Dataset Information

An investigation into the performance of the Adjuvant! Online prognostic programme in early breast cancer for a cohort of patients in the United Kingdom.

ABSTRACT: Adjuvant! Online is an internet-based computer programme providing 10-year prognosis predictions for early breast cancer patients. It was developed in the United States, has been successfully validated in Canada, and is used in the United Kingdom and elsewhere. This study investigates the performance of Adjuvant! in a cohort of patients from the United Kingdom.Data on the prognostic factors and management of 1065 women with early breast cancer diagnosed consecutively at the Churchill Hospital in Oxford between 1986 and 1996 were entered into Adjuvant! to generate predictions of overall survival (OS), breast cancer-specific survival (BCSS), and event-free survival (EFS) at 10 years. Such predictions were compared with the observed 10-year outcomes of these patients.For the whole cohort, Adjuvant! significantly overestimated OS (by 5.54%, P<0.001), BCSS (by 4.53%, P<0.001), and EFS (by 3.51%, P=0.001). For OS and BCSS, overestimation persisted across most demographic, pathologic, and treatment subgroups investigated. Differences between Adjuvant! predicted and observed EFS appeared smaller, and were significant for far fewer subgroups, only 5 out of the 28. The likely explanation for such discordance is that US breast cancer mortality rates (upon which Adjuvant! is based) appear to be systematically lower than breast cancer mortality rates in the United Kingdom. Differences in survival after recurrence would seem to be one contributory factor, with data suggesting that prognosis after relapse appears poorer in the United Kingdom. This may reflect the fact that new and more effective cancer drugs are often only approved for use in the United Kingdom many years after their adoption in the United States.The use of Adjuvant! by clinicians within the UK National Health Service is increasing, under the assumption that the programme is transferrable to the United Kingdom. At least for women treated for breast cancer at the Churchill Hospital in Oxford, however, Adjuvant!'s predictions were on the whole overoptimistic. If the findings reported here could be shown to be generalisable to other areas of the United Kingdom, then thought should perhaps be given to the development of a UK-specific version of the programme.

SUBMITTER: Campbell HE

PROVIDER: S-EPMC2768087 | biostudies-other | 2009 Oct

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:BackgroundEarly uptake of multiple risky behaviours during adolescence, such as substance use, antisocial and sexual behaviours, can lead to poor health outcomes without timely interventions. This study investigated how early-life maternal attachment, or emotional bonds between mothers and infants, influenced later risky behaviours in adolescence alongside other potential explanatory pathways using the United Kingdom Millennium Cohort Study.MethodsTotal maternal attachment scores measured at 9 months using the Condon (1998) Maternal Postnatal Attachment Scale compared higher and lower attachment, where mothers in the lowest 10th percentile represented lower attachment. Multiple risky behaviours, defined as two or more risky behaviours (including smoking cigarettes, vaping, alcohol consumption, illegal drug use, antisocial behaviour, criminal engagement, unsafe sex, and gambling), were scored from 0 to 8 at age 17. Five multivariate logistic regression models examined associations between maternal attachment and multiple risky behaviours among Millennium Cohort Study members (n = 7796). Mediation analysis sequentially adjusted for blocks of explanatory mechanisms, including low attachment mechanisms (multiple births, infant prematurity, sex, breastfeeding, unplanned pregnancy and maternal age at birth), maternal depression, and social inequalities (single-parent status, socioeconomic circumstance by maternal education and household income) at 9 months and poor adolescent mental health at 14 years.ResultsChildren of mothers with lower maternal attachment at 9 months had 23% increased odds of multiple risky behaviours at 17 years (OR: 1.23, 95% CI: 1.00-1.50) in the unadjusted baseline model. All five explanatory blocks attenuated baseline odds. Low attachment mechanisms attenuated 13%, social inequalities 17%, and poor mental health 17%. Maternal depression attenuated the highest proportion (26%) after fully adjusting for all factors (30%).ConclusionsLower maternal attachment in early life predicted increased adolescent multiple risky behaviours. Almost a third of the excess risk was attributable to child, maternal and socioeconomic factors, with over a quarter explained by maternal depression. Recognising the influence of early-life risk factors on adolescent health could innovate current policies and interventions addressing multiple risky behaviour uptake affecting health inequalities across the life course.

Dataset Information

An investigation into the performance of the Adjuvant! Online prognostic programme in early breast cancer for a cohort of patients in the United Kingdom.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets