Applying the healthcare failure mode and effects analysis approach to improve the quality of an organised colorectal cancer screening programme

Abstract

Objective

The first level of a colorectal cancer (CRC) screening process was systematically analysed using the Healthcare Failure Mode and Effects Analysis (HFMEA) approach by a multidisciplinary team aiming to improve the programme quality.

Setting

The study was conducted at the Local Health Authority of Bologna, Northern Italy.

Methods

Seven brainstorming sessions were conducted and all the activities performed were recorded on a FMEA worksheet consisting of individual records reporting the specific phases of the analysed process along with associated activities, possible failure modes, their causes and effects, the obtained risk priority numbers (RPNs) and the control measures to plan.

Results

Twenty-three failure modes, 14 effects and 12 possible causes were identified. Nine failure modes were prioritised according to the RPN obtained; most resulted in possible false-negative faecal immunochemical test (FIT) results (66.7%), followed by sample loss (22.2%) and not reaching the entire target population (11.1%). This leads to 66.7% of corrective/preventive actions being applied to the phase of returning the stool sample by the citizen. For this phase reorganisation, the local pharmacies were involved not only as FIT kit delivery points but also as specimen collection and sending points to the laboratory. These organisational changes allowed the introduction of complete traceability of kits and specimens flow, as well as temperature control. A re-evaluation of the prioritised failure modes 6 months after launching the implemented screening process showed that HFMEA application decreased the risk of potential errors by 75.9%.

Conclusion

HFMEA application in CRC screening programme is a useful tool to reduce potential errors.

Keywords

Organised colorectal cancer screening HFMEA process quality faecal immunochemical test local pharmacies

Introduction

According to the most recent Global Cancer Observatory data, colorectal cancer (CRC) ranks third for incidence and second for mortality.¹ It is known that organised screening programmes are able to reduce the CRC burden in the involved population,² as it is a highly preventable cancer that can be reliably detected at an early stage, when treatment is significantly associated with better clinical outcomes.^3,4 A screening programme is effective depending on the quality of each subprocess (phase), i.e. from the identification of the population to invite up to the eventual treatment and follow-up.⁵ Given the complexity of the entire screening process, there is a high likelihood of adverse events occurring.⁶

To date, the Local Health Authority (LHA) of Bologna (the capital of the Emilia-Romagna Region, Northern Italy) offers to males and females at average risk for CRC, that is aged 50 to 69 years, a biennial faecal immunochemical test (FIT); approximately 120,000 citizens are invited to undergo screening and 60,000 FITs are performed annually. Following one sentinel event that occurred in 2018 and two near misses in early 2019, the Screening Centre decided to systematically analyse the first level of the screening process using the Healthcare Failure Mode and Effects Analysis (HFMEA) approach^6,7 with the aim of improving the quality of the programme. This paper describes the rationale for the definition of the HFMEA topic and outlines all the key aspects of the analysis from team assembly to the impact of the planned control measures.

Methods

The HFMEA was rolled out during the period from 01 March 2019 to 30 April 2020 following the methodology introduced by DeRosier and colleagues in 2001, when they developed the FMEA model for proactive risk assessment within the healthcare community. The model consists of five steps,⁷ which are described below. For the activities falling under step 4, the hazard score matrix adopted was the one used by Deandrea et al. in a recent quality improvement project performed in the Milan CRC screening programme.⁶

Step 1. Definition of the HFMEA topic

The experienced sentinel event consisted of the failure in communication of a positive FIT result due to misidentification of the citizen who performed the test; this was attributable to the screening organisation that, for citizen identification, involved the use of labels with finite number combinations restarting approximately within four screening rounds (i.e. 8 years). The two near misses consisted of a sample loss and a citizen misidentification. The first near miss was intercepted as a result of the citizen's warning of non-receipt of the test result 3 months after the test had been performed. Of note, the warning occurred close to the summer period, when the first step of the CRC screening activity was usually interrupted due to high environmental temperatures, as it has been shown that temperature during transport and storage impacts the stability of haemoglobin in stool samples.⁸ Consequently, the citizen was offered to repeat the test approximately 6 months after the first FIT specimen collection. The second near miss was intercepted following a warning by laboratory staff of no matching between the ID on the barcode label and the citizen's anagraphic data.

The HFMEA focused on the phases of the process from the invitation of the target population to the communication of the FIT result to the citizen.

Step 2. Assembling of team

The team was assembled as per indications of the methodology.⁷ Specifically, the team's size adhered to the recommended range of 6–10 individuals, the composition respected the multidisciplinary approach for ensuring that various viewpoints are considered, and subject-matter experts were included. In particular, the team comprised the following members: (a) two experts in the screening field working daily for the CRC screening programme (i.e. the director of the programme and the coordinator of the Screening Centre), (b) two experts in the laboratory activities of performing daily FIT; (c) an expert in the informatics systems; and (d) a member of the Operative Unit of the Forensic Medicine and Clinical Risk Management, who was trained in HFMEA and served as team facilitator and leader.

Steps 3 and 4. Graphical description of the process and conduction of the hazard analysis

During seven brainstorming meetings, the team graphically described the chosen process by breaking it down into phases, performed the analysis of the failure modes for each phase and calculated the risk priority number (RPN) for each failure mode. The RPN was calculated by multiplying three factors such as the severity of the failure effect (S), the probability of occurrence of the failure mode (O) and the possibility of its detection and interception before the occurrence (D). In accordance with DeRosier et al.,⁷ failure modes were operationally defined as ‘the different ways that a particular process or subprocess step can fail to accomplish its intended purpose’.⁷ According to Deandrea et al.,⁶ to compute the RPN, numerical scores from 1 to 4 were associated to the severity and detection factors, i.e. meaning from ‘minor event’ to ‘catastrophic event’ and from ‘certain’ to ‘remote’, respectively. Numerical scores from 1 to 6 were associated to the occurrence factor, i.e. meaning from ‘remote’ to ‘very high’. The numerical scores were attributed to the three factors using the Centre Screening's informatics system or the experience of the multidisciplinary team members. In the latter case, any disagreements among members were solved by discussion to reach consensus.

Step 5. Identification of actions and outcome measures

Similarly to Deandrea et al., corrective/preventive actions were planned for the failure modes with RPN higher than half of the highest RPN obtained.⁶ The implementation of the identified actions lead to a reorganisation of the process, which was launched in July 2020. After 6 months, the prioritised failure modes were revaluated.

During the brainstorming sessions, all the activities were recorded on a FMEA worksheet consisting of individual records reporting the specific phase of the analysed process along with the associated activities, possible failure modes, their causes and effects, the obtained RPN (S × O × D) and the control measures to plan.

The study did not require approval by the institutional review board as no human subjects were involved in the project.

Results

The flow-chart of the entire process analysed by the multidisciplinary team, with reference to the failure modes identified for each phase, is shown in Figure 1.

Figure 1.

Flow-chart of the analysed process with reference to the failure mode(s) (FM) identified for each phase. LHA: Local health authority; EHR: Electronic health record.

The analysis identified a total of 23 failure modes, 14 effects and 12 possible causes. The RPNs obtained ranged from 4 to 60 (mean value, 21; median value, 18). All the data are reported in Table 1. A total of nine failure modes had a RPN > 30 and as per protocol were prioritised. In 66.7% of these cases (n = 6), the failure modes resulted in a possible false-negative FIT result (i.e. #6, #13, #14, #15, #18, #19), followed by sample loss (22.2%; n = 2), and not reaching the entire target population (11.1%; n = 1) (i.e. #16 and #17, and #2, respectively). As a result, 66.7% of the corrective/preventive actions (6/9 failure modes) were to be applied to the phase of return of stool specimen by the citizen. In particular, the reorganisation of this phase consisted of involving the local pharmacies not only as FIT kit delivery points, but also as the points of stool sample collection and sending to the laboratory.

Table 1.

Failure mode and effect analysis worksheet results.

Phase	Failure mode #Number	Effects	Possible causes	S	O	D	RPN
Identification of the target population	#1: Missing or incomplete data on medical history leading to temporary or permanent exclusion from the programme	Misidentification of the target population	No information on tests performed outside the organised screening programme and lack of cancer registry	1	1	4	4
Sending via mail the invitation letter with a barcode label identifying the invited citizen	#2: Non-receipt of the invitation letter	Non-reach of entire target population	Postal dispersion	3	5	4	60
	#3: Non-use of a unique barcode	Misidentification of the citizen who is performing FIT	The laboratory informatics system recognises a finite number of barcode labels (i.e. 100,000) and the laboratory identifies the anagraphic data for which the barcode label is most recently generated The numbering restarts approximately within 8 years (i.e. 4 screening rounds); therefore, the presence of two identical valid barcode labels in the invited population is a possible event if the citizen participates in the screening programme with a latency of 8 years from receipt of the invitation	4	1	4	16
	#4: Non-use of the last invitation letter, but one of the previous ones	Misidentification of the citizen who is performing FIT	The text of the letter does not make it sufficiently clear that only the latest invitation is the correct one to use	4	1	4	16
Delivery of the FIT kit at a local pharmacy	#5: The citizen loses/damages the invitation letter with his/her own barcode label	Impossibility to provide the kit to the citizen	Citizen disregard	3	2	4	24
	#6: The pharmacist provides the citizen with an expired sample tube	Possible false-negative FIT result	Non-check by the pharmacist of the kit's expiry date	4	2	4	32
	#7: The pharmacist misapplies barcode label to the sample tube	Unreadable barcode label by the laboratory reading device	Non-check by the pharmacist or pharmacist forgetfulness	1	3	3	9
	#8: The pharmacist forgets to apply the barcode label to the sample tube	Unlabelled (anonymous) sample with the laboratory being unable to investigate the sample	Pharmacist forgetfulness	4	2	3	24
	#9: The pharmacist does not collect the informed consent from the citizen	Specimen without informed consent	Pharmacist forgetfulness	1	2	3	6
	#10: The citizen does not fill in the informed consent correctly	Incomplete or wrong information associated to the citizen	Citizen disregard	1	2	2	4
	#11: The compiled informed consent does not arrive at the Screening Centre	Sample investigation without informed consent	Postal dispersion or pharmacist forgetfulness	1	2	4	8
	#12: Missing or incorrect manual citizen data entry into the screening's informatics system	Incomplete or wrong information associated to the citizen	Operator manual error	1	2	2	4
Returning the stool specimen at the sample collection points	#13: The timeframe from FIT kit pick-up to sample collection leads to expiration of the tube	Possible false-negative FIT result	Non-careful reading of the instructions by the citizen	4	2	4	32
	#14: The returned specimens are not properly stored before analysis, in terms of temperature and elapsed time between sample collection and returning (i.e. at room temperature for > 7 days)	Possible false-negative FIT result	Non-careful reading of the instructions by the citizen	4	2	4	32
	#15: The citizen returns the FIT kit on an unscheduled day	Delay in sending the stool sample to the laboratory with possible false-negative FIT result	Non-careful reading of the instructions by the citizen	4	2	4	32
	#16: The citizen does not properly identify the container set up for the specimen collection	Sample loss	Non-careful reading of the instructions by the citizen	4	2	4	32
	#17: The container is not guarded	Sample loss	Organisational issues	4	2	4	32
	#18: The container is not refrigerated	Possible false-negative FIT results in case of delay in sample sending	Organisational issues	4	2	4	32
Sending the stool specimen to the laboratory	#19: Inappropriate transportation (time, temperature)	Possible false-negative FIT result	Organisational issues	4	2	4	32
Sending the stool specimen to the laboratory	#20: The specimen does not arrive at the laboratory	Sample loss with no possibilities to attribution to a specific citizen	Organisational issues	4	1	4	16
Sample investigation by the laboratory	#21: The stool sample is not suitable for investigation	The specimen is not analysed by the laboratory	Insufficient/too much sample quantity	1	2	3	6
Visualisation of FIT result by Screening Centre	#22: Unreachable citizen with an inadequate FIT result	Non-communication of the FIT result to citizen needing FIT repetition	Postal dispersion and non-use of the EHR by the citizen	3	2	3	18
Visualisation of FIT result by Screening Centre	#23: Unreachable citizen with a positive FIT result	Non-communication of the FIT result to citizen needing further investigation	Postal dispersion, non-response by telephone and non-use of the EHR by the citizen	4	1	3	12

S: severity; O: occurrence; D: detection; RPN: risk priority number; FIT: faecal immunochemical test; EHR: electronic health record.

At various stages of the reorganisation process, several meetings were conducted with the four associations representing the local pharmacies that would be involved in the screening programme. Initially, these meetings focused on evaluating the feasibility and acceptability of the reorganisation plan by the pharmacies. Subsequently, the meetings aimed to provide pharmacists with the necessary knowledge to ensure the correct operation of the new phases of the screening programme. Additionally, the meetings served as an opportunity to encourage better awareness of the pharmacists for participation in the kit delivery phase.

The corrective/preventive actions identified for each phase, with the re-evaluation of the RPNs after their application, are reported in Table 2. It can be observed that efforts were made to improve communication with the citizens. This included introducing the use of the citizen's electronic health record for sending both invitation and reminder letters for attending the screening. Moreover, a mobile short message service was planned to be introduced and modifications to the invitation letter text were considered. Of note, all the re-evaluated RPNs were reduced compared to their initial values, with a percentage reduction ranging from 40% to 100% (n = 5 failure modes were eliminated, with RPNs = 0).

Table 2.

Prioritised failure modes: implementation of corrective/preventive actions and recalculation of the risk priority number.

Phase	Failure mode #Number	RPN initial state	Corrective/preventive actions	RPN re-evaluation	RPN reduction (%)
Sending via mail the invitation letter with a barcode label identifying the invited citizen	#2: Non-receipt of the invitation letter	60	Sending the invitation letter also on the citizen's EHR. After 1 month from the invitation letter, sending text message reminder for attending CRC screening to citizen's mobile telephone After 6 months from the invitation letter, sending a letter reminder on the citizen's EHR or via mail for attending CRC screening	36	40
Delivery of the FIT kit at a local pharmacy	#6: The pharmacist provides the citizen with an expired sample tube	32	Increasing the pharmacist's awareness in checking the FIT kit's expiry date at the moment of delivery to the citizen Adding a statement on the invitation letter with the indication for the citizen to check the FIT kit's expire date and not using an expired kit but picking-up another one at a local pharmacy. This statement is highlighted both on the invitation letter's main text and on the stool sample collection instruction	12	62.5
Returning the stool specimen at the sample collection points	#13: The timeframe from FIT kit pick-up to sample collection leads to expiration of the tube	32	In the invitation letter, highlighting the statement indicating that the stool sample has to be collected as soon as possible after the FIT kit pick-up Two months after FIT kit pick-up, sending text message reminder to citizen's mobile telephone for collecting the stool sample and returning it to a local pharmacy	12	62.5
	#14: The returned specimens are not properly stored before analysis, in terms of temperature and elapsed time between sample collection and returning (i.e. at room temperature for > 7 days)	32	Highlighting in the invitation letter the statements indicating the correct timing and method of stool storing Increasing the pharmacist's awareness to check the correctness of the citizen's actions	16	50
	#15: The citizen returns the FIT kit on an unscheduled day	32	Replacing the sample collection points set up in different districts of the LHA with the local pharmacies where the citizen can return the kit anytime during the opening hours of the pharmacy, even in summer	0	100
	#16: The citizen does not properly identify the container set up for the specimen collection	32		0	100
	#17: The container is not guarded	32		0	100
	#18: The container is not refrigerated	32	The stool sample is stored in pharmacy at a temperature of +4°C until sent to laboratory. The samples transportation to the laboratory is also refrigerated	0	100
Sending the stool specimen to the laboratory	#19: Inappropriate transportation (time, temperature)	32	Implementation of an informatics system to track the entire system from FIT kit delivery to the citizen to the reporting of results by the laboratory. Using a temperature logger placed in the sample transportation bag	0	100
Total		316		76	75.9

RPN: risk priority number; EHR: electronic health record; CRC: colorectal cancer; FIT: faecal immunochemical test.

In addition to the aforementioned prioritised failure modes, five failure modes (#3, #4, #8, #20, #23, with RPNs ranging from 12 to 24) with remote/very low probability of occurrence but with effects defined as ‘catastrophic’ were identified (Table 1) and organisational changes were applied to eliminate/reduce them. Of note, the sentinel event and the two near misses leading to the decision to roll out HFMEA were due to failure modes identified within this group (i.e. #3, #4, #20). Four (80%) of these failure modes (i.e. #3, #4, #8, #20) were eliminated by introducing in the screening programme the use of pre-labelled barcode tubes, with an infinite number combination, that are attributed by the pharmacist to the specific citizen by means of a barcode scanner on the occasion of FIT kit delivery. To track this step of the process and all the following steps until the visualisation of the test results at the Screening Centre, an informatics tool was created. In particular, an online portal was designed for registering at the local pharmacy both the FIT kit delivery and the stool sample collection for each citizen participating in the screening. The online portal is integrated with the screening programme software, which was implemented for checking the kits and the specimen flow. In particular, a series of pre-set queries allows for the identification of any failures in kit delivery, stool sample return and arrival at the laboratory, as well as too long time spent in investigating stool samples (i.e. >14 days between sampling and processing), leading to the repetition of the test by the citizen. The probability of occurrence of the remaining failure mode (i.e. #23) was minimised by requiring pharmacists, at FIT kit delivery and stool sample return, to check the telephone number visible on the online portal that is associated with the citizen and, if necessary, modify it or add a further telephone number. At the same time, the pharmacist reminds the citizen that in the case of a positive FIT result, there will be a telephone communication by the Screening Centre's health staff. Lastly, if a citizen cannot be reached, their General Practitioner is alerted on the FIT result by the Screening Centre's health staff.

Before introducing the new path, a cost-benefit analysis was performed. The additional costs due to the personnel assigned to monitor the start-up phase (4800 Euros for 3 hours of work/day for 2 months) and for the ongoing process control (6240 Euros for 3 hours of work/week for 12 months), the implementation of the new IT procedures (10,000 Euros), as well as the cost of the local pharmacies’ service (3.66 Euros/sample) were taking into account. On the other hand, the cost of outpatient procedures for treatment of pre-cancerous lesions identified by the screening (126 Euros/procedure)⁹ and the average cost per patient with CRC in the initial and final phases of care (18,100 Euros/year)¹⁰ were evaluated in the context of an increasing screening attendance.

Discussion

To date, several studies reporting the application of FMEA in various areas such as healthcare process,^11,12 hospital management,^13,14 hospital informatisation^15,16 and medical equipment/production^17,18 have been published. However, to the best of our knowledge, only one paper discussed the use of the FMEA technique in a population-based screening programme for CRC.⁶

Here, we have described the application of HFMEA, which resulted in an improvement in the organisation of the CRC screening programme, with complete traceability of the flow of kits and stool samples, as well as temperature control. The membership of the multidisciplinary team that performed the HFMEA, the map of the analysed process with the identified potential failure modes and their causes and effects, the control measures applied, and the impact of each intervention are here reported.

In our setting, by focusing on the prioritised failure modes, the application of HFMEA decreased the risk of potential errors by 75.9%. The reorganisation of the process mainly consisted of changes to the return of stool samples phase up to their transportation to the laboratory. Of note, the corrective/preventive actions applied to these two phases had the greatest impact, with the elimination of more than half (55.5%) of the prioritised failure modes identified. Moreover, unlike before, storing stool samples in the pharmacy at a temperature of +4°C until they are sent to the laboratory, along with refrigerated sample transportation, allows the CRC screening activity to continue even during the summer. Similarly, the organisational changes resulted in informatics control over the process steps that were previously lacking and that allows to identify the occurrence of any failures, with the possibility of identifying the involved citizen who can then be contacted by the LHA Screening Centre's health staff to repeat the test. Besides the decision to include local pharmacies in the organisational model, aiming to eliminate or reduce the vulnerabilities identified in the analysed process, we also assumed that since pharmacists are accessible and trusted members of the healthcare team and routinely encounter citizens at their local pharmacies,¹⁹ a more extended role of pharmacies in the screening path could have facilitated citizen attendance.

A lower percentage reduction in the RPN (i.e. 40%) was observed for the failure mode consisting of the non-receipt of the invitation letter via mail by the citizen. Since there are no possible corrective actions to increase the possibility of postal dispersion being detected and intercepted before it occurs, besides the introduction of the use of further communication tools such as the citizen's electronic health record and short messaging service, an internal control system is applied in our centre allowing identification of undelivered letters, leading in fact to the monitoring of this failure mode. Finally, despite the previous misidentification of a citizen performing FIT not being an effect of a prioritised failure mode, on the basis of this event's high severity, the multidisciplinary team proposed and implemented organisational changes that led to its elimination.

Of note, the high degree of automation of the laboratory processes, which is known to improve performance through minimisation of human error,²⁰ resulted in a unique failure mode detected in the sample investigation and report of test results phase; specifically, the failure mode was related to specimen suitability.

Both the new organisation and informatics tools were welcomed by the Screening Centre's staff. However, implications for the sustainability of the process in terms of work and costs are to be noted. In particular, an intensive time investment for monitoring the process is required by the personnel and the organisational changes in the stool samples return phase result in additional costs related to the involvement of the local pharmacies of approximately 215,000 Euros/year. Nevertheless, we believe that the increase in costs is largely offset by an increase in the safety and quality of the process as well as in attendance to the organised screening. When comparing local data recorded in 2019 (i.e. the base year under the previous organisational model) with the latest data in 2022, an increase in screening attendance of 4.6 percentage points (i.e. equivalent to 2729 more FITs performed, for a total of 60,031 FITs) was observed. Considering that this CRC screening programme identifies 1 pre-cancerous lesion/130 screened individuals,²¹ the new path allowed for the identification of 21 more pre-cancerous lesions that were treated by outpatient procedures, preventing possible development of CRC. By comparing the value of the emerging costs/year equal to 238,199 Euros with the mean cost per patient/year in the initial and final phases of CRC care (i.e. 380,100 Euros), a potential saving of approximately 141,900 Euros for cancer cost may be estimated. Of note, the yearly average cost per patient in the continuing phase of CRC care (864 Euros/year)¹⁰ was not considered in this analysis, as well as the indirect costs (reduced or lost productivity due to morbidity or mortality) of disease.²² Not least, profound physical and emotional side-effects are recognised among the CRC survivors, impacting their overall quality of life.²³

The need for payment to the pharmacies is expected only for the stool return phase. We believe that this organisational decision could enhance the participation of pharmacists in encouraging citizens to return their samples. Moreover, health counselling is known to be essential in promoting citizen participation in screening, and the invitation letter is more effective when accompanied by the advice of a health professional.²⁴ Not least, the new organisation allowed for a timely resumption of activities after the COVID-19 pandemic-related stoppage. Of note in 2020, despite the 3-month stoppage, a total of 51,295 FITs (including 11,609 FITs during the summer period) were performed in our Centre, compared to 57,302 FITs in 2019. All these findings suggest that the offer of proximity and wide daily and hourly availability provided by the pharmacies facilitated screening attendance, as we had hypothesised. These aspects were secondary to the implementation of the new model and should be considered by those programme organisers who decide to apply a similar path.

Conclusions

Despite this work being a single-centre experience and that the organisation of CRC screening programmes worldwide differs making the results difficult to generalise, it strengthens the findings from the previous literature study by Deandrea et al.⁶ showing that the application of HFMEA in a CRC screening programme is a useful tool to reduce potential errors leading to an improvement in the safety and quality of the healthcare service provided. Furthermore, our work provides new insights into how an organised CRC screening programme can benefit from collaborating with local pharmacies in terms of service quality and screening attendance. Indeed, involving the local pharmacies as CRC screening kit delivery points as well as sample collection and sending points may allow for complete traceability of the kit-sample flow and temperature control. In addition, the convenience of using a local pharmacy may facilitate screening attendance as well as the participation of the pharmacists as stakeholders who can provide health advice to citizens that may promote participation in screening.

Further studies focusing on the application of HFMEA in organised screening programme settings are advocated.

Footnotes

Acknowledgements

Authors would like to thank the IT engineer, Dr Gianfranco Valastro, for his assistance in developing the informatics tools, and all the members of the Screening Centre's staff of the Local Health Authority of Bologna for their work.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Angela Chiereghin

Lorenzo Pizzi

Francesca Mezzetti

References

Sung

Ferlay

Siegel

, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209–249.

Roselló

Simón

Cervantes

. Programmed colorectal cancer screening decreases incidence and mortality. Transl Gastroenterol Hepatol 2019; 4: 84.

Ladabaum

Dominitz

Kahi

, et al. Strategies for colorectal cancer screening. Gastroenterology 2020; 158: 418–432.

Stoffel

McGregor

Hirst

, et al. Evaluation of the call for a kit intervention to increase bowel cancer screening uptake in Lancashire, England. J Med Screen 2022; 29: 166–171.

European Colorectal Cancer Screening Guidelines Working Group, von Karsa

Patnick

, et al. European guidelines for quality assurance in colorectal cancer screening and diagnosis: overview and introduction to the full supplement publication. Endoscopy 2013; 45: 51–59.

Deandrea

Tidone

Bellini

, et al. Implementation of failure mode and effects analysis to the specimens flow in a population-based colorectal cancer screening programme using immunochemical faecal occult blood tests: a quality improvement project in the Milan colorectal cancer screening programme. BMJ Open Qual 2018; 7: e000299.

DeRosier

Stalhandske

Bagian

, et al. Using health care failure mode and effect analysis: the VA National Centre for patient safety’s prospective risk analysis system. Jt Comm J Qual Improv 2002; 28: 248–267.

Reid

Paul

Mostoufi

, et al. Evaluation of the stability of faecal immunochemical test specimens. Clin Biochem 2023; 115: 92–96.

Regione Emilia-Romagna. Nomenclatore tariffario regionale della specialistica, https://salute.regione.emilia-romagna.it/ssr/strumenti-e-informazioni/nomenclatore-tariffario-rer/ottobre-2022 (2022, accessed 01 August 2023).

10.

Francisci

Capodaglio

Gigli

, et al. Cancer cost profiles: the epicost estimation approach. Front Public Health 2022; 10: 974505.

11.

Mora

Ayala

Bielza

, et al. Improving safety in blood transfusion using failure mode and effect analysis. Transfusion 2019; 59: 516–523.

12.

Shariff

Stillkrieg

Lotter

, et al. Dosimetry, optimization and FMEA of total skin electron irradiation (TSEI). Z Med Phys 2022; 32: 228–239.

13.

Jain

. Use of failure mode effect analysis (FMEA) to improve medication management process. Int J Health Care Qual Assur 2017; 30: 175–186.

14.

Lin

Wang

Chen

, et al. Failure mode and effects analysis on the control effect of multi-drug-resistant bacteria in ICU patients. Am J Transl Res 2021; 13: 10777–10784.

15.

Cimadomo

Ubaldi

Capalbo

, et al. Failure mode and effects analysis of witnessing protocols for ensuring traceability during PGD/PGS cycles. Reprod Biomed Online 2016; 33: 360–369.

16.

Frewen

Brown

Jenkins

, et al. Failure mode and effects analysis in a paperless radiotherapy department. J Med Imaging Radiat Oncol 2018; 22: 707–715.

17.

Frosini

Miniati

Grillone

, et al. Integrated HTA-FMEA/FMECA methodology for the evaluation of robotic system in urology and general surgery. Technol Health Care 2016; 24: 873–887.

18.

Manrique-Rodríguez

Sánchez-Galindo

López-Herce

, et al. Risks in the implementation and use of smart pumps in a pediatric intensive care unit: application of the failure mode and effects analysis. Int J Technol Assess Health Care 2014; 30: 210–217.

19.

Berenbrok

Gabriel

Coley

, et al. Evaluation of frequency of encounters with primary care physicians vs visits to community pharmacies among medicare beneficiaries. JAMA Netw Open 2020; 3: e209132.

20.

Armbruster

Overcash

Reyes

. Clinical chemistry laboratory automation in the 21st century - amat Victoria curam (victory loves careful preparation). Clin Biochem Rev 2014; 35: 143–153.

21.

Regione Emilia-Romagna. Screening tumore colon retto. Infografica percorso e risultati, https://salute.regione.emilia-romagna.it/screening/colon/percorso-risultati (2023, accessed 01 August 2023).

22.

Darbà

Marsà

. Evaluation of productivity losses due to premature mortality from colorectal cancer. PLOS ONE 2020; 15: e0244375.

23.

Blum-Barnett

Madrid

Burnett-Hartman

, et al. Financial burden and quality of life among early-onset colorectal cancer survivors: a qualitative analysis. Health Expect 2019; 22: 1050–1057.

24.

EpiCentro-Istituto Superiore di Sanità. Gli screening oncologici e l’impatto della pandemia: i dati dalla sorveglianza PASSI, https://www.epicentro.iss.it/passi/focus/screening-oncologici-impatto-pandemia-dati-passi-2020 (2020, accessed 21 May 2023).