NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Cancer (UK). Early and Locally Advanced Breast Cancer: Diagnosis and Treatment [Internet]. Cardiff (UK): National Collaborating Centre for Cancer (UK); 2009 Feb. (NICE Clinical Guidelines, No. 80.)

Cover of Early and Locally Advanced Breast Cancer

Early and Locally Advanced Breast Cancer: Diagnosis and Treatment [Internet].

Show details

Appendix 1Adjuvant! Online: review of evidence concerning its validity, and other considerations relating to its use in the NHS

A report by Jonathan Gribbin & Robyn Dewis

Introduction

The Guideline Development Group (GDG) for early and locally advanced breast cancer proposed a piece of work to assess the validity of Adjuvant! Online as a tool to assist with clinical decisions, about adjuvant therapy, in patients with early invasive breast cancer. This document summarises the methodology used to assess this, and the key findings including a description of the Adjuvant! Online product, the methods used to develop it, and commercial issues associated with recommending its use.

Adjuvant! Online can be accessed at www.adjuvantonline.com. It is a tool for assessing the risks of an individual patient developing recurrent disease and/or dying within 10 years, when receiving specific treatment (on the basis of well validated factors such as age, menopausal status, oestrogen receptor (ER) status, number of involved axillary lymph nodes etc.). Doctor and patient can use the tool together to decide on the most appropriate adjuvant treatment regimen (chemotherapy, endocrine therapy, or none). Adjuvant! Online is a decision aid and does not direct towards a specific treatment regimen.

This appraisal has been proposed as an alternative to a question that had been framed in the PICO question ‘What are the indications for adjuvant chemotherapy in patients with early invasive breast cancer?’. The GDG agreed that this PICO question covered huge topic areas and would need to be addressed using a very long list of search terms which the group were unable to specify satisfactorily.

Noting that Adjuvant! Online is already in use in the UK and is designed to incorporate the Oxford Overview meta-analyses, an alternative, pragmatic approach was proposed, namely of undertaking an appraisal of evidence about the validity of Adjuvant! Online. Two SpR/SpTs providing support to the GDG were asked to undertake this appraisal by reviewing what is known about the tool. The following represents their understanding of the research question, and the approach they took in addressing it.

Research question

The primary purpose of the appraisal was to summarise and critique what is known about Adjuvant! Online, and its validity as a tool for supporting clinical decisions about adjuvant chemotherapy, in UK patients, with early invasive breast cancer. Where it exists, evidence regarding its usefulness is also included.

This is a narrative report incorporating a formally referenced review of the published literature, together with other information provided by Adjuvant!. It addresses:

  • A description of Adjuvant! Online: its intended purpose and use
  • Current usage in the NHS
  • Methodology underpinning Adjuvant! Online, including how it was developed and how it is updated.
  • Any caveats/issues/known shortcomings highlighted to Adjuvant! Online users
  • An appraisal of published evidence about Adjuvant! Online’s validity
  • An appraisal of any published evidence regarding its usefulness
  • General assumptions/issues/uncertainties in applying this tool based on USA data to NHS patients
  • Commercial considerations – implications for Adjuvant! Online’s validity and/or practical use
  • Licensing considerations – implications for unrestricted access to Adjuvant! Online
  • Any other practical considerations relating to Adjuvant! Online’s use in the NHS.

Specific questions raised by GDG members that were included within the appraisal

  1. To what extent does the SEER database on which the tool is based consider adverse reactions?
  2. What is the applicability of the USA data in the SEER database to UK patients in the NHS?
  3. What commercial relationships underpin the design and maintenance of the system?
  4. Are there any current/future licensing considerations for NHS users?
  5. What are the key practical considerations relating to its use?

Excluded from the appraisal

The decision about which chemotherapy or endocrine therapy regimen to recommend are separate questions, which fall outside the scope of this appraisal. This appraisal focuses on the validity of the Adjuvant! Online tool itself.

This approach highlights issues relating to major assumptions inherent in the methodology which are apparent from a consideration of Adjuvant! Online’s methodology and the published literature. However, it does not provide a systematic, exhaustive breakdown of all the individual factors, algorithms and statistical models on which the Adjuvant! Online model may be based (except where these are appraised in the published literature). Similarly, this relatively short piece of work is not intended to be a critical appraisal of the Oxford Overviews (whose meta-analyses are fundamental to Adjuvant! Online).

Search strategy

Sources

The Ovid search engine was used to interrogate MEDLINE database (1950 to October 2007) and EMBASE. A subsequent search was also made against SIGLE for relevant grey literature.

Search parameters

A pilot search experimented with a number of synonyms for Adjuvant! Online. The final definitive search was executed using the following search criteria (Table A1.1) and the above source.

Table A1.1

Table A1.1

Parameters and logic used in the automated search

Further screening and supplementary information

The results of the automated search were manually screened, by reading the abstracts, in order to identify relevant articles and to exclude all other papers that were not reporting research into Adjuvant! Online or similar decision support tools.

Adjuvant Inc. was invited to respond directly to specific questions that the literature does not address. These responses are incorporated in the findings.

Search results

Executing the automated search strategy resulted in the identification of 615 papers satisfying the search parameters. Manual screening of abstracts resulted in the exclusion of all but 9 of these papers. Excluded papers included studies of specific treatments, risk communication and other methods of displaying outcomes for example prognostic tables.

Findings

Adjuvant! Online tool

The purpose of Adjuvant! Online is to assist healthcare professionals and patients with early stage breast cancer to discuss the risks and benefits of adjuvant therapy after surgery. It does this by presenting estimates of the risk of cancer-related mortality or relapse, which can be used in consultations. It is intended to be operated and interpreted by oncologists and suitably qualified healthcare professionals. It is not intended to replace clinical judgement and is not designed to be used by patients alone.

Conceptual design

The concept behind Adjuvant! Online is that the quality of decision-making about adjuvant therapy is enhanced in consultations where clinicians can communicate to patients the net benefit of various adjuvant therapies (Ravdin et al., 2001). Therefore Adjuvant! Online is designed to:

  1. Estimate the ‘baseline’ risk of mortality or relapse for patients without adjuvant therapy
  2. Estimate the proportion of negative events that given therapies are known to prevent
  3. Epply this effect to the baseline risk so that direct comparisons can be made of the estimated risks of mortality or relapse between treatments and with no treatment.

User functionality

The current version of Adjuvant! is version 8. User functionality comprises facilities to:

  1. Enter patient information including age, comorbidities plus tumour information including size, oestrogen receptor status and number of involved lymph nodes. This is used to estimate risk at 10 years of breast cancer related death or relapse without additional therapy
  2. Display information about the efficacy of different therapy options, with the option of overriding the estimated efficacies
  3. Derive estimates of risk at 10 years of breast cancer related death or relapse for the treatments selected by the user
  4. Print results, access on-line help and links to sources of evidence.

Underlying this user functionality there are tables and algorithms, which aim to encapsulate evidence of effectiveness according to the Oxford Overviews. These are maintained by Adjuvant! Inc.

User access to these is limited to that described above. User access to Adjuvant! Online is controlled via a logon screen requiring a username and password. Registration for a username and password is open to users willing to sign a license agreement. In doing so they agree that they are a suitably qualified medical professional. There is no additional authentication of this at registration.

Technological implementation

Users access Adjuvant! Online via a desktop browser with an Internet connection to www.adjuvantonline.com. User functionality is implemented in a Java-based program which is only present for the duration of the user’s session. Some functionality also requires Adobe Acrobat and/or a printer. The server functionality runs under a Unix operating system. No patient identifiers are entered into Adjuvant! Online, thereby avoiding any risk or concern relating to patient confidentiality.

Further evaluation of the physical implementation is beyond the scope of this study.

There are also versions of Adjuvant! Online designed to run on Palmtop or PocketPC. These are also beyond the scope of this study.

Control and licensing

Adjuvant! Online is owned by a US-based company called Adjuvant Inc. Adjuvant Inc. and all IP rights in the Adjuvant! Online tool are owned by Dr Ravdin, who has created and developed the tool over a period of more than 10 years. Dr Ravdin’s stated motivation is academic; the venture has not been for the purpose of realising financial profit (Ravdin, 2008).

Over the years, funding has been secured from government, industry and research foundations. None of these sources of funding exercise editorial purview over the content of releases. Adjuvant! Online carries no advertising and there are no other sources of revenue.

Licenses to use Adjuvant! Online are free of charge. Dr Ravdin states they will remain free of charge indefinitely; there is no plan to charge a license fee either now or in the long term (Ravdin, 2008).

Maintenance and development

Maintenance of functionality in the current version of the tool is undertaken by Adjuvant! Inc., which secures part-time or occasional assistance from a small group of relevant specialists.

Help files are updated to reflect the current literature. The user functionality and underlying methodology is updated less frequently; recent versions of the tool have incorporated only minor changes.

The direction and timing of these developments is determined by Dr Ravdin, according to the publication of new evidence, requests from users, and the availability of personnel to implement the changes. In the past, new versions have been released around the time of major research meetings, for example ASCO, San Antonio Breast Cancer Symposium.

Currently efforts are focussed on developing the next major release of Adjuvant! Online, which will incorporate recent trial evidence relating to human epidermal growth factor receptor 2 (HER2) and trastuzumab. Beyond this, there is no formally documented plan describing the development path for the product.

Users are not required to undertake any maintenance.

Current usage in the NHS

Dr Ravdin reports that there were 2,978 registered active users in the UK as at July 2007 (which represents about 7% of the total registered user base of more than 42,000). This is based on information supplied at registration which is not authenticated.

Estimates of frequency of usage are derived from the number of Adjuvant! Online sessions that ran in a given period of time. In the first six months of 2007 the Adjuvant! Online platform delivered 110,800 user sessions. Based on the crude assumption that frequency of usage is the same across all users, this represents an estimated 8,000 user sessions in the same period for users registered in the UK. It is not possible to determine how many of these sessions supported actual consultations with NHS patients.

A survey of usage amongst oncologists in the UK is planned but will not report before July 2008 at the earliest (Agarwal, 2008).

Underlying methodology - derivation of baseline risk estimate

Population

The data used for the baseline risk estimate was derived from the SEER database (Surveillance, Epidemiology and End Results Program in the USA) (US National Cancer Institute). Adjuvant! Online was based upon database 9 which covered 14% of the US population (Warren et al., 2002). Detailed information was not available on the breakdown for the SEER 9 population but studies have assessed its similarity to the US population:

  1. The SEER population is similar to the US population in terms of age and sex distribution. The US population has a larger percentage of the population in the under 55 age groups and fewer in the over 55 age groups, when compared to the population of England and Wales (Office of National Statistics).
  2. The SEER population over represents certain ethnic groups, for example Native American/Hawaiian and some South East Asian groups compared to the US population. This is related to the States that are included in the database e.g. Alaska and Hawaii (US National Cancer Institute).
  3. The ethnic mix of the US population differs from that of England and Wales. Only broad categories can be considered due to differences in categorising ethnicity, but broadly speaking in the US there are lower percentages of white and mixed races, with higher percentages of black and other races (US National Cancer Institute)
  4. Socioeconomic data in the SEER database is of poor quality.
  5. Date and cause of death are recorded. Date of death is considered robust, however cause of death is of poor quality (Warren et al., 2002).

As survival is analysed in terms of age group the differences in the age of the population is unlikely to affect the generaliseability of the data. The difference in ethnicity, however, is likely to affect this. The incidence of breast cancer is highest in the white population, but mortality is highest in the black population. A program based on this data, that does not take ethnicity into account, will tend to overestimate survival in the black population and underestimate in the white. It is difficult to assess what effect this would have on other ethnic groups or to know if survival differs in these ethnic groups in the United Kingdom.

Selection

Ravdin et al. (2001) selected a population from the SEER database for the development of Adjuvant!. Women who met the following criteria were included in the calculations of baseline risk:

  1. Had invasive, unilateral and non-inflammatory breast cancer
  2. Had received definitive surgery and axillary staging with at least 6 lymph nodes
  3. Had data on tumour size, number of lymph nodes sampled and the number of positive lymph nodes.

Women were specifically excluded from the calculations of baseline risk for the following reasons:

  1. Those aged under 35 years. This group of young women were observed to have a worse prognosis than the other age groups. (A correction applied to allow for this group of women is described below.)
  2. Those aged over 59 years. This group of women was believed to be healthier and have better access to health care. Analysis of this group revealed that women with breast cancer appeared to have better survival than the general US population of the same age.

Survival

The SEER data were then used to calculate survival. This was observed survival for 5 years that was then extrapolated to 10 years, as the data were insufficient to cover this period. Relative survival was used, which makes an adjustment for age specific death rates from other causes. This survival estimate is based upon the tumour size, the number of positive lymph nodes and the oestrogen receptor status of the tumour. There are some assumptions made in calculating survival for Adjuvant! Online.

  1. Impact of ER status. There were data issues around ER status that led to estimates inconsistent with what would be expected from the literature. For this reason a relative risk of 1.3 was applied to predict survival in ER-positive and negative individuals (based on evidence from long-term studies of lymph node-negative patients).
  2. The effect of stage of tumour and adjuvant therapy received. An assumption was made that a percentage of the population would have received adjuvant therapy. In order to find the ‘baseline risk’, the survival without the use of adjuvant therapy, it was assumed that at stage one the adjuvant therapy would have improved outcomes by 15% and at all other stages by 30%.
  3. Constant hazard. Survival calculations assume that the risk of death/recurrence remain constant throughout the study period considered.

Relapse

The SEER database does not hold information on relapse of disease. An assumption is made that, on average, individuals survive for three years after relapse of breast cancer in order to calculate the risk of relapse.

Other issues with UK/US comparisons

Other differences between the US and UK population were also considered. There is a lack of universal access to healthcare in the US, which may affect the survival of certain groups within the US. However, individuals’ data were only entered into the study when they had received initial surgery and staging and so should not affect applicability to the UK population. There are also differences in attitudes towards healthcare between the two countries, for example the UK population tend to choose less radical surgery than the US population (Locker et al., 2004). Although this may lead to differences in decisions made when using the tool it does not affects its validity for the UK.

Estimating negative outcomes averted

Adjuvant! Online applies an estimation of negative outcomes averted to the baseline survival to give an estimation of survival following one or more adjuvant therapies (Ravdin et al., 2001). Estimation of negative outcomes averted is quantified in terms of the proportion risk reduction (PRR), i.e. the proportion of the baseline risk, which is reduced by each therapy.

PRR for specific therapies are derived from the Overviews. They are incorporated into Adjuvant! Online to derive estimates of breast cancer specific mortality. To avoid the possibility of gross error in estimating the breast cancer specific mortality of over 70 year olds (in which group most mortality is probably non-breast cancer specific), Adjuvant! applies the PRR for 50–69 years for women 70 years or older. When the operator is using the tool to model outcomes for patients over 70 years of age, Adjuvant! Online warns the user about the possible effect of this simplifying assumption.

To model the relative value of various chemotherapy regimens Adjuvant! Online groups treatments into three distinct “generations”, based on their perceived efficacy and toxicity. Prompts appear on screen at relevant points in the user session with details of the basis on which this grouping has been done. The prompts also outline the key inferences that Adjuvant! Online makes to estimate relative efficacy (for example of a third generation regimen compared to none) and points the user to further information contained in the Help files.

Applying calculation to previous baseline

The Oxford Overviews report the results of clinical trials. Few trials for cancer therapy consider the effect of one treatment against placebo/no treatment. The majority report the risk reduction of using one treatment over another. According to Ravdin et al. (2001), the Overviews suggest that treatment effects are independent of other treatment used. Adjuvant! uses this assumption, through the following formula, to calculate the proportionate risk reduction achieved by the use of a specific adjuvant therapy:

PRR combined therapy=1-[(1-PRR therapy 1)×(1-PRR therapy 2)]

Validation

Since Ravdin et al.’s 2001 paper describing the tool and its methodology, there have been two further published studies that assess the validity of the Adjuvant! Online tool. The tool is currently being compared against two further European registers (Ravdin, 2008).

Prospective population-based validation

Olivotto et al. (Olivotto et al., 2005) set out to independently validate Adjuvant! Online by comparing the observed 10 year outcome of each of 4083 patients with stage 1 and 2 breast cancer on a British Columbian register with the outcome predicted by Adjuvant! Online.

Taking the cohort as a whole, they found a high degree of agreement between the predicted and observed overall survival and breast cancer specific survival. They also analysed the differences between observed and predicted outcomes for specific subgroups which in most cases were within 2% or not significantly different (at P>0.05).

For patients younger than 35 years of age or with lymphatic or vascular invasion (LVI) Adjuvant! over-estimated the survival. After the operators applied their judgement to adjust for LVI using the prognostic factor impact calculator tool within Adjuvant! (PFIC), the 10 year predictions were no longer significantly different.

The strength of this study is that it provides validation of Adjuvant! Online predictions using an external reference population. The strength of evidence it provides in this assessment is limited by the following factors:

  • The study was undertaken on version 5 of Adjuvant! Online
  • It is implicit that the operators were very familiar with the tool, and may have included its author. It is not clear whether an “average” operator would achieve the same level of agreement when making adjustments using the prognostic factor impact calculator (PFIC).

In summary, the study provides independent validation of an earlier version of Adjuvant! Online. For women aged 30 to 59 years of age whose adverse prognostic factors are automatically accounted for within the tool, Adjuvant! Online provides reliable predictions of the benefits of adjuvant therapy. The reliability of predictions for other groups depends in part on the knowledge and judgement of the operator in making adjustments using the PFIC.

It should be noted that more recent versions of Adjuvant! Online incorporate an adjustment to “correct” the overestimation of survival for young ER-positive patients (Ravdin, 2008).

Clinician-based validation

Loprinzi et al. (2001) describe the development of an algorithm to calculate 10-year outcomes for breast cancer patients. As part of this, they asked 11 US oncologists for their estimates of 10-year disease-free survival. The mean of these estimates were compared to predictions generated by Adjuvant! Online. The degree of correlation was not measured formally; the graphical representation of the correlation suggests a reasonable degree of agreement.

These published data provides weak evidence for the validity of Adjuvant! Online. However, the fact that the predictions of oncologists vary supports the rationale that there is a need for a tool, which provides evidence-based predictions in an understandable format.

Impact and usefulness

The purpose of Adjuvant! Online is to provide predictions of risk that support dialogue between clinician and patient about the most appropriate adjuvant therapies for that patient. There is little published literature evaluating the impact of Adjuvant! on these interactions, nor on the degree to which clinicians correctly handle the tool or what meaning patients ascribe to the predictions. A USA study (Siminoff et al., 2006) of the effects on treatment choices of Adjuvant! Online compared to well presented information pamphlets did not find statistically significant differences between the groups. After adjusting for disease-related and socio-demographic confounders, they found that those who used Adjuvant! Online were less likely to choose adjuvant treatment (OR 0.32 95%CI 0.12–0.84). This is broadly consistent with the findings of an apparently related study (Peele et al., 2005).

A study of 102 treatment management decisions in a Hong Kong oncology centre (Epstein et al., 2006) found that clinicians changed their decision in 13 instances after taking into consideration the predictions made by Adjuvant! Online. Based on analysis of this decision-making, Adjuvant! Online’s impact was attributed to: the distinction it makes between the marginal benefits of intervention compared to prognosis per se, the deeper consideration of therapeutic goals and costs for individuals which it enables, a comparison of the relative benefits of different treatments, the quantification of iatrogenic risks. The study found that treatment decisions continued to be strongly influenced by factors omitted from the version of Adjuvant! Online used in the study (for example lymphovascular invasion and HER2 expression). Clinicians in this study tended to ignore the adjustments to risk recommended by the programme on the basis of low tumour grade when these adjustments were perceived to conflict with other indicators such as lymph node-positivity. Clinicians’ attitudes to the utility of Adjuvant! Online were varied but the study authors formed the impression that, in the context of case discussions, the tool enabled groups to achieve consensus more quickly.

There is a body of literature concerning the impact of other decision tools on a range of patient-clinician interactions. For example, a systematic review (O’Connor et al., 1999) of 17 RCTs did not show a consistent impact on patient knowledge and satisfaction. More recently, there has been at least one trial to evaluate the effect of a decision support tool on the knowledge and satisfaction of breast cancer patients in particular (Whelan et al., 2003). A full review of this literature is beyond the scope of this assessment.

Summary/Conclusions

The predictions made by Adjuvant! Online are based on a published methodology, which has been updated periodically as evidence of treatment effectiveness and data on risk factors becomes available.

Help files and published descriptions of the tool make clear some of the assumptions and limitations that underpin the methodology. The impact of these individual assumptions is difficult to assess, and beyond the scope of this paper. Adjuvant! Online deals with key uncertainties by alerting the user to them at relevant points.

Survival estimates are derived from a US population. Quantifying the impact on survival of socio-economic background and of ethnic differences between the US and UK populations is difficult.

The strongest evidence of Adjuvant! Online’s validity for the UK is derived from comparisons between its predictions and observed outcomes using a Canadian population. This study found its predictions to be reliable for most groups. Since that study, an adjustment has been applied to ‘correct’ the predictions made for a subset of younger patients.

Further validation is under way using European populations. Dr Ravdin would welcome similar validation against a UK population.

Weaker evidence for its validity includes comparisons of its predictions with the predictions of clinicians. The development path for Adjuvant! Online appears to be consistent with a product which intends to remain evidence-based.

Dr Ravdin’s stated intention is that license to use Adjuvant! Online will remain free of charge. This together with its web-based design means that the cost to users of using Adjuvant! Online should remain very low.

There are only two trials assessing the impact of Adjuvant! Online in patient and clinician interactions. These indicate that in a USA setting patients considering adjuvant treatment were less likely to select adjuvant treatment if their consultation involved use of Adjuvant! Online instead of an information pamphlet. A third study of 102 clinician decisions about patient management found that using Adjuvant! Online resulted in a change of decision in 13 cases, and that clinicians’ views of the tool’s utility were varied.

References

  1. Adjuvant! Online program www​.adjuvantonline.com.
  2. Ravdin PM. A computer program to assist in making breast cancer adjuvant therapy decisions. Seminars in Oncology. 1996;23(1 Suppl 2):43–50. [PubMed: 8614844]
  3. Ravdin PM, Siminoff LA, Davis GJ, Mercer MB, Hewlett J, Gerson N, et al. Computer program to assist in making decisions about adjuvant therapy for women with early breast cancer. Journal of Clinical Oncology. 2001;19(4):980–991. [PubMed: 11181660]
  4. Loprinzi CL, Thome SD. Understanding the utility of adjuvant systemic therapy for primary breast cancer. Journal of Clinical Oncology. 2001;19(4):972–979. [PubMed: 11181659]
  5. Ravdin PM, Davis GJ. A method for making estimates of the benefit of the late use of letrozole in patients completing 5 years of tamoxifen. Clinical Breast Cancer. 2004;5(4):313–316. [PubMed: 15507180]
  6. Whelan TJ, Loprinzi C. Physician/patient decision aids for adjuvant therapy. Journal of Clinical Oncology. 2005;23(8):1627–30. [PubMed: 15755969]
  7. Siminoff LA, Gordon NH, Silverman P, Budd T, Ravdin PM. A decision aid to assist in adjuvant therapy choices for breast cancer. Psycho-Oncology. 2006;15(11):1001–1013. [PubMed: 16511899]
  8. Olivotto IA, Bajdik CD, Ravdin PM, Speers CH, Coldman AJ, Norris BD, et al. Population-based validation of the prognostic model ADJUVANT! for early breast cancer. Journal of Clinical Oncology. 2005;23(12):2716–25. [PubMed: 15837986]
  9. Peele PB, Siminoff LA, Xu Y, Ravdin PM. Decreased use of adjuvant breast cancer therapy in a randomized controlled trial of a decision aid with individualized risk information. Medical Decision Making. 2005;25(3):301–307. [PubMed: 15951457]
  10. Epstein RJ, Leung TW, Mak J, Cheung PS. Utility of a web-based breast cancer predictive algorithm for adjuvant chemotherapeutic decision making in a multidisciplinary oncology center. Cancer Investigation. 2006;24(4):367–373. [PubMed: 16777688]
  11. Ravdin P. Personal communication to: J. Gribbin regarding Adjuvant! tool. Feb, 2008.
  12. Agarwal V. Personal communication to J. Gribbin regarding plans for surveying usage of Adjuvant Online. Mar, 2008.
    US National Cancer Institute. Surveillance, Epidemiology and End Results program
  13. Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Medical Care. 2002;80(8 Suppl IV):3–18. [PubMed: 12187163]
  14. Office of National Statistics
  15. Locker GY, Sainsbury JR, Cuzick J, Group AT. Breast surgery in the ‘Arimidex, Tamoxifen Alone or in Combination’ (ATAC) trial: American women are more likely than women from the United Kingdom to undergo mastectomy. Cancer. 2004;101(4):735–740. [PubMed: 15305403]
  16. O’Connor AM, Rostom A, Fiset V, Tetroe J, Entwistle V, Llewellyn-Thomas HA, et al. Decision aids for patients facing treatment or screening decisions: systematic review. British Medical Journal. 1999;319(7212):731–734. [PMC free article: PMC28223] [PubMed: 10487995]
  17. Whelan T, Sawka C, Levine M, Gafni A, Reyno L, Willan A, et al. Helping patients make informed choices: a randomized trial of a decision aid for adjuvant chemotherapy in lymph node-negative breast cancer[see comment] Journal of the National Cancer Institute. 2003;95(8):581–587. [PubMed: 12697850]
Copyright © 2009, National Collaborating Centre for Cancer.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licenses issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK11639

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.0M)
  • Disable Glossary Links

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...