Table 2

Risk-stratified treatment recommendations of the included guidelines.

Guideline title

NCEP

NICE1

AHA1

AHA2


    Risk assessment tools

Risk prediction model

Framingham Risk Score

10-year risk of developing CVD, not referring to a specific risk model

Framingham Risk Score

Framingham Risk Score

Outcome of interest and its timeframe

CHD (10 years)

CVD (CHD and stroke, 10 years)

CHD (10 years)

CHD (10 years)

Information on validation of the model provided in the guideline

Yes

Unclear

Yes

Yes

    Evidence of treatment effects

Treatment considered

LDL-lowering therapy, therapeutic lifestyle change and LDL goals

Statin

Diet, weight management, physical activity, drug therapy and LDL-C goals

Lifestyle management, pharmacotherapy and LDL-C target levels

Target population

Adults

Adults at risk of CVD

Adult patients at increased risk of stroke

Adult women 20 years and older

Type of studies considered in the evidence of treatment benefits

Single or several RCTs

Meta-analyses

Single or several RCTs

Meta-analyses

Single or several RCTs

Meta-analyses

Single or several RCTs

Meta-analyses

Type of studies considered in the evidence of treatment harms

Observational studies

Single or several RCTs

Single or several RCTs

Meta-analyses

Treatment harms not reported

Treatment harms not reported

Heterogeneity of treatment effects assessed in the guideline

Yes

Yes

No

No

    Application of treatment evidence to baseline risks

Methods to apply treatment evidence to baseline risks

Used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Unclear, presumably used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Not reported

Not reported

Assumptions specified when applying treatment evidence

'For every 30-mg/dL change in LDL-C, the relative risk for CHD is changed in proportion by about 30%, and the relative risk is set at 1.0 for LDL-C = 40 mg/dL.'

'For every 1% reduction in LDL-C levels, relative risk for major CHD events is reduced by approximately 1%.'

'Statins do not differ in their relative effectiveness in a number of subgroups: in women compared with men at a similar level of cardiovascular risk; in people with diabetes compared with people without diabetes; or in people aged over 65 years compared with people aged under 65 years.'

Not reported

Not reported

    Development of treatment thresholds

Risk stratification in which different treatments were recommended

•10-year CHD risk > 20%

•10-year CHD risk 10 to 20%

•10-year CHD risk < 10%

•10-year CVD risk ≥20%

•0 to 1 CHD risk factor

•≥2 CHD risk factors and 10-year CHD risk < 20%

•≥2 CHD risk factors and 10-year CHD risk 10% to 20%

•CHD or CHD risk equivalent (10-year risk > 20%)

•10-year CHD absolute risk > 20%

•10-year CHD absolute risk 10% to 20%

•10-year CHD absolute risk < 10%

Methods to develop treatment thresholds

Unclear

Expert consensus

Referring to NCEP ATP-III guideline

Not reported

Explicitly planned benefit and harm assessment as the basis for making recommendations

No

No

No

No

Patient preferences considered when developing recommendations

No

No

No

No


Guideline title

MSC1

NICE2

ACCP

MSC2


    Risk assessment tools

Risk prediction model

Framingham Risk Score (for patients without diabetes) and UKPDS Risk Engine (for patients with diabetes)

Framingham Risk Score

Framingham Risk Score

Framingham Risk Score or UKPDS Risk Engine for patients with diabetes

Outcome of interest and its timeframe

CHD (10 years)

CVD (CHD and stroke, 10 years)

CHD (10 years)

CHD (10 years)

Information on validation of the model provided in the guideline

No

Yes

No

No

    Evidence of treatment effects

Treatment considered

Lifestyle management, pharmacologic treatment and desirable lipid results

Lifestyle advice and statin

Aspirin and vitamin K antagonists

Lifestyle management and antihypertensive drugs

Target population

Men aged > 40 years and women aged > 50 years

Adults aged 18 and older and who have established CVD or who are at high risk of developing CVD

Patients at risk for coronary artery disease

Non-pregnant adults (age 19 years and older) with hypertension

Type of studies considered in the evidence of treatment benefits

Other guidelines

Meta-analyses

Meta-analyses

Meta-analyses

Type of studies considered in the evidence of treatment harms

Treatment harms not reported

Meta-analyses

Single or several RCTs

Meta-analyses

Treatment harms not reported

Heterogeneity of treatment effects assessed in the guideline

No

Yes

Yes

No

    Application of treatment evidence to baseline risks

Methods to apply treatment evidence to baseline risks

Not reported

Not reported

Unclear, presumably used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Assumptions specified when applying treatment evidence

Not reported

Not reported

Not reported

'This assumes 20% risk reduction of CHD based on average outcomes for appropriately used blood pressure lowering medications and statin medications.'

    Development of treatment thresholds

Risk stratification in which different treatments were recommended

•Framingham CHD risk ≥20% without CHD

•Framingham CHD risk 10% to 19%

•Framingham CHD risk < 10%

•CVD risk < 20%

•CVD risk ≥20%

Moderate risk for a coronary event (10-year risk of a cardiac event > 10%)

Diagnosis of hypertension confirmed and CHD risk ≥20% over 10 years

Method to develop treatment thresholds

Referring to 2005 British Columbia guideline Diabetes Care

Referring to the NICE technology appraisal Statins for the Prevention of Cardiovascular Events

Unclear, presumably putting benefits and harms on the same scale and find a balance between them

Unclear

Explicitly planned benefit and harm assessment as the basis for making recommendations

No

Unclear

No

Unclear

Patient preferences considered when developing recommendations

No

No

No

No


Guideline title

USPSTF

UMHS1

ISCI

MQIC


    Risk assessment tools

Risk prediction model

Framingham Risk Score

Framingham Risk Score

Framingham Risk Score

Framingham Risk Score

Outcome of interest and its timeframe

CHD (10 years) in men and stroke (10 years) in women

Hard CHD (myocardial infarction and coronary death, 10 years)

CHD (10 years)

CHD (10 years)

Information on validation of the model provided in the guideline

No

No

No

No

    Evidence of treatment effects

Treatment considered

Aspirin

Lifestyle changes, drug therapy and LDL-C goals

Drug therapy and LDL goals

Drug therapy and goal for LDL-C

Target population

Men aged 45 to 79 years and women aged 55 to 79 years

Adults 20 to 75 years of age without familial or severe dyslipidemias

Adults 20 years and older and who are dyslipidemic

Adults ≥18 years

Type of studies considered in the evidence of treatment benefits

Meta-analyses

Treatment benefits reported but study type unclear

Single or several RCTs

Meta-analyses

Treatment benefits not reported

Type of studies considered in the evidence of treatment harms

Observational studies

Treatment harms reported but study type unclear

Single or several RCTs

Meta-analyses

Treatment harms not reported

Heterogeneity of treatment effects assessed in the guideline

Yes

Yes

No

No

    Application of treatment evidence to baseline risks

Methods to apply treatment evidence to baseline risks

Used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Not reported

Assumptions specified when applying treatment evidence

There is 'a 32% risk reduction of MIs with regular aspirin use' (in men) and 'a 17% risk reduction of strokes with regular aspirin use' (in women).

'The risk for gastrointestinal bleeding increases with age.'

Not reported

Not reported

Not reported

    Development of treatment thresholds

Risk stratification in which different treatments were recommended

•Men aged 45 to 59 years and 10-year CHD risk ≥4%; men aged 60 to 69 years and 10-year CHD risk ≥9%; men aged 70 to 79 years and 10-year CHD risk ≥12%

•Women aged 55 to 59 years and 10-year stroke risk ≥3%; women aged 60 to 69 years and 10-year stroke risk ≥8%; women aged 70 to 79 years and 10-year stroke risk ≥11%

•0 to 1 risk factors

•2+ risk factors and 10-year CHD risk < 10%

•2+ risk factors and 10-year CHD risk 10% to 20%

•0 to 1 risk factor and 10-year CHD risk < 10%

•2+ risk factors and 10-year CHD risk < 10%

•2+ risk factors and 10-year CHD risk 10% to 20%

•CHD or CHD equivalent and/or 10-year risk > 20%

•CHD or CHD risk equivalents 10-year risk > 20%

•2+ risk factors 10-year CHD risk ≤20%

•0 to 1 risk factor

Method to develop treatment thresholds

Putting benefits and harms on the same scale (events saved/in excess per 1,000 people) and find a balance between them

Expert consensus and referring to NCEP ATP-III guideline

Referring to NCEP ATP-III guideline

Referring to ICSI Lipid Management in Adults guideline

Explicitly planned benefit and harm assessment as the basis for making recommendations

Yes

No

No

No

Patient preferences considered for the development of recommendations

Yes

No

No

No


Guideline title

ES

NICE3

MSC3

ADA


    Risk assessment tools

Risk prediction model

Framingham Risk Score, PROCAM and SCORE

UKPDS Risk Engine

UKPDS Risk Engine

Not specified, presumably Framingham Risk Score

Outcome of interest and its timeframe

10-year CHD risk (Framingham and PROCAM) and 10-year total cardiovascular mortality (SCORE)

CHD (10 years) in patients with diabetes

CHD (10 years) in patients with diabetes

CVD (CHD and stroke, 10 years)

Information on validation of the model provided in the guideline

Yes

Yes

No

No

    Evidence of treatment effects

Treatment considered

Aspirin, LDL-C goals and non-HDL-C goals

Simvastatin and statin

Statin and lipid targets

Aspirin

Target population

Patients at high metabolic risk for CVD

People with type 2 diabetes

Non-pregnant adults with type 2 diabetes

Patients with type 1 or type 2 diabetes mellitus

Type of studies considered in the evidence of treatment benefits

Single or several RCTs

Meta-analyses

Single or several RCTs

Meta-analyses

Single or several RCTs

Meta-analyses

Type of studies considered in the evidence of treatment harms

Treatment harms reported but study type unclear

Single or several RCTs

Treatment harms not reported

Treatment harms reported but study type unclear

Heterogeneity of treatment effects assessed in the guideline

No

No

No

Yes

    Application of treatment evidence to baseline risks

Methods to apply treatment evidence to baseline risks

Not reported

Not reported

Not reported

Unclear, presumably used evidence of relative risk reduction from RCT/meta-analysis and applied it to different absolute risks

Assumptions specified when applying treatment evidence

Not reported

Not reported

Not reported

Not reported

    Development of treatment thresholds

Risk stratification in which different treatments were recommended

•Individuals over age 40 and 10-year risk for CHD > 10%

•10-year risk for CHD > 20%

•10-year risk for CHD 10% to 20%

•At least two major risk factors and 10-year risk for CHD < 10%

The cardiovascular risk exceeds 20% over 10 years

•Moderate risk (< 20% 10-year CHD risk)

•High risk (≥20% 10-year CHD risk)

•Adults with type 1 or type 2 diabetes at increased cardiovascular risk (10-year CVD risk > 10%)

•Adults with diabetes and 10-year CVD risk < 5%

•Adults with 10-year CVD risk 5% to 10%

Methods to develop treatment thresholds

Unclear, presumably putting benefits and harms on the same scale and find a balance between them to recommend using aspirin; referring to NCEP ATP-III guideline on LDL-C and non-HDL-C goals

Not reported

Not reported

Unclear, presumably putting benefits and harms on the same scale and find a balance between them

Explicitly planned benefit and harm assessment as the basis for making recommendations

No

No

No

No

Patient preferences considered when developing recommendations

No

Yes

No

No


Guideline title

UMHS2

NSGC

ASCO

NICE4


    Risk assessment tools

Risk prediction model

NCI Breast Cancer Risk Assessment Tool

The guideline mentioned different models

NCI Breast Cancer Risk Assessment Tool

Nottingham Prognostic Index

Outcome of interest and its timeframe

Invasive breast cancer (5 years)

Absolute risk of developing breast cancer or the likelihood of carrying a BRCA1 or BRCA2 mutation (unclear timeframe)

Invasive breast cancer during the next 5-year period and up to age 90 (lifetime risk)

Survival (10 years)

Information on validation of the model provided in the guideline

No

No

Yes

No

    Evidence of treatment effects

Treatment considered

Tamoxifen and raloxifene

Tamoxifen; oral contraceptives; prophylactic mastectomy, prophylactic bilateral salpingo-oophorectomy

Tamoxifen and raloxifene

Aromatase inhibitors

Target population

Adults age 18 and older (non-pregnant)

Individuals at risk for hereditary breast and ovarian cancer

Women at increased risk of breast cancer

Women with breast cancer

Type of studies considered in the evidence of treatment benefits

Treatment benefits reported but study type unclear

Treatment benefits reported but study type unclear

Single or several RCTs

Meta-analyses

Single or several RCTs

Type of studies considered in the evidence of treatment harms

Treatment harms reported but study type unclear

Treatment harms not reported

Single or several RCTs

Meta-analyses

Single or several RCTs

Heterogeneity of treatment effects assessed in the guideline

No

No

Yes

No

    Application of treatment evidence to baseline risks

Methods to apply treatment evidence to baseline risks

Used evidence of relative risk reduction from the same risk profile population for which the recommendation was made

Not reported

Unclear

Unclear

Assumptions when applying treatment evidence

Not reported

Not reported

Not reported

Not reported

    Development of treatment thresholds

Risk stratification in which different treatments were recommended

Women at high risk (5-year risk of invasive cancer ≥1.7%)

The guideline made risk-stratified recommendations, but it is unclear how they defined high risk, moderate risk and low risk

Premenopausal and postmenopausal women with a 5-year projected breast cancer risk ≥1.66% or with lobular carcinoma in situ

•Postmenopausal women with estrogen-receptor-positive early invasive breast cancer not at low risk (those in the Excellent Prognosis Group or Good Prognosis Group in the Nottingham Prognostic Index)

•Postmenopausal women with estrogen-receptor-positive early invasive breast cancer not at low risk and who have been treated with tamoxifen for 2 to 3 years

Method to develop treatment thresholds

Expert consensus

Not reported

Expert consensus

Unclear

Explicitly planned benefit and harm assessment as the basis for making recommendations

No

No

Yes

No

Patient preferences considered when developing recommendations

No

No

No

Yes


CHD: coronary heart disease; CVD: cardiovascular disease; LDL: low-density lipoprotein; LDL-C: low-density lipoprotein cholesterol; MI: myocardial infarction; NCEP ATP-III: National Cholesterol Education Program Adult Treatment Panel III; NCI: National Cancer Institute; NICE: National Institute for Health and Clinical Excellence; HDL-C: high-density lipoprotein cholesterol; PROCAM: Prospective Cardiovascular Münster; RCT: randomized clinical trial; SCORE: Systematic Coronary Risk Evaluation; UKPDS: United Kingdom Prospective Diabetes Study.

Yu et al. BMC Medicine 2013 11:7   doi:10.1186/1741-7015-11-7

Open Data