NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Collaborating Centre for Cancer (UK). Neutropenic Sepsis: Prevention and Management of Neutropenic Sepsis in Cancer Patients. London: National Institute for Health and Clinical Excellence (NICE); 2012 Sep. (NICE Clinical Guidelines, No. 151.)
Neutropenic Sepsis: Prevention and Management of Neutropenic Sepsis in Cancer Patients.
Show detailsEconomic Plan
This document identifies the priorities for economic analysis and the proposed methods for addressing these questions as described in section 7.1.3 of the Guidelines Manual (2009).
Guideline
Full title of guideline: Neutropenic sepsis: Prevention and management of neutropenic sepsis in cancer patients (short: Neutropenic sepsis)
Process for agreement
The economic plan was prepared by the guideline economist in consultation with the rest of the NCC technical team and GDG. It was discussed and agreed on 23/03/2011 by the following people a:
For the NCC and GDG
| NCC economist: | Huajie Jin |
|---|---|
| NCC representative(s) b: | John Graham, Lianne Black, Nathan Bromham |
| GDG representative(s) c: | Barry Hancock, Bob Phillips |
- b
May be the project manager, a systematic reviewer or research fellow and/or the centre director or manager, as appropriate for the NCC and guideline.
- c
May be GDG chair, clinical lead and/or other members as appropriate.
For NICE (completed by NICE)
| CCP lead: | Sharon Summers-Ma |
|---|---|
| Commissioning manager: | Claire Turner |
| Economic lead: | Prashanth Kandaswamy |
| Costing lead: |
Proposals for any changes to the agreed priorities will be circulated by email to this group. If substansive revisions are agreed, they will require to be recorded as addenda to this document (section 7) or as an updated version of the document.d
Topic priorities identified in the Scope
This section contains all topics covered by the scope. These topics usually reflect selected clinical issues. Please indicate if an area is relevant for economic consideration and if modelling is deemed appropriate to address it.
| Aread | Relevant?e | Appropriate for modelling?f |
|---|---|---|
| Topic A: | Not applicable | N/A |
| Signs and symptoms in people with suspected neutropenic sepsis in the community that necessitate referral to secondary/tertiary care. | This topic does not lend itself to economic evaluation (no comparative analysis of cost and outcomes). | |
| Topic B: | Not applicable | N/A |
| Education and support for patients and carers on the identification of neutropenic sepsis. | This topic does not lend itself to economic evaluation (no comparative analysis of cost and outcomes). | |
| Topic C: | Medium | The feasibility of building a model on this topic is hampered by
|
| Emergency assessment in secondary/tertiary care of a person with suspected neutropenic sepsis. | This question is about patients in secondary or tertiary care with suspected neutropenic sepsis There is uncertainty over the usefulness of emergency assessment before treatment. Doing an assessment first could avoid over-treatment and guide the subsequent treatment strategy; but it may also cause a delay in treatment and thus increase the risk to the patients. Despite the importance of this topic, it would be impossible to measure the cost of treatments because there is no clear definition of ‘treatment’. The GDG thinks that the choice of treatment will depend on each patient's individual health status so it would be difficult to define a standard treatment for all patients. Therefore no economic analysis will be done for this topic. A cost impact analysis will be conducted at the completion of this guideline. Unit data cost will be presented during the GDG meeting if appropriate. | |
| Topic D: | D1: Not applicable D2: Low | D1: This topic is about the definition of neutropenic sepsis and does not lend itself to economic evaluation (no comparative analysis of cost and outcomes). D2: This topic is about identification of patients who are at high risk of an adverse outcome. Patients with poor prognosis outcome will be provided with more aggressive management and intensive monitoring. However, management of high-risk patients by intensive/critical care units is beyond the remit of this guideline. Therefore no economic analysis will be conducted for this topic. Unit cost of each test will be presented during GDG meeting if appropriate. |
| Appropriate initial investigations of suspected infection in a neutropenic patient in secondary care | This topic does not lend itself to economic evaluation (no comparative analysis of cost and outcomes). | |
| Topic E: | Medium | A preliminary search of the economic literature suggested a small number of economic evaluations have been published for E2 and E8, but not all may be relevant to the UK healthcare setting. Few papers have been identified for other questions within topic E. Therefore the feasibility of doing any models for topic E will be hampered by lack of evidence. Summary of approach for Topic E:
|
| Risk stratification and management of suspected bacterial infection | Topic E covers a range of clinical questions related to the management of neutropenic sepsis. Many of the specific clinical questions within this topic are unlikely to be answered by existing economic studies in the literature. Question E1 on the use of risk stratification algorithms is one such example. While it may be possible to evaluate different risk stratification algorithms based on ease of use in clinical practice and accuracy of predicting patient prognosis, a comparative analysis of the impact of choosing one risk stratification algorithm on actual patient outcomes (such as mortality or QALYs) would require data not only on the accuracy of the risk stratification algorithm, but also on the case-mix of patients and their long-term health outcomes. This type of analysis is of questionable relevance as well as feasibility for de novo modelling. Several clinical questions within Topic E relate specifically to optimal timing of a change in management strategy (E4, E5, E6). The difference between strategies being compared in each of these questions are unlikely to lead to large differences in cost, but rather may be guided by differences in patient outcomes and other considerations such as service configuration that may be difficult to accurately capture using economic modelling. The questions within Topic E that were considered to have the most relevance for cost and healthcare resource use are those related to alternative management strategies involving inpatient care (E2, E3 and E8). There was considerable discussion with the GDG about the potential for undertaking economic modelling for E2 and E8. Importantly, it was noted that such studies examine different definitions of what constitutes inpatient care or duration of care, making it difficult to generalise findings across studies. As there is no definition of what constitutes a specific inpatient management strategy for this question, costing and evaluating health outcomes using economic modelling will not be feasible. Rather, the GDG anticipated that the different management strategies are unlikely to result in large differences in terms of patient outcomes and those strategies that minimise or reduce the duration of inpatient care will generally be less costly, therefore the level of uncertainty surrounding this question is low and may be adequately answered by a simple cost impact analysis rather than formal economic modelling. There is uncertainty over the use of monotherapy or combination therapy for patients with neutropenic sepsis. Monotherapy has potential advantages over combination therapy. These could include cost, avoidance of the side effects and need for monitoring of drug levels associated with aminoglycosides (aminoglycosides is one important component of combination therapy, and is associated with kidney or inner ear toxicity). Despite this, combination regimens are still widely employed. There are additional reasons why aminoglycosides may still be used, including concerns about secondary infection with clostridium difficile and emerging forms of antibiotic resistance. In addition, particular subgroups of patients may fare better with combination therapy and local knowledge of microbiological flora may also affect treatment choices.There is relatively small difference in cost between the competing alternatives. Therefore on balance, this topic is considered a medium priority for economic analysis. Unit data cost of relevant topics of E will be presented during the GDG meeting if appropriate. | |
| Topic F: | High | Separate models will be built for F1 (primary) and F2 (secondary) prophylaxis, as the GDG think the targeted population and interventions of interest are different for these two questions; and the primary strategy prophylaxis won't affect the choice of secondary prophylaxis strategy. Many economic analyses have been identified for both primary and secondary prophylaxis. However the validity of applicability of those analyses would need to be confirmed with the GDG. If the GDG thinks none of them could be directly used/adapted to the NHS setting, then a de novo decision tree will be developed for this topic. |
| Primary and secondary prophylaxis with GCSF | There is great uncertainty over the use of G-CSF and/or antibiotics in the prevention of neutropenic sepsis. G-CSF is used to raise neutrophil counts, and shorten the duration of neutropaenia, by stimulating the bone marrow to produce neutrophils. However, adverse effects include diarrhoea, weakness, a flu-like syndrome, and rarely more serious complications such as clotting disorders and capillary leak syndrome. What's more, GCSF must be given by injection, and this may lead to local reactions at the site of administration, and repeated injections may not be desired by patients. Depot formulations are available but expensive. Pre-emptive use of oral antibiotics could reduce the likelihood of infection, but may incur patient-related risks of gut disturbance, allergy, etc and more general risks related to the development of antibiotic resistance in populations. Patients with a prior episode of significant neutropaenia are likely to become more neutropaenic with repeated doses of chemotherapy, putting them at greater risk of neutropaenic sepsis than patients who have never experienced this complication. The trade-off of using G-CSF and antibiotics as secondary prophylaxis is similar to primary prophylaxis. Considering the overall importance of this topic, characterized by a large patient group and potentially significant difference in cost, this topic is highlighted as high priority. | |
| Topic G: | High | For topic G, no economic or clinical evidence has been identified from a cursory search. The GDG is not aware of any direct relevant economic or clinical evidence either. In the absence of direct relevant evidence, the GDG has been asked if they feel confident to make assumptions of key parameters; or can we ‘borrow’ data from similar settings. Their answers to both questions are ‘No’. Therefore despite the importance of topic G, no economic models will be built for this topic due to paucity of data. |
| Empiric glycopeptides antibiotics | Central venous catheter (central line) is commonly used in cancer patients, but may introduce bacteria into the bloodstream and thus cause potentially life-threatening infection. The difficult question for the clinician is: for cancer patients with central line who are suspicious of/ diagnosed as neutropenic sepsis with unknown bacteria, should empiric antibiotics be added in addition to first line antibiotics? Trade off could be important because there are monitoring costs and toxicities associated with glycopeptides antibiotics. Considering the overall importance of this topic, characterized by a large patient subgroup and potentially significant difference in cost, this topic is highlighted as high priority. | |
| Topic H: | Not applicable | N/A |
| Indications for removing central line | This topic does not lend itself to economic evaluation (no comparative analysis of cost and outcomes). | |
| Topic I: | Not applicable | N/A |
| Information for patients and carers | This topic does not lend itself to economic evaluation (no comparative analysis of cost and outcomes). | |
| Topic J: | Low | N/A |
| Training of healthcare professionals | Whilst there are potential implications for health benefits from the interventions of interest, these are likely to be small and will be difficult to attribute to the training of healthcare professionals. Economic analysis is therefore not appropriate for this question. | |
- d
This corresponds to the “Key clinical issues that will be covered ” section in the scope
- e
Please state if this area is deemed relevant for considering opportunity costs and likely disinvestments. Areas might pose a decision problem directly or implicitly infomr the choice between options. Categories should include information on relevance and if of high or low priority for health economic work (see below).
- f
Health economic work comprises literature reviews, qualitative consideration of expected costs and effects and/or formal decision modelling. Decision modelling is particularly useful where it can reduce uncertainty over cost effectiveness and/or where a recommendation is likely to result in considerable changes in health and/or costs. For further details please see section 7.1 of the Guidelines Manual (2009). It may not be feasible or efficient to address every relevant decision problem by de novo work. There rationale for choosing areas for cost effectiveness modelling should be discussed in detail in Section 5.
List of clinical questions
Insert a list of all clinical questions in a ‘PICO’ format that are covered by the guideline.g
| # | Clinical questions by scope area |
|---|---|
| Area 1 (Diagnosis of neutropenic sepsis) | |
| 1 | Question A Which symptoms and/or signs experienced by patients in the community predict neutropenic sepsis? |
| Area 2 (Education and support for reducing adverse effects) | |
| 2 | Question B What information and support for patients receiving anti-cancer treatment and their carers reduces the adverse effects of neutropenic sepsis? |
| Area C (Emergency assessment) | |
| 3 | Question C Which test should be used in the emergency empirical assessment of a person with suspected neutropenic sepsis? |
| Area D (Risk of complications) | |
| 4 | Question D1 How do neutrophil count and temperature relate to the risk of complications of sepsis, in cancer patients with suspected neutropenic sepsis? |
| 5 | Question D2 Which tests predict outcome and response to treatment in patients with suspected neutropenic sepsis? |
| Area E (Management of neutropenic sepsis) | |
| 7 | Question E1 Which is the most valid published risk stratification score or algorithm for influencing management and predicting outcome in patients with neutropenic sepsis? |
| 8 | Question E2 Is there any difference between the outcome of patients with neutropenic sepsis managed in hospital and those managed as outpatients? |
| 9 | Question E3 Is there a difference in the effectiveness of empiric intravenous antibiotic monotherapy and empiric dual therapy in the treatment of patients with neutropenic sepsis. |
| 10 | Question E4 Does the length of time before empiric antibiotics are given influence patient outcomes? |
| 11 | Question E5 When is the optimal time to switch (step down) from intravenous to oral antibiotic therapy? |
| 12 | Question E6 What is the optimal time to change the primary empiric treatment in unresponsive fever? |
| 13 | Question E7 What is the optimal duration of empiric antibiotic therapy in patients with neutropenic sepsis? |
| 14 | Question E8 What is the optimal duration of inpatient care for patients receiving empiric treatment for neutropenic sepsis? |
| Area F (Prophylaxis of neutropenic sepsis) | |
| 15 | Question F1 Does prophylactic treatment with growth factors, granulocyte infusion and/or antibiotics improve outcomes in patients receiving anti-cancer treatment? |
| 16 | Question F2 Does prophylactic treatment with growth factors, granulocyte infusion and/or antibiotics improve outcomes in patients with a prior episode of neutropenic sepsis? |
| Area G (Empirical antibiotic for patients with central line) | |
| 17 | Question G In patients with a central venous access device with no external signs of line infection but with suspected neutropenia or neutropenic sepsis, what are the benefits and risks of adding vancomycin, teicoplanin or linezolid to first-line antibiotics? |
| Area H (Removal of central line) | |
| 18 | Question H Which patients with central venous access devices and neutropenic sepsis will benefit from removal of their central line? |
| Area I (General support and information) | |
| 19 | Question I What types of support and information have patients with neutropenic sepsis (and their carers) have found useful or requested? |
| Area J (Training of healthcare professionals) | |
| Question J Does training healthcare professionals on the identification and management of neutropenic sepsis improve outcomes for patients receiving anti-cancer treatment? |
Planned de novo modelling
This section will specify modelling work prioritised by the GDG. It will provide details on how cost effectiveness will be considered for relevant, prioritised clinical areas/decision problems. Proposed modelling work should be listed in chronological order. For each decision model, please state the proposed analytical methods, relevant references and any comments on, for example, possible diversions from the reference case.
| Scope area h (clinical question(s) i ) | Outline proposed analysis |
|---|---|
| Topic F1 and F2 | Background: F1: There is great uncertainty over the use of Granulocyte colony-stimulating factor (G-CSF), and/or antibiotics in the prevention of neutropenic sepsis. G-CSF is used to raise neutrophil counts, and shorten the duration of neutropenia, by stimulating the bone marrow to produce neutrophils. However, adverse effects include diarrhoea, weakness, a flu-like syndrome, and rarely more serious complications such as clotting disorders and capillary leak syndrome. What's more, G-CSF must be given by injection, and this may lead to local reactions at the site of administration, and repeated injections may not be desired by patients. Depot formulations are available but expensive. Pre-emptive use of oral antibiotics could reduce the likelihood of infection, but may incur patient-related risks of gut disturbance, allergy, etc and more general risks related to the development of antibiotic resistance in populations. Therefore the question is whether the use of growth factors and/or antibiotics in patients on chemotherapy may improve patient overall outcomes within a reasonable cost. F2: Patients with a prior episode of significant neutropenia are likely to become more neutropenic with repeated doses of chemotherapy, putting them at greater risk of neutropenic sepsis than patients who have never experienced this complication. The trade-off of using G-CSF and antibiotics are similar to F1. Separate models will be built for F1 (primary) and F2 (secondary) prophylaxis, as the GDG think the targeted population and interventions of interest are different for these two questions; and the primary strategy prophylaxis won't affect the choice of secondary prophylaxis strategy. Aim of analysis: To assess the cost effectiveness of several primary and secondary prophylaxis strategies to prevent first and secondary neutropenic sepsis for cancer patients undergoing anti-cancer therapy. Patient population: F1: All patients receiving anti-cancer therapy F2: Patients receiving anti-cancer therapy, with a prior episode of neutropenic sepsis. Intervention: F1:
Same for both F1 and F2.
Same for both F1 and F2.
Same for both F1 and F2: Within one course of chemotherapy. (The length of chemotherapy course may differ for different types of cancer, ranged from 5-12 cycles) The GDG are not interested in long-term outcomes, such as overall death rate caused by cancer/chemotherapy. The GDG are aware of long-term impacts of using different prophylaxis strategy, such as delay/dose deduction of chemotherapy, long-term complications from both GCSF and more widespread use of prophylactic antibiotics, etc. However the GDG don't think these long-term impacts could be captured in the model due to data paucity.Sensitivity analysis will be conducted to explore if the final result is sensitive to reoccurrence risk of NS, expected years of life etc. Analysis methods: A decision tree approach will be adopted to model the clinical pathway and a cost-utility analysis will be performed using QALYs as the measure of health outcomes. Clinical evidence: The clinical data used to populate the model will be mainly derived from the systematic reviews conducted to identify clinical and cost-effectiveness evidence for the topic. To populate the model the following data will be required: (for both primary and secondary prophylaxis)
To populate the model the following data will be required:
National reference costs of PbR tariff will be used as sources of unit costs. Feasibility issues: F1: A cursory search of NHS EED and HTA has identified many economic studies on this topic; five of them were conducted in the U.K. The most commonly used model is a decision-analytic model that was developed to assess the relative clinical outcomes and costs of primary prophylaxis with pegfilgrastim compared with filgrastim. The base case was for a 45-year-old woman with Stage II breast cancer receiving four cycles of chemotherapy with a ≥20% risk of neutropenic sepsis. The model simulated clinical outcomes and life expectancy in a cohort of women with breast cancer and follows them until death (either from cancer or other causes). The model also included the probability of receiving standard or extensive chemotherapy based upon the RDI. The problems of this model are: firstly, it only looks at patients with breast cancer. The GDG need to make a decision about whether the clinical pathway for women with breast cancer could represent the pathways for patients with various kinds of cancer. Secondly, this model takes the death rate caused by cancer into consideration. The GDG have confirmed that they are not interested in the death rate caused by cancer. Thirdly, this model didn't take reoccurrence of neutropenic sepsis within one course of chemotherapy into consideration. Therefore the overall survival rate of each prophylaxis strategy will be falsely increased, while the total cost associated with neutropenic sepsis will be falsely decreased. F2: Several economic papers have been identified for topic F2. All of them take re-occurrence of neutropenic sepsis into consideration. Two of the papers also provided decision-tree. The assumptions of the model built by Timmer-Bonte are:
|
- h
This should be the key areas relevant for considering opportunity costs and high priority for de novo modelling, as identified in section 3.
- i
Two or more questions may be addressed by a single analysis if appropriate.
References
- 1.
- Borget I. Pegfilgrastim: a health economic model to assess overall cost-effectiveness. 2009;15(5):58–61.
- 2.
- Timmer-Bonte Jn, Adang EMT. Modeling the cost effectiveness of secondary febrile neutropenia prophylaxis during standard-dose chemotherapy. 2008;26(2):290–96. [PubMed: 18182670]
Addenda to economic plan
Please state any changes that have been made to the above agreed plan, together with date. If clinical questions have changed since the economic plan was signed off, include a new list with all clinical questions as part of the addenda, together with a comment where questions were inserted, deleted or altered and an explanation.
| Scope area 10 (clinical question(s) 11 ) | Proposed changes | Date agreed |
|---|---|---|
| Topic F1 and F2 | Granulocyte infusion was taken out from the protocol. So the inventions of interest for topic F1 and F2 become the same. | 7th Feb 2011 |
| Topic F1 and F2 | Topic F1 and F2 were combined into one topic. Therefore instead of building two separate economic models for primary and secondary prophylaxis, only one economic model will be built to cover both primary and secondary prophylaxis. | 29th March 2011 |
| Topic F | Two decision trees were built: Model A assumes patients will continue to receive full-dose chemotherapy regardless of previous episodes of neutropenic sepsis. Model B assumes that if patients develop one episode of neutropenic sepsis, they will then receive dose-reduction chemotherapy; if they develop two episodes of neutropenic sepsis chemotherapy will be discontinued. | 26th May 2011 |
| Topic F | Co-trimoxazole was taken out form PICO. | 9th Sep 2011 |
| Topic F |
| 18th Nov 2011 |
- 10
This should be the key areas relevant for considering opportunity costs and high priority for de novo modelling, as identified in section 3.
- 11
Two or more questions may be addressed by a single analysis if appropriate.
Footnotes
- a
This may be done by face-to-face meeting, teleconference, or email as convenient.
- d
In case clinical questions are changed, for example, section 4 requires updating as well as other sections if modelling priorities are affected.
- g
This is the list of clinical questions to be covered by the guideline.
- PubMedLinks to PubMed
- Health economics plan - Neutropenic Sepsis: Prevention and Management of Neutrop...Health economics plan - Neutropenic Sepsis: Prevention and Management of Neutropenic Sepsis in Cancer Patients
Your browsing activity is empty.
Activity recording is turned off.
See more...