Sunday, June 27, 2004

Alternatives to EBM 

I was cleaning out my office lately and found this list from a few years ago. The source is unknown:

1. Eminence based medicine
2. Vehemence based medicine
3. Eloquence based medicine
4. Providence based medicine
5. Diffidence based medicine
6. Nervousness based medicine
7. Confidence based medicine
8. Opulence based medicine
9. Annoyance based medicine
10. Arrogance based medicine
11. Propaganda based medicine

And, my favorite...
12. Webidence based medicine

I figure we could add:
13. Surrogate marker based medicine
14. Pharmaceutical Rep based medicine

Wednesday, June 23, 2004

Simple Breast Cancer Screening Algorithm 

As a family physician, I share many female patients with a gynecologist. Most of the time it works. Sometimes it doesn't.

I recently had a 48 year old HMO patient contact my office needing a referral to a breast surgeon for a "second opinion" regarding a "nodule".

I asked my staff to get more information. Turns out, it wasn't a lump but rather an abnormal mammogram.

I asked her to come in so we could talk and she agreed.

Here's the actual sequence:
1. My patient can self refer for an annual "well woman" exam. Never mind that this is a political/benefit issue > medical issue. The benefit does not require the patient nor the gyn to let me know she has been in or what they did. Some gyn's keep me in the loop by way of a letter or courtesy copy of any tests but most don't. It sort of works as long as the patient does not have a problem and the tests are normal. In this case, I did not get a letter from the gyn about my patient's clinical breast exam status (lump or no lump?). I was able to obtain a Birads 3 report that included a comparison with three other normal mammograms over the past 8 years.
2. My position is to offer mammograms in women between the ages of 40 and 49 but not recommend them until age 50. I also use the American Academy of Family Physicians Breast Cancer Screening Counseling Tool handout and strive to keep my language neutral. I think of it as seeking her permission but only after she has the necessary information upon which to decide. In this case, she told me the gyn recommended the test just like the other times without any discussion or tools.
3. The patient confirmed at the time of the well woman visit neither of them thought that she had a breast lump.
4. The mammogram report said:
Small density laterally in the left breast. This appears benign but cannot be confirmed on previous studies due to the difference in technique. A repeat left cc view for this region is recommended in six months time. Probably benign, Birads, category 3.
5. The gyn recommended she see a breast surgeon so I got involved when the patient called my office to arrange what I like to call "the hearsay referral". Namely, the specialist tells the patient to tell me.
6. Anyway, she comes to see me and I examine her. We both agree there is not a palpable lump.
7. I go into shared decision making mode. For example: "I would like you to have some additional information, then whatever you chose will be fine with me". We review false positive, false negative and low positive predictive values for Birads 3 mammograms in a screening situation for women under 50. I ask her: "If the surgeon recommends a biopsy, would you have it?" or "Are you comfortable honoring the radiologist's recommendation to repeat an Xray in 6 months?". She surprises me when she states her initial reaction to the gyn's recommendation was to call the gyn and seek a more conservative option. She was told by the gyn's nurse: "We send all abnormal mammogram patients to a surgeon".
8. I'm thinking, must be nice...no time spent explaining anything before or after a test and no tracking/follow-up/reminders.

On that basis, this appears to be the gyn's breast cancer screening algorithm:
1. Recommend mammograms early and often.
2. If normal, done. See #1...
3. If anything but normal, send to a surgeon.

The gyn's system is designed to meet her needs but not mine and certainly not the patient's.

Tuesday, June 22, 2004

Turn Left Right Here 

Or...two wrongs don't make a right but three rights make a left.

Wrong side surgery continues to be a problem: Yahoo! News - Doctors Must Double-Check Before Surgery.

My hospital committee was recently asked to update our existing policy to meet JCAHO standards (Patient Safety, Universal Protocol for Wrong Site, Wrong Procedure and Wrong Person Surgery).

As a family physician, I've always been aware of the issue but had not thought much about it.

Until recently, each institution could have their own policy. Naturally, this led to confusion. I think this will help but won't solve the problem until the culture of medicine changes away from provider centered to patient centered.

Monday, June 21, 2004

Due Remember Me 

I was looking something else up and came across this: The Journal of Urology - Abstract: Volume 170(6, Part 1 of 2) December 2003 p 2356-2358 Salvage of Sildenafil Failures Referred From Primary Care Physicians.The first paragraph has an error that makes for a fun malapropism considering the article is about "inadequate ... education" [my bold].
Purpose: Sildenafil citrate is an effective first line agent for most causes of erectile dysfunction. Primary care providers (PCPs) write the majority of these prescriptions and most failures of sildenafil therapy are subsequently referred to urologists for alternative therapies. Often it is concluded that the drug is ineffective when in actuality the failure is do to inadequate patient education. We examined patients referred from PCPs who were nonresponders to sildenafil therapy and attempted to convert them to responders through reeducation.
Even the title is fun: Salvage of Sildenafil Failures ...

I think of the word "salvage" in the context of oncology therapy. I'm not sure it applies for Viagra failures according to the MedlinePlus: Medical Dictionary that says:
Main Entry 2: salvage
Function: transitive verb ... to save (an organ, tissue, or patient) by preventive or therapeutic measures. Examples: ... a salvaged cancer patient, or ... salvaged lung tissue.

Friday, June 11, 2004

House Painter Says: "House Needs Painting" 

Yet another "expert" is describing how they would like the world to be instead of how it is...
Sinus headaches are just another medical meme. This (Yahoo! News - Sinus Headaches Often Really Migraines: Study) research only validates something obvious. Which came first? Patients deciding most of the headaches were "sinus" or passive medical providers not taking the time to challenge this widespread belief? The article says:
In the study, over-the-counter pain relievers were the most commonly used medications, followed by non-prescription antihistamines. Only about 10 percent of headache sufferers had been prescribed drugs called triptans, which Eross described as the "gold standard" for treating migraine.

Although triptans were used the least, people who took them were most satisfied with the treatment, according to Eross.
I'm thinking of course the treatment is unsatisfactory when the diagnosis is wrong. That does not make triptans a "gold standard".

The article concludes with the usual self endorsement:
Eross encouraged people who think they are having sinus headaches to ask their doctor if they might have a migraine. In many cases, it may be a good idea to see a headache specialist, he said.

Thursday, June 10, 2004

My Mail Bin 

I'm on an electronic medical record (EMR) so most of my day is moving electrons around. Dealing with paper is an ongoing problem. We have a good lab and Xray interface as long as the data is within my integrated health care system. This means I get those reports automatically tagged to the patient account and that makes me spoiled. My patients get Xray results the same day and most lab results the next day.

Next month will be three years on the EMR. To date, only a handful of the specialists that I work with have agreed to change how they send outpatient consult information. We don't even need the other doctor to do anything different but continue to dictate. What happens behind the scenes is my group practice works with the specialist's transcription service to include the necessary patient identification codes.

The doctors generally agree to convert (after all, I'm not asking them to do anything) but I understand from my people that we get undermined by the office managers of the specialists because they don't want to change.

Today I got two progress notes about two different patients by traditional paper and U.S. Mail. The first was 16 working days ago. At least the patient is stable.

The other was 12 working days ago and recommended surgery. Turns out, the patient had the surgery one week ago. I actually found out earlier this week when I heard about a post operative complication that resulted in an emergency room visit when I needed to arrange some follow up medical testing.

Let's not forget, reducing delay = care.

Number Not Needed to Barf 

How fun is this? From today's headlines comes Cheap Drugs Work Against Post-Op Nausea per the NEJM -- A Factorial Trial of Six Interventions for the Prevention of Postoperative Nausea and Vomiting.

First some facts that this family physician found interesting from the article:
1. Untreated, one third of surgery patients will have post operative nausea, vomiting, or both.
2. The problem is so big, they even have an acronym, postoperative nausea and vomiting (PONV).
3. There is literature to identify high risk patients Anesthesiology, V 91, No. 3, Sept 1999. The four predictors are: female, history of motion sickness or PONV, nonsmoking and use of postoperative opioids. If none, one, two, three, or four of the risk factors are present, the incidence of PONV is 10%, 21%, 39%, 61% and 79% respectively.

The trial included patients at high risk for PONV. The primary endpoint was nausea and vomiting within 24 hours of surgery.
Three drugs were compared: 4 mg dexamethasone, 4 mg ondansetron (Zofran), and 1.25 mg droperidol (Inapsine).

Nausea occurred in 31% and vomiting in 14%. The relative risk reduction (RRR) for all three drugs was about 26% and the absolute risk reduction (ARR) was about 10% so the number needed to treat (NNT) to prevent one primary endpoint (combined nausea or vomiting) within 24 hours of surgery = 10. Drug combinations worked even better resulting in additional RRR of 26% or ARR of 12% and NNT for 2nd drug = 8.

I was not aware dexamethasone was useful in this situation so now I'm wondering how often it is tried in my hospital. In the meantime, I've been aware that Zofran is the de facto standard "because it was better". My hospital's pharmacist tells me their cost is $ 0.81 per 5 mg Inapsine dose and $16.94 per 4 mg Zofran dose. Since the two drugs are equally good at preventing nausea and vomiting and NNT = 10, it takes $161.30 more using Zofran routinely instead of Inapsine.

Inapsine does have a black box (Torsades) but the anti-emetic dose is lower than the chemical restraint dose by about 1/4 and is dependent on drug-drug interactions.

Table 3 reminds us the absolute benefit is small (10%) unless baseline risk is large (80%). For example: If baseline = 10% and one chooses Zofran as first line, then the ARR = 3% and NNT = 33 so $/episode prevented = $559.

I don't like to generalize research done on one clinical situation (PONV) to another (hospital inpatients in general ) but I do get a lot of calls from the floor needing an order for nausea or vomiting. The nurses steer me to IV > po, IM, or pr route and their first choice is Zofran. When I ask why, they say, "it works better (how do they know?) and it is IV (patient advocacy?)". I asked a hospital pharmacist if there were other IV options? Answer is yes, Compazine (prochlorperazine), but the staff doesn't like the slow administration rate. Also, there was a recent Compazine shortage but the pharmacist reassures me this is resolved.

It is worth remembering newer is not always better. I plan to visit with an anesthesiologist to understand how they do things now and how this information may change their practice.

Tuesday, June 08, 2004

The Enemy is Us 

The PSA issues never seem to go away. Today's Medscape e-mail update says PSA For Prostate Cancer Detection but I'm disappointed when it is only the results of a poll triggered by the recent NEJM -- Prevalence of Prostate Cancer among Men with a Prostate-Specific Antigen Level <=4.0 ng per Milliliter article.

Here's the question:
Prostate cancer is not rare in men with prostate-specific antigen (PSA) levels in the normal range, according to a study in the May 27 issue of the New England Journal of Medicine. How useful do you think PSA levels are as a marker for prostate cancer?
Here's the answers (n = 633):
Very useful. 32% (358)
Somewhat useful. 52% (578)
Not very useful. 12% (132)
Not at all useful. 2% (27)
Until today, I held off on posting about this dubious article since DB's Medical Rants � Testing for prostate cancer already got it right, especially pointing out the importance of the ROC curve.

It's still about sensitivity and specificity. All tests have inherent limitations. A desirable property for a screening test is high sensitivity but it comes with a price, i.e. low specificity. Combine this with the lack of a "gold standard" and the authors sum it up nicely when they say:.
Given the lack of a rigorous evaluation of the optimal PSA level for the detection of prostate cancer and the changes in biopsy technique, it is not surprising that the predictive value of the PSA level is not known.
A discouraging 84 % (32% + 52%) of doctors find the PSA either very or somewhat useful.

No wonder the patients don't get it. The doctors don't either. We control the message that we send to patients. We doctors aren't held to a high enough standard regarding the medical literature.

Sunday, June 06, 2004

When You Can't Find a Specialist When You Need One 

Yahoo! News - Boston Patients Wait Long Time for Doctors and this Boston.com / Business / Boston found to endure longest wait for doctors (courtesy of Kevin, M.D. - Medical Weblog) validate something that I struggle with on a daily basis. In fact, just this week I got a letter from a local dermatologist announcing the addition of a physician assistant. It says (my italics):
When patients call for their appointment, it is typical that they may see Mr. X the next day or possibly even the same day as opposed to the usual 2 to 6 week waiting period for a dermatologist.
I'm not impressed since I already have Same-Day Appointments: Exploding the Access Paradigm - September 2000 - Family Practice Management but I can only control my schedule.

Right now the game is played as follows:
1. Access to me is as close to frictionless as I can make it (open access style scheduling plus e-mail availability).
2. Unfortunately, when we (me and the patient) decide we need something extra done (Xrays, cardiovascular testing, or specialty care) the delay starts.
3. The contracts all say about the same thing about access. Namely, emergency = same day, urgent/sick = a few days, and routine = thirty days. Every time I ask the insurance people, they say the access standards are the same for primary care as for specialists. The problem is the standards are not enforced.
4. My people call your people and the fun begins. The initial offered appointment is too far out for the clinical situation. The specialists have a standing rule for their front office staff that meets their needs but not mine or the patient's. If I need the patient in sooner then I'm supposed to call their doctor. In my experience, the answer is always yes. So...why do I have to call? Years ago, I could direct the referrals elsewhere. These days, network obligations limit me from redirecting the referral to another provider.
5. PCP's can make the situation worse. My spies (hospital Xray tech's and patients that work for the specialists) tell me certain PCP's are notorious for always calling to get their patients in sooner. For us system thinkers this is just aggravating the tragedy of the commons problem. Ultimately, the beleaguered specialist staff learn certain PCP's will call no matter what so they might as well just give in. Those PCP's feel the advocacy on behalf of their patient is helping but I think they make the situation worse. Calling to pass on important clinical information is advocacy. Begging for appointments is unbecoming and a waste of both doctor's time.

Rather than just complain, I have a solution from the recent Institute for Healthcare Improvement (IHI) 5th Annual International Summit on Redesigning the Clinical Office Practice from March 2004. From Mike Davies, MD talk I learned about a "Service Agreement" which is an, "understanding or agreement between any 2 parties, one of which sends work to another, defining work flow rules".

This is what patients want (in order):
1. Reassurance
2. Expectation for the process
3. Knowledge of the process
4. Certainty
5. Speed
6. Quality of the care and the experience
7. Appropriate referral
This is what PCP's want:
1. Access to specialty care
2. Speed
3. Know specialty care expectations
4. Simple process
5. Certainty of appointment
6. Questions answered
7. Let my people go
This is what specialists want:
1. Knowledge of which doctor is referring
2. Right patients sent with right information
3. Patient expectations
4. PCP expectations
5. When do I let the people go?
In other words, PCP's list access as number one. A specialist in attendance said is best when he said the most important thing to him was "octane", i.e. cases (doing heart caths instead of reassuring MVP).

I have an obligation to "package" the patient the right way:
1. Refer the right problem
2. Sell the consult to the patient
3. Clarify expectations
4. Provide the right information to the specialist
The ideal patient experience with the specialist is:
A Specialty clerk
- warmly welcomes patient
- mentions who referred the patient
- reinforces reason for and expectations from referral
The Specialty provider
- is familiar with key historical facts
- is able to provide needed care in 1 or 2 visits
- facilitates smooth hand back to PCP
I look forward to future service agreements.

This page is powered by Blogger. Isn't yours?