Impact of fast-track intervention systems on time intervals
| First author, year [ref.] | Study design, intervention and setting | Sample size without intervention/control group (group C) n | Sample size with intervention (group I) n | GP–LCS interval days | Treatment interval days | Other intervals from figure 1 days | New intervals described days | Author conclusions | Statistical significance |
| Brocken, 2011 [34] | Retrospective study comparing delays in a RODP (including PET-CT) for suspected lung cancer patients with delays described in literature and guideline recommendations (the Netherlands, 1999–2009) | 280 | Median (IQR) 7 (5–9) days n=236 | Median (IQR) for group I 19 (6.5–27) days n=215 | Median (IQR) primary care interval 18 (6–46) days; median (IQR) secondary care interval 36 (26–46) days | Median (IQR) interval from LCS to diagnosis (“diagnostic delay”) 2 (1–17.5) days | The RODP including PET-CT resulted in timely care, with strongest effect on diagnostic and secondary care intervals | N/T | |
| Prades, 2011 [35] | Mixed–methods study including prospective data analysing a cancer fast-track programme's impact on reducing the time that elapsed between clinical suspicion of breast, colorectal and lung cancer and treatment start (Spain, 2006–2009) | 3481 (for year 2009) | Mean total interval 36.7 days | Approximately half of all new patients with breast, lung or colorectal cancer were diagnosed via the fast track programme, although the cancer detection rate declined across the period | N/T | ||||
| Murphy, 2015 [36] | Prospective cluster randomised trial assessing use of electronic health record-based trigger algorithms to identify patients at risk of diagnostic delays (USA, 2015) | Unknown (total sample 19) | Unknown (total sample 19) | Median interval from scan to diagnosis 65 days in group I versus 93 days in group C (p=0.59) | No statistical difference was observed in the time to diagnostic evaluation between the intervention and control groups | Nonsignificant | |||
| Leiro-Fernandez, 2014 [37] | Prospective analysis of effectiveness of an email alert system to a pulmonologist attached to a lung cancer rapid diagnostic unit (Spain, 2008–2010) | 47 | Median (IQR) interval from scan to diagnosis 13 (7.3–30) days | This strategy for radiological suspicion of lung cancer improves diagnostic efficacy and the communication between GPs, radiologists and pulmonologists | N/T | ||||
| Iachina, 2017 [38] | Retrospective cohort study evaluating effect of hospital transfers on the delay in diagnosis and treatment using 2009 national fast track cancer care pathways initiative (Denmark, 2008–2012) and data from the Danish Lung Cancer Registry | 11 273 | Mean±sd for group I 16.9±10.64 days | Mean±sd secondary care interval 38.4±15.42 days | Transfer between hospitals during the care pathway might cause delay from diagnosis to treatment as well as in the total time from referral to treatment in patients with NSCLC | N/T | |||
| Alsamarai, 2013 [39] | Retrospective cohort study analysing effect of the CCCP at a Veterans Affairs hospital (USA, 2005–2010) | 163 | 189 | Median (range) 28 (0–265) days; mean 40 days for total sample n=352 Mean for n=163 versus n=189: 46 days versus 43 days (p=0.6) | Mean system interval in group C versus group I 126 days versus 101 days (p=0.015) | Mean interval from scan to diagnosis in group C 76 days versus group I 53 days (p=0.016) | A centralised, hospital-based CCCP can significantly reduce times to diagnosis of cancers that are early stage/incidentally found and reduce system interval by 25 days | Significant reduction for system and scan to diagnosis intervals; not for treatment interval | |
| Cattaneo, 2015 [40] | Report on effect of RACLAP in one medical centre RACLAP includes rapid thoracic nurse consultation, navigation and triage referral system (USA, 2010) | 121 | Median interval from scan to diagnosis 16 days | RACLAP provided rapid and evidence-based evaluation and management of patients resulting in a short time to diagnosis | N/T | ||||
| Murray, 2003 [41] | Multisite prospective randomised pilot study to test feasibility of two-step rapid diagnostic system (Royal Marsden Hospital) compared to conventional diagnostic workup in three local district hospital chest clinics (UK, 1998–2001) | 45 | 43 | Median total interval in group C versus group I 49 days versus 21 days (p=0.0025) | There are several advantages to investigations and diagnosis in the intervention arm, particularly in time to treatment initiation, patient satisfaction and rate of radical treatments | Significant reduction in total interval | |||
| Lo, 2007 [42] | Retrospective cohort study of waiting times pre- and post-implementation of TTT programme: streamlined referral system from GPs to LCS (Canada, 2004–2005) | 52 | 430 | Median for group C 17 days versus group I 4 days | Median times from scan to diagnosis in group C versus group I 39 versus 6; “suspicion” to LCS referral in group C versus group I 20 versus 6; LCS visit to CT in group C versus group I 52 versus 3; “suspicion” to diagnosis in group C versus group I 128 versus 20 | TTT programme was effective in shortening the time from suspicion of lung cancer to diagnosis and reduced time intervals at each step in the process | N/T | ||
| Dransfield, 2006 [43] | Retrospective cohort study of timeliness for patients referred to specialised lung mass clinic (USA, 1999–2003) | 31 (resected), 125 (nonresected) | Median time from LCS to diagnosis in resected patients versus nonresected patients 70 days versus 8 days (p<0.001) Median time from LCS to resection in resected patients 104 days | Since the inception of the lung mass clinic, the resection rate at Birmingham VA Medical Center has improved | Significant reduction only for LCS to diagnosis | ||||
| Laroche, 1995 [44] | Prospective review of a new quick access “two-stop” multidisciplinary investigation service at Papworth Hospital (UK, 1995) | 209 | Median (range) time from LCS to surgical resection 35 (7–81) days | The two-stop investigation service led to higher rates of histological confirmation, routine CT scanning and review of every patient with confirmed lung cancer by a thoracic surgeon. This resulted in a substantial increase in the successful surgical resection rate | N/T | ||||
| Spurgeon, 2000 [45] | Retrospective tracking cohort study assessing impact of TWW system (UK, 1997–1998) | Unknown (total sample 767) | Unknown (total sample 767) | Median (IQR) before and after 12 (7–22) days versus 7 (3–13) days, respectively | Median (IQR) secondary care interval in group C versus group I 47 (28–77) days versus 39 (21–61) days, respectively | Waiting times for urgent appointments were significantly less than the waiting times for nonurgent appointments for all 10 types of cancer | N/T | ||
| Jiwa, 2004 [46] | Retrospective review of impact of urgent (TWW or marked “urgent”) system (UK, 1990) | Unknown (total sample 6) | Unknown (total sample 6) | Mean primary care interval 40 days; mean diagnostic interval 95 days | Mean time from GP referral to diagnosis 55 days | Patients referred as “urgent” were diagnosed soonest | Nonsignificant | ||
| Neal, 2014 [47] | Retrospective cohort study of diagnostic intervals between two cancer cohorts, defined before and after the implementation of the 2005 NICE referral guidelines for suspected cancer and by NICE-qualifying presenting symptoms (UK, 2001–2008) | 1816 | 2851 | Median (IQR) diagnostic interval in group C versus group I 114 (48–238) days versus 112 (45–251) days (p=0.47) | Fast-track referrals may prioritise those with advanced disease in lung cancer, who are more likely to have “red flag” symptoms | Nonsignificant | |||
| Neal, 2007 [48] | Retrospective cohort study comparing outcomes of cancer patients referred through the urgent TWW referral guidance with those who were not (UK, 2000–2001) | 313 | 96 | Median (IQR) for group C 10 (4–17) days versus group I 10 (6–13) days, respectively | Median LCS to diagnosis in group C versus group I 15 (4–28) days versus 18 (8–36) days, respectively | Urgent guideline referrals had later-stage diagnosis compared with patients diagnosed through other routes. There was some evidence for differences in outcomes for lung cancer between urgent guideline referrals (and all referrals marked as urgent) and those diagnosed through other routes | Nonsignificant | ||
| Forrest, 2015 [49] | Retrospective data linkage study investigating factors impacting timely care in the setting of NHS Cancer Plan diagnostic pathways, including the TWW system | Unknown (total sample 28 733) | Unknown (total sample 28 733) | Median (IQR) 10 (6–17) days (n=14 507) | Median (IQR) 35 (21–55) days (n=14 692) | Median (IQR) secondary care interval 56 (39–79) days | Median (IQR) time from GP referral to diagnosis 13 (7–24) days and from LCS to diagnosis 0 (0–0) days | No detail of proportion of urgent referrals, but 70% of patients referred by GP saw a LCS within target interval of 14 days and 61% within secondary care target interval of 62 days | N/T |
| Devbhandari, 2008 [50] | Prospective tracking cohort study of how bronchoscopy results affected waiting times to lung cancer treatment in patients referred by standard (via urgent GP TWW referral) and nonstandard referral pathways (UK, 2003–2005) | 149 | 193 | Median for group C 1 day | Range of medians in group C 8–12 days | Range of median secondary care intervals 45–75 days | Range of median times from LCS to diagnosis 33–57 days | Delays persist despite TWW fast-track system due to hospital barriers Treatment, secondary care and LCS diagnosis intervals significantly longer for bronchoscopy-negative groups | |
| Bowen, 2002 [51] | Prospective pilot study evaluating time between occurrence of symptoms and presentation to GP for patients presenting with lung cancer to two NHS trusts with “rapid access clinics” (UK, 2002) | 37 | Median (range) interval from first GP visit to first LCS visit 56 (0–175+) days | There were delays in assessment and referral in primary care | N/T | ||||
| Hunnibell, 2012[52] | Prospective tracking cohort study to investigate timeliness of lung cancer care before and after creation of a CT-VAHCS nurse navigator position (USA, 2007–2010) | 57 | 66 | Median system interval in group C versus group I 40 and 45 days | Median scan to LCS group C versus group I 13 and 10 days, respectively | CT-VAHCS created and modified several processes to improve timeliness and quality of cancer care as soon as a patient's imaging suggested a new diagnosis of malignancy. The cancer care coordinator effected a measurable improvement in timeliness | N/T | ||
| Lal, 2011 [53] | Retrospective comparative cohort study of patients referred by GPs to lung cancer clinics for investigation of suspicious imaging before and after introduction of fast-track CT pathway (UK, 2006–2007) | 124 | 86 | Median secondary care interval in group C versus group I 55 and 49 days, respectively (p=0.095) | Median referral to decision to treat l in group C versus group I 42 and 35 days, respectively (p<0.05) | Fast-tracking outpatients with suspicious chest radiographs straight to CT results in more effective use of clinic appointments, reduced diagnostic delay and more rapid treatment decision times | Significant reduction only for interval from referral to diagnosis | ||
| Aasebo, 2012 [54] | Retrospective cohort study of workup times for patients with lung cancer using the “Lean” quality improvement process (using mechanisms to identify and sustain high-value encounters and eliminate obstacles) to improve patient flow (Norway, 2006–2009) | 40 | 33 | Median time to surgery/chemo/XRT=26.5/6/5.5 days, respectively Median/mean time to surgery for intervention group 15/17 days (n=14) | Median scan to diagnosis in group C versus group I 64 versus 16 days, respectively Median time from chest radiography to CT in group C versus group I 10 versus 5.5 days, respectively | It is feasible to improve patient flow for patients with lung cancer by employing the Lean method as a pathway instrument | N/T | ||
| Lewis, 2005 [55] | Retrospective comparative cohort study examining the impact of TWW referral pathway for lung cancer over three different time periods, presented here as three separate samples: 1) 1999–2000; 2) 2000–2001; and 3) 2001–2002 (UK, 1999–2002) | Sample (1) n=286 | Sample (2) n=352 Sample (3) n=404 | Median (range): 1) 7 (0–124) days 2) 8 (0–101) days 3) 9 (0–98) days (p=0.0009 for (1) versus (3)) | Median (range) secondary care interval: 1) 37 (2–228) days; 2) 41 (2–307) days; 3) 42 (0–239) days (p<0.04 for (1) versus (2) versus (3)) | Median (range) GP referral to diagnosis: 1) 26 (0–228) days; 2) 33 (2–307) days; 3) 27 (0–300) days (p<0.00001 for (1) versus (2); p=0.0003 for (2) versus (3)) Median (range) LCS to diagnosis: 1) 15 (0–219) days; 2) 21 (0–294) days; 3) 15 (0–300) days (p<0.00001 for (1) versus (2) versus (3)) | The TWW system failed to reduce waiting times for lung cancer in this study due to urgent referral routes used outside the TWW scheme and a large increase (42%) in referrals. Patients referred outside the TWW appear to be disadvantaged | Significant increase in all waiting times | |
| Larsen, 2013 [56] | Retrospective population-based study of changes in secondary care intervals in two hospital groups (Vejle versus other) after 2008 introduction of urgent referral scheme for cancer (Denmark, 2007–2009) | Vejle n=387; other n=3131 | Vejle n=388; other n=2612 | Median (IQR) secondary care interval in group C versus group I for Vejle 31 (20–41) days versus 29 (23–65) days (p=0.39) Median (IQR) secondary care interval in group C versus group I for other 37 (21–64) days versus 33 (16–53) days (p=0.008) | Urgent referral systems had a positive effect on secondary care intervals, although location-specific factors played a role | Significant reduction in secondary care interval | |||
| Riedel, 2006 [57] | Retrospective sequential single-institution (Veterans Affairs) cohort study evaluating the impact of a MTOC (USA 1999–2003) pre- and post-implementation | 101 | 244 | Median before (n=89) versus after (n=205) 23 versus 21 days, respectively (p=0.38) | Median diagnostic interval in group C versus group I 47 (n=89) versus 45 days (n=201), respectively (p=0.12) | Median GP visit to LCS visit interval in group C versus group I 22 (n=90) versus 25 days (n=162), respectively (p=0.01) Median LCS to diagnosis interval in group C versus group I 14 (n= 90) versus 12 days (n=166), respectively (p=0.97) Median LCS to surgery interval in group C versus group I 40 (n=30) versus 50 days (n=56), respectively (p=0.21) | Retrospective comparison with attendant confounders failed to reveal benefit of a MTOC as an intervention for timely lung cancer care | Significant reduction only for interval from first GP to first LCS visit |
Group C: control group: group I: intervention group; GP: general practitioner; LCS: lung cancer specialist; RODP: rapid outpatient diagnostic programme; PET-CT: positron emission tomography-computed tomography; IQR: interquartile range; N/T: not tested; NSCLC: nonsmall cell lung cancer; CCCP: cancer care coordination programme; RACLAP: rapid access chest and lung assessment programme; TTT: time to treat; TWW: 2-week wait; NICE: National Institute for Health and Clinical Excellence; NHS: National Health Service; CT-VAHCS: Connecticut Veterans Affairs Healthcare System; XRT: radiation therapy; MTOC: multidisciplinary thoracic oncology clinic.