there isn’t anything else that works which was covered.
UPDATE 1/15/2012Thanks for the votes and comments.
When thinking about this situation, three issues came to my mind.
First, Tamiflu is approved by the FDA for acute uncomplicated influenza when symptoms have been present for not more than two days. It is also approved for prevention of influenza. From a technical perspective, a Tamiflu prescription was indicated if the child had influenza or if the child was being prophylaxed against influenza. The indications for use do not contain a description of any symptoms that must be present before Tamiflu is prescribed. If influenza is diagnosed or being prevented, Tamiflu is indicated.
Second, diagnosis of influenza in an inexact science. Rapid influenza testing is notoriously inaccurate. If the test is positive, there is a high likelihood that influenza is present (although about 1 in 50 patients with positive results do not have influenza). However, if the test is negative, one third of children and about one half of adults will still have influenza. A negative influenza test by no means excludes a diagnosis of influenza. Between 30 and 50 percent of influenza cases are asymptomatic and it is difficult to distinguish between a common cold and influenza early in the course of the disease.
Third, Tamiflu probably doesn’t work. A Cochrane review found that Roche funded all of the studies demonstrating Tamiflu’s efficacy, that Roche hid a majority of the data from studies about the effectiveness of Tamiflu (and still won’t disclose it), and that the study data that was released had significant bias. See also this article from Forbes.
Putting all of these facts together, then we need to consider why most people in this case thought that prescribing Tamiflu was “unnecessary”.
Was it unnecessary because Tamiflu doesn’t work? Then anyone who prescribes Tamiflu is providing unnecessary care.
Was it because the patient tested negative for influenza? How do we accurately diagnose influenza in order to prescribe the medications, then?
Was it because the patient’s symptoms weren’t severe enough? What symptoms are needed before Tamiflu can be appropriately prescribed?
In the end, when dealing with this patient’s mother, I decided that the prescription of Tamiflu ends up being an issue of medical discretion. Most people probably wouldn’t have given the medication to their children for those symptoms, but if she wanted it that bad, the medication was technically medically indicated and not prescribing the medication would have resulted in even lower patient satisfaction scores. So I gave her the prescription.
The incentive to improve patient satisfaction favored prescribing Tamiflu and I had no disincentive for doing so. If the patient’s mother wants to pay for a useless medication that has many side effects in order to treat mild symptoms – after knowing these facts – why should she be prevented from doing so. This is just one example of how emphasizing patient satisfaction adversely affects medical judgment, encourages testing and treatment which is of questionable benefit, and drives up the cost of care.
“Unnecessary care” is going to be a catch phrase in the next few years as government and insurers try to decrease expenditures in medical care by refusing to pay for testing or treatment which is deemed “unnecessary.” When you hear this phrase, ask yourself who is making the determination, how the determination is being made, and who stands to benefit.