Table 1.

General characteristics of articles included in systematic review

Author/yearType of cancerCountryArticle typeAim — text taken from paperStudy designSettingPopulation characteristics
Banks et al, 201433Lung, pancreatic, colorectalEnglandResearchTo establish the likelihood that individuals would choose to be tested for cancer at various levels of riskVignette-based study26 general practices across England3469 individuals (≥40 years)
Birt et al, 201434LungEnglandResearchTo understand the symptom evaluation, or ‘appraisal’, and help-seeking decisions of patients with symptoms suggestive of lung cancerProspective cohort face-to-face interview studySpecialist respiratory clinics in secondary and tertiary care, England35 patients (male = 20, female = 15; mean age 58.5 years; range 41–88 years; 17 with lung cancer, 18 with other conditions)
Black et al, 201535Colorectal, upper GI, lung, haematological, head and neck, gynaecological, unknown primaryEnglandResearchTo capture in detail the experiences of patients whose cancer was diagnosed following an ED visit to understand how emergency presentations arise, and identify where there is scope to improve outcomesInterview11 EDs across nine acute NHS trusts (participants came from seven NHS trusts)27 patients (male = 15, female = 12; median age 59 years; range 18–92 years)
Evans et al, 200636OvarianUKResearch1) To use accounts of provider delay from a sample of British women with ovarian cancer to enhance Andersen’s model of total patient delay; 2) To suggest what GPs might do to minimise delaysSemi-structured interviews with a maximum variation sampleRecruitment through GPs, clinicians, support organisations, and personal contacts43 women diagnosed with ovarian cancer (mean age 54 years; range 33–80 years)
Jensen et al, 201437Bladder, breast, colon, lung, malignant melanoma, others, ovarian, prostate, rectal, uterusDenmarkResearch1) To describe the prevalence and types of QDs that arose during the diagnostic pathway for Danish patients with cancer as assessed by the GPs; 2) To analyse the associations between QDs and the type of cancer; between QDs and the GP interpretation of presenting symptoms; 3) Impact of QDs on diagnostic interval lengthNationwide retrospective cohort questionnaire-based study1446 GPs identified through Danish national patient registry4036 newly diagnosed cancer patients (male = 2176, female = 1860)
Lim et al, 201438CervicalEnglandResearchTo collect data on nature and duration of symptoms and risk factors for delay in presentation and diagnosis in young females with cervical cancer. To inform approaches to promote early presentation and prompt referral among young females with symptoms of cervical cancerSemi-structured interview-based measureEligible participants identified by 116 hospitals and publicising of study through charity websites128 women (18–29 years) with recent diagnosis of cervical cancer
Lyratzopoulos et al, 201529Not specifiedN/APerspective1) To highlight factors involved in missed opportunities for cancer diagnosis among symptomatic patients; 2) Discuss potential mechanisms and approaches to accelerating progress towards minimising diagnostic delays post-presentationN/AAll settings involved in diagnostic process including general practice and primary careN/A
Mitchell et al, 201226Lung, ovarian, upper GIEnglandToolkitDeveloped as a resource for use when considering ways to improve the diagnosis of cancer in primary care1) National audit of cancer diagnosis in primary care;
2) Analysis of SEA;
3) Action plans for early cancer diagnosis derived using cancer-related practice profiles		Primary care1) 1170 practices across 20 cancer networks;
2) SEA reports for 132 lung, 68 ovarian, and 78 upper GI cancer diagnoses;
3) general practices across England
Mitchell et al, 20156Brain, breast, carcinoid, CNS, colorectal, gynaecological, haematological, head and neck, lung, melanoma, sarcoma, upper GI, urological, unknown primaryEnglandResearch1) Understanding the cause of emergency presentations; 2) Determining the degree to which earlier intervention by general practice was possibleQualitative synthesis of SEAs203 primary care practices across North of England Cancer Network222 SEAs for most recent patients diagnosed with cancer as a result of an emergency presentation
Molassiotis et al, 201039Breast, GI, gynaecological, lung, brain, lymphoma, head and neckUKResearch1) To map the pathway from initial persistent change in health to diagnosis of cancer in a sample of patients from seven diagnostic groups in the UK; 2) Explore the patient and system factors mediating this processIn-depth qualitative interviewsOutpatient clinics at one large cancer hospital74 patients (mean age 58.5 years; range 18–93 years)
National Reporting and Learning Service, 201040All cancersUKThematic review1) To explore issues of patient safety around delayed diagnosis of cancer; 2) Provide the NHS with potential solutions1) Thematic review;
2) Review of incidents reported to the NPSA’s NRLS;
3) Consultation with stakeholders (focus groups, discussions, presentations);
4) Consideration of other patient safety data, such as complaints, litigation, and audits		Primary and secondary care1) N/A; 2) 508 patient safety incidents from random sample of 1500; 3) Two focus groups (n= 50); 4) NHS Litigation Authority; MDU Report (2003); MPS (analysis of 1000 cases of delayed or missed diagnosis); Scottish Primary Care Cancer Group reports of 4181 cancer cases
Round et al, 201330Not specifiedUKEssayTo highlight clinician- and system-related factors that may be contributing to a delayed diagnosis of cancer in primary careN/APrimary careN/A
Rubin et al, 201531Not specifiedHigh-income countries/countries with universal healthcare systemsCommissionTo distil the evidence for the effectiveness of interventions for cancer control based in primary care at each stage of the cancer journey, and to consider how cancer care might be delivered differently in the future. Discuss how and whether health policy for cancer control will help or hinder such change. Examines the implications for the future education and training of doctors, and identifies emerging examples of good practice worldwideN/APrimary and specialist careN/A
Siminoff et al, 201141ColorectalUSResearch1) To describe the communication and symptom appraisal strategies of CRC patients;
2) Determine the patient and physician characteristics, consultation communication factors, follow-up recommendations, and the referral delay factors associated with DD		Semi-structured interview and review of medical records for verification of patient-provided informationFive academic and community oncology practices in two states242 patients of any age (mean = 58.07 years) or sex (male = 126, female = 116)
Stevens et al, 201242LungNew ZealandReportTo develop recommendations to improve the clinical pathway from presentation to diagnosis for people with suspected lung cancerMixed methods utilising both quantitative and qualitative data:
1) Literature review;
2) National Stocktake;
3) Clinical audit;
4) Practice survey;
5) GP survey;
6) GP focus group;
7) Patient interviews;
8) Patient and family focus groups		Four primary care organisations and three DHBs within the Northern and Midland Cancer Networks1) N/A; 2) Primary or secondary care services that could be relevant to people with lung cancer; 3) Clinical pathway from initial presentation to diagnosis (272 cases); 4) All general practices across two regions (n= 170); 5) 102 GPs across two regions; 6) 33 GPs across two regions; 7) 19 people suspected of having lung cancer who presented to a hospital ED; 8) 20 people (11 patients with lung cancer diagnosis [male = 5; female = 6] and nine family members)
Walter et al, 201432CervicalUK and internationalEditorialNot specifiedN/APrimary careN/A

  • CNS = central nervous system. CRC = colorectal cancer. DD = diagnostic delay. DHB = district health board. ED = emergency department. GI = gastrointestinal. MDU = Medical Defence Union. MPS = Medical Protection Society. N/A = not applicable. NPSA = National Patient Safety Agency. NRLS = National Reporting and Learning System. QD = quality deviation. SEA = significant event audit.