Table 2 Overview of data sources used to populate the models

Staging of breast cancer

Sloka et al. 2005 [19]

Data sources

Accuracy data: From literature (MEDLINE, Current Contents, EMBASE; completed in December 2003; details of literature search given); the authors conducted a meta-analysis using 12 published PET studies of staging axillary lymph node metastases; Life expectancy: From literature; life expectancies calculated by authors using the DEALE method; Costs: From literature; Mortality: Mortality estimated by authors; Quality of life: Not relevant.

Dates to which data relate

The PET accuracy data were derived from studies published between 1989 and 2001. The price year was 2000.

Staging of liver metastases from colorectal cancer

Lejeune et al. 2005 [15]

Data sources

Accuracy data: From literature, second liver biopsy sensitivity from expert opinion, PET accuracy data were taken from three published studies; Life expectancy: Burgundy Digestive Cancer Registry, life expectancies calculated by authors using the DEALE method; Costs: The costs of the diagnostic procedures were obtained from 'Nomenclature Générale des Actes Professionnels'. The cost of hospital stay required for liver biopsy and the cost of treatment procedures were based on the national reimbursement rate for the medical department of the Dijon University Hospital. The length of hospitalization for symptomatic treatment was estimated on the basis of the national hospital database on diagnosis-related groups. Other resource quantities were based on expert opinion; Mortality: From literature; Quality of life: Not relevant.

Dates to which data relate

The PET accuracy data were derived from three studies published between 1997 and 2002. The price year was 2004.

Staging of pulmonary metastases from malignant melanoma

Krug et al. 2010 [16]

Data sources

Accuracy data: From literature, PET accuracy data were taken from three published studies; Life expectancy: The life expectancies were calculated by running the model over different time periods and until the entire cohort died; Costs: The unit cost values of hospitalization, drugs, surgery, diagnostic procedures, and the direct health care-related out-of-pocket expenses of patients were obtained from the current public prices published by the RIZIV/INAMI (Health Insurance Institution in Belgium). A micro costing approach was used to calculate the true actual costs of performing a PET/CT study. Resource quantities were based on the resources used by a cohort of patients followed in standardized administrative databases of 19 hospitals between 2005 and 2006 identified by the Diagnosis Related Groups (DRG) codes; Mortality: From literature; Quality of life: Not relevant.

Dates to which data relate

The PET accuracy data were taken from three studies published between 2006 and 2007. The price year was 2009.

Staging of recurrent nasopharyngeal carcinoma (NPC)

Yen et al. 2009 [17]

Data sources

Accuracy data: From literature, probability of an uncertain MRI result estimated by authors, PET accuracy data were taken from four published studies; Life expectancy: From literature and from life tables from the Department of Health, Executive Yuan, Taiwan, 2002; Life expectancies calculated by authors using the DEALE method; Costs: Based on reimbursements from National Health Insurance, Taiwan; Mortality: From literature; Quality of life: Utilities elicited from 10 recurrent NPC patients and 10 oncologists involved in the management of these patients.

Dates to which data relate

The PET accuracy data were taken from four studies published between 2002 and 2004. The price year was not reported.

Staging of head and neck cancer

Sher et al. 2010 [18]

Data sources

Accuracy data: From literature, the PET accuracy data were taken from a meta-analysis of the role of PET in the follow-up of head and neck squamous cell carcinoma following radiotherapy or chemoradiotherapy; Life expectancy: The (quality-adjusted) life expectancy was calculated by running the model; Costs: Costs were taken from the published literature and publicly available Medicare payment schedules; Mortality: From literature; Quality of life: QALYs were calculated using utilities that were taken from literature or based on authors' assumptions.

Dates to which data relate

The meta-analysis of PET was published in 2008. The price year was 2006.

Staging of recurrent ovarian cancer

Mansueto et al. 2009 [14]

Data sources

Accuracy data/Life expectancy/Mortality: From Mangili et al. [36]; Costs: The costs were based on DRG tariffs; Quality of life: Not relevant.

Dates to which data relate

The clinical study by Mangili et al. was published in 2007.The price year was 2006.

Staging of non-small cell lung cancer (NSCLC)

Alzahouri et al. 2005 [21]

Data sources

Accuracy data: From literature, PET accuracy data were taken from the cost-effectiveness analysis of Dietlein et al. [31] and from a meta-analysis by Gould et al. [91]; Life expectancy: From the cost-effectiveness analysis by Dietlein et al.; Costs: PET and chemoradiation from literature, CT and cervical mediastinoscopy from 'Nomenclature Générale des Actes Professionnels', surgery from 'Echelle Nationale des Coûts'; Mortality: From literature; Quality of life: Not relevant.

Dates to which data relate

The PET accuracy data were derived from two studies published between 2000 and 2003. The price year was not reported.

Bird et al. 2007 [24]

Data sources

Accuracy data: From literature, PET accuracy data were taken from the HTBS [29] and from the NCCAC model [26]; Life expectancy: Life expectancies were taken from the cost-effectiveness analysis by Dietlein et al. [31]; Costs: From literature and from the Australian National Hospital Cost Data Collection; the chemotherapy costs were estimated using the HTBS model, the Rosenthal study [92], and Australian data; Mortality: Mortality rates were taken from the cost-effectiveness analysis by Dietlein et al.; Quality of life: QALYs were based upon the values used in the HTBS model.

Dates to which data relate

The PET accuracy data were derived from two HTA reports published between 2002 and 2005. The price year was 2006.

Kee et al. 2010 [25]

Data sources

Accuracy data/Life expectancy/Costs/Mortality: From the HTBS model [29]; Quality of life: Utilities elicited from 75 NSCLC patients.

Dates to which data relate

The PET accuracy data were derived from a HTA report published in 2002. The price year was not reported.

Mansueto et al. 2007 [23]

Data sources

Accuracy data: From a meta-analysis by Gould et al. [91]; Life expectancy: From the cost-effectiveness analysis by Dietlein et al. [31]; Costs: From values of diagnosis-related groups and tariffs of the regional health care system; Mortality: From literature; Quality of life: Not relevant.

Dates to which data relate

The accuracy data were taken from a study published in 2003. The price year was 2005/6.

National Collaborating Centre for Acute Care

2005 [26]

Data sources

Accuracy data: From the HTBS model [29]; Life expectancy: From the cost-effectiveness analysis by Dietlein et al. [31]; Costs: From the NHS reference costs 2002, from literature, or from individual NHS trusts; Mortality: From the cost-effectiveness analysis by Dietlein et al.; Quality of life: From the HTBS model.

Dates to which data relate

The PET accuracy data were taken from the HTBS model published in 2002. The price year was 2002/3.

Nguyen et al. 2005 [22]

Data sources

Accuracy data: From Pieterman et al. [93]; Life expectancy: From literature; Costs: The costs of PET were derived from the costs of PET scanning at the Centre Hospitalier Universitaire de Sherbrooke, Quebec. The costs of hospital stay were determined by the Ministère de la Santé et des Services Sociaux's 1998-1999 diagnosis-related groups database. The figures for physicians' fees were taken from the Medical Specialists' Manual; Mortality: From literature; Quality of life: Not relevant.

Dates to which data relate

The PET accuracy data were derived from a study published in 2000. The price year was 1998/99.

Follow-up of non-small cell lung cancer (NSCLC)

Van Loon et al. 2010 [20]

Data sources

Accuracy data: Data on the ability of follow-up strategies to detect progression were taken from a previous prospective study [94]; Life expectancy: The (quality-adjusted) life expectancy was calculated by running the model; Costs: The costs of diagnostic procedures and treatment were taken from the Dutch Health Insurance Board. Other costs were based on the guideline costing statement provided by the UK National Institute for Health and Clinical Excellence (NICE) and a published randomized controlled trial; Mortality: From literature, supplemented by expert opinion; Quality of life: Utility values were taken from a published cross-sectional study, supplemented by expert opinion.

Dates to which data relate

The prospective study was published in 2009. The price year was 2007.

Diagnosis of a solitary pulmonary nodule (SPN)

Lejeune et al. 2005 [13]

Data sources

Accuracy data: From literature, the PET accuracy data were taken from five published studies; Life expectancy: From literature; Costs: The costs of radiography and CT were obtained from the 'Nomenclature Générale des Actes Professionnels'. The cost of PET was derived from two French studies. The costs of treatment procedures were derived from diagnosis-related groups for 2003 in the public health care sector. The cost of the patient's hospital stay was derived from the French national public cost scale, which was derived from a representative sample of French hospitals; Mortality: From literature; Quality of life: Not relevant.

Dates to which data relate

The PET accuracy data were derived from different studies published between 1990 and 1998. The price year was not reported.