Competing Demands or Clinical Inertia: The Case of Elevated Glycosylated Hemoglobin

To answer the first question raised by Dr. Phillips will require a much more in-depth analysis of our rich data set than we have been able to do thus far. However, two points are worth mentioning. First, there was no relationship between time from the most recent A1c measurement and the likelihood of a change in medication for an elevated A1c. So physicians were not delaying a change in therapy because the A1c was more than 3 months ago and checking a more recent value before intensifying therapy.

But far more importantly, Dr. Phillips comment continues to ignore the important context in which decisions are made about intensifying medication, both the context of the encounter, with its multiple competing demands, and the context of the clinic. As we will show in a recently accepted manuscript to Medical Care, control of A1c is significantly associated with characteristics of the clinic, not the physician. Clinics in which the structure of the care, as well as care processes, are more consistent with the chronic care model, have patients whose A1c is significantly better. Dr. Phillips dogged persistence in narrowing the focus to just improving clinician behavior without improving overall clinic structures and processes in a manner that will overcome competing demands during the encounter has, in my mind, little hope or promise of success. For example, Krein and colleagues found more variation in diabetes care at the facility level than at the level of the primary care physician in the VA. ( Health Serv Res. 2002 Oct;37(5):1159-80)

Regarding the recent NEJM paper showing that objective measures of quality imiproved as number of medical problems increased, Dr. Phillips is comparing apples with oranges. In our study, we found no relationship between number of co-morbidities and number of questions or problems patients raised during the visit, thus increasing the level of competing demands. Furthremore, the outcome measure in this study was a compostie quality score derived from a remarkably heterogeneous set of technical process indicators that do not directly address the phenomenon of why no change in therapy was made if the A1c was elevated, although this is one of several hundred indicators that went into the calculation of this score.

Finally, this is a stimulating and vital discussion and I applaud Dr. Lawrence for his insight, experiences and his passion about how we can improve clinical outcomes for our patients with type 2 diabetes. Thank you for your response.

Competing interests:   None declared

Was diabetes an appropriate target, and on the table?

Clinical inertia implies that a change in therapy is clinically indicated. To better understand the behavior of the providers studied by Parchman and colleagues, we need to know whether a change in therapy indeed was appropriate; in evaluating failure to intensify therapy for diabetes when A1c is high, it's important to have the right denominator. Was the analysis restricted to visits in which the available A1c would have been expected to reflect previous management – say, 1-3 months after the previous change in therapy – and where life expectancy was such that benefit from intensifying therapy would have been expected?

Then, in analyzing provider actions for those visits in which therapy was not intensified, but would have been justified, it’s important to know whether failure to intensify therapy was an active decision – or whether diabetes wasn’t on the table at all. First, was diabetes or A1c or self- monitored glucose mentioned by physician or patient? Second, if any of these were mentioned, was there a reason given for not intensifying therapy – such as hypoglycemia (which should at least have resulted in a change in therapy), or a decision to address diabetes issues at the next visit?

The authors might also wish to discuss their findings in light of the recent New England Journal of Medicine paper (vol. 356, p. 2496) showing that objective measures of quality of care increased as the number of medical problems increased.

Competing interests:   None declared

Concerning languages preference, many patients and physicians were Spanish speaking in this sample.

Dr. Hicks makes an excellent point about why there was no change in medication when the most recent A1c was above goal. It is entirely possible that some of the observed visits occured shortly after a visit where therapy was intensified for A1c above goal. But the observed visit was too soon to judge the full impact of the change in therapy so no changes were made. Thanks for pointing this out!

Michael L. Parchman

Competing interests:   None declared

Physicians take care of patients---not checkmarks on a to-do list of care management chores. If family physicians believe in patient-centered medicine, then we need to allow the patient a role in setting the agenda for the visit. The article by Haidet, “Jazz and the Art of Medicine”(1), in the previous Annals made a wonderful point about why physicians can‘t just read the script (musical score) as written. We must use improvisation to flex and adjust during the visit to meet the patient’s needs.

But family physicians are in conflict. Often our bonuses and even our basic compensation often depend on the checkmarks which intrude upon our visits. Our time with a patient is limited by factors outside of our control. Our nurses take more time because they are required make their checkmarks, which might make the difference between a 90213 and a 90214 on the billing form. Technology has inserted a computer screen between the physician and the patient. Next we must add a list of care management criteria and another list of new “care initiatives”. We know we must “squeeze in” the smoking question, and ask about advanced directives on each visit. The A1C and microalbumen for diabetics must be addressed to score high on our care management/bonus incentives. Then there are still co-morbidities to be addressed.(2) Many of us fall into the trap of considering all of this before we bother to ask, “How can I help you today?”

If the patient and physician together decide that the priorities at this visit promote the patient’s depression, her alcohol problem and a new, itchy rash to a higher rung than an A1C of 7.2%, then those problems can be aggressively attacked. If we dare to consciously drop the A1C on our priority list, our statistics will suffer, but the patient will receive better care.

This Parchman article points out very well that good primary care physicians recognize these difficulties and have adopted strategies to compensate. They plan ahead to future visits. There is no emergency room pressure to do it all at one visit. As stated by Jerome Groopman, M.D. (How Doctor’s Think), “A discerning doctor will recognize when more time is needed to ask questions and explain his thinking. In such instances, the appointment may need to be extended or a follow-up visit scheduled as soon as feasible(3)”. This article confirms that we do utilize this strategy.

1. Paul Haidet, Jazz and the ‘Art’ of Medicine: Improvisation in the Medical Encounter, Ann. Fam. Med, Mar 2007; 5: 164 - 169.

2. John W. Beasley, Terry H. Hankey, Rodney Erickson, Kurt C. Stange, Marlon Mundt, Marguerite Elliott, Pamela Wiesen, and James Bobula, How Many Problems Do Family Physicians Manage at Each Encounter? A WReN Study, Ann. Fam. Med, Sep 2004; 2: 405 - 410.

3. Jerome Groopman, How Doctor’s Think, Houghton Mifflin Company, Boston, 2007.

Competing interests:   None declared

Dr. Parchman

I saw your article in the Annals. Great one. Timely. Useful. Good work.

I had a couple questions…

1. I am assuming that these were English-speaking patients? Not mentioned in the methods.

2. It does not seem like you accounted for recent changes in medications. As I am sure is common in your practice, it is rare that a HgA1c comes back as a surprise to me. After hearing their FSBSs at home and seeing the one in the office, I usually will titrate up their meds in that visit, will draw the HgA1c to confirm that we are as out of contol as suspected and give us a benchmark by which we will judge future levels. Then I see folks back in a shorter time frame to make sure we are not dropping too low and if still high to titrate up again.

So, the HgA1c is more of an afterthought and not a driver of action. If it is surprisingly high, I call the pt and titrate up the meds prior to the next visit and then reassess clinical response in that next visit as above. It seems to me that all those visits would be described as no change in therapy.

Thanks again for this article. It is a good contribution to the literature.

P

Competing interests:   None declared

This article by Dr. Michael Parchman offers a new look at how family doctors make decisions. This new perspective is extremely important in this day of care management and pay for performance. It demonstrates how careful we must be in choosing one, isolated measurement to judge a physician’s care as good or poor care. A positive decision NOT to change a medication at the current encounter is not merely “clinical inertia”.

An important aspect of the Parchman study is the use of direct observation of the physician-patient interaction. We still won’t have the perfect answer until we can actually get inside the physician’s head. Perhaps someday we can put both the patient and physician inside a functional MRI for the interview. We will see much taking place when an A1C is reviewed and the physician also sees 3 or 4 other problems looming. The actual average number of problems dealt with per visit for a diabetic patient is 4.6.(1) I’ll wager that the integrative functions in the parietal and prefrontal areas are going wild. This supercharged cognitive activity could hardly be classified as “inertia”.

As the article demonstrates, family physicians actually do shorten up the time interval until the next visit if we see an A1C too high and can’t deal with it today. We bring the patient back sooner if we have several concurrent problems to address.(2) We understand that we can’t assimilate too many changes into the patient’s daily life all at once. New medications must fit into the patient’s budget. We must be cognizant of reimbursement mechanisms, prior authorization, pre-certification and of other rules which challenge our ability to get things done. We must realize that competing demands come not only from co-morbidities, but also from technology, insurance companies, our employers and other reimbursement mechanisms. Any tool to evaluate care must take all of these competing demands into account.

This article by Parchman shows that when we veer from “the checklist” and appear to neglect a check-mark, we are not just inert lumps doing nothing, but rather we are actively interacting, prioritizing, balancing, and planning on the patient’s behalf. The integrative parts of our brains are working in high gear. We must use the new evidence provided in this article to create a rational “care evaluation” system, which takes patient concerns, co-morbidities, system complexity and other competing demands into consideration.

1. John W. Beasley, Terry H. Hankey, Rodney Erickson, Kurt C. Stange, Marlon Mundt, Marguerite Elliott, Pamela Wiesen, and James Bobula, How Many Problems Do Family Physicians Manage at Each Encounter? A WReN Study, Ann. Fam. Med, Sep 2004; 2: 405 - 410.

2. Jerome Groopman, How Doctor’s Think, Houghton Mifflin Company, Boston, 2007.

Competing interests:   None declared

Clinical inertia is recognizing that the patient is not at goal, knowing what can be done to reach goal, and not doing it. Some of apparent clinical inertia is an active decision (limited life expectancy, etc.) but most is probably more passive (didn't get around to it, etc. -- which might be due in part to competing demands and/or feeling that the asymptomatic problem generally is of low priority). But the choice is largely the provider's (most patients don't say, "Let's not worry about the diabetes or hypertension today -- other issues are more important to me today...").

To the extent that the choice is the provider's, then a focus on provider behavior might be expected to improve outcomes. In a site where there were provider-directed interventions (feedback on performance and computerized reminders), there were marked improvements in A1c and lipids, and smaller but highly significant improvements in blood pressure -- compared to no improvement in A1c at all in another site serving a comparable patient population (1); within the study site, the feedback on performance intervention was more effective than the reminders. Similarly, a feedback on performance intervention also improved A1c, blood pressure, and lipids in a group of primary care practices (2).

A1c, blood pressure, and lipids will improve little without action by the provider -- overcoming clinical inertia, and figuring out how to improve outcomes despite the presence of lots of competing demands.

(1) Phillips, L.S., Ziemer, D.C., Doyle, J.P., Barnes, C.S., Kolm, P., Branch, W.T., Caudle, J.M., Cook, C.B., Dunbar, V.G., El-Kebbi, I.M., Gallina, D.L., Hayes, R.P., Miller, C.D., Rhee, M.K., Thompson, D.M., Watkins C. An endocrinologist-supported intervention aimed at providers improves diabetes management in a primary care site. IPCAAD 7. Diab Care 28: 2352-2360, 2005.

(2) Ornstein S. Nietert PJ. Jenkins RG. Wessell AM. Nemeth LS. Feifer C. Corley ST. Improving diabetes care through a multicomponent quality improvement model in a practice-based research network Am J Med Qual 2007;22:34-41.

Competing interests:   None declared

We would like to respond to several interesting points made by Dr. Phillips. All of us have observed the phenomenon of a failure to intensify therapy when glucose or blood pressure is poorly controlled. As physician -scientists, we all recognize the importance of using terminology that accurately describes the phenomenon we are investigating: “plantar fasciitis” rather than “foot pain.” The choice of terms is important because it reflects our underlying assumptions about the nature of the phenomenon and informs efforts to intervene on behalf of the patient. The use of the term “clinical inertia,” as coined by Dr. Phillips and colleagues, leads to the assumptions reflected in their as yet unproven hypotheses (1) about the cause of the phenomenon: 1) use of “soft” reasons to delay a change in treatment; 2) over-estimation of the quality of care provided; 3) lack of knowledge about appropriate goals for therapy. If these hypotheses are true, then one should develop interventions using the approach described by Dr. Phillips in his letter of response to our study: “…it is logical to deconstruct clinical encounters to see if problems can be identified and, ideally, fixed.” The logic behind this approach is that delivery of patient care in primary care settings is like an automobile engine, it is best understood by reducing it to its component parts, identify the defective part and fix it. By definitiion, use of the term clinical inertia identifies the defective part as physician behavior and it is this part that needs to be fixed.

Unfortunately, a number of studies have demonstrated that a focus on changing physician behavior is rarely effective in improving patient outcomes. (2-4) Moreover, little of the observed variation in diabetes care practices is attributable to individual primary care clinicians. (5) In their own research, Phillips and colleagues tested an intervention based on the clinical inertia theory, but were only able to improve 1 out of 3 intermediate clinical endpoints for patients with diabetes. (6) In contrast, a study conducted across 66 primary care practices across the United States demonstrated significant improvements in all intermediate clinical endpoints when, instead of focusing on the physician, they allowed clinic teams to tailor a combination of organizational approaches that work best in their setting to overcome competing demands. (7)

Without a doubt, as physicians our primary concern should be to apply evidence-based findings concerning prevention of diabetes complications to our patients who entrust their care to us. The question is how best to “close the gap” and apply the evidence in a sustainable fashion. One might argue that by addressing patient concerns, primary care clinicians are working to gain patient trust. Trust in the physician has been shown to be a significant predictor of patient adherence to medication regimens. (8) Thus, by allowing patient concerns to be a high priority on the visit agenda, primary care physicians may be insuring that any intensification of treatment for poorly controlled diabetes is adhered to. It is likely that this is the phenomenon we are observing in our cross-sectional data since others have found that two-thirds of patients seen in primary care with poorly controlled A1c do have intensification of their treatment over 6 months. (9)

So while Dr. Phillips insists that "clinical inertia" on the part of the physician is the central issue, evidence to date supports the conclusion that approaches to overcome competing demands that are multi- faceted and occur at the level of the practice are more effective strategies to improve outcomes for our patients with diabetes than merely focusing on changing physician behavior.

1. O’Conner PJ. Commentary-Improving diabetes care by combating clinical inertia. Health Serv Res 2005;1854-1861.

2. Oxman AD, Thomson MA, Davis DA, Haynes RB. No magic bullets: a systematic review of 102 trials of interventions to improve professional practice. Canadian Med Assoc J. 1995;153:1423-1431.

3. Cabana MD. Rand CS. Powe NR. Wu AW. Wilson MH. Abboud PA. Rubin HR. Why don't physicians follow clinical practice guidelines? A framework for improvement. JAMA. 282(15):1458-65, 1999 Oct 20.

4. Bero LA, Grilli R, Grimshaw JM, Harvey E, Oxman AD, Thomson MA. Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. BMJ 1998;317:465-468.

5. Krein SL, Hoer TP, Kerr EA, Hayward RA. Whom should we profile? Examining diabetes care practice variation among primary care providers, provider groups, and health care facilties. Health Serv Res 2002;37:1159- 1180.

6. Ziemer DC. Doyle JP. Barnes CS. Branch WT Jr. Cook CB. El-Kebbi IM. Gallina DL. Kolm P. Rhee MK. Phillips LS. An intervention to overcome clinical inertia and improve diabetes mellitus control in a primary care setting: Improving Primary Care of African Americans with Diabetes (IPCAAD). Archives of Internal Medicine. 166(5):507-13, 2006 Mar 13.

7. Ornstein S. Nietert PJ. Jenkins RG. Wessell AM. Nemeth LS. Feifer C. Corley ST. Improving diabetes care through a multicomponent quality improvement model in a practice-based research network Am J Med Qual 2007;22:34-41.

8. Piette JD. Heisler M. Krein S. Kerr EA. The role of patient- physician trust in moderating medication nonadherence due to cost pressures. Arch Intern Med 2005;165:1749-55.

9. Rodondi N, Peng T, Karter AJ, Bauer DC, Vittinghoff E, et al. Therapy modifications in response to poorly controlled hypertension, dsylipidemia, and diabetes mellitus. Ann Intern Med 2006;144:475-484.

Competing interests:   None declared

This paper by Parchman et al is an important contribution to the literature on the management of diabetes mellitus in the primary care setting. The authors present convincing evidence that there are a number of factors that may inhibit primary care clinicians from responding immediately to a higher-than-target HbA1c level by changing the patient's diabetes medications. Competeing demands such as other patient concerns are likely mitigating factors. We have previously described the challenge of achieving combined target levels of HbA1c, blood pressure and LDL cholesterol in diabetices (only 7% of diabetics in our large cross- sectional study of diabetes management in primary care achieved target levels for all three measures). (1) To imply that primary care physicians don't change medications in diabetic patients whose HbA1c levels are higher than target because of clinical inertia ignores the complexity of primary care practice and the reality of multiple competing demands, as well as the importance of considering the patient's own goals and preferences for care.

References: 1. Stephen J. Spann, Paul A. Nutting, James M. Galliher, Kevin A. Peterson, Valory N. Pavlik, L. Miriam Dickinson, and Robert J. Volk. Management of Type 2 Diabetes in the Primary Care Setting: A Practice- Based Research Network Study. Ann. Fam. Med, Jan 2006; 4: 23 - 31.

Competing interests:   None declared

We know that family physicians address an average of 4.6 problems per visit when meeting with patients with DM. (1) In this issue, Parchman et.al. elegantly illustrate that, at times, some of these competing demands are associated with a lack of immediate intensification of diabetes treatment.

My own qualitative interviews with patients with diabetes as well as other conditions (unpublished) suggest that patients with multiple medical conditions want (among other things) relevant health information, clear treatment plans, and to have their own competing demands and priorities solicited and acknowledged. They also want communication with their provider outside of the office visit. Patients who do not feel that their primary concerns are addressed (if not necessarily solved) may be less inclined to adhere to other treatment recommendations.

As is often the case, exploration of complex situations reveals parallel processes: in this case competing demands for both patients and their providers. The good news is that the encounter is part of a larger collaboration in which patient and provider share a common goal of benefiting the patient’s health. It will take increased study of the process of care (as compared to the content of care) in order to design solutions. For starters, we might talk to both patients and providers and focus on what they have in common: a need to prioritize concerns for a common goal. One part of this is to emphasize the larger relationship and step out of the office encounter as the primary vehicle for medical recommendations.

(1) Beasley JW, Hankey TH, Erickson R, Stange KC, Mundt M, Elliott M, Wiesen P, Bobula J. How many problems do family physicians manage at each encounter? A WReN study. Ann Fam Med. 2004 Sep-Oct;2(5):405-10.

Competing interests:   None declared

Bravo for this well designed and clinically sound study. I completely agree that competing demands is a principle for constructing models of the primary care encounter that is more congruent with reality.

The practice of patient-centered care implies that the patient has the room to discuss his or her complaints with the physician.(1)Together, they have to find common ground and decide upon which problems should be prioritized for discussion and care. This may imply delaying discussion of an intervention that is required but can reasonably be postponed without any lasting harm done to the patient. This is exactly the case with A1c levels. Although it is "proven" important to reach target levels in the long run, it may be worth waiting till the next appointment if other concerns have been judged a priority. We should find it reassuring that the issue of priority is not solely dependant upon the doctor’s point of view. In certain situations clinical judgement that respects the patient’s needs may supersede the automatic application of practice guidelines. This is exactly what we teach our students when we teach patient-centered medicine.

On the other hand, not responding to the patient’s complaints may consititute clinical inertia.

1. Stewart MA, Brown JB, Weston WW, McWhinney IR, McWilliam CL, Freeman TR. Patient-centered medicine - Transforming the clinical method. Abingdon: Radcliffe Medical Press; 2003.

Competing interests:   None declared

The evidence for benefit of good glycemic control in type 2 diabetes is strong -- just not based on direct, randomized, controlled trials. See Stratton's epidemiologic analysis in the UKPDS, the Kumamoto study, and Nichols' analysis of the natural history of attaining A1c <6%. All physicians should consider such evidence, and decide what to do for their patients.

What I think is more of a problem is education of physicians as to how to monitor their own performance, and what the results show. Physicians have also not been taught how to manage asymptomatic problems efficiently. Toward this objective, we have developed a paradigm that may help: Barnes,CS, Ziemer,DC, Miller,CD, Doyle,JP, Watkins,C, Jr., Cook,CB, Gallina,DL, El-Kebbi,I, Branch,WT, Jr., Phillips,LS: Little time for diabetes management in the primary care setting. Diabetes Educ 30:126-135, 2004.

Competing interests:   None declared

Responding to patient complaints is being patient-centered. It would be most inappropriate if I pontificated, "your injured ankle is important, but we really have to work on your diabetes today."

In an ideal world, we would have time enough in any patient encounter for acute issues of the patient's concern and disease management and prevention as best judged and desired by the physician who cares about the patient's future. That's what Dr. Phillips advocates.

But in the real world, busy primary care doctors do not have that sufficient amount of time. In a parallel example buttressed by data, I recently enjoyed reading discussions and research studies arguing back and forth at the Annals of Internal Medicine documenting how ideally we should do disease prevention and screening at every visit, but in the real world it just doesn't happen. This is one sound reason for annual "complete physicals", because it carves out time specifically for preventive health care which otherwise gets short shrift when competing against pt. acute complaints and even against chronic disease management.

Anyway, all good physicians want the best for their patients' longterm future. I'm sure Dr. Phillips and I are in complete agreement on that. And we all want to give the patients both what they request acutely and what they need over the longterm. The trick is to balance what to work on so that all needs are met and all parties are happy. It is a shame that our "system" or current circumstances do not allow or encourage a doctor to easily do both the acute issues and the chronic issues at every visit as Dr. Phillips advocates. Another reason to change the system!

-- Josh

Competing interests:   None declared

Dear Editors,

I cannot help but respond to the lead article. "Competing Demands or Clinical Inertia" is written based on the assumption that improved glycemic control equals better treatment of our diabetics. I reject this notion. The references cited (citations 1 through 3) do not in fact support this contention. The 1st reference, UKPDS, showed that retinophotocoagulation treatments were improved with better glycemic control, and little else was improved. The second citation is a meta-analysis of observational studies. Observational studies have a built in tendency to overestimate benefit, and this does not form the sound basis for aggressive glycemic control. The third citation is just the expert opinion of an endocrinology association which is in the business of treating blood sugar as if it is itself an important patient outcome. Patient oriented evidence based practice which family physicians specialize in would suggest we could do otherwise.

I like to think the inertia and distraction is probably a good thing. Since the evidence supports the benefit of blood pressure control in diabetes, I don't mind being distracted by hypertension management. Ditto goes for lipid management and exercise and other health habits and lifestyle efforts. I also like to think that since pouring attention and resources into glycemic control may not improve patient outcomes, turning our attention to nearly anything else in our patient's lives is likely to be an improvement that they care about and benefit from, not a "distraction"!

EBM contrarian gadfly, -- Joshua Steinberg MD

Competing interests:   None declared

Parchman et al, nicely demonstrate the effect that competing demnads place on family physicians (1). However, their assertion as to the minimal effect clinical inertia has in the practice of medicine is overstated.

Although their study provides an alternate hypothesis for physician behavior when compared to two of the three componenets of clinical inertia (physicians overestimating the amount of care they provide, and “soft” reasons to avoid titration of therapy) they fail to account for the third, and most important component of the clinical inertia theory: lack of education, training, and practice organization which hinder the ability of physicians to achieve the desired treatment goals (2).

Blood pressure control in the United States is poor (3). This is in part due to the fact that forty per cent of primary care physicians, are not familiar with, or have not heard of the Joint National Committee on Prevention, Detection, Evaluation, and Management of High Blood Pressure (JNC) guidelines (4). There was a direct correlation between familiarity with JNC guidelines, and use of those guidelines (including achievement of blood pressure goal). As such, it would appear that not following the guidelines has less to do with disagreements in treatment options and more to do with education about hypertension. As suggested by the concept of clinical inertia this would appear to be due to a “lack of education, and training”.

In an unpublished study administered to a Department of Family Medicine, twenty-eight Family Physicians felt they did a good job treating hypertension, and twenty-seven responded that they were familiar with JNC guidelines (5). However, when asked what level of blood pressure to target for patients with essential hypertension or related co-morbidities (heart disease, diabetes, or kidney disease) half of the respondents were incorrect. Therefore, not only was lack of education or training likely a factor, but many of these physicians overestimated the level of care they were providing as it related to hypertension.

Competing demands are a significant patient related factor in reaching a desired treatment goal. However, physician factors, as suggested by clinical inertia, also have an impact, and should not be discounted.

1. Parchman ML, Pugh JA, Romero RL, Bowers KW. Competing demands or clinical inertia: the case of elevated glycosylated hemoglobin. Annals of Family Medicine. 2007;5:196-201. 2. Phillips LS, Branch WT, Book CB, Doyle JP, El-Kebbi IM, Gallina DL, et al, Clinical Inertia, Annals of Internal Medicine, 2001;135:825-834. 3. Chobanian A, Bakris G, Black H, Cushman W, Green L, Izzo J, Jones DW, Marerson BJ, Oparil S, Wright JT, Roccella EJ. The Seventh Report of The Joint National Committee on Prevention, Detection, Evaluation, and Management of High Blood Pressure (JNC-7). Hypertension. 2003;42:1206- 1252. 4. Hyman DJ, Pavlik VN. Self-reported Hypertension treatment practices among primary care physicians: Blood pressure thresholds, drug choices, and the role of guidelines and evidence based medicine. Archives of Internal Medicine. 2000;160:2281-2286. 5. Wexler RK. Unpublished data. The Ohio State University, Columbus, Ohio.

Competing interests:   None declared