ALLHAT – So What? – James McCormack
Email to the Editors
Readers respond to “Pharmacist Scope of Practice: A Response to the 2002 ACP-ASIM Position Paper”
Times have changed since pharmacists were primarily concerned with licking, sticking, counting and pouring pharmaceuticals. A shift from a product-centered profession to a patient-centered, outcome-oriented profession is well underway and is already firmly established in institutional practice. In recent years several studies have demonstrated that pharmacists practicing in direct patient care roles as part of multidisciplinary care teams not only reduce overall health care costs but morbidity and mortality as well (1-3). It is perhaps in response to this “pharmacy movement” that in January 2002 the American College of Physicians-American Society of Internal Medicine (ACP-ASIM) published a position paper describing their perceptions of the pharmacist’s scope of practice.
The physician-authored paper accurately presents available data supporting the roles of pharmacists in various patient care settings. It acknowledges the aspirations of pharmacists as described in the American Pharmaceutical Association’s definition of pharmaceutical care and upholds the necessity that pharmacists be remunerated for their cognitive services. Wisely, the paper supports that pharmacists take an active role in educating physicians about drug use. Pharmacists as immunization providers is also advocated. The authors even provide a detailed account of the multitude of progressive roles undertaken by pharmacists with the support of state legislatures.
The paper outlines five positions on the scope of pharmacist practice starting with support for research into the effects of pharmacy automation and the move to the all-Pharm.D. degree on pharmacy practice, and concludes with a position reiterating support for therapeutic interchange programs which is similar to the position of the American College of Clinical Pharmacy (ACCP). What lies between these benign stands are more contentious positions concerning collaborative care and the value of pharmacists intervention.
The ACP-ASIM position on collaborative care supports “physician-directed, pharmacist-physician collaborative practice agreements limited to pharmacists involvement in patient education and hospital rounds under the following conditions:
(i) expanded roles of pharmacists should not be solely based on cost-savings;
(ii) the responsible physician and pharmacist should be compensated for their time spent on collaborative services;
(iii) the physician should solely determine if a relationship will be formed with the pharmacist;
(iv) the physician should solely and individually refer a patient to a pharmacist; and
(v) only the physician can and will diagnose the patient’s condition prior to any referral.”
The idea asserted by the ACP-ASIM that a patient shall not form a relationship with a pharmacist unless authorized to do so by a physician is reduced to absurdity when one considers the following:
The sheer number and accessibility of pharmacists far exceeds that of physicians.
Patients have always sought the advice of pharmacists on a broad range of health issues. In this context, pharmacists have been recommending, initiating, monitoring and modifying drug therapy for decades. They have been independently educating patients about drug therapy all along. Patients have been making informed decisions about prescription and non-prescription drugs based on this information. The pharmacy profession’s efforts over the last decade have been aimed at constantly improving pharmacists’ skills in providing this type of care. Patients decide whom they wish to consult for health advice and the value of that service to them.
The very act of writing a prescription invokes the formation of a professional relationship between pharmacist and patient and all of the legislated, mandated, ethical and professional responsibilities inherent therein.
Do physicians long for the days past when they could prescribe an NSAID to a patient for joint pain without informing them of the risk of death due to upper gastrointestinal bleeding, expect that the pharmacist won’t “undermine” their prescription by doing the same, and assume the patient will faithfully take the drug as directed? The pharmacist’s foremost duty is to the patient, not the physician.
In most institutions, the pharmacist’s duty to independently assess and strive to ensure appropriate drug therapy on a patient-specific basis is a Medical Advisory Committee decreed policy.
Such policies partially arose from the need to reduce the risk of avoidable adverse drug events resulting from mis-prescribing by physicians.
Experience dictates that physicians lacking appropriate drug therapy skills are frequently unaware of this deficit.
That these physicians should be relied upon to seek advice and refer patients for pharmacotherapeutic consultation is not in the best interests of patients.
Though we do not assert that all pharmacists are trained as diagnosticians, the reality that pharmacists interpret the complaints of scores of patients every day, form correct diagnoses and initiate and monitor appropriate drug therapy must be acknowledged.
Patients would not tolerate any attempt to limit the pharmacist’s ability to provide this advice, nor could any nation’s health system bear the cost of having these complaints referred to physicians, as the ASP-ASIM apparently desire.
Pharmacists have already established themselves as direct patient care providers in many settings where justification based on cost savings alone is not necessary (1-3). Despite this, the necessity for provision of cost-effective and safe care in an era of skyrocketing cost and complexity of drug therapy has allowed motivated pharmacists to make great strides in expanding their scope of practice. No data is presented by the authors of the ACP-ASIM paper which assert that the wide variety of initiatives undertaken by pharmacists has resulted in harm or have in any way been detrimental to patients.
Much of the ACP-ASIM’s concern centers on the perceived differential between “hands-on” patient care training in PharmD programs and medical schools. The debate about the sufficiency of the pharmacotherapy training imparted in most medical schools is one for another day. The ACP-ASIM’s statement that “physicians are qualified to…treat patients, while pharmacists expertise lies with pharmaceuticals” is cryptic, and their position that “decisions about the most appropriate drug therapy for a patient’s condition are often subtle and require a level of experience and training that is not provided in obtaining a PharmD degree” is objectionable. Neither group has sufficient evidence to properly support or refute ACP/ASIM’s antiquated opinions. However, thousands of highly trained, experienced, patient-oriented pharmacists adept at interpreting the subtleties of patients’ drug therapy needs are disproving this statement every day.
ACP-ASIM also opposes independent pharmacist prescriptive privileges and initiation of drug therapy. This is not surprising given that this is one of the most sensitive areas of the “turf battle” between the professions, including that brewing between nurses and physicians. As already stated, pharmacists already have prescriptive authority for patients seeking therapy in a community pharmacy in which the pharmacist can prescribe from a wide array of over-the-counter medications. The ACP-ASIM paper nicely catalogues the many other areas where prescriptive authority is being exercised or evaluated, such as emergency contraception, anticoagulation, diabetes management, and numerous other “disease management” programs. The experience being accrued in these and other jurisdictions will ultimately provide the data required to define what initiatives work best, what patients prefer and which result in the best outcomes. The ACP-ASIM’s lack of support for these initiatives is out of step with the demand by patients for such services and the reality that patients are already benefiting from them.
ACP-ASIM’s statements such as “pharmacists can educate physicians on drug interactions and cost savings and educate patients on drug safety, while physicians provide safe, effective care to patients”, fuels exactly the “turf battle” the physicians claim in their paper that they want to avoid. Such marginalizing statements belie the underlying aim of the ACP-ASIM’s position, which seems to be to protect the domain of the physician rather than to support a multidisciplinary and balanced approach to patient care. This ridiculous attempt to reduce a profession to discrete tasks such as those named while ascribing to physicians the exclusive right to “provide safe, effective care to patients” shows ACP-ASIM’s lack of insight about the rapidly evolving multidisciplinary health system.
What the ACP-ASIM position paper stridently aims to assert is that physicians are already providing high-quality safe, effective drug therapy to their patients and that pharmacists, while potentially helpful (though annoyingly time-consuming) assistants in this regard, ought not to interfere in that provision. The alarming rates of prescription drug-related morbidity and death in North America are a stark reminder that this is not the case. Informed consumers demand that this situation be improved. Neither physicians nor pharmacists are providing the care necessary to avoid this needless suffering. Turning back the clock on pharmacist roles and limiting access by patients to pharmacists, even if physicians had the authority to do so, would not serve the best interests of our patients.
The ACP-ASIM could be forgiven for their extreme traditionalist positions if they were aimed at the lowest level of pharmacist practice. Indeed, not all pharmacists are sufficiently trained, motivated, experienced, ethical, or skilled enough to provide high-level pharmaceutical care to patients. Neither, for that matter, are all physicians. The profession of pharmacy, like medicine, has evolved to include a wide array of professionals with diverse and varied levels of training, skill and experience. Fortunately, enhanced responsibilities for pharmacists have been limited mainly to those who obtain specialized training and achieve certification in various forms. The position paper; however, purports to describe the “scope of practice” of pharmacists. In this attempt, they have accurately described that scope as it existed 20 years ago. The ACP-ASIM appears to desire a return to a time when pharmacists dutifully told patients in response to their drug therapy questions, “I shouldn’t answer that, you’ll have to discuss that with your doctor.”
It is doubtful that the conservative and demeaning ACP-ASIM position statement will have much impact on the pharmacy profession who, supported by numerous state and provincial legislatures, have embraced the role of drug therapy providers and competently provide for the needs of their patients. They do this within a complex array of collaborations with physicians ranging from total independence to direct supervision. Ultimately patients, not physicians, will decide which services they value and how much.
Issues in Pharmacotherapy Practice
contemporary issues by clinical practitioners
Is There an Association between Antipsychotic Medications and Venous Thromboembolism?
Background: Antipsychotic agents are widely used medications. An association between antipsychotic agents and venous thromboembolism has been proposed in the literature.
Methods: A literature search was conducted using MEDLINE, EMBASE, PubMed, and national and international adverse drug reaction reporting programs.
Results: Several case reports, case series, observational and retrospective studies have been published. Investigators of a large case control study involving 30,000 current antipsychotic users found an increased risk of venous thromboembolism OR 7.1 (95% confidence interval 2.3-21.97). Agents most commonly implicated include low potency conventional antipsychotics and clozapine. Proposed mechanisms include drug-induced effects of platelet aggregation, anticardiolipin antibodies or weight gain, sedation and diabetes.
Conclusion: A possible association between antipsychotics and venous thromboembolism, although rare, does exist. Clinicians should be aware of this potential adverse drug reaction.
Antipsychotic agents are widely used for many indications in medicine and psychiatry with therapeutic indications ranging from treatment of nausea to refractory schizophrenia. Patients receiving antipsychotics are usually monitored closely for many known adverse effects including movement disorders, seizures and blood dyscrasias. Case reports, retrospective analyses of cause of death in psychiatric patients and national adverse drug reaction reporting programs suggest that an association between antipsychotic medications and venous thromboembolism (VTE) may exist. The purpose of this paper is to review and summarize the published literature regarding this apparently rare, but potentially dangerous drug reaction.
Searches through MEDLINE (1966 – April 2002) and EMBASE (1980 – April 2002), were conducted using the search terms “antipsychotic, antipsychotics, neuroleptic, neuroleptics, antipsychotic agents” and “thrombosis, venous thrombosis, thromboembolism” and “pulmonary embolism.” Observational studies (cohort and case control) and randomized controlled trials reporting VTE outcome data with exposure to antipsychotics were sought. References from articles identified through the MEDLINE search were manually reviewed to identify for further reports.
Several reports suggesting a possible association between antipsychotic medications and VTE have been published. Cases involved different types, doses and durations of exposure to antipsychotics. Where possible, cases were evaluated using a standard method for estimating the probability of adverse drug reactions. Reported cases were considered to reflect a possible association between antipsychotic drug ingestion and the development of VTE. While many cases appeared to involve objective confirmation of the diagnosis, none involved a drug rechallenge procedure.
One case series identified 49 cases of VTE (3.1%) in 1,590 patients treated with antipsychotics from 1958-61 at a single centre, as compared to 20 cases (0.3%) of VTE in patients on very low doses or no antipsychotics. This report raised the possibility of a higher than expected incidence of VTE in this population. Like other reported cases, many patients were elderly with concomitant medical conditions or medications. Although a comparator group was indicated, it was not stated if the controls were matched for any characteristics. “Low dose” antipsychotic exposure was not defined or described, thus limiting the utility of the control group and the interpretation of this report.
Retrospective cohort analyses of autopsies have also suggested an increased incidence of VTE in patients with psychosis, and patients treated with antipsychotics. In a study that compared causes of death in 67,072 current and former users of clozapine, it was noted that current clozapine use was associated with an increased risk of pulmonary embolism (RR 5.2 for current compared with past clozapine use). At a hospital in the Netherlands a review of 14,000 autopsies showed that 10 of 27 patients with idiopathic pulmonary embolism were psychiatric patients and five were treated with antipsychotics.
Several national adverse drug reaction (ADR) reporting programs have described reports of VTE possibly associated with antipsychotics. Twelve cases of VTE (5 fatal) associated with clozapine that were reported to the Swedish Adverse Reactions Advisory Committee from 1989-2000 are described in Table 1. Based on an estimate derived from Swedish pharmacy statistics, the authors proposed an incidence of VTE associated with clozapine of 0.017 – 0.05%. A German ADR surveillance program that assessed severe ADRs in all inpatients of psychiatric hospitals in Switzerland and Germany (103,000 patients from 1993-99), reported that the incidence of VTE was 0.038%, 0.029% and 0.026% in patients treated with clozapine, other antipsychotics and no antipsychotics, respectively, although there was no statistically significant difference between the relative incidences of VTE in these groups. From 1990-99, the Food and Drug Administration (FDA) in the United States received 99 reports of VTE (39 with objectively confirmed diagnosis) possibly associated with clozapine. From 1964 to 2002, 27 cases of VTE (13 fatal) associated with clozapine and one case of VTE associated with risperidone were reported to the Canadian Adverse Drug Reactions Database (CADRAMP). From 1988 to 1998 there were 137 reports of VTE associated with antipsychotics made to the World Health Organization (WHO) database of ADRs (clozapine 111, chlorpromazine 4, haloperidol 18, thioridazine 4). It is possible that due to the spontaneous nature of adverse drug reaction reporting programs that these cases represent a duplication of previously published reports.
A recent study assessed the incidence of VTE in a group of nearly 30,000 patients prescribed an antipsychotic from 1991-8. The patients were part of the UK based General Practice Research Database, a computerized database of physician visits. Data were collected at the point of visit including: patient demographics, characteristics, symptoms, medical diagnoses, hospital admissions and prescriptions. Patients had used at least one antipsychotic during the study period, and were followed for a minimum of six months. Cases had objectively diagnosed first-time VTE, were admitted to hospital and treated with anticoagulants (identified through computer recorded medical diagnosis and confirmed by discharge records). All patients were less than 60 years and free of medical conditions potentially associated with VTE including: history of VTE, trauma, pregnancy or surgery within six months, history of an acute psychotic episode within two months, or history of coagulopathy, congestive heart failure, myocardial infarction, cancer, renal failure, epilepsy, diabetes mellitus, cystic fibrosis, multiple sclerosis, alcohol and substance use disorders. For each case, four controls were matched for age, sex, general practitioner and years in the database.
During a mean follow-up of 6.8 years, 42 cases of idiopathic VTE were identified. The odds ratio (OR) for VTE (adjusted for smoking, body mass index, hypertension, estrogen and antidepressant use) for current antipsychotic use (within 60 days) was 7.1 (2.3-21.9). The OR for VTE was for recent use (61-120 days) was not significant 2.1 (0.4-11.8). The OR for VTE was greater for low potency agents (chlorpromazine, thioridazine, mesoridazine, pericyazine, methotrimeprazine, pipothiazine, sulpiride) as compared to high potency agents (haloperidol, benperidol, pimozide, trifluoperazine, fluphenazine, perphenaxine, zuclopenthixol, flupenthixol, thiothixene, lozapine); 24.1 (3.3-172.7) and 3.3 (0.8-13.2) for low and high potency agents, respectively. The OR for VTE was 12.4 (3.2-48.3) and 2.3 (0.4-14.9) for doses less than and greater than 100 mg chlorpromazine equivalents, respectively. The OR for VTE was 28.7 (4.9-169.5) and 1.0 (0.1-7.3) for use within 11 months, or greater than or equal to 12 months, respectively. There was no difference between phenothiazines, thiothixines and heterocyclic agents. There were no cases of VTE reported with use of atypical antipsychotics (quetiapine, olanzapine clozapine and risperidone), although only 3% of the patients were actually receiving these agents. Patients and controls were well matched for factors including body mass index, smoking status, history of diabetes or hypertension, estrogen and antidepressant use. Stratification for age, sex and outcome did not change the study results. The authors proposed an incidence for VTE associated with antipsychotic use of 0.14%. Although case reports and autopsy data suggest there may be an association with clozapine and VTE, (9,12-17,27-30), the results of this study do not add to the available literature as only 3% of patients were taking atypical antipsychotics and no VTE occurred in these patients.
The following case control studies add to the results of earlier observational studies in Germany. (32-34) One study identified an increased incidence of VTE (3.3%) among 338 phenothiazine users between 1954-57, as compared to a non-phenothiazine control group (< 0.5%). The scarce available information about patient diagnoses, medical conditions and medications limits the interpretation of this study. Another study has reported an increased incidence of VTE in psychiatric and neurological patients from 1958-61(following the introduction of chlorpromazine) as compared to historical controls from 1915-22 (18% vs. 6% in males and 29% vs. 10% in females). The use of historic controls, as well as minimal reported patient information, limits the interpretation of this study. An early case control study, assessed 2,344 psychiatric patients and noted an increased incidence of VTE in patients treated with antipsychotic or antidepressant medications (chlorpromazine, amitriptyline, imipramine) as compared to patients with similar psychiatric patients who did not take these medications (2.9% vs. 0.6%). Most (22/34) patients with VTE had concurrent medical conditions, limiting the interpretation of this data.
Table 1: Case reports of venous thromboembolism associated with antipsychotic medications.
* From Naranjo CA, Busto U, Sellers EM, et al. Clin Pharmacol Ther 1981;30:239-45. Naranjo score 1-4 suggest possible association between drug and adverse reaction.
The mechanism for antipsychotic-induced VTE is unclear, although several mechanisms have been proposed. One hypothesis is based upon an enhanced platelet response to 5-hydroxytriptamine in the presence of antipsychotics. An alternate hypothesis suggests that increased anticardiolipin (Ac) levels increase the risk of VTE. However, Ac antibodies have heterogeneous specificity, may be associated with psychosis, and are not clearly associated with antipsychotics or thrombosis. A third hypothesis suggests that venous stasis secondary to antipsychotic associated sedation or weight gain causes VTE. Finally, obesity and diabetes are possible adverse effects associated with atypical antipsychotics, and have been identified as important risks for venous thrombosis. These hypotheses lend biological plausibility to a possible association. Only one large well designed case control study suggests an association between VTE and antipsychotic use. Although this study suggested an incidence of VTE associated with antipsychotics is to be 0.14%, the relative risk observed was 7.1 (2.3-21.9) for current antipsychotic use as compared to no antipsychotic use. The magnitude of the OR is suggestive of an association between VTE and antipsychotics. The objective nature of the mode of VTE diagnosis, as well as the systematic collection of patient data make it unlikely that cases would be missed or that prescriptions for antipsychotics would be undetected. Selection bias was thus minimized. There may have been differences between cases and controls with respect to unmeasured coagulation abnormalities, stress , alcohol , sedation , or immobility related to psychiatric illness.
Although case reports suggest that VTE can occur after taking antipsychotics for many years , the risk for VTE was increased only during the first two months of therapy in the UK case control study. This is suggestive of a temporal association. Further support of an association is offered by the objective diagnostic criteria for acute VTE, which minimized the likelihood that VTE may have occurred before exposure to antipsychotics.
There does not appear to be a relationship between antipsychotic dose and incidence of VTE. The OR for VTE at doses of antipsychotic greater than 100 mg of chlorpromazine equivalents was not statistically significant. This argues against an association between antipsychotics and VTE. Only a small number of cases were identified, suggesting that the study may not have had enough cases to detect differences that may exist between different classes of antipsychotics. Additionally, only small numbers of patients were taking atypical antipsychotics in this study, The association between these agents and VTE thus remains uncertain.
Many case reports, adverse drug reaction monitoring system reports and retrospective autopsy data suggest that there is an increased risk of VTE with clozapine. (9,12-17,27-30) This might be explained by weight gain, sedation or diabetes associated with clozapine, or that these patients are followed very closely due to the clozapine-associated risk of agranulocytosis. There are no formal observational studies assessing the potential for VTE associated with clozapine or other atypical antipsychotics. The proposed association between clozapine and VTE has some biological plausibility and temporal correlation, but lacks consistent assessment of data, controlled studies, or obvious relation to clozapine dose.
It is important for clinicians to be aware of the possibility of VTEs associated with antipsychotic use. Current manufacturer labeling for clozapine includes VTE as a possible adverse effect. Management of patients who have experienced VTE associated with antipsychotics should be evaluated on an individual basis. One author has suggested discontinuing the antipsychotic should a patient develop VTE. Most authors offered no management suggestions. For any case of VTE potentially associated with antipsychotics, it would seem reasonable to reassess the risks and benefits of continued therapy with antipsychotic medication for that patient in light of an increased thrombotic risk. Consideration should be made for other causes of a hypercoagulable state, both congenital and acquired. Although there is no information to suggest an effect of estrogen use, smoking, obesity or physical activity on VTE risk reduction in this patient population, it would be prudent to suggest modification of existing risks for VTE if therapy with antipsychotic medication is to be continued. Any cases of VTE possibly associated with antipsychotics should be reported to an ADR reporting program.
Although rare, it seems that a possible association between antipsychotics and VTE exists. While suggestive of an association, the available published data suffers from methodological limitations. Case reports, retrospective cohorts and spontaneous adverse drug reaction reporting databases lack control groups, and therefore the ability to draw firm conclusions about an association between drug and adverse effect. This association for conventional antipsychotics is supported by one large case control study, and for clozapine by case reports and retrospective data. Further study of atypical antipsychotics is warranted.
Evidence-Based Snapshots with emphasis on rapid review of results
- Major Outcomes in High-risk Hypertensive Patients Randomized to ACE Inhibitor or Calcium Channel Blocker vs. Diuretic
- Major Outcomes in Moderately Hypercholesterolemic, Hypertensive Patients Randomized to Pravastatin verus Usual Care
- Celecoxib versus Diclofenac and Omeprazole in reducing the risk of Recurrent Ulcer Bleeding in Patients with Arthritis
Evidence-Based Reviews of the Pharmacotherapeutic Literature with emphasis on interpretation of the findings, designed to help practitioners put the findings into proper perspective
- Can Vitamins Improve Clinical Outcomes after Percutaneous Coronary Intervention?
- Heparin Alone or Combined with Alteplase for Patients with Submassive Pulmonary Embolism: Are Two Agents Better than One?
- The Atrial Fibrillation Follow-up Investigation of Rhythm Management: Does this Study AFFIRM the Need for Rate or Rhythm Control?
- Treatment of the Common Cold with Unrefined Echinacea A: Better than Chicken Soup?
Technology in Practice helping pharmacotherapy practitioners use technology to enhance their practice
- Maximizing Your Continuing Education Yield: Incorporating Streaming Media Into Your Oral Presentations