The full content of Annals is available to subscribers

Subscribe/Learn More  >
Articles |

Cost-Effectiveness of Using Pharmacogenetic Information in Warfarin Dosing for Patients With Nonvalvular Atrial Fibrillation

Mark H. Eckman, MD, MS; Jonathan Rosand, MD, MSc; Steven M. Greenberg, MD, PhD; and Brian F. Gage, MD, MSc
[+] Article, Author, and Disclosure Information

From the University of Cincinnati, Cincinnati, Ohio; Massachusetts General Hospital, Boston, Massachusetts; and Washington University in St. Louis, St. Louis, Missouri.

Grant Support: By National Institute of Diabetes and Digestive and Kidney Diseases grant K23 DK075599, National Heart, Lung, and Blood Institute grant K30 HL078581-01, and the Foundation for Informed Medical Decision Making (Dr. Eckman); National Institutes of Neurological Disorders and Stroke grants K23 NS42695-01 and R01 NS04217 and the Deane Institute for Integrative Study of Atrial Fibrillation and Stroke (Dr. Rosand); National Institutes of Neurological Disorders and Stroke grant R01 NS04217 (Dr. Greenberg); and the Established Investigator Award from the American Heart Association (Dr. Gage).

Potential Financial Conflicts of Interest:Consultancies: B.F. Gage (Bristol-Myers Squibb), Grants received: B.F. Gage (Osmetech).

Reproducible Research Statement:Study protocol and data set: Not available. Statistical code: The decision model is available from Dr. Eckman (e-mail, mark.eckman@uc.edu).

Requests for Single Reprints: Mark H. Eckman, MD, MS, University of Cincinnati Medical Center, PO Box 670535, Cincinnati, OH 45267-0535; e-mail, mark.eckman@uc.edu.

Current Author Addresses: Dr. Eckman: University of Cincinnati Medical Center, PO Box 670535, Cincinnati, OH 45267-0535.

Dr. Rosand: Center for Human Genetic Research, Richard B. Simches Research Center, Massachusetts General Hospital, CPZN-6810, 185 Cambridge Street, Boston, MA 02114.

Dr. Greenberg: Massachusetts General Hospital, 175 Cambridge Street, Suite 300, Boston, MA 02114

Dr. Gage: Washington University, 660 South Euclid Avenue, Suite CB8005, St. Louis, MO 63110.

Author Contributions: Conception and design: M.H. Eckman,

Analysis and interpretation of the data: M.H. Eckman, B.F. Gage.

Drafting of the article: M.H. Eckman.

Critical revision of the article for important intellectual content: M.H. Eckman, J. Rosand, S.M. Greenberg, B.F. Gage.

Final approval of the article: M.H. Eckman, J. Rosand, S.M. Greenberg, B.F. Gage.

Statistical expertise: M.H. Eckman, B.F. Gage.

Collection and assembly of data: M.H. Eckman.

Ann Intern Med. 2009;150(2):73-83. doi:10.7326/0003-4819-150-2-200901200-00005
Text Size: A A A

Background: Variants in genes involved in warfarin metabolism and sensitivity affect individual warfarin requirements and the risk for bleeding. Testing for these variant alleles might allow more personalized dosing of warfarin during the induction phase. In 2007, the U.S. Food and Drug Administration changed the labeling for warfarin (Coumadin, Bristol-Myers Squibb, Princeton, New Jersey), suggesting that clinicians consider genetic testing before initiating therapy.

Objective: To examine the cost-effectiveness of genotype-guided dosing versus standard induction of warfarin therapy for patients with nonvalvular atrial fibrillation.

Design: Markov state transition decision model.

Data Sources: MEDLINE searches and bibliographies from relevant articles of literature published in English.

Target Population: Outpatients or inpatients requiring initiation of warfarin therapy. The base case was a man age 69 years with newly diagnosed nonvalvular atrial fibrillation and no contraindications to warfarin therapy.

Time Horizon: Lifetime.

Perspective: Societal.

Intervention: Genotype-guided dosing consisting of genotyping for CYP2C9*2, CYP2C9*3, and/or VKORC1 versus standard warfarin induction.

Outcome Measures: Effectiveness was measured in quality-adjusted life-years (QALYs), and costs were in 2007 U.S. dollars.

Results: In the base case, genotype-guided dosing resulted in better outcomes, but at a relatively high cost. Overall, the marginal cost-effectiveness of testing exceeded $170 000 per QALY. On the basis of current data and cost of testing (about $400), there is only a 10% chance that genotype-guided dosing is likely to be cost-effective (that is, <$50 000 per QALY). Sensitivity analyses revealed that for genetic testing to cost less than $50 000 per QALY, it would have to be restricted to patients at high risk for hemorrhage or meet the following optimistic criteria: prevent greater than 32% of major bleeding events, be available within 24 hours, and cost less than $200.

Limitation: Few published studies describe the effect of genotype-guided dosing on major bleeding events, and although these studies show a trend toward decreased bleeding, the results are not statistically significant.

Conclusion: Warfarin-related genotyping is unlikely to be cost-effective for typical patients with nonvalvular atrial fibrillation, but may be cost-effective in patients at high risk for hemorrhage who are starting warfarin therapy.

Funding: The National Institute of Diabetes and Digestive and Kidney Diseases; National Heart, Lung, and Blood Institute; Foundation for Informed Medical Decision Making; National Institutes of Neurological Disorders and Stroke; Deane Institute for Integrative Study of Atrial Fibrillation and Stroke; and American Heart Association.


Grahic Jump Location
Appendix Figure 1.
Decision model.

CNS = central nervous system; ICH = intracerebral hemorrhage.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 2.
Cost-effectiveness acceptability curve.

QALY = quality-adjusted life-year.

Grahic Jump Location
Grahic Jump Location
Figure 1.
Efficacy of pharmacogenetic dosing strategy.

The efficacy of genotype-guided dosing in preventing major bleeding during the induction phase of warfarin therapy is shown on the x-axis and ranges from 0 to 1. The upper curve represents the base-case situation in which genotyping must be done as a “send-out” with a 3-day turnaround and a cost of $400. The marginal cost-effectiveness ratio is less than $50 000 per QALY only if the efficacy of genotype-guided dosing exceeds 90%. The lower curve represents a scenario in which in-hospital genotyping can be done at a lower cost of $200 and without any delay in initiating warfarin. The marginal cost-effectiveness ratio of genotype-guided dosing is less than $50 000 per QALY if efficacy exceeds 32%. QALY = quality-adjusted life-year.

Grahic Jump Location
Grahic Jump Location
Figure 2.
Tornado diagram of 1-way sensitivity analyses.

The marginal cost-effectiveness ratio in dollars per QALY is shown on the x-axis and ranges from $0 to $250 000 per QALY. For each value examined, the upper and lower limits of the sensitivity analysis (labels appear at either end of each band) are based on either the 95% CIs or a clinically reasonable range. The analyses with arrows at the right end of the band indicate that the results have been truncated because we only plotted marginal cost-effectiveness ratios up to $250 000 per QALY. Values at the top of the figure, such as efficacy of gene-based dosing, show a larger effect on cost-effectiveness across their CIs; thus, resolving uncertainty in these measurements would improve predictions of cost-effectiveness. Uncertainty in the value of measurements at the base of the figure has a smaller effect on the results of the analysis. INR = international normalized ratio; QALY = quality-adjusted life-year; RR = relative risk.

* Differences in overall bleeding risk that are attributable to patient-to-patient variability.

Grahic Jump Location
Grahic Jump Location
Figure 3.
Three-way sensitivity analysis: duration of benefit from pharmacogenetic-based dosing and patient-specific relative risk for major bleeding events.

The left y-axis describes the risk for major bleeding events relative to the base-case value. This is the patient-specific bleeding risk associated with clinical factors, such as those captured by HEMMOR2HAGES score. The 2 secondary y-axes show the annual rate of major bleeding events (with warfarin) and the corresponding HEMMOR2HAGES score ranging from 0 to 2. The 3 lines represent $50 000-per-QALY willingness-to-pay thresholds for different efficacies of pharmacogenetic-based dosing: the base-case value of 32% (solid line), 50% (dotted line), and the upper confidence limit of 78% (dashed line). For each of the 3 values of efficacy, points above each threshold line have a marginal cost-effectiveness less than $50 000 per QALY, whereas points below each threshold line have a marginal cost-effectiveness greater than $50 000 per QALY. The base-case value is shown at the lower left of the figure. QALY = quality-adjusted life-year.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 3.
Whisker plot of meta-analysis results.

RR = relative risk.

Grahic Jump Location




Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).


Submit a Comment/Letter
Alternate Reasons Why Genetic Testing Will Not Be Cost-Effective
Posted on January 21, 2009
Al Lodwick
Conflict of Interest: None Declared

Many people starting warfarin have friends or relatives who are eager to tell "how it really is". This includes scaring the new warfarin patient with statements like, "You know it is really rat poison" or "Remember that Aunt Minnie bled to death on that drug". These are strong disincentives to properly starting warfarin therapy.

Many professional warfarin managers are overly cautious and as a result scare the new warfarin patient into being reluctant to start warfarin. Statements such as, "You have to be careful about green vegetables," "Shave only with an electric razor", "Be careful not to cut yourself" are powerful disincentives to taking warfarin that are of questionable value.

No amount of expensive testing will overcome what friends or relatives say. Until health-care professionals realize that the green vegetable statement is the equivalent of saying, "I have determined your proper warfarin dose and now it is your duty to change your lifestyle to prove that I am correct" we cannot expect people who need warfarin to be enthusiastic about initiating therapy. Thus genetic testing is unlikely to be cost effective in today's environment.

Conflict of Interest:

None declared

Clinical And Economic Factors That Might Change the ICER of Genotyping Before Warfarin Therapy
Posted on February 6, 2009
Benjamin P Geisler
Harvard School of Public Health
Conflict of Interest: None Declared
We were surprised that Eckman et al. found through comprehensive modeling (1) that genotyping before warfarin therapy instead of standard care without genotyping is not cost-effective when compared with other, well- accepted medical interventions.

We suggest that additional factors may improve the cost-effectiveness of warfarin genotyping. For example, if patients with a short-term warfarin indication (i.e. uncomplicated deep vein thrombosis) had their genotype recorded in an interoperable electronic health record, this test would not need to be repeated prior to subsequent warfarin use. Because genotype testing is the largest component of costs, additional benefit would accrue with limited or no incremental costs. In addition, restricting testing to those most likely to have complications of supra-therapeutic warfarin levels (i.e. the elderly at a high risk for falls or patients with a known history of gastrointestinal bleeding), may improve cost-effectiveness. The incremental costs of avoiding complications may be lower and the incremental health benefits may be higher.

Although restricting testing to those most likely to benefit may improve cost-effectiveness, diagnostic tests tend to be used more broadly than the initial research indications, potentially diluting the average health benefits. However, widespread adoption of warfarin genotype testing or technologic advancements may decrease the unit cost of testing beyond the range evaluated in the sensitivity analysis. If so, the overall cost- effectiveness may improve. These clinical and economic factors may change the incremental cost- effectiveness ratio, although the direction and magnitude is unclear without full model evaluation.


1. Eckmann MH, Rosand J, Greenberg SM, Gage BF. Cost-Effectiveness of Using Pharmacogenetic Information in Warfarin Dosing for Patients With Nonvalvular Atrial Fibrillation. Ann Intern Med. 2009;150(2):73-83.

Conflict of Interest:

None declared

Potential cost savings of fewer visits and less time for patients should be included in analysis
Posted on February 17, 2009
Daniel E. Jonas
University of North Carolina Chapel Hill, Institute for Pharmacogenomics and Individualized Therapy
Conflict of Interest: None Declared


In their recent article on the cost-effectiveness of using pharmacogenetic information in warfarin dosing, Eckman and colleagues present the most complete and practical cost-effectiveness analysis of this issue to date (1). However, their model did not consider some cost savings that could result from pharmacogenetic testing. If genotype-guided warfarin dosing results in shorter time to stable anticoagulation or the need for fewer INR measurements and fewer visits, then important cost savings would result. There is some evidence to suggest this may be the case. First, a controlled trial in Israel comparing genotype-guided dosing with clinical management reported that stable anticoagulation was reached 18.1 days earlier for those in the genotype-guided group (2). Second, an RCT in the US comparing genotype-guided dosing with standard dosing reported no significant difference in their primary outcome (% out-of-range INRs), but reported fewer dose adjustments (mean 3.0 vs. 3.6, p=0.035) and fewer INRs needed (mean 7.2 vs. 8.1, p=0.06) for those in the genotype-guided group (3). The difference suggests that there could be an average savings of one visit with genotype-guided dosing. From the societal perspective, the value of this would include the cost savings of the visit and one INR ($18.95 and $5.49 according to the Eckman article, respectively) as well as the cost of patient time.

The Panel on Cost-effectiveness in Health and Medicine recommends including patient time costs in analyses conducted from the societal perspective (4). It has been estimated that one anticoagulation clinic visit requires an average of 147 minutes for travel, waiting, and visit time (5). Using the human capital method with the hourly wage rate of $19.29 (from the National Compensation Survey, June 2006), the time patients spend for one visit is worth $47.26 (5).

Inclusion of the potential cost savings of fewer visits and less time for patients would decrease the cost per QALY to less than $171,800 as calculated in their base case. In addition, for centers where in-hospital testing is available at a cost of less than $200 with a turnaround time under 24 hours, the more favorable marginal cost-effectiveness ratio of $51,000 per QALY (from their sensitivity analyses) could decrease to a number that would make genotype-guided dosing cost-effective, perhaps even if the efficacy for reduction in hemorrhage is worse than their base-case value of 32%. It is also worth noting that the cost-effectiveness threshold of $50,000 per QALY may significantly undervalue QALYs, with some economists estimating the value of a human life year around $129,000 (6).


1. Eckman MH, Rosand J, Greenberg SM, Gage BF. Cost-effectiveness of using pharmacogenetic information in warfarin dosing for patients with nonvalvular atrial fibrillation. Ann Intern Med. 2009;150(2):73-83.

2. Caraco Y, Blotnick S, Muszkat M. CYP2C9 genotype-guided warfarin prescribing enhances the efficacy and safety of anticoagulation: a prospective randomized controlled study. Clin Pharmacol Ther. 2008;83(3):460-70.

3. Anderson JL, Horne BD, Stevens SM, et al. Randomized trial of genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation. Circulation. 2007;116(22):2563-70.

4. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-Effectiveness in Health and Medicine. New York: Oxford University Press; 1996:xxiii, 425.

5. Jonas DE, Bryant Shilliday ME, Laundon WR, Pignone M. Patient Time Requirements for Anticoagulation Care. Society of General Internal Medicine, 31st Annual Meeting. Pittsburgh, Pennsylvania; 2008.

6. Kingsbury K. The Value of a Human Life: $129,000. Time.com May 20, 2008. Available at: http://www.time.com/time/health/article/0,8599,1808049,00.html.

Conflict of Interest:

None declared

Study Misinterpreted Meta-Data - Raises Concern on the Validity of the Findings
Posted on February 18, 2009
Yusuke Tsukahara
Conflict of Interest: None Declared

To the Editor:

There are several issues in the recent article on the cost-effectiveness of using pharmacogenetic information in warfarin dosing that raise concern about the adequacy of the analysis and hence the significance of the findings (1).

Dr. Anderson and his colleagues reported the proportion of patients with SERIOUS ADVERSE CLINICAL EVENTS in each group during the 90 days clinical study (2). Those clinical events included 7 different categories (INR>3.9, use of vitamin K, major bleeding events, thromboembolic events, stroke, myocardial infarction, and death). Dr. Eckman and his colleagues used those results as if they were all MAJOR BLEEDING EVENTS. This is clearly a mistake since 5 events in the control group and 4 events in the study group (5.1% and 4.0%, respectively, P=0.71, OR=0.78 [95% CI 0.20-2.98]) are much higher rates of major bleeding events reported in other clinical studies (3, 4).

Dr. Eckman and his colleagues also argue that if 300,000 patients start warfarin therapy each year in the U.S., a genotype-guided dosing would prevent 300 major bleeding events during the first year of therapy. This conclusion is contradicting their own meta-data analysis. Drs. Landefeld and Goldman reported that the first month risk of major bleeding events is 3% (3). A projected 32% reduction in major bleeding events from the use of genotype-guided dosing would suggest that the 3% first month risk would be reduced to 2% and 1% would be prevented, which is 3,000 major bleeding events. This is a factor of 10 mistake!

Analyzing the results of just the other two clinical studies by Caraco et al.(5) and Hillman et al.(6) used by Dr. Eckman and his colleagues in the meta-analysis suggest that genotype-guided dosing efficacy in preventing major bleeding is 100%. According to this article analysis, the marginal cost-effectiveness of testing in the base case would actually be below $50,000 per QALY. However, these cohorts are too small for the underlying statistics for a meaningful pharmacoeconomics analysis.


1. Eckman MH, Rosand J, Greenberg SM, Gage BF. Cost-Effectiveness of Using Pharmacogenetic Information in Warfarin Dosing for Patients With Nonvalvular Atrial Fibrillation. Ann Intern Med. 2009;150:73-83.

2. Anderson JL, Horne BD, Stevens SM, Grove AS, et al. Randomized Trial of Genotype-Guided Versus Standard Warfarin Dosing in Patients Initiating Oral Anticoagulation. Circulation. 2007;116:2563-70.

3. Landefeld CS, Goldman L. Major bleeding in outpatients treated with warfarin: incidence and prediction by factors known at the start of outpatient therapy. Am J Med. 1989;87(2):144-52.

4. Wadelius M, Chen LY, Lindh JD, Eriksson N, et al. The largest prospective warfarin-treated cohort supports genetic forecasting. Blood. 2009;113:784-92.

5. Caraco Y, Blotnick S, Muszkat M. CYP2C9 Genotype-guided Warfarin Prescribing Enhances the Efficacy and Safety of Anticoagulation: A Prospective Randomized Controlled Study. Clin Pharmacol Ther. 2008;83:460-70.

6. Hillman MA, Wilke RA, Yale SH, Vidaillet HJ, et al. A Prospective, Randomized Pilot Trial of Model-Based Warfarin Dose Initiation using CYP2C9 Genotype and Clinical Data. Clin Med Res. 2005;3:137-45.

Conflict of Interest:

None declared

No Title
Posted on February 27, 2009
John R Schmelzer
Marshfield Clinic Research Foundation
Conflict of Interest: None Declared
No Comment
Health Care System Issues & Patient Characteristics Impact Cost Effectiveness of PG-Guided Dosing
Posted on February 27, 2009
Mark H Eckman
University of Cincinnati
Conflict of Interest: None Declared

We appreciate the comments from your readers. In fact, we agree with many of their observations. The results of our analysis are more nuanced than a simple assertion that pharmacogenetic-guided (PG) dosing is not "cost effective." Indeed, we used sensitivity analyses to explore how changes in either health care delivery systems or patient characteristics might affect the cost-effectiveness of PG-guided dosing. With regards to systems changes that might make this strategy "cost-effective," we found that the marginal cost-effectiveness ratio drops below a societal willingness-to-pay threshold of $50K per quality adjusted life year (QALY), when the cost of genotyping is less than $200 and results were available at the time of warfarin initiation. Regarding patient characteristics, we found that patients at higher risk for bleeding would benefit more from PG-guided dosing. In particular, PG-guided dosing would cost less than $50K/QALY for patients having more than a two-fold increase, risk of major bleeding compared to a patient without risk factors for hemorrhage. This corresponds to a HEMORR2HAGES score between 1 and 2, or an absolute risk of major hemorrhage greater than ~4% per year(1). Note that all of these findings are predicated on the assumption that PG-guided dosing reduces adverse events. The 32% reduction in major hemorrhage that we estimated was not statistically significant [relative risk "“ 0.68, 95% CI, 0.22 to 2.06]. Thus, additional randomized trials are needed to quantify this reduction.

Finally, the point is made that use of genotype might be helpful in the future, such as if a patient requires a second course of warfarin therapy at a later date. However, in our experience when both genotype and a prior therapeutic dose are known, knowledge of the prior dose trumps genotype.


1. Gage BF, Yan Y, Milligan PE, et al. Clinical classification schemes for predicting hemorrhage: results from the National Registry of Atrial Fibrillation (NRAF). Am Heart J. 2006;151(3):713-9.

Conflict of Interest:

None declared

New Developments May Change Study Results
Posted on July 10, 2009
Charis Eng
Cleveland Clinic
Conflict of Interest: None Declared

Given the FDA's decision to include on warfarin labels a suggestion to consider genetic testing prior to dosing, we read with interest the thoughtful analysis of studies to date on genetic testing for warfarin sensitivity ("Cost- Effectiveness of Using Pharmacogenetic Information in Warfarin Dosing for Patients With Nonvalvular Atrial Fibrillation," 1/20/09, Volume 150, Issue 2, pp. 73-83). The authors' finding that warfarin testing is not cost- effective is reasonable based on the studies they examined, and we appreciate the inclusion of caveats that could potentially change that conclusion. We would like to highlight a few of those points, based on rapidly emerging developments in this field.

Cost. The authors assumed that genetic tests would cost $400, the price that might be paid by a consumer who seeks an independent test from a lab. With more widespread use of the tests, costs for hospitals or clinics are now much lower, as low as $50-$75 per test, or about $150 to $225 if positive and negative controls are run with each test. As the authors note, a $50 test is in fact cost-effective. Given the advances in technology, we suspect the price point will drop further and therefore, make this test not only cost- effective but also beneficial ("clinically worthy") to the patient.

Timing. In their analysis, the authors assume that it would take about three days to get a result from a genetic test. This time frame was reasonable when the underlying studies were done, given that few labs processed the tests at that time. Today, the tests can be turned around in as little as 45 minutes from patient to calculated dose, with most in the range of two to six hours and results obtainable in no more than 24 hours. While it is difficult to calculate the cost-effectiveness of a quicker test, some studies suggest that it will result in fewer adverse events, simply because patients could be given a more appropriate dose more quickly (1, 2, 3).

Finally, we are concerned that the underlying studies, among others, examine the number of bleeding events as a way of determining whether treatment is effective. Bleeding is a relatively rare event, even more so in many of these studies, since it is unlikely that caregivers in state-of-the-art anti- coagulation clinics or clinical research (academic medical centers of excellence) settings would have failed to adjust a patient's warfarin dose before bleeding could occur. Alternative measurements, that are possibly superior, are the number of dose adjustments (times a patient must get a new prescription filled or visit a clinic for an INR measurement), change between initial and final warfarin dose, or time to stable INR, in order to measure whether the genetic test resulted in more accurate initial dosing.

As we are all aware, these are serious issues. Errors in warfarin dosing result in 17,000 strokes and 85,000 serious bleeding events each year, and as many as 43,000 emergency room visits. If the 2 million people who start taking warfarin each year are able to receive a more accurate dose more quickly, the benefit to patients might be immense, with substantial healthcare cost savings.


1. Influence of CYP2C9 and VKORC1 on warfarin response during initiation of therapy. Blood Cells Mos Dis. 2009 Mar 16

2. Effects of CYP2C9 and VKORC1 on INR variations and dose requirements during initial phase of anticoagulant therapy. Pharmacogenomics. 2008 Sept;9(9):1237-50

3. Frequency of adverse events in patients with poor anticoagulation: a meta-analysis. CMAJ. 2007 May 22; 176(11): 1589"“1594

Conflict of Interest:

None declared

Submit a Comment/Letter

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.


Buy Now for $32.00

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Related Articles
Related Point of Care
Topic Collections
PubMed Articles
Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.