Performance Management in the Oncology Hospital: From KPI to Management Decision (Introduction 101)

Abstract. This chapter introduces, at beginner level, time to treatment initiation, multidisciplinary tumour board (MDT) coverage, radiotherapy waiting time, and capacity and quality indicators for physicians who manage or are preparing to manage an oncology hospital, and ties these indicators to management decisions through the chain of target, threshold, variance and managerial intervention. The core message is that an indicator is an examination finding, variance analysis is a differential diagnosis, and a managerial intervention is a treatment. The chapter contains current evidence selected from seven Web of Science searches, a worked case and a pocket card for physicians.

Key concepts: key performance indicator, time to treatment initiation, MDT coverage, radiotherapy waiting time, capacity utilization, statistical process control, variance analysis, patient navigation

Chapter plan and reading guide

This chapter covers four core indicator families in an oncology hospital (time to treatment initiation, multidisciplinary tumour board coverage, radiotherapy waiting time, and capacity and quality indicators) and the chain of “target, threshold, variance, intervention” that ties them to management decisions. It moves from simple to complex: first the concepts, then the indicators one by one, then how to read a variance and choose an intervention, and finally a worked case and a pocket card for physicians.

Section	What the reader takes from it
1. Introduction: the clock is also a clinical finding	Why time is a managerial indicator in oncology
2. Core concepts	Indicator, KPI, target, threshold, variance and intervention in the physician’s language
3. Choosing indicators	How we decide what to measure, and the “indicator passport”
4. Time to treatment initiation	Definition, evidence, the waiting time paradox and splitting the time into intervals
5. MDT coverage	Coverage, timeliness and the rate at which decisions are implemented
6. Radiotherapy waiting time	The chain from simulation to first fraction
7. Capacity indicators	Demand, capacity, utilization and spare capacity
8. Quality indicators	Structure, process, outcome and balancing measures
9. Target, threshold and variance	Where targets come from, threshold gaming and control chart logic
10. From variance to managerial intervention	A six-step decision cycle and an evidence-based menu of interventions
11. Worked case	Step-by-step solution of the Module II workshop case
12. Common mistakes	Eleven pitfalls and how to avoid them
13. Pocket card for physicians	Monday morning questions, templates, golden rules
14. Conclusion	A brief appraisal

How to read it. A reader in a hurry can build the framework from Sections 2, 9 and 13. A manager working on a specific indicator can go straight to the relevant section (4, 5, 6, 7 or 8). For a participant preparing for the workshop, Sections 10 and 11 should be read together.

Note on sources. The chapter was written by selecting the directly relevant studies from the corpus obtained in seven separate Web of Science (WoS) searches (time to treatment initiation, waiting times, MDT coverage, oncology quality indicators, oncology capacity indicators, control charts and variance analysis). A few classic works from the quality management literature were added for the conceptual foundation. The figures given in the text belong to the context of the study concerned; they should not be carried over unchanged to another hospital but read as pointers.

1. Introduction: the clock is also a clinical finding

In oncology, time is a finding that describes the workings of the institution as much as the biology of the tumour. The number of days in which a patient gets from diagnosis to treatment tells us about both the course of the disease and the hospital’s appointment, diagnostic, tumour board and treatment capacity. The first task of the oncology hospital manager is therefore to learn to read the clock like a clinical finding.

The evidence shows that delay has a real price in some cancers. In a meta-analysis in colorectal cancer, treatment delay was associated with an increase in all-cause mortality, with hazard ratios (HR) of 1.12 at 4 weeks, 1.24 at 8 weeks and 1.39 at 12 weeks (Ungvari et al., 2025). A similar meta-analysis in breast cancer reported that the HR for breast cancer-specific death rose to 1.20, 1.43 and 1.71 at 4, 8 and 12 weeks respectively (Ungvari, Fekete, Buda, Lehoczki, Munkácsy, et al., 2026). A microsimulation for Canada projected that disruptions to diagnosis and treatment during the pandemic could lead to about 21,247 (2.0%) additional cancer deaths between 2020 and 2030 (Malagón et al., 2022).

The picture is not the same in every cancer, however. In lung cancer, short delays of 4 to 12 weeks showed no significant relationship with survival (Ungvari, Fekete, Buda, Lehoczki, Fekete, et al., 2026). There are even findings in which patients with advanced non-small cell lung cancer who waited longer had a lower risk of death, as if “waiting were good for them” (Klarenbeek et al., 2023). The reason is that sicker patients are treated faster anyway; this is known in the literature as the waiting time paradox (Tope et al., 2025). The lesson for the physician is clear: time is an important indicator, but it is never read alone; it is interpreted together with tumour type, stage and treatment pathway.

The starting point of this chapter is the case used in Module II of the Oncology Hospitals Executive Development Programme. In the case, one of the hospital’s strategic goals is to reduce the time from diagnosis to the start of treatment. Over the last three months, however, this time has lengthened, radiotherapy waiting time has risen, the rate at which MDT decisions are implemented on time has fallen, and capacity in some branches has begun to be under-used. The surgical, medical oncology and radiation oncology teams disagree about the source of the problem, and patient complaints are increasing. This chapter offers the tools needed to read such a picture and derive a management decision from it.

After reading this chapter, the reader is expected to be able to:

• Distinguish the concepts of indicator, KPI, target, threshold and variance.
• Define time to treatment initiation, MDT coverage and radiotherapy waiting time correctly and read them by splitting them into parts.
• Use capacity and quality indicators together with time indicators in a balanced dashboard.
• Use control chart logic to tell whether a variance is real or random.
• Choose an evidence-based managerial intervention matched to the root cause of the variance and set up a monitoring plan.

2. Core concepts: a glossary from indicator to decision

Performance management is the institution-level counterpart of the clinical reasoning a physician uses every day. An indicator is what an examination finding is; variance analysis is what a differential diagnosis is; a managerial intervention is what a treatment is; monitoring is what a follow-up visit is. This similarity helps physicians see performance management not as a “foreign language” but as the application of a familiar logic to a new field.

Concept	Short definition	In the physician’s language	Oncology example
Indicator	A measure that expresses the state of a process or an outcome as a number	Vital sign	Number of days from diagnosis to treatment
Key performance indicator (KPI)	One of a small number of critical indicators tied directly to a strategic goal	A parameter monitored continuously in intensive care	Share of new patients starting treatment within 30 days
Target	The level to be reached	Target HbA1c or target blood pressure	“At least 80% of new patients start treatment within 30 days” (example)
Threshold	The limit beyond which action is required	Critical value notification	If the rate falls below 70%, a root cause review begins (example)
Tolerance band	The range within which natural fluctuation is accepted	Reference range	Ups and downs of a few days from week to week
Variance (deviation)	The difference between the actual value and the target	Departure from the expected clinical course	Median time 9 days longer than the target
Common-cause variation	Ever-present fluctuation arising from the structure of the system itself	Physiological variability	Seasonal patient volume
Special-cause variation	A signal from outside the system, arising from a specific event	A newly developed complication	A one-week backlog after a LINAC breakdown
Root cause	The underlying mechanism producing the variance	Aetiology	Nobody being responsible for the appointment after the MDT
Managerial intervention	A change designed to address the root cause	Treatment	Assigning a patient navigator
Balancing measure	An indicator that monitors whether the intervention causes harm elsewhere	Monitoring for side effects	Tracking guideline adherence while time is shortened

There is an order among these concepts. First indicators are defined; then a few that connect to strategy are chosen as KPIs. A target and a threshold are set for each KPI. When the actual value departs from the target, a variance arises. The manager first decides whether this variance is common cause or special cause, then finds the root cause and chooses a matching intervention. Finally the effect and the side effects of the intervention are monitored. The distinction between structure, process and outcome made by Donabedian (1988), one of the founders of quality management, forms the background of this chain: structure (people, machines, beds) shapes process (appointments, tumour board, treatment), and process affects outcome (survival, quality of life, experience).

Figure 1 shows the cycle that forms the backbone of this chapter. The cycle works with the same logic as the physician’s sequence of “finding, differential diagnosis, treatment, follow-up”.

3. Choosing indicators: what do we measure, and why?

A good dashboard consists of a small number of well-defined indicators for which data can actually be collected. The oncology literature has no shortage of indicators; the problem is choosing. From 15 quality indicator frameworks used for breast and colorectal cancer in the G7 countries, a total of 512 indicators were extracted, the great majority of which were process indicators derived from clinical guidelines (T. Nash et al., 2026). In a Delphi study for thoracic oncology in Australia and New Zealand, 422 international indicators were screened, 71 were assessed and consensus was reached on only 27 (J. Nash et al., 2024). These proportions tell the manager something: the task is not to measure everything that can be measured but to choose what will produce decisions.

Three questions guide the choice:

1. Which strategic goal does this indicator serve? For an indicator to be a KPI it has to be tied to a goal. For the goal “shorten the time from diagnosis to treatment”, time to treatment initiation is a natural KPI; operating theatre cleaning time can only be an intermediate indicator.
2. Can its data be produced reliably? When electronic health record data from 63 oncology practices in the United States were examined, only 2 of 19 oncology quality measures tied to the payment system could be calculated for more than 1% of patients (Schorer et al., 2022). An indicator without data remains an empty box on the dashboard or, worse, a misleading number.
3. Is it clear who will do what when its value changes? An indicator without an owner and an action plan is watched but not managed.

It also helps to know that quality indicators are used for different purposes. In joint work by the US Veterans Affairs Radiation Oncology programme and ASTRO, 18 common measures were developed for five tumour sites and classified as 13 quality, 4 surveillance and 1 aspirational measure (Park et al., 2022). This distinction can be carried over to the dashboard: some indicators are managed today, some are only watched, and some express the institution’s ambition for tomorrow. In value-based healthcare it is recommended that, alongside survival and disease control, patient-reported outcome measures (PROMs) be included among the indicators (Abdalla et al., 2022). Indicators that measure avoidance of low-value care are still a minority; only 18% of 313 cancer quality measures compiled from six organisations serve this purpose (Ellsworth et al., 2022).

The indicator passport

Preparing a one-page “indicator passport” for each KPI ends definitional arguments before they start. The example below is completed for time to treatment initiation; the target and threshold values are examples only, and each institution should set its own.

Field	Example content for time to treatment initiation
Definition	Calendar days from the date of pathological diagnosis to the day the first definitive oncological treatment starts
Numerator and denominator	Numerator: new patients starting treatment within 30 days. Denominator: new patients starting their first definitive treatment in the period
Exclusions	Cases postponed at the patient’s own request where this is documented (reported separately, not deleted)
Reporting format	Median, 90th percentile, share of patients within target
Stratification	Tumour type, stage, first treatment modality (surgery, systemic, radiotherapy), branch
Data source	Pathology system, hospital information system, radiotherapy and chemotherapy scheduling records
Frequency	Weekly monitoring, monthly report to the management board
Owner	Deputy chief physician responsible for oncology services
Target and thresholds	Example: target 80%, warning 75%, action threshold 70%
Balancing measures	Adherence to MDT recommendation, first-time pass rate of plan quality assurance
If the threshold is crossed	Root cause review within 10 working days and a short note to the management board

Four layers of a balanced oncology dashboard

A single indicator means seeing the institution with one eye. The dashboard proposed in this chapter has four layers, each balancing the others:

• Time: time to treatment initiation, time from diagnosis to MDT, time from MDT decision to treatment, radiotherapy waiting time.
• Access and coverage: MDT coverage rate, rate of implementation of MDT recommendations.
• Capacity: demand, capacity, utilization and backlog (LINAC, chemotherapy chairs, operating theatres, beds, diagnostic units, workforce).
• Quality and safety: guideline adherence, monitoring of patient-reported outcomes, share of diagnoses made through emergency presentation, composite outcome measures.

If a quality indicator worsens while a time indicator improves, or if waiting grows while capacity utilization looks high, the manager should be able to see this contradiction on the dashboard.

4. Time to treatment initiation

Time to treatment initiation (TTI) is the single strongest indicator of whether all parts of an oncology hospital are working together. A blockage in any one of appointments, imaging, pathology, molecular testing, the tumour board, planning or the treatment units is reflected in this time.

4.1 Definition: when does the clock start and stop?

The most widely used definition is the number of calendar days from the date of pathological diagnosis to the day the first definitive oncological treatment starts. In the literature, however, start and end points vary widely. A systematic review of 130 articles on timing in colorectal cancer showed that time intervals were categorised arbitrarily in 83% of studies, that there was no adjustment for confounders in 50%, and that immortal time bias was present in 46% of the articles with survival as the outcome (Drosdowsky et al., 2023).

The choice of definition directly changes the number the manager sees. In a breast cancer cohort from Türkiye, “time from admission to surgery” and “time from biopsy result to surgery” could not be used interchangeably; cases in which the two measurements disagreed pointed to different bottlenecks. The authors recommended monitoring both intervals together and putting discordant cases through a short review with reason codes (Tandogan et al., 2025). In a study with kidney cancer data in England, applying the waiting time adjustment rule, which corrects for postponement at the patient’s own request, reduced the mean time at patient level from 29.2 to 27.3 days (Varma et al., 2026). In short, the figures of two branches should not be compared before the definition and adjustment rules are written in the indicator passport.

4.2 Typical values and trend

National registries show that time to treatment initiation is around one month in many countries and has lengthened over time. In Japan, the median time in the national hospital-based cancer registries for six cancer types is 33 days (Watanabe et al., 2023). In the US National Cancer Database, the median time for the 30 most common cancer types is 26 days; it rose from 21 days in 2004 to 28 days in 2015 and did not shorten in any cancer type. In stage I patients, the median time doubled from 14 to 28 days (Dafflisio et al., 2025). In the Swiss sarcoma network, the median time was reported as 30 days (Schärer et al., 2023). In the Netherlands, a six-week norm has been adopted for surgical oncology, and in 2020, 76% of patients started treatment within this period (van Vuren et al., 2024). Whether rapid growth in demand can strain the capacity for timely treatment, especially in referral centres, has also been examined (Frosch et al., 2021).

4.3 Evidence: delay does not carry the same price in every cancer

As noted in the introduction, in colorectal and breast cancer there is a relationship between delay and risk of death that grows by the week (Ungvari et al., 2025; Ungvari, Fekete, Buda, Lehoczki, Munkácsy, et al., 2026). In head and neck cancers, each additional day until the start of treatment raised the hazard of death by a small but significant amount (HR 1.003 per day), and the rise was steeper in oral cavity cancer (Frank et al., 2024). In early-stage lung cancer, a time above 50 days was associated with lower one-year and five-year survival (Tang et al., 2023). In breast cancer, starting treatment after 12 weeks was associated with a 24% higher risk of death than starting within 8 weeks (Wang et al., 2025).

Some studies, by contrast, give opposite or neutral results. In the Canadian province of Quebec, longer waiting did not worsen survival in lung cancer (Denault et al., 2022); in a national population-based cervical cancer cohort, time to curative treatment likewise showed no significant relationship with survival (Hack et al., 2022). An important reason for this difference is the waiting time paradox: physicians treat sicker patients faster, so the group treated quickly is in fact the group with the worse prognosis (Tope et al., 2025; J. R. Zhang et al., 2022). A decision model developed during the pandemic also showed that the effect of delay varies with patient, tumour, treatment and community characteristics and that uniform prioritisation rules cannot capture this variation (Hartman et al., 2020).

In recent years, pathway-specific thresholds have been proposed in place of a single general threshold. In colorectal cancer, even delays as short as four days were associated with a higher risk of metastasis in patients receiving adjuvant chemotherapy after surgery, whereas no meaningful threshold was found in patients treated with surgery alone (Nguyen et al., 2026). In Ontario, Canada, beyond the traditional 28-day target for colorectal cancer surgery, a data-derived inflection point was found at about 45 days, where the hazard of death begins to rise (Bondzi-Simpson et al., 2025).

Implication for the manager: time to treatment initiation is a system indicator; it passes no verdict on the prognosis of an individual patient. It should therefore be monitored stratified by tumour type, stage and treatment pathway, reported as median and 90th percentile rather than as a mean, and compared with pathway-specific targets set together with the clinical teams.

4.4 Splitting the time into parts: where is the time lost?

Total time is a symptom; the diagnosis is made only when the time is split into parts. At Moffitt Cancer Center, when the mean 6.9 weeks from breast biopsy to surgery was divided into three intervals, 35% of the time fell between biopsy and first consultation, 13% between first consultation and tumour board, and 52% between tumour board and surgery; the longest interval (3.6 weeks on average) lay after the tumour board, and physician capacity was identified as the main cause of delay (Loftus et al., 2014). Modelling the lung cancer diagnostic pathway in Wales with discrete event simulation showed that giving the first outpatient appointment within 7 days and streamlining investigations could gain about 20 days and raise the share of patients treated within 62 days by 20 to 25% (England et al., 2021). In early-stage lung cancer, preoperative physiological assessment and multidisciplinary evaluation were the main determinants of the time (Tang et al., 2023); in advanced disease, complex diagnostic procedures delayed treatment (Klarenbeek et al., 2023).

The table below divides the patient journey into manageable intervals. Figure 2 summarises the same journey visually.

Interval	Start and end	Common bottleneck	Suggested intermediate indicator
A1. Access	Referral or suspicion → first oncology appointment	Appointment capacity, incomplete referral documents	Median days to first appointment
A2. Diagnosis	First appointment → pathological diagnosis	Imaging and biopsy queues, pathology reporting time	Imaging waiting time, pathology reporting time
A3. Molecular profile	Diagnosis → biomarker result	Specimen transfer, sequential test workflow	Molecular test turnaround time
A4. Tumour board	Diagnosis → MDT decision	Board frequency, incomplete files	Median days from diagnosis to MDT
A5. Implementation	MDT decision → first definitive treatment	Theatre, LINAC or chair capacity, appointments with no owner	Median days from MDT decision to treatment

5. MDT coverage: who comes to the table, and where does the decision go?

The multidisciplinary tumour board (multidisciplinary team, MDT) is where the treatment decision is produced in modern oncology. From the manager’s standpoint, the MDT is measured with three questions: how many patients come to the table (coverage), how quickly they reach the table and move from the table to treatment (timeliness), and to what extent the decision made at the table is carried out (implementation and adherence).

5.1 Coverage

The MDT coverage rate is the number of newly diagnosed patients discussed at the board before treatment in a given period, divided by all newly diagnosed patients in the same period who should have been discussed. Defining the denominator correctly matters: the rate reflects reality better if cases for which the guideline does not recommend discussion before treatment are left out of the denominator.

Although guidelines treat the MDT as standard, coverage in real life is incomplete. In a health services study of patients with metastatic colorectal cancer in Germany, guideline-concordant MDT discussion took place in 63% of operated colon cancer patients, 38% in stage IV colon cancer and 47% in rectal cancer (Lowes et al., 2017). When a cloud-based, asynchronous MDT platform was set up for digestive system cancers in China, coverage rose from 47.85% to 79.99%; each case was discussed in 7.68 minutes on average, and the mean response time of physicians was 27.21 hours (Y. Zhang et al., 2024). In Sweden, participation of district hospitals in the head and neck board by telemedicine was found to cost slightly less than attendance in person (Stalfors et al., 2005). These findings show that coverage depends largely on organisational design and can be extended with technology.

Who sits at the table is also an indicator. After a clinical geneticist began to attend the gynaecological oncology board in Australia, the referral rate of eligible ovarian cancer patients to genetic counselling rose from 26.7% to 51.7% (Cohen et al., 2016). In rectal cancer, mandatory MDT attendance was associated with improvement in 12 of 14 preoperative process variables and a marked increase in the completeness of total mesorectal excision (Richardson et al., 2016). In the USA, weekly board attendance by physicians caring for patients with lung and colorectal cancer was not associated with overall survival, but it raised the likelihood of clinical trial enrolment (OR 1.6) and of curative surgery in stage I to II non-small cell lung cancer (OR 2.9) (Kehl et al., 2015).

5.2 Timeliness: is the board a door or a wall?

The MDT can lengthen the time while raising the quality of the decision. In oral cavity cancer, patients discussed at the board before treatment had higher guideline adherence (79.6% against 68%) but also a longer time to the start of treatment (33.5 against 20.1 days); there was no significant difference in three-year survival (Burkhardt et al., 2024). In contrast, a well-designed multidisciplinary breast clinic in a safety-net hospital reduced time from diagnosis to treatment from 46.9 to 35.0 days (Kobzeva-Herzog et al., 2026). The difference lies in how the board is organised: a board that meets once a week and defers incomplete files to the following week becomes a wall; a board that assesses on the same day and has missing investigations completed beforehand works as a door.

5.3 Implementation and adherence: does the decision stay on paper?

The rate at which board decisions are carried out (MDT adherence or implementation rate) is monitored less than coverage but is at least as important. In Germany, 75.5% of the recommendations of a lung cancer board were guideline-concordant and 15.6% went beyond the guideline, and about 90% of recommendations were put into practice (Walter et al., 2023). In digestive system cancers, although 76% of recommendations were fully guideline-concordant, the full implementation rate was only 64%; the main reasons for deviation were comorbidities and a different preference on the part of the patient (Krause et al., 2023).

The relationship between unimplemented decisions and outcome is striking. Among 4,501 patients with early breast cancer in China, 81.8% followed the MDT recommendation; disease-free survival (HR 1.813) and overall survival (HR 2.478) were worse in the non-adherent group (Yang et al., 2020). In upper gastrointestinal cancers with a chance of curative treatment, 41.8% of patients did not follow the recommendation, the most frequent reason being patient refusal (50%) (Soon et al., 2023). In stage III non-small cell lung cancer in Switzerland, adherence was 70.4%, and non-adherence was associated with worse event-free and overall survival (Koster et al., 2025). In a study of gynaecological cancer boards, the recommendation was not followed in 19.5% of board decisions, and non-adherence was associated with shorter survival (Naessens et al., 2023). In colorectal cancer too, the most frequent reason for non-adherence was patient preference (Kandemir et al., 2025).

These relationships should not be read as causal; patients who fall outside the recommendation are often older, frailer or at a more advanced stage. The message for the manager is nevertheless clear: the reason for non-adherence should be recorded and classified with codes such as “patient preference”, “change in clinical condition”, “capacity” and “communication breakdown”. Non-adherence caused by capacity or communication is the direct responsibility of management. A consensus text on surgical oncology networks likewise describes the MDT as the “cognitive backbone” of the network while warning about the risks of hierarchy, protocol-bound decisions and diffusion of responsibility, and recommends structured audit and transparent documentation (Nicolai et al., 2026). Population-based findings showing that patients who receive multidisciplinary consultation are more likely to receive guideline-recommended adjuvant treatment also support this approach (Sanchez et al., 2024).

5.4 A suggested MDT indicator set

Indicator	Definition	What does it tell us?
MDT coverage rate	New patients discussed at the board before treatment / new patients who should be discussed	Whether the decision is made collectively
Time from diagnosis to MDT	Median days from pathological diagnosis to board decision	Whether the board is a door or a wall
Time from MDT decision to treatment	Median days from board decision to first definitive treatment	How quickly the decision is put into practice
On-time implementation rate	Patients whose board decision is implemented within the set period / patients with a board decision	Whether the decision has an owner
Distribution of reasons for non-adherence	Unimplemented decisions by reason code	Whether the problem lies with the patient, the clinic or the system
Completeness of the decision record	Share of decisions that include stage, recommendation, responsible physician and target date	The reliability of the measurement itself

6. Radiotherapy waiting time

Radiotherapy waiting time shows how well the most expensive and least flexible resource of the oncology hospital (the linear accelerator, the physics and dosimetry team, the radiation oncologist’s time) is managed. The number of LINACs cannot be increased quickly; waiting in radiotherapy is therefore usually reduced by reorganising the workflow before buying a new machine.

6.1 Definition: not one wait but a chain

Starting radiotherapy consists of linked steps: the treatment decision, computed tomography simulation, contouring of targets and organs at risk, dosimetric planning, plan-specific quality assurance, plan verification and the first fraction. Process modelling of a clinic with a single physician and a single LINAC showed that when the weekly number of new patients rose from 1 to 10, mean time from simulation to treatment increased from 4 to 7 days and the maximum from 6 to 12 days (Chowdhry & Simpson, 2023). This result recalls a basic fact of queueing theory: as resource utilization approaches one hundred per cent, waiting grows not linearly but rapidly. The manager’s reading “the machine is completely full, so we are efficient” is therefore dangerous.

The recommended main indicator is the time from the radiotherapy decision (or referral to the radiotherapy department) to the first fraction. Intermediate indicators are time from simulation to first fraction, planning turnaround time and the first-time pass rate of plan quality assurance. In adjuvant radiotherapy, the time from the end of surgery or systemic treatment to radiotherapy should also be monitored.

6.2 Evidence and trend

During a period in Ontario, Canada, when radiotherapy waiting times were unusually long because of capacity problems, event-free survival was worse in patients with early breast cancer who had not received chemotherapy and who waited 12 weeks or more from surgery to radiotherapy (Raphael et al., 2020). In US Commission on Cancer centres, the median time from breast-conserving surgery to radiotherapy lengthened from 43 days in 2004 to 48 days in 2020; among the new breast cancer quality measures, radiotherapy timing was the one met least often, and only 35.2% of centres reached the 80% performance level (Fefferman et al., 2025).

Behind capacity there is usually the workforce. In the United Kingdom, the shortfall of consultant clinical oncologists is currently reported as 17% and is expected to reach 26% in 2030; the authors stress that this shortfall increases radiotherapy delays (Vulpe et al., 2026).

6.3 How is waiting reduced?

• Process management: in a radiation oncology department in Spain, moving to a process management model shortened waiting times for first consultation and for the start of treatment, lowered cost per patient and, thanks to hypofractionated schedules, saved 6,035 sessions in one year (Lupiáñez-Pérez et al., 2017). Hypofractionation is a powerful lever that expands capacity in clinically eligible patients, but it must always be considered together with the clinical indication.
• Clearing administrative bottlenecks: a radiation oncology department in the USA carried out a root cause analysis of insurance prior authorisation delays and redesigned clinical treatment plan templates and the prior authorisation workflow using Plan-Do-Check-Act and A3 methods (Tran et al., 2025). Even where there is no prior authorisation, similar administrative steps (reports, approvals, procurement of materials) are the invisible waiting areas of radiotherapy.
• Patient navigation: in Brazil, the national breast cancer patient navigation programme shortened time from diagnosis to radiotherapy by about 10 days compared with the comparison group; in the same period, radiotherapy costs rose by about 16.5%. The authors interpreted this rise as previously unmet demand becoming visible (da Silva & Postali, 2026).
• Keeping the balance between quality and speed: in thoracic stereotactic radiotherapy, pre-treatment peer review at two meetings a week aimed to raise quality without creating routine treatment delays, and the level of compliance was monitored with a statistical process control chart (Kut et al., 2023).
• Communication: interviews with radiation oncologists in India showed that unavoidable waits can be made more bearable for patients through two-way communication and organisation of the waiting area (Saini et al., 2022).

6.4 A suggested radiotherapy indicator set

Indicator	What does it tell us?
Median and 90th percentile time from radiotherapy decision to first fraction	How long the patient actually waits
Time from simulation to first fraction	The speed of the planning chain
First-time pass rate of plan quality assurance	Whether speed is achieved without compromising quality
LINAC utilization and daily operating hours	How much of the machine capacity is used
Hypofractionation rate in eligible patients	Whether the capacity lever is being used
Difference in LINAC utilization between branches	Where the spare capacity is

7. Capacity indicators: demand, capacity and spare capacity

Capacity indicators answer the “why” of waiting times. Time indicators say that there is a problem; capacity indicators show whether the problem stems from excess demand, from insufficient resources or from resources standing in the wrong place.

7.1 Four core concepts

• Demand: the number of patients or procedures needing service in a given period (for example new radiotherapy patients per week).
• Capacity: the maximum amount of service that can be produced in the same period with the available people, machines, beds, chairs and time.
• Utilization: the ratio of service delivered to capacity.
• Backlog: the number of patients in the queue who have not yet received service.

These four concepts should be read together. If demand persistently exceeds capacity, the backlog grows and waiting lengthens. If waiting lengthens although demand is below capacity, the problem usually lies in flow, appointment design or the misallocation of resources.

7.2 The spare capacity paradox

Spare capacity can exist even in a system with long waiting lists. In England, the “maximum surgical capacity” of each hospital performing breast and colorectal cancer surgery was defined as the highest value of its six-month moving average volume. By this measure, 40.4% of hospitals performing breast surgery and 50.3% of those performing colorectal surgery used less than 80% of their maximum capacity. Spare capacity was greater in centres that lost their patients to more distant hospitals (Aggarwal et al., 2023). This finding corresponds directly to the problem of “under-use of capacity in some branches” in the Module II case: the main site can overflow while a branch stands empty, and the solution is to redesign patient referral before making new investments.

Capacity can also be tracked not only by volume but by the number of patients served within target. The case throughput metric developed in England relates the number of patients meeting the waiting standard to the pre-pandemic level, providing a measure of resilience that is comparable between institutions; it showed that diagnostic and referral pathways suffered greater and more lasting losses than treatment pathways (Lam et al., 2026).

7.3 Demand is not fixed

When capacity is increased, demand can change too. In a pilot in England that shortened waiting times for urgent diagnosis of suspected cancer, the share of patients with suspected colorectal cancer exceeding the waiting target fell by 3.9 points, but referrals from general practitioners referring to the pilot hospitals rose by 10.8% (Hayes et al., 2024). Shorter waits attract new demand; the effect of a capacity increase can therefore partly melt away after a while. In the USA, time from diagnosis to treatment shortened in 2020 during the pandemic and lengthened again in 2021 and 2022; the authors believe workforce constraints played a part in the later lengthening (Fan et al., 2025).

7.4 Capacity indicators at unit level

Resource	Example capacity indicators	Interpretation tip and example from the literature
Linear accelerator	Daily operating hours, fractions per day, utilization	Waiting grows rapidly as utilization approaches one hundred per cent (Chowdhry & Simpson, 2023)
Day chemotherapy unit	Chair occupancy, time from appointment to start of infusion, drug preparation time	With design thinking and regimens prepared in advance, the time to start of morning treatments fell from 83 to 45 minutes (Jen et al., 2023); a lean and DMAIC approach reduced waiting by 19% (Radheshyam et al., 2024)
Oncology pharmacy	Drug preparation turnaround time, share of early prescriptions	Early prescribing, advance preparation, robotic compounding and lean management are recommended (Giraud et al., 2026)
Systemic treatment flow	Share of delays of more than 7 days between cycles	In early breast cancer, 17% of patients had a delay of more than 7 days between cycles, and this delay was associated with five-year mortality (HR 1.33) (Steventon et al., 2024)
Operating theatre and beds	Planned theatre utilization, cancellation rate, length of stay	Minimally invasive surgery shortened length of stay in lobectomy by 1.87 days compared with open surgery (Ricciardi et al., 2022)
Bed turnover	30-day unplanned readmission rate	Palliative care consultation was associated with a lower rate of readmissions in ovarian cancer (Francoeur et al., 2025)
Imaging	Waiting time for CT and MRI	With targeted additional funding, waiting time for CT and MRI in suspected cancer fell from 17.2 to 4.2 days (Papp et al., 2023)
Molecular pathology	Turnaround time from specimen collection to report	With parallel specimen processing and closed-loop communication, median time to report fell from 22 to 16 days (Yeo et al., 2025)
Workforce	Workload per physician, nurse, physicist and dosimetrist	Workforce shortage is one of the main causes of radiotherapy delay (Vulpe et al., 2026)

7.5 Simulation: trying the decision on the computer before trying it on the ground

Capacity decisions are expensive and hard to reverse. Discrete event and agent-based simulation models are therefore used to test in advance questions such as “should we buy a second LINAC or strengthen the planning team?” A configurable simulation model developed for an oncology department in Italy working with an external pharmacy helped choose the configuration that would most reduce patient waiting time by comparing different departmental layouts (Corsini et al., 2023). The lung cancer model in Wales tested “what if” scenarios in a similar way (England et al., 2021).

8. Quality indicators: accuracy alongside speed

Time and capacity indicators show how fast the hospital works; quality indicators show whether it does the right thing in the right way. In oncology, speed is not a value in itself; starting the wrong treatment quickly can be more harmful than starting the right treatment a few days late. Every time indicator should therefore be accompanied by at least one quality indicator as a balancing measure.

8.1 Oncology quality indicators through the Donabedian triad

Dimension	Oncology example	A finding from the literature
Structure	Accreditation, existence of a tumour board, registry system, number of specialists	In the US state of Iowa, Commission on Cancer accreditation was the strongest independent predictor of high hospital performance on four quality measures (Schroeder et al., 2022)
Process	Guideline-concordant treatment, PRO monitoring, adequate lymph node sampling	Among patients with lung cancer receiving systemic treatment, adherence to completing PRO forms was 49.4% overall and ranged from 27.2% to 70.0% between centres (Takvorian et al., 2022)
Diagnostic process	Share of cancer diagnoses made through emergency presentation	A digital quality measure showed that in the Veterans system 20.9% of lung cancers and 22.4% of colorectal cancers were diagnosed after an emergency presentation; missed diagnostic opportunities were found in a substantial share of these cases (Kapadia et al., 2024)
Outcome	Composite outcome measures, risk-adjusted mortality	In endometrial cancer surgery, the “textbook oncologic outcome”, meaning that all six criteria are met, was achieved in 81% of patients and was associated with better five-year survival (adjusted HR 0.46) (Caruso et al., 2025)

8.2 Four principles for choosing quality indicators

1. No comparison without risk adjustment. In surgery for central nervous system tumours, the risk-standardised mortality ratio and hospital volume classified different centres as “good”; high volume does not always mean low risk (Chalif et al., 2024).
2. Feedback, not measurement, brings improvement. In a regional surgical quality programme in the USA, after institutions received detailed feedback on rectal cancer measures, use of preoperative MRI rose from 32% first to 88% and then to 92%, and the rate of total mesorectal excision from 47% to 68% and 80% (Charboneau et al., 2025). In the Australian lung cancer clinical quality registry, 25 indicators in 12,121 patients were met across a very wide range between institutions (2.1% to 90%), and meeting most of the indicators was associated with a lower hazard of death (Stirling et al., 2025).
3. Indicators can hide inequality. Most national quality frameworks do not report results broken down by socioeconomic group; stratified reporting of indicator results is recommended (T. Nash et al., 2026). In a public referral hospital in Brazil, only 35% of patients with hormone receptor-positive breast cancer started endocrine therapy within the recommended time (Lima et al., 2026).
4. The quality indicator is the counterweight of speed. In projects to shorten time, indicators such as guideline adherence, plan quality assurance, unplanned readmission and patient experience should be monitored on the same dashboard. Even in countries where quality measurement in oncology is tied to payment and reporting programmes, many patients have been shown not to receive optimal care (Kline et al., 2024).

9. Target, threshold and variance: the grammar of reading a number

An indicator gains meaning only when it is compared with a target; a variance becomes a managerial signal only when it is separated from natural fluctuation. This section deals with the two questions a manager should ask when reading a number: where does our target come from, and is the change we see real?

9.1 Where does a target come from?

Targets draw on four sources, and a good target usually uses several together:

• National or international standards. The six-week norm for surgical oncology in the Netherlands (van Vuren et al., 2024) and, in England, the standards of a specialist appointment within 2 weeks, 31 days after the decision and 62 days after referral (Quinn et al., 2024) are of this kind.
• Evidence-based thresholds. In Ontario, alongside the traditional 28-day target for colorectal cancer surgery, an inflection point of about 45 days derived from survival data has been defined (Bondzi-Simpson et al., 2025). Pathway-specific thresholds are proposed in colorectal cancer (Nguyen et al., 2026).
• The institution’s own history. The institution’s median value over the last 12 months is taken as the baseline, and a realistic rate of improvement is targeted.
• Benchmarking. As in the Swiss sarcoma network, comparison is made with real-world data from similar centres (Schärer et al., 2023).

An unrealistic target is as harmful as no target. In Northern Ireland, across a three-cycle improvement project in patients undergoing lung lobectomy, only 5.8% of patients overall met the 62-day target and 16.3% the 31-day target (Campbell et al., 2026). A target that seems impossible to reach breeds indifference in the team; intermediate targets and a step-by-step improvement plan are important for this reason.

9.2 Threshold gaming: when the target starts to measure itself

When a measure becomes a target, it changes behaviour. When the performance of hospital trusts in England against cancer waiting standards was examined, strong statistical evidence was found of a “threshold effect”, a clustering just above the target, especially for the 62-day standard tied to financial sanctions (Quinn et al., 2024). Some hospitals, that is, may have prioritised according to the target itself rather than clinical need. A classic study of the target regime in the English health system likewise showed that targets can improve performance but are open to gaming through the perception that “what is measured matters, what is not measured does not” (Bevan & Hood, 2006). For this reason, the whole distribution (median, 90th percentile, waiting time of patients exceeding the target) should be monitored rather than a single threshold, and cases “fitted” within the target should be reviewed with reason codes.

9.3 Not every change is a signal: control chart logic

Statistical process control (SPC) was developed to separate the natural fluctuation of a process (common cause) from the signals that depart from it (special cause), and it has been adapted successfully to healthcare (Benneyan et al., 2003). A control chart consists of data plotted over time, a centre line, and control limits usually placed three standard deviations above and below the mean (Mohammed et al., 2008). A physician can read this chart like the consecutive laboratory values of a patient: a single value crossing the control limit, many consecutive values lining up on the same side of the centre line, or a steadily rising trend shows that something in the system has changed.

Control charts are used in oncology in both clinical and managerial areas:

• Improving waiting time: in a high-volume paediatric neuro-oncology outpatient clinic, a process map and Pareto charts identified physician and room availability as the main causes of delay; successive PDSA cycles were monitored with run charts and Shewhart charts. The share of patients waiting more than 30 minutes to be roomed fell from 21% to 13% and mean waiting from 21 to 11 minutes (Vinitsky et al., 2022).
• Radiotherapy quality assurance: in plan-specific quality assurance, control charts are used to set the tolerance limit and the action limit (H. Y. Zhang et al., 2022; Singh et al., 2026). Patient set-up and brachytherapy plan quality are assessed with the process capability index (Cpk) (Li et al., 2021; Chen et al., 2025). In prostate brachytherapy, examining out-of-control points revealed oedema as a cause explaining dosimetric variability (Nasser et al., 2022).
• Diagnostic variability: differences in breast cancer grading between pathologists have been made visible with control charts (Moodley et al., 2020).
• Advanced methods: risk-adjusted cumulative sum (CUSUM) charts, which take patient risk into account, are being developed for monitoring breast cancer treatment processes (Rafiei, 2026).

The distinction between “tolerance limit and action limit”, borrowed from radiotherapy physics, is a powerful model for managerial indicators as well. Managers generally use three zones on their dashboards: a target zone, a warning zone and an action zone (Figure 3).

Zone	Meaning	What does the manager do?
Green: target zone	The value is on target and within control limits; fluctuation is common cause	Does not intervene, keeps monitoring, recognises the team
Amber: warning zone	The value has moved away from the target but has not crossed the action threshold, or a trend has begun	Verifies the data, splits into intervals, increases the frequency of monitoring
Red: action zone	The action threshold has been crossed, or the control chart has given a special-cause signal	Starts a root cause analysis, sets an owner and a deadline, informs the management board

9.4 Not every variance is bad, and not every difference matters

Variance analysis has long been used in the management of clinical pathways. In a clinical pathway for video-assisted lung lobectomy, variances occurred in 1.6% to 34.4% of cases depending on the activity, and complications and abnormal body mass index were independent factors that increased variance (Okita et al., 2009). An analysis of this kind helps separate the patient groups in which a variance is to be expected from those in which it points to a system problem.

Sometimes large differences make no meaningful difference to the result. In the Finnish cervical cancer screening programme, laboratories differed by factors of 3.6 to 4 in follow-up recommendation and referral rates, yet these differences did not noticeably change the effectiveness of the programme (Lönnberg et al., 2012). The manager’s task is not to intervene in every difference but to find the difference that affects the outcome.

10. From variance to managerial intervention: a six-step decision cycle

A variance is not an end but a beginning. The manager’s strength comes not from seeing the variance but from choosing the intervention that fits it and monitoring the result. The six steps below carry the path a physician follows at the bedside to the management table.

1. Verify the variance. First ask whether the number is real. Has the definition changed, are data missing, has a new branch been added? It should be remembered that electronic records are often insufficient for calculating quality measures (Schorer et al., 2022) and that adjustment rules change the result (Varma et al., 2026).
2. Disaggregate. The total value is broken down by tumour type, stage, treatment pathway, branch and interval of the patient journey (A1 to A5). The problem is usually concentrated in a single interval or a single branch.
3. Determine the type of variation. If the control chart gives a special-cause signal (a sudden jump, a one-off event), a local fix may be enough. If the variation is common cause, that is, if the system produces this result by design, the system should be redesigned rather than individuals blamed (Benneyan et al., 2003).
4. Find the root cause. The Module II programme recommends examining root causes in five dimensions: process, people, communication, capacity and resources. The fishbone diagram, the “5 whys” and the process map are the basic tools of this step. Where clinical teams blame one another, the process map moves the discussion from persons to flow.
5. Select the intervention. The intervention should fit the root cause, have evidence behind it, be feasible and be affordable. The table below offers a menu compiled from the literature.
6. Set the owner, the deadline and the monitoring. Every intervention should have an owner, an end date, a target indicator and at least one balancing measure. The control chart continues to be updated in order to see the effect.

10.1 An evidence-based menu of interventions

Root cause	Intervention	Example of effect from the literature
Delay in referral and first appointment	Fast-track diagnostic pathway	In head and neck cancer, median referral time fell from 9 to 5 days and time from referral to treatment from 50 to 43 days; no difference in survival was seen (Lorenzana et al., 2026)
The patient getting lost in the system, lack of coordination	Patient navigation	In gastrointestinal cancers in Rwanda, median time from pathology report to treatment fell from 53.5 to 32 days (Kwizera et al., 2026); in Brazil, time from diagnosis to radiotherapy shortened by about 10 days (da Silva & Postali, 2026)
Sequential, disconnected appointments across specialties	Same-day multidisciplinary clinic	In breast cancer, time from diagnosis to treatment fell from 46.9 to 35.0 days (Kobzeva-Herzog et al., 2026)
Disconnection between surgery and oncology	Joint clinic and patient pooling	Mean time from referral to lobectomy fell from 150.47 to 109.67 days, a reduction of 27% (Campbell et al., 2026)
Delay in molecular testing	Parallel specimen processing, closed-loop communication	Median time from specimen collection to report fell from 22 to 16 days (Yeo et al., 2025)
Biomarker result requested after diagnosis	Liquid biopsy before diagnosis	In advanced lung cancer, median time from diagnosis to treatment fell from 35 to 21 days (Russo et al., 2024); the question has also been tested in a prospective clinical trial (García-Pardo et al., 2023)
Comprehensive panel taking too long	Rapid single-gene test	A rapid test for EGFR shortened time to the start of treatment by 10.1 days on average (25%) (Banyi et al., 2022)
Waste and waiting in the diagnostic pathway	Lean patient pathway	In breast cancer surgery, waiting on the lean pathway was 9.3 days shorter on average than on the traditional pathway (Freeze et al., 2025); in prostate cancer, median time from biopsy to result fell from 21 to 15 days and time to the treatment consultation from 63 to 38 days (Soroush et al., 2025)
Imaging queue	Targeted additional funding and capacity	Waiting time for CT and MRI in suspected cancer fell from 17.2 to 4.2 days (Papp et al., 2023)
Morning peak in day treatment	Advance preparation and appointment redesign	Time to start of morning treatments fell from 83 to 45 minutes (Jen et al., 2023)
Low board coverage	Cloud-based, asynchronous MDT	Coverage rose from 47.85% to 79.99% (Y. Zhang et al., 2024)
Differences between institutions in quality indicators	Data reporting and feedback	Use of preoperative MRI in rectal cancer rose from 32% to 92% (Charboneau et al., 2025)
Uncertainty in capacity decisions	Scenario testing with simulation	A gain of about 20 days was projected in the lung cancer diagnostic pathway (England et al., 2021)

Two lessons follow from this table. First, most interventions that shorten time are not major investments but changes in flow and coordination. Second, process indicators improve much faster than outcome indicators: the fast-track pathway in head and neck cancer shortened times but made no significant difference to survival (Lorenzana et al., 2026). When judging the success of an intervention, the manager should allow for this time lag and follow outcome indicators over a longer window.

10.2 The output of the management decision: a one-page decision chain

The expected output of the Module II workshop is the completion, on a single page, of the following chain: Problem → Strategic goal → KPI → Root cause → Stakeholders → Intervention → Owner → Communication approach → Timeline → Expected result. This chain forms the skeleton of the note to be presented to the management board. Every link must rest on the one before it; for a problem whose root cause is defined as “communication”, for example, the only intervention proposed cannot be “purchase of a new machine”. The next section shows how this chain is completed for the case.

11. Worked case: from a missed target to a note for the management board

This case is a version of the scenario used in the Module II workshop, enriched with numbers. All values in the table are fictional, constructed for teaching, and do not belong to any real hospital.

11.1 The situation

One of the hospital’s strategic goals this year is to reduce the time from diagnosis to the start of treatment. Over the last three quarters, however, the picture is as follows:

Indicator	Quarter 1	Quarter 2	Quarter 3	Example target
Median time from diagnosis to treatment (days)	28	34	41	30 or less
Share of new patients starting treatment within 30 days	62%	51%	42%	75% or more
MDT coverage rate	88%	86%	85%	90% or more
Median time from diagnosis to MDT decision (days)	10	11	11	10 or less
Median time from MDT decision to treatment (days)	12	17	24	14 or less
On-time implementation rate of MDT decisions	81%	72%	63%	85% or more
Median time from radiotherapy decision to first fraction (days)	18	24	31	21 or less
LINAC utilization, main campus	91%	95%	98%	Set by the institution
LINAC utilization, Branch A	72%	66%	61%	Set by the institution
Patient complaints about waiting (number)	12	19	31	Decrease

The surgical team believes the delay comes from radiotherapy, the radiation oncology team from the tumour board, and medical oncology from operating theatre capacity. Branch managers say they are not involved in decisions.

11.2 Solution in six steps

1. Verify the variance. The data team confirms that definition and data source did not change over the three quarters and that the data of Branch A arrived complete. The variance is real. Time from diagnosis to treatment and the 30-day rate have moved into the action (red) zone.

2. Disaggregate. When the total time is split into intervals, the time from diagnosis to MDT decision is almost constant (10 to 11 days). The MDT coverage rate, although slightly below target, is also stable. Almost all of the increase lies in the interval from MDT decision to treatment (from 12 to 24 days). Breaking down by treatment modality shows that the greatest lengthening is in patients receiving radiotherapy. The problem, then, is not “the board does not meet” but “the board’s decision is put into practice late”. This finding ends the blame debate between the teams with data.

3. Determine the type of variation. On the weekly control chart, all of the last eight weeks of the third quarter lie above the centre line, and there is a sustained rise. This is not a one-off failure but a signal that the system has settled at a new level. The focus is therefore on system design rather than individual blame.

4. Find the root cause. The five-dimension review gives the following picture:

Dimension	Finding (fictional)
Process	It is not defined who books the radiotherapy simulation appointment after the MDT decision; the decision form has no target date field
People	A dosimetrist left the main campus and was not replaced; planning turnaround time has lengthened
Communication	Branch A physicians do not join the board; physicians at the main campus are unaware of Branch A’s capacity
Capacity	LINAC utilization at the main campus has reached 98% and the queue is growing fast; at Branch A the machine is used at 61%
Resources	There is no regular shuttle for patient transport to Branch A; patients prefer the main campus

This picture mirrors the spare capacity finding in England: a queue grows in one place while capacity stands idle in another (Aggarwal et al., 2023). Utilization approaching one hundred per cent at the main campus explains, by queueing logic, the rapid lengthening of waits (Chowdhry & Simpson, 2023).

5. Select the intervention. In terms of impact and feasibility, the team chooses the following package:

• Decision card and ownership: “responsible physician” and “target treatment date” fields are added to the MDT decision form; an oncology nurse is assigned as navigator for each decision (Kwizera et al., 2026).
• Radiotherapy referral protocol: a rule for referring clinically suitable patients to Branch A is written, a shuttle is arranged for patient transport, and Branch A physicians join the board remotely (Y. Zhang et al., 2024).
• Planning capacity: the vacant dosimetrist post is filled; temporarily, the planning team of Branch A works jointly with the main campus.
• Weekly capacity huddle: a 15-minute weekly capacity meeting is held with representatives of radiotherapy, surgery and medical oncology; the control chart is read at this meeting.
• Hypofractionation review: the use of hypofractionated schedules in clinically suitable patients is reassessed by the radiation oncology team (Lupiáñez-Pérez et al., 2017).

6. Owner, deadline and monitoring. An owner and a date are set for each intervention. The main indicator is “median time from MDT decision to treatment”. The balancing measures are the first-time pass rate of plan quality assurance at Branch A, the rate of adherence to MDT recommendations, and patient experience.

11.3 A one-page note for the management board

Link	Content
Problem	Median time from diagnosis to treatment lengthened from 28 to 41 days over three quarters; most of the increase lies in the interval from MDT decision to treatment and in patients receiving radiotherapy
Strategic goal	At least 75% of new patients start treatment within 30 days
KPI	Share starting treatment within 30 days; median time from MDT decision to treatment; median time from radiotherapy decision to first fraction
Root cause	Decisions without an owner, loss of planning workforce, and over-utilization at the main campus coexisting with spare capacity at the branch
Stakeholders	Radiation oncology, surgery, medical oncology, Branch A management, nursing, patient transport unit, patients and relatives
Intervention	Decision card and navigator, referral protocol and shuttle to Branch A, dosimetrist recruitment, weekly capacity huddle, hypofractionation review
Owners	Deputy chief physician (overall coordination), head of radiation oncology, Branch A manager, head of oncology nursing
Communication approach	Data, not blame: the disaggregated time chart is shared with all teams; the Branch A team is made a partner in the decision; the reason for referral to the branch is explained clearly to patients
Timeline	First 30 days: decision card, navigator, weekly huddle. 60 days: referral protocol and shuttle. 90 days: first evaluation
Expected result	Median time from MDT decision to treatment approaching 14 days at 90 days, and the 30-day rate approaching 75% at 6 months; higher LINAC utilization at Branch A and a shorter queue at the main campus

This note is also the template for the “Executive Improvement Project” that participants in Module II will prepare for their own hospitals.

12. Common mistakes and how to avoid them

Pitfall	What happens?	How to avoid it
1. The mean trap	Waiting times are right-skewed; a few patients with very long waits distort the mean and hide the majority	Report median, 90th percentile and share of patients within target together
2. Definition drift	The moment the clock starts and stops differs from branch to branch; numbers become incomparable (Drosdowsky et al., 2023; Tandogan et al., 2025)	Write an indicator passport and record any change of definition with its date
3. The waiting time paradox	Because sicker patients are treated quickly, the false conclusion emerges that “waiting is good” (Tope et al., 2025)	Use the time indicator as a system indicator and read its relationship with outcome stratified and risk-adjusted
4. Threshold gaming	The target overrides clinical priority; patients are “fitted” just under the target (Quinn et al., 2024; Bevan & Hood, 2006)	Monitor the whole distribution and also report the waiting time of cases exceeding the target
5. Demand response	When waiting shortens, referrals rise and the gain melts away (Hayes et al., 2024)	Include demand growth as a scenario in the capacity plan
6. Loss of quality for the sake of speed	The board is skipped or plan checks are shortened (Burkhardt et al., 2024)	Place a quality indicator next to every time indicator
7. An indicator without data	There is a number on the dashboard but it is not reliable (Schorer et al., 2022)	Test the data source and completion rate before choosing the indicator
8. Billing a common cause to a person	Individuals are blamed for a problem that stems from system design, and the team becomes defensive (Benneyan et al., 2003)	Separate common and special cause with a control chart; solve system problems with the system
9. The volume equals quality fallacy	A high-volume unit is automatically assumed to be good (Chalif et al., 2024)	Use risk-adjusted outcome indicators
10. Inequality behind the average	The delay of patients who live far away or are disadvantaged is invisible in the total (Wang et al., 2025; T. Nash et al., 2026)	Break results down by groups such as place of residence, insurance status and age
11. Impatience for outcomes	The process improves but survival does not change at once, and the project is declared a “failure” (Lorenzana et al., 2026)	Set different evaluation windows for process and outcome indicators

13. Pocket card for physicians: a gift package from performance management

This section is the summary of the chapter that can be cut out and kept on the desk.

Monday morning questions

1. Of the patients diagnosed last week, how many will start treatment within 30 days and how many are at risk?
2. Which is the longest interval in our time to treatment initiation: diagnosis, the board, or after the board?
3. How many of the decisions we made at the board have an owner and a target date?
4. Where is our LINAC, chair and theatre utilization approaching one hundred per cent, and where is it standing idle?
5. Where we speeded up this week, did we compromise on quality?

Indicator passport template

Field	To be completed
Name of the indicator
Which strategic goal is it tied to?
Definition (when does the clock start and stop?)
Numerator and denominator
Exclusions and adjustment rules
Data source and data owner
Reporting format (median, 90th percentile, rate)
Stratification (tumour, stage, treatment pathway, branch)
Target, warning limit, action limit
Balancing measure
If the action limit is crossed, what will be done and by whom?

Ten golden rules

1. The clock is a clinical finding; read it together with tumour type, stage and treatment pathway.
2. Choose few indicators that produce decisions, and give every indicator an owner.
3. Use the median and the 90th percentile, not the mean.
4. Do not make a diagnosis before splitting the total time into parts.
5. Measure the MDT not only by coverage but by time after the decision and by implementation rate.
6. When utilization approaches one hundred per cent, waiting grows rapidly; “completely full” does not always mean “efficient”.
7. Look for spare capacity first, then ask for new investment.
8. Place a quality indicator next to every time indicator.
9. For a common cause fix the system, for a special cause fix the event; target the process, not the person.
10. Process changes fast and outcome slowly; monitor patiently and feed the results back to your team.

Mini glossary

Term	Meaning in brief
Time to treatment initiation (TTI)	Days from pathological diagnosis to first definitive treatment
Key performance indicator (KPI)	A critical indicator tied directly to a strategic goal
Multidisciplinary team (MDT), tumour board	The meeting at which the treatment decision is made collectively
MDT coverage rate	Share of eligible new patients discussed at the board before treatment
MDT adherence, implementation rate	Share of board decisions actually carried out
Utilization	Service delivered as a share of capacity
Backlog	Patients in the queue who have not yet been served
Balancing measure	An indicator that checks for harm elsewhere
Common-cause and special-cause variation	Natural fluctuation of the system versus a signal from a specific event
Statistical process control (SPC)	Methods that separate signal from noise over time
Tolerance limit and action limit	The limits that mark the warning and action zones
Patient navigation	A named person guiding the patient through the pathway
Waiting time paradox	Sicker patients are treated faster, so short waits can look harmful

14. Conclusion

Performance management in an oncology hospital is the work a physician does at the bedside carried to the scale of the institution: measuring the finding correctly, making the differential diagnosis correctly, choosing the appropriate treatment and monitoring the result. Time to treatment initiation, MDT coverage and radiotherapy waiting time are the most visible findings of this work; capacity and quality indicators show the cause and the cost of those findings.

The literature is clear on three points. First, the price of delay varies with cancer type and treatment pathway; pathway-specific targets set together with the clinical teams are therefore needed in place of a single universal threshold. Second, the value of the MDT lies not only in its meeting but in the timely and complete implementation of its decision. Third, most interventions that shorten waiting times are not major investments but changes in flow, coordination and ownership.

The manager’s task is not to react to every number but to separate signal from noise and tie the right intervention to the right signal. Once this relationship between target, threshold, variance and intervention is established, the indicator ceases to be a column in a report and becomes a management decision that touches the day on which the patient reaches treatment.

References

Abdalla, R., Pavlova, M., Hussein, M., & Groot, W. (2022). Quality measurement for cardiovascular diseases and cancer in hospital value-based healthcare: A systematic review of the literature. BMC Health Services Research, 22(1), Article 979. https://doi.org/10.1186/s12913-022-08347-x

Aggarwal, A., Han, L., Sullivan, R., Haire, K., Sangar, V., & van der Meulen, J. (2023). Managing the cancer backlog: A national population-based study of patient mobility, waiting times and ‘spare capacity’ for cancer surgery. The Lancet Regional Health – Europe, 30, Article 100642. https://doi.org/10.1016/j.lanepe.2023.100642

Banyi, N., Alex, D., Hughesman, C., McNeil, K., Ionescu, D. N., Ma, C., Yip, S., & Melosky, B. (2022). Improving time-to-treatment for advanced non-small cell lung cancer patients through faster single gene EGFR testing using the Idylla EGFR testing platform. Current Oncology, 29(10), 7900-7911. https://doi.org/10.3390/curroncol29100624

Benneyan, J. C., Lloyd, R. C., & Plsek, P. E. (2003). Statistical process control as a tool for research and healthcare improvement. Quality and Safety in Health Care, 12(6), 458-464. https://doi.org/10.1136/qhc.12.6.458

Bevan, G., & Hood, C. (2006). What’s measured is what matters: Targets and gaming in the English public health care system. Public Administration, 84(3), 517-538. https://doi.org/10.1111/j.1467-9299.2006.00600.x

Bondzi-Simpson, A., Sutradhar, R., Wright, F. C., Ribeiro, T., Chan, W. C., Perera, S., Covelli, A., Lofters, A., Snyder, R. A., Clarke, C. N., Coburn, N. G., & Hallet, J. (2025). Association of traditional and new surgical wait time targets and survival for curative-intent colorectal cancer surgery. Annals of Surgery, 282(4), 608-619. https://doi.org/10.1097/SLA.0000000000006846

Burkhardt, V., El-Shabrawi, K., Riemann, S., Voss, P., & Becker, C. (2024). Multidisciplinary tumor boards in oral cavity cancer: Survival effect due to balancing guideline adherence and treatment delays. Frontiers in Oral Health, 5, Article 1493319. https://doi.org/10.3389/froh.2024.1493319

Campbell, D., Calvert, G. H. M., Mhandu, P., & Beattie, R. (2026). Countdown to lobectomy: Interventions to improve waiting times for lung cancer resection. International Journal for Quality in Health Care, 38(1), Article mzag035. https://doi.org/10.1093/intqhc/mzag035

Caruso, G., Ainio, C., Fumagalli, D., Reynolds, E. A., Nasioudis, D., Gree, M. E. M., Fought, A. J., Giuntoli, R. L., Mariani, A., & Cliby, W. A. (2025). Textbook oncologic outcome in endometrial cancer surgery: A quality measure to improve survival and address disparities. Gynecologic Oncology, 200, 33-43. https://doi.org/10.1016/j.ygyno.2025.07.009

Chalif, E. J., Young, J. S., Villa, G. R., Aghi, M. K., Lenzi, J., & Berger, M. S. (2024). High-volume facilities are not always low risk: Comparing risk-standardized mortality rates versus facility volume as quality measures in surgical neuro-oncology. Journal of Neurosurgery, 140(1), 10-17. https://doi.org/10.3171/2023.5.JNS222913

Charboneau, A. J., Cragle, C., Frankhouse, J., Kanneganti, S., Kaplan, J. A., Moonka, R., Rashidi, L., & Simianu, V. V. (2025). Impact of regional data reporting and feedback on rectal cancer surgery quality metrics in the Surgical Care Outcomes Assessment Program (SCOAP). Surgery Open Science, 24, 74-79. https://doi.org/10.1016/j.sopen.2025.03.002

Chen, X., Shi, X. X., Zhang, H. W., & Pang, H. W. (2025). Quality control study of cervical cancer interstitial brachytherapy treatment plans using statistical process control. Brachytherapy, 24(3), 371-378. https://doi.org/10.1016/j.brachy.2024.12.007

Chowdhry, V. K., & Simpson, N. C. (2023). Process modeling a radiation oncology clinic workflow from therapeutic simulation to treatment: Identifying impending strain and possible treatment delays. Advances in Radiation Oncology, 8(6), Article 101261. https://doi.org/10.1016/j.adro.2023.101261

Cohen, P. A., Nichols, C. B., Schofield, L., Van Der Werf, S., & Pachter, N. (2016). Impact of clinical genetics attendance at a gynecologic oncology tumor board on referrals for genetic counseling and BRCA mutation testing. International Journal of Gynecological Cancer, 26(5), 892-897. https://doi.org/10.1097/IGC.0000000000000689

Corsini, R. R., Costa, A., Fichera, S., & Pluchino, A. (2023). A configurable computer simulation model for reducing patient waiting time in oncology departments. Health Systems, 12(2), 208-222. https://doi.org/10.1080/20476965.2022.2030655

da Silva, A. R., & Postali, F. A. S. (2026). The Brazilian breast cancer patient navigation program: Impact on waiting times and treatment costs. Health Policy and Planning. Advance online publication. https://doi.org/10.1093/heapol/czag105

Dafflisio, G. J., Wang, M., Wang, X., Spratt, D. E., Zuhour, R., Khorana, A. A., Bilimoria, K. Y., & Zaorsky, N. G. (2025). Time to treatment initiation for the 30 most prevalent cancer types: Trends and predictors of change. JCO Oncology Practice, 22(1). https://doi.org/10.1200/OP.23.00614

Denault, M. H., Labbé, C., St-Pierre, C., Fournier, B., Gagné, A., Morillon, C., Joubert, P., Simard, S., & Martel, S. (2022). Wait times and survival in lung cancer patients across the province of Quebec, Canada. Current Oncology, 29(5), 3187-3199. https://doi.org/10.3390/curroncol29050259

Donabedian, A. (1988). The quality of care: How can it be assessed? JAMA, 260(12), 1743-1748. https://doi.org/10.1001/jama.1988.03410120089033

Drosdowsky, A., Lamb, K. E., Bergin, R. J., Boyd, L., Milley, K., IJzerman, M. J., & Emery, J. D. (2023). A systematic review of methodological considerations in time to diagnosis and treatment in colorectal cancer research. Cancer Epidemiology, 83, Article 102323. https://doi.org/10.1016/j.canep.2023.102323

Ellsworth, B. L., Metz, A. K., Mott, N. M., Kazemi, R., Stover, M., Hughes, T., & Dossett, L. A. (2022). Review of cancer-specific quality measures promoting the avoidance of low-value care. Annals of Surgical Oncology, 29(6), 3750-3762. https://doi.org/10.1245/s10434-021-11303-4

England, T. J., Harper, P. R., Crosby, T., Gartner, D., Arruda, E. F., Foley, K. G., & Williamson, I. J. (2021). Examining the diagnostic pathway for lung cancer patients in Wales using discrete event simulation. Translational Lung Cancer Research, 10(3), 1368-1382. https://doi.org/10.21037/tlcr-20-919

Fan, Q. J., Dong, W. C., Schafer, E. J., Wagle, N. S., Zhao, J. X., Shi, K. S., Han, X. S., Yabroff, K. R., & Nogueira, L. M. (2025). Changes in time-to-treatment initiation for breast, non-small cell lung, colon, or rectal cancers throughout the COVID-19 pandemic in the United States. JNCI: Journal of the National Cancer Institute, 117(7), 1506-1511. https://doi.org/10.1093/jnci/djaf011

Fefferman, M. L., Thompson, D. M., Wilke, L. G., Hwang, S., Bleicher, R., Freedman, L. M., Meisel, J. L., Kuchta, K., & Yao, K. (2025). Commission on Cancer center performance with the new breast cancer quality measures: A review of historical data. Annals of Surgical Oncology, 32(3), 2045-2055. https://doi.org/10.1245/s10434-024-16594-x

Francoeur, A. A., Chervu, N., Mardock, A. L., Solaru, S., Sakowitz, S., Benharash, P., & Lai, T. S. (2025). Association of palliative care with readmission and resource utilization in patients with ovarian cancer: A national perspective. American Journal of Hospice & Palliative Medicine, 42(10), 997-1004. https://doi.org/10.1177/10499091241301753

Frank, M. H., van Dijk, B. A. C., Schoonbeek, R. C., Zindler, J., Devriese, L. A., van Es, R. J. J., Merkx, M. A. W., & de Bree, R. (2024). Differences in the association of time to treatment initiation and survival according to various head and neck cancer sites in a nationwide cohort. Radiotherapy and Oncology, 192, Article 110107. https://doi.org/10.1016/j.radonc.2024.110107

Freeze, T. A., Hanson, N. E., Skerry, L. L., Nesbitt, M. E., Bryden, P. L., Smith, S. A., & Chiu, S. S. Y. (2025). Wait times for breast cancer surgery in lean and traditional care pathways: A multi-group path analysis. Canadian Journal of Surgery, 68(3), E175-E181. https://doi.org/10.1503/cjs.005524

Frosch, Z. A. K., Illenberger, N., Mitra, N., Boffa, D. J., Facktor, M. A., Nelson, H., Palis, B. E., Bekelman, J. E., Shulman, L. N., & Takvorian, S. U. (2021). Trends in patient volume by hospital type and the association of these trends with time to cancer treatment initiation. JAMA Network Open, 4(7), Article e2115675. https://doi.org/10.1001/jamanetworkopen.2021.15675

García-Pardo, M., Czarnecka-Kujawa, K., Law, J. H., Salvarrey, A. M., Fernandes, R., Fan, Z. J., Waddell, T. K., Yasufuku, K., Liu, G., Donahoe, L. L., Pierre, A., Le, L. W., Gunasegaran, T., Ghumman, N., Shepherd, F. A., Bradbury, P. A., Sacher, A. G., Schmid, S., Corke, L., … Leighl, N. B. (2023). Association of circulating tumor DNA testing before tissue diagnosis with time to treatment among patients with suspected advanced lung cancer: The ACCELERATE nonrandomized clinical trial. JAMA Network Open, 6(7), Article e2325332. https://doi.org/10.1001/jamanetworkopen.2023.25332

Giraud, J. S., Brun, M., Desmaris, R., Cottu, P., & Acramel, A. (2026). Reducing waiting time of injectable anticancer drug preparations in day care oncology units: A summary of evidence, implemented strategies, and perspectives for hospital pharmacists. Journal of Oncology Pharmacy Practice, 32(5), 934-945. https://doi.org/10.1177/10781552261422457

Hack, A. P., Zweemer, R. P., Jonges, T. N., van der Leij, F., Gerestein, C. G., Peters, M., Jürgenliemk-Schulz, I. M., & van Rossum, P. S. N. (2022). Prognostic impact of waiting time between diagnosis and treatment in patients with cervical cancer: A nationwide population-based study. Gynecologic Oncology, 165(2), 339-346. https://doi.org/10.1016/j.ygyno.2022.03.006

Hartman, H. E., Sun, Y. L., Devasia, T. P., Chase, E. C., Jairath, N. K., Dess, R. T., Jackson, W. C., Morris, E., Li, P., Hochstedler, K. A., Abbott, M. R., Kidwell, K. M., Walter, V., Wang, M., Wang, X., Zaorsky, N. G., Schipper, M. J., & Spratt, D. E. (2020). Integrated survival estimates for cancer treatment delay among adults with cancer during the COVID-19 pandemic. JAMA Oncology, 6(12), 1881-1889. https://doi.org/10.1001/jamaoncol.2020.5403

Hayes, H., Meacock, R., Stokes, J., & Sutton, M. (2024). How do family doctors respond to reduced waiting times for cancer diagnosis in secondary care? The European Journal of Health Economics, 25(5), 813-828. https://doi.org/10.1007/s10198-023-01626-2

Jen, W. Y., Chan, Z. Y., Lee, Y. M., Ng, N., Tan, B. L. D., Teo, C., Wong, Y. P., Chee, C. E., & Chee, Y. L. (2023). Reducing chemotherapy waiting times in the ambulatory setting of a tertiary cancer centre using a design thinking approach. Cancers, 15(18), Article 4625. https://doi.org/10.3390/cancers15184625

Kandemir, E. A., Roeper, J., Zimmermann, H., Ansmann, L., Hülper, P., Bockhorn, M., Köhne, C. H., & Griesinger, F. (2025). Adherence to multidisciplinary tumor board recommendations and its association with survival: A retrospective observational study of colorectal cancer patients. Journal of Gastrointestinal Cancer, 56(1), Article 124. https://doi.org/10.1007/s12029-025-01246-4

Kapadia, P., Zimolzak, A. J., Upadhyay, D. K., Korukonda, S., Rekha, R. M., Mushtaq, U., Mir, U., Murphy, D. R., Offner, A., Abel, G. A., Lyratzopoulos, G., Mounce, L. T. A., & Singh, H. (2024). Development and implementation of a digital quality measure of emergency cancer diagnosis. Journal of Clinical Oncology, 42(21). https://doi.org/10.1200/JCO.23.01523

Kehl, K. L., Landrum, M. B., Kahn, K. L., Gray, S. W., Chen, A. B., & Keating, N. L. (2015). Tumor board participation among physicians caring for patients with lung or colorectal cancer. Journal of Oncology Practice, 11(3), e267-e278. https://doi.org/10.1200/JOP.2015.003673

Klarenbeek, S. E., Aarts, M. J., van den Heuvel, M. M., Prokop, M., Tummers, M., & Schuurbiers, O. C. J. (2023). Impact of time-to-treatment on survival for advanced non-small cell lung cancer patients in the Netherlands: A nationwide observational cohort study. Thorax, 78(5), 467-475. https://doi.org/10.1136/thoraxjnl-2021-218059

Kline, R. M., Hughes, D. L., & Schreiber, M. (2024). Quality measurement in oncology: Time to take the next step! JCO Oncology Practice, 20(12). https://doi.org/10.1200/OP.24.00348

Kobzeva-Herzog, A. J., Palaniappan, S., Jiangliu, Y., Kraus, E., Elassar, H., Merrill, A., Oshry, L., Acevedo, J., Ko, N., & Cassidy, M. R. (2026). A multidisciplinary breast cancer clinic improves time to treatment at an urban, safety net hospital. Annals of Surgical Oncology, 33(2), 1171-1179. https://doi.org/10.1245/s10434-025-18497-x

Koster, K. L., Yohasenan, S., Pakmak, A. S., Mark, M., Metaxas, Y., Lips, C., Hitz, F., Leskow, P., Ludwig, C., Putora, P. M., Glatzer, M., Schneider, T., Caviezel, C., Mader, T., Mousavi, M., Blum, M., Früh, M., & Joerger, M. (2025). Predictors of multidisciplinary tumor board adherence in stage III non-small-cell lung cancer patients from a large multicenter study. Clinical Lung Cancer, 26(7), e487-e495. https://doi.org/10.1016/j.clc.2025.05.010

Krause, A., Stocker, G., Gockel, I., Seehofer, D., Hoffmeister, A., Bläker, H., Denecke, T., Kluge, R., Lordick, F., & Knödler, M. (2023). Guideline adherence and implementation of tumor board therapy recommendations for patients with gastrointestinal cancer. Journal of Cancer Research and Clinical Oncology, 149(3), 1231-1240. https://doi.org/10.1007/s00432-022-03991-6

Kut, C., Chang, L., Hales, R. K., Voong, K. R., Greco, S., Halthore, A., Alcorn, S. R., Song, D., Briner, V., McNutt, T. R., Viswanathan, A. N., & Wright, J. L. (2023). Improving quality metrics in radiation oncology: Implementation of pretreatment peer review for stereotactic body radiation therapy in patients with thoracic cancer. Advances in Radiation Oncology, 8(2), Article 101004. https://doi.org/10.1016/j.adro.2022.101004

Kwizera, V., Aihnoa, C. C., Rutikanga, K. K., Kakwezi, J., Nyiraneza, M., Mizero, A. F., Mutetiwabo, I., Izabiriza, C., Rugengamanzi, E., Saeed, S., & Rubagumya, F. (2026). Impact of patient navigation on time to treatment among patients with GI cancers at Rwanda Military Referral and Teaching Hospital. JCO Global Oncology, 12, Article e2500339. https://doi.org/10.1200/GO-25-00339

Lam, C. H. M., Mason, T., & Hollingsworth, B. (2026). Measuring performance in cancer services during COVID-19 and beyond: A novel case throughput metric using NHS Cancer Waiting Times data. Journal of Public Health, 48(3), 754-758. https://doi.org/10.1093/pubmed/fdag041

Li, Z. B., Xiao, Q., Li, G. J., Wu, X. H., Zhang, Y. J., Wang, G. Y., Dai, G. Y., Bai, L., Wang, Q., & Bai, S. (2021). Performance assessment of surface-guided radiation therapy and patient setup in head-and-neck and breast cancer patients based on statistical process control. Physica Medica, 89, 243-249. https://doi.org/10.1016/j.ejmp.2021.08.007

Lima, A. D. N., de Aguiar, S. S., Bergmann, A., Bines, J., & Thuler, L. C. S. (2026). Breast cancer treatment delays in the Brazilian public health system: Evaluation using international and national quality metrics. Cancer Epidemiology, 104, Article 103163. https://doi.org/10.1016/j.canep.2026.103163

Loftus, L., Laronga, C., Coyne, K., & Hildreth, L. (2014). Race of the clock: Reducing delay to curative breast cancer surgery. Journal of the National Comprehensive Cancer Network, 12, S13-S15.

Lönnberg, S., Nieminen, P., Kotaniemi-Talonen, L., Kujari, H., Melkko, J., Granroth, G., Vornanen, M., Pietiläinen, T., Arola, J., Tarkkanen, J., Luostarinen, T., & Anttila, A. (2012). Large performance variation does not affect outcome in the Finnish cervical cancer screening programme. Cytopathology, 23(3), 172-180. https://doi.org/10.1111/j.1365-2303.2011.00849.x

Lorenzana, P. R., Poggioli, N. G., Soilán, L. A., Nogueira, A. C., Rodríguez, H. G., Yáñez, M. M., & Arias, P. P. (2026). Impact of fast-track pathway implementation on time to diagnosis, stage at diagnosis, and overall survival in head and neck cancer. Clinical & Translational Oncology. Advance online publication. https://doi.org/10.1007/s12094-026-04438-x

Lowes, M., Kleiss, M., Lueck, R., Detken, S., Koenig, A., Nietert, M., Beissbarth, T., Stanek, K., Langer, C., Ghadimi, M., Conradi, L. C., & Homayounfar, K. (2017). The utilization of multidisciplinary tumor boards (MDT) in clinical routine: Results of a health care research study focusing on patients with metastasized colorectal cancer. International Journal of Colorectal Disease, 32(10), 1463-1469. https://doi.org/10.1007/s00384-017-2871-z

Lupiáñez-Pérez, Y., Gómez-Millán, J., Lobato, M., Pedrosa, P., Lupiáñez-Pérez, I., & Medina, J. A. (2017). Improving efficacy and efficiency through the implementation of a new organisational model in a radiation oncology department. Clinical & Translational Oncology, 19(11), 1375-1381. https://doi.org/10.1007/s12094-017-1678-2

Malagón, T., Yong, J. H. E., Tope, P., Miller, W. H., Jr., & Franco, E. L. (2022). Predicted long-term impact of COVID-19 pandemic-related care delays on cancer mortality in Canada. International Journal of Cancer, 150(8), 1244-1254. https://doi.org/10.1002/ijc.33884

Mohammed, M. A., Worthington, P., & Woodall, W. H. (2008). Plotting basic control charts: Tutorial notes for healthcare practitioners. Quality and Safety in Health Care, 17(2), 137-145. https://doi.org/10.1136/qshc.2004.012047

Moodley, J., Williams, P., Gohla, G., Major, P., & Bonert, M. (2020). Variation in breast cancer grading in 1,636 resections assessed using control charts and in silico kappa. PLOS ONE, 15(12), Article e0242656. https://doi.org/10.1371/journal.pone.0242656

Naessens, C., Laloze, J., Leobon, S., Gauthier, T., Lacorre, A., Monteil, J., Venat, L., & Deluche, E. (2023). Physician compliance with multidisciplinary tumor board recommendations for managing gynecological cancers. Future Oncology, 19(13), 897-908. https://doi.org/10.2217/fon-2022-1183

Nash, J., Stone, E., Vinod, S., Leong, T., Dawkins, P., Stirling, R. G., Harden, S., Bolton, A., McWilliams, A., O’Byrne, K., Wright, G. M., Brunelli, V. N., Guan, T. C., Philpot, S., Navani, N., & Brims, F. (2024). Lung cancer (internet-based) Delphi (LUCiD): A modified eDelphi consensus process to establish Australasian clinical quality indicators for thoracic cancer. Respirology, 29(12), 1085-1094. https://doi.org/10.1111/resp.14812

Nash, T., Lemmon, E., Katikireddi, V., & Hall, P. (2026). Use of routine quality metrics to describe disparities in cancer care: A narrative review & thematic analysis. Journal of Cancer Policy, 50, Article 100797. https://doi.org/10.1016/j.jcpo.2026.100797

Nasser, N. J., Saibishkumar, E. P., Wang, Y. K., Chung, P. W. M., & Breen, S. L. (2022). Control charts for evaluation of quality of low-dose-rate brachytherapy for prostate cancer. Journal of Contemporary Brachytherapy, 14(4), 354-363. https://doi.org/10.5114/jcb.2022.119513

Nguyen, C. M., Skaar, T. C., Imperiale, T. F., Johnson, T. S., Monahan, P. O., Turk, A. A., & Nguyen, M. P. (2026). Colorectal cancer treatment delay thresholds and metastasis risk. JAMA Network Open, 9(7), Article e2623057. https://doi.org/10.1001/jamanetworkopen.2026.23057

Nicolai, N., Bogani, G., Agresti, R., Audisio, R. A., Beretta, P., Biondi, A., Bocciolone, L., Boni, L., Cabitza, F., Callegaro, D., Cananzi, F. C., Carnelli, M., Casali, P. G., Casarin, J., Castiglioni, I., Coccolini, F., Cortinovis, U., D’Amico, F. E., De Braud, F., … Gronchi, A. (2026). Cutting-edge organization: A consensus-informed framework in surgical oncology. EJSO, 52(12), Article 112127. https://doi.org/10.1016/j.ejso.2026.112127

Okita, A., Yamashita, M., Abe, K., Nagai, C., Matsumoto, A., Akehi, M., Yamashita, R., Ishida, N., Seike, M., Yokota, S., Umekawa, N., Matsumoto, Y., Kishimoto, Y., Okazaki, A., Komori, E., Sawada, S., & Takashima, S. (2009). Variance analysis of a clinical pathway of video-assisted single lobectomy for lung cancer. Surgery Today, 39(2), 104-109. https://doi.org/10.1007/s00595-008-3821-8

Papp, Z. M., Szakács, L., Hajivandi, S. S., Kalina, I., Varga, E., Kiss, G., Solymos, F., Takács, I., Dank, M., Dudás, I., Szanka, T., Dózsa, C. L., Rékassy, B., Merkely, B., & Maurovich-Horvat, P. (2023). Impact of a targeted project for shortening of imaging diagnostic waiting time in patients with suspected oncological diseases in Hungary. Medicina, 59(1), Article 153. https://doi.org/10.3390/medicina59010153

Park, J., Puckett, L. L., Katsoulakis, E., Venkatesulu, B. P., Kujundzic, K., Solanki, A. A., Movsas, B., Simone, C. B., II, Sandler, H., Lawton, C. A., Das, P., Wo, J. Y., Buchholz, T. A., Fisher, C. M., Harrison, L. B., Sher, D. J., Kapoor, R., Chapman, C. H., Dawes, S., … Kelly, M. D. (2022). Veterans Affairs Radiation Oncology Quality Surveillance Program and American Society for Radiation Oncology Quality Measures Initiative. Practical Radiation Oncology, 12(6), 468-474. https://doi.org/10.1016/j.prro.2022.05.015

Quinn, L., Bird, P., & Lilford, R. (2024). Effect of cancer waiting time standards in the English National Health Service: A threshold analysis. BMC Health Services Research, 24(1), Article 929. https://doi.org/10.1186/s12913-024-11350-z

Radheshyam, A., Ramani, V. K., Mhaske, M., & Naik, R. (2024). Decreasing patient wait time by process level interventions in an oncology daycare unit: A quasi-experimental study. Health Science Reports, 7(12), Article e70224. https://doi.org/10.1002/hsr2.70224

Rafiei, N. (2026). A novel log-logistic risk-adjusted CUSUM control chart for monitoring therapeutic processes in breast cancer patients. Sequential Analysis, 45(3), 533-559. https://doi.org/10.1080/07474946.2026.2634402

Raphael, M. J., Saskin, R., & Singh, S. (2020). Association between waiting time for radiotherapy after surgery for early-stage breast cancer and survival outcomes in Ontario: A population-based outcomes study. Current Oncology, 27(2), e216-e221. https://doi.org/10.3747/co.27.5629

Ricciardi, R., Goldstone, R. N., Francone, T., Wszolek, M., Auchincloss, H., de Groot, A., Shih, I. F., & Li, Y. L. (2022). Healthcare resource utilization after surgical treatment of cancer: Value of minimally invasive surgery. Surgical Endoscopy, 36(10), 7549-7560. https://doi.org/10.1007/s00464-022-09189-8

Richardson, B., Preskitt, J., Lichliter, W., Peschka, S., Carmack, S., de Prisco, G., & Fleshman, J. (2016). The effect of multidisciplinary teams for rectal cancer on delivery of care and patient outcome: Has the use of multidisciplinary teams for rectal cancer affected the utilization of available resources, proportion of patients meeting the standard of care, and does this translate into changes in patient outcome? The American Journal of Surgery, 211(1), 46-52. https://doi.org/10.1016/j.amjsurg.2015.08.015

Russo, A., Lee, J. K., Pasquina, L. W., Del Re, M., Dilks, H. H., Murugesan, K., Madison, R. W., Lee, Y., Schrock, A. B., Comment, L., Dietrich, M., Oxnard, G. R., & Rolfo, C. (2024). Liquid biopsy of lung cancer before pathological diagnosis is associated with shorter time to treatment. JCO Precision Oncology, 8, Article e2300535. https://doi.org/10.1200/PO.23.00535

Saini, G., Padhi, S., Patro, K. C., Shukla, R., Shukla, S. K., Arora, D., Singh, T. R., Kundu, C., Bhattacharya, P. S., Krishna, V., & Madhur, P. (2022). Daily waiting time management for modern radiation oncology department in Indian perspective. Journal of Cancer Research and Therapeutics, 18(6), 1796-1800. https://doi.org/10.4103/jcrt.JCRT_1481_20

Sanchez, J. I., Doose, M., Zeruto, C., Chollette, V., Gasca, N., Singla, A., & Weaver, S. J. (2024). Assessing the impact of multidisciplinary cancer consultations on cancer quality metrics. American Journal of Clinical Oncology, 47(12), 595-606. https://doi.org/10.1097/COC.0000000000001136

Schärer, M., Heesen, P., Bode-Lesniewska, B., Studer, G., & Fuchs, B. (2023). Benchmarking time-to-treatment initiation in sarcoma care using real-world-time data. Cancers, 15(24), Article 5849. https://doi.org/10.3390/cancers15245849

Schorer, A. E., Moldwin, R., Koskimaki, J., Bernstam, E. V., Venepalli, N. K., Miller, R. S., & Chen, J. L. (2022). Chasm between cancer quality measures and electronic health record data quality. JCO Clinical Cancer Informatics, 6, Article e2100128. https://doi.org/10.1200/CCI.21.00128

Schroeder, M. C., Gao, X., Lizarraga, I., Kahl, A. R., & Charlton, M. E. (2022). The impact of Commission on Cancer accreditation status, hospital rurality and hospital size on quality measure performance rates. Annals of Surgical Oncology, 29(4), 2527-2536. https://doi.org/10.1245/s10434-021-11304-3

Singh, P., Singh, M. K., & Mishra, A. (2026). Use of statistical process control to study the patient-specific quality assurance in head and neck cancer. World Journal of Clinical Oncology, 17(4), Article 115851. https://doi.org/10.5306/wjco.v17.i4.115851

Soon, J. J. Y., Zhao, Y., Shannon, N. B., & Tan, J. T. H. (2023). Adherence to multidisciplinary tumor board recommendations in patients with curable esophageal and gastric cancers. Journal of Gastrointestinal Cancer, 54(2), 614-622. https://doi.org/10.1007/s12029-022-00847-7

Soroush, S., Lim, S., Beniwal, P., Wei, G. V., Lu, Y., Lim, K. Y. Y., Holden, K., Harper, M., Donnellan, S., & Ranasinghe, W. (2025). Optimising prostate cancer pathways: Improving post-biopsy waiting times in a tertiary centre. BJUI Compass, 6(7), Article e70045. https://doi.org/10.1002/bco2.70045

Stalfors, J., Björholt, I., & Westin, T. (2005). A cost analysis of participation via personal attendance versus telemedicine at a head and neck oncology multidisciplinary team meeting. Journal of Telemedicine and Telecare, 11(4), 205-210. https://doi.org/10.1258/1357633054068892

Steventon, L., Kipps, E., Man, K. K. C., Roylance, R., Forster, M. D., Wong, I. C. K., Baser, M., Miller, R. E., Nicum, S., Shah, S., Almossawi, O., & Chambers, P. (2024). The impact of inter-cycle treatment delays on 5-year all-cause mortality in early-stage breast cancer: A retrospective cohort study. European Journal of Cancer, 210, Article 114301. https://doi.org/10.1016/j.ejca.2024.114301

Stirling, R. G., Samankula, U., Lloyd, M., Brand, M., Briggs, L., Lesage, J., Wood, T., Underhill, C., Parakh, S., Faisal, W., Blum, R., Richardson, G., Parente, P., Caldecott, M., Olesen, I., Torres, J., Samuel, E., Jennings, B., See, K., … Zalcberg, J. (2025). Impacts of a clinical quality registry on lung cancer quality measures: A retrospective observational study of the Victorian Lung Cancer Registry. Clinical Oncology, 44, Article 103878. https://doi.org/10.1016/j.clon.2025.103878

Takvorian, S. U., Anderson, R. T., Gabriel, P. E., Poznyak, D., Lee, S., Simon, S., Barrett, K., & Shulman, L. N. (2022). Real-world adherence to patient-reported outcome monitoring as a cancer care quality metric. JCO Oncology Practice, 18(9). https://doi.org/10.1200/OP.21.00855

Tandogan, C., Berkesoglu, M., Tuncel, F., Yildirim, D. D., Özcan, C., Benli, S., Güler, E., & Yilmaz, E. B. (2025). Measuring surgical waiting times in breast cancer: Admission to surgery versus biopsy result to surgery. Healthcare, 13(23), Article 3010. https://doi.org/10.3390/healthcare13233010

Tang, A. D., Ahmad, U., Raja, S., Bribriesco, A. C., Sudarshan, M., Rappaport, J., Khorana, A., Blackstone, E. H., Murthy, S. C., & Raymond, D. P. (2023). How much delay matters? How time to treatment impacts overall survival in early stage lung cancer. Annals of Surgery, 277(4), e941-e947. https://doi.org/10.1097/SLA.0000000000005307

Tope, P. R., Goncalves, B. P., El-Zein, M., & Franco, E. L. (2025). The health-related impact of disruptions in cancer care and the Waiting Time Paradox. American Journal of Epidemiology, 194(10), 2928-2935. https://doi.org/10.1093/aje/kwaf128

Tran, M., Chera, B., Shrake, K., Depaoli, B., Duffy, E., III, Hall, M., Steinman, J., Myers, S., Igiebor, O., Callahan, J., McDonald, D., Harper, J., & Cooper, S. L. (2025). Reducing patient care delays in radiation oncology via optimization of insurance preauthorization. Practical Radiation Oncology, 15(2), e100-e108. https://doi.org/10.1016/j.prro.2024.10.002

Ungvari, Z., Fekete, M., Buda, A., Lehoczki, A., Fekete, J. T., Munkácsy, G., Varga, P., Ungvari, A., & Győrffy, B. (2026). No detectable impact of short-term treatment delays on lung cancer survival. GeroScience, 48(1), 793-805. https://doi.org/10.1007/s11357-025-01684-9

Ungvari, Z., Fekete, M., Buda, A., Lehoczki, A., Munkácsy, G., Scaffidi, P., Bonaldi, T., Fekete, J. T., Bianchini, G., Varga, P., Ungvari, A., & Győrffy, B. (2026). Quantifying the impact of treatment delays on breast cancer survival outcomes: A comprehensive meta-analysis. GeroScience, 48(1), 1173-1187. https://doi.org/10.1007/s11357-025-01719-1

Ungvari, Z., Fekete, M., Fekete, J. T., Lehoczki, A., Buda, A., Munkácsy, G., Varga, P., Ungvari, A., & Győrffy, B. (2025). Treatment delay significantly increases mortality in colorectal cancer: A meta-analysis. GeroScience, 47(3), 5337-5353. https://doi.org/10.1007/s11357-025-01648-z

van Vuren, R. M. G., Janssen, Y. F., Hogenbirk, R. N. M., de Graaff, M. R., van den Hoek, R., Kruijff, S., Heineman, D. J., van der Plas, W. Y., & Wouters, M. W. J. M. (2024). The impact of the COVID-19 pandemic on time to treatment in surgical oncology: A national registry study in the Netherlands. Cancers, 16(9), Article 1738. https://doi.org/10.3390/cancers16091738

Varma, R., Mayne, E., Parry, M. A., Nathan, A., van der Meulen, J., Bahl, A., Stewart, G. D., & Cowling, T. E. (2026). Comparing the national cancer waiting times dataset with other linked electronic health records to measure treatment timeliness: A national cohort study of kidney cancer in England. Cancer Epidemiology, 102, Article 103030. https://doi.org/10.1016/j.canep.2026.103030

Vinitsky, A., David, B., Michalik, L., Ramirez, N., Risinger, A., Burlison, J. D., Zanders, J., Mans, B., Heady, K., Holdiness, J., Qaddoumi, I., Robinson, G. W., Moreira, D., Upadhyaya, S. A., & Gajjar, A. (2022). Reducing wait time in a high-volume pediatric neuro-oncology clinic by optimizing process flow: A quality improvement project. Pediatric Quality & Safety, 7(3), Article e557. https://doi.org/10.1097/pq9.0000000000000557

Vulpe, H., Tree, A., Price, P., Abelson, D., Aggarwal, A., & Sullivan, R. (2026). The UK’s radiotherapy workforce shortage: Reforming regulation to address the gap. European Journal of Cancer, 248, Article 117043. https://doi.org/10.1016/j.ejca.2026.117043

Walter, J., Moeller, C., Resuli, B., Kauffmann-Guerrero, D., Manapov, F., Dinkel, J., Neumann, J., Kovacs, J., Schneider, C., Huber, R. M., & Tufman, A. (2023). Guideline adherence of tumor board recommendations in lung cancer and transfer into clinical practice. Journal of Cancer Research and Clinical Oncology, 149(13), 11679-11688. https://doi.org/10.1007/s00432-023-05025-1

Wang, Y. Z., Quinones, C. M., Gonzalez, E., Farmah, P., Schoellhammer, H. F., Gonzalez, L., Shah, N., Schulz-Costello, K., Tseng, J., & Jones, V. C. (2025). Impact of travel burden on timeliness of care and overall survival for breast cancer: A National Cancer Database analysis. Cancer Medicine, 14(21), Article e71354. https://doi.org/10.1002/cam4.71354

Watanabe, T., Rikitake, R., Kakuwa, T., Ichinose, Y., Niino, M., Mizushima, Y., Ota, M., Fujishita, M., Tsukada, Y., & Higashi, T. (2023). Time to treatment initiation for six cancer types: An analysis of data from a nationwide registry in Japan. World Journal of Surgery, 47(4), 877-886. https://doi.org/10.1007/s00268-022-06883-5

Yang, X. X., Huang, J. H., Zhu, X. P., Shen, K. W., Zhu, J. Y., & Chen, X. S. (2020). Compliance with multidisciplinary team recommendations and disease outcomes in early breast cancer patients: An analysis of 4501 consecutive patients. The Breast, 52, 135-145. https://doi.org/10.1016/j.breast.2020.05.008

Yeo, S., Sun, J., Zibrik, K., Shaipanich, T., Carolan, H., Melosky, B., Ionescu, D., Alex, D., Yip, S., & Wang, Y. (2025). Reduced wait times for molecular-biomarker testing among patients with advanced lung cancer using parallel sample processing and closed-loop communication: The Continuous Oncopanel and ALK Status Tracking (COAST) Project. BMJ Open Quality, 14(1), Article e003001. https://doi.org/10.1136/bmjoq-2024-003001

Zhang, H. Y., Lu, W. L., Cui, H. X., Li, Y., & Yi, X. (2022). Assessment of statistical process control based DVH action levels for systematic multi-leaf collimator errors in cervical cancer RapidArc plans. Frontiers in Oncology, 12, Article 862635. https://doi.org/10.3389/fonc.2022.862635

Zhang, J. R., IJzerman, M. J., Oberoi, J., Karnchanachari, N., Bergin, R. J., Franchini, F., Druce, P., Wang, X. F., & Emery, J. D. (2022). Time to diagnosis and treatment of lung cancer: A systematic overview of risk factors, interventions and impact on patient outcomes. Lung Cancer, 166, 27-39. https://doi.org/10.1016/j.lungcan.2022.01.015

Zhang, Y., Li, J., Liao, M., Yang, Y. L., He, G., Zhou, Z. H., Feng, G., Gao, F., Liu, L. H., Xue, X. J., Liu, Z. L., Wang, X. Y., Shi, Q. L., & Du, X. B. (2024). Cloud platform to improve efficiency and coverage of asynchronous multidisciplinary team meetings for patients with digestive tract cancer. Frontiers in Oncology, 13, Article 1301781. https://doi.org/10.3389/fonc.2023.1301781