Table 2 Statements for which consensus was reached.

Utilisation of BC risk assessment tools

When using a validated breast cancer risk assessment tool, there should be national consensus on which tool to use?

(51% Strongly Agree; 45% Agree (96% consensus, n = 47))

Which breast cancer risk assessment tool do you think should be used?

(CanRisk; 100% consensus, n = 47)

Providing information to patients from CanRisk

When using CanRisk, it is best practice to explain that the risk assessment could alter depending on accuracy and extent of the information input into the model and/or changes in understanding of how these factors influence risk.

(48% Strongly Agree; 52% Agree (100% consensus, n = 48))

It is best practice to provide patients with information explaining their risk assessment in simple lay terms and in comparison to population breast cancer risk.

70% Strongly Agree; 30% Agree (100% consensus, n = 43)

Use of CanRisk for breast surveillance recommendations and appropriate timing

Where appropriate infrastructure is available, use of CanRisk is the preferred method to make breast surveillance recommendations for women unaffected with breast cancer and relevant family history (where no known monogenic cause).

(36% Strongly Agree; 55% Agree (91% consensus, n = 47))

When using CanRisk, for the purpose of a risk assessment to inform breast surveillance recommendations, it is best practice to undertake the assessment close to the age at which screening would commence i.e approaching age 40 (where no known monogenic cause).

(14% Strongly Agree; 73% Agree (87% consensus, n = 44))

When using CanRisk, for the purpose of a risk assessment to inform breast surveillance recommendations in women with a Likely Pathogenic/Pathogenic variant in a breast cancer susceptibility gene it is best practice to undertake the assessment close to the age at which screening would commence (e.g. approaching age 25 for BRCA1).

(19% Strongly Agree; 79% Agree (98% consensus, n = 42))

Where appropriate resource is available, it is best practice to use CanRisk to provide information on 5/10 year/lifetime breast cancer risks to women unaffected with cancer considering bilateral risk reducing mastectomy as part of broader consultation and shared decision making.

(53% Strongly Agree; 45% Agree (98% consensus, n = 47))

Where appropriate resource is available, it is best practice to use CanRisk to provide information on 5/10 year/lifetime breast cancer risks to women with breast cancer considering risk reducing mastectomy as part of broader consultation and shared decision making.

(19% Strongly Agree; 67% Agree (86% consensus, n = 42))

Use of CanRisk for genetic testing eligibility

It is best practice to use CanRisk, where the Manchester score is borderline to determine eligibility for genetic testing, where national test directory criteria  are not reached based on personal history alone [33].

(30% Strongly Agree; 68% Agree (98% consensus, n = 44)

When using CanRisk, it is appropriate to round up to the nearest whole number for the purpose of determining genetic test eligibility.

(13% Strongly Agree; 70% Agree (83% consensus, n = 46)

CanRisk model inputs

Where family history is utilised, it is best practice to include at least a three-generation family tree and ensure that information on both relatives affected and unaffected with cancer is included

(52% Strongly Agree; 48% Agree (100% consensus, n = 40)

When utilising CanRisk to provide an individualised risk assessment, what should represent the minimum inputs to the model for recommendations on breast surveillance (vote on all options you consider should always be used in the assessment).

Personal History of cancer; 100% consensus, n = 46)

Family History of cancer; 98% consensus, n = 46)

Genetic test results (patient); 93% consensus, n = 46)

Genetic test results (family member); 93% consensus, n = 46)

Where additional information is available for a woman beyond the agreed minimum inputs, it is best practice to utilise this information in a risk assessment.

(33% Strongly Agree; 63% Agree (96% consensus, n = 46))

Where possible, all available information should be incorporated when making a risk assessment.

(38% Strongly Agree; 58% Agree (96% consensus, n = 45))